From 026593f316d8596d197bbb522a1fe33ab1a94785 Mon Sep 17 00:00:00 2001
From: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
Date: Sat, 22 Feb 2025 21:15:13 +0000
Subject: [PATCH] Built site for AMR@2.1.1.9154: 226d10f

---
 404.html                                 |   2 +-
 LICENSE-text.html                        |   2 +-
 articles/AMR.html                        |   2 +-
 articles/AMR_for_Python.html             |   2 +-
 articles/AMR_with_tidymodels.html        |   2 +-
 articles/EUCAST.html                     |   2 +-
 articles/MDR.html                        |   2 +-
 articles/PCA.html                        |   2 +-
 articles/WHONET.html                     |   2 +-
 articles/datasets.html                   |   2 +-
 articles/index.html                      |   2 +-
 articles/resistance_predict.html         |   2 +-
 articles/welcome_to_AMR.html             |   2 +-
 authors.html                             |   2 +-
 index.html                               |   2 +-
 news/index.html                          |  16 +-
 pkgdown.yml                              |   2 +-
 reference/AMR-deprecated.html            |   2 +-
 reference/AMR-options.html               |   2 +-
 reference/AMR.html                       |   2 +-
 reference/WHOCC.html                     |   2 +-
 reference/WHONET.html                    |   2 +-
 reference/ab_from_text.html              |   2 +-
 reference/ab_property.html               |   2 +-
 reference/add_custom_antimicrobials.html |   2 +-
 reference/add_custom_microorganisms.html |   2 +-
 reference/age.html                       |   2 +-
 reference/age_groups.html                |   2 +-
 reference/antibiogram.html               |   2 +-
 reference/antibiotics.html               |   2 +-
 reference/antimicrobial_selectors.html   |   2 +-
 reference/as.ab.html                     |   2 +-
 reference/as.av.html                     |   2 +-
 reference/as.disk.html                   |   2 +-
 reference/as.mic.html                    |   2 +-
 reference/as.mo.html                     |   2 +-
 reference/as.sir.html                    |  22 +--
 reference/atc_online.html                |   2 +-
 reference/av_from_text.html              |   2 +-
 reference/av_property.html               |   2 +-
 reference/availability.html              |   2 +-
 reference/bug_drug_combinations.html     |   2 +-
 reference/clinical_breakpoints.html      |   2 +-
 reference/count.html                     |   2 +-
 reference/custom_eucast_rules.html       |   2 +-
 reference/dosage.html                    |   2 +-
 reference/eucast_rules.html              |   2 +-
 reference/example_isolates.html          |   2 +-
 reference/example_isolates_unclean.html  |   2 +-
 reference/export_ncbi_biosample.html     |   2 +-
 reference/first_isolate.html             |   2 +-
 reference/g.test.html                    |   2 +-
 reference/get_episode.html               |   2 +-
 reference/ggplot_pca.html                |   2 +-
 reference/ggplot_sir.html                |   2 +-
 reference/guess_ab_col.html              |   2 +-
 reference/index.html                     |   2 +-
 reference/intrinsic_resistant.html       |   2 +-
 reference/italicise_taxonomy.html        |   2 +-
 reference/join.html                      |   2 +-
 reference/key_antimicrobials.html        |   2 +-
 reference/kurtosis.html                  |   2 +-
 reference/like.html                      |   2 +-
 reference/mdro.html                      |   2 +-
 reference/mean_amr_distance.html         |   2 +-
 reference/microorganisms.codes.html      |   2 +-
 reference/microorganisms.groups.html     |   2 +-
 reference/microorganisms.html            |   2 +-
 reference/mo_matching_score.html         |   2 +-
 reference/mo_property.html               |   2 +-
 reference/mo_source.html                 |   2 +-
 reference/pca.html                       |   2 +-
 reference/plot-1.png                     | Bin 26471 -> 0 bytes
 reference/plot-10.png                    | Bin 27160 -> 0 bytes
 reference/plot-11.png                    | Bin 33179 -> 0 bytes
 reference/plot-12.png                    | Bin 80659 -> 0 bytes
 reference/plot-13.png                    | Bin 68687 -> 0 bytes
 reference/plot-14.png                    | Bin 26842 -> 0 bytes
 reference/plot-15.png                    | Bin 19275 -> 0 bytes
 reference/plot-16.png                    | Bin 73236 -> 0 bytes
 reference/plot-17.png                    | Bin 93183 -> 0 bytes
 reference/plot-18.png                    | Bin 86607 -> 0 bytes
 reference/plot-19.png                    | Bin 38148 -> 0 bytes
 reference/plot-2.png                     | Bin 27051 -> 0 bytes
 reference/plot-20.png                    | Bin 57251 -> 0 bytes
 reference/plot-21.png                    | Bin 25154 -> 0 bytes
 reference/plot-22.png                    | Bin 28723 -> 0 bytes
 reference/plot-23.png                    | Bin 19541 -> 0 bytes
 reference/plot-24.png                    | Bin 29791 -> 0 bytes
 reference/plot-3.png                     | Bin 26115 -> 0 bytes
 reference/plot-4.png                     | Bin 40312 -> 0 bytes
 reference/plot-5.png                     | Bin 40720 -> 0 bytes
 reference/plot-6.png                     | Bin 39878 -> 0 bytes
 reference/plot-7.png                     | Bin 24961 -> 0 bytes
 reference/plot-8.png                     | Bin 27928 -> 0 bytes
 reference/plot-9.png                     | Bin 30774 -> 0 bytes
 reference/plot.html                      | 188 +----------------------
 reference/proportion.html                |   2 +-
 reference/random.html                    |  34 ++--
 reference/resistance_predict.html        |   2 +-
 reference/skewness.html                  |   4 +-
 reference/top_n_microorganisms.html      |   2 +-
 reference/translate.html                 |   2 +-
 search.json                              |   2 +-
 104 files changed, 113 insertions(+), 301 deletions(-)
 delete mode 100644 reference/plot-1.png
 delete mode 100644 reference/plot-10.png
 delete mode 100644 reference/plot-11.png
 delete mode 100644 reference/plot-12.png
 delete mode 100644 reference/plot-13.png
 delete mode 100644 reference/plot-14.png
 delete mode 100644 reference/plot-15.png
 delete mode 100644 reference/plot-16.png
 delete mode 100644 reference/plot-17.png
 delete mode 100644 reference/plot-18.png
 delete mode 100644 reference/plot-19.png
 delete mode 100644 reference/plot-2.png
 delete mode 100644 reference/plot-20.png
 delete mode 100644 reference/plot-21.png
 delete mode 100644 reference/plot-22.png
 delete mode 100644 reference/plot-23.png
 delete mode 100644 reference/plot-24.png
 delete mode 100644 reference/plot-3.png
 delete mode 100644 reference/plot-4.png
 delete mode 100644 reference/plot-5.png
 delete mode 100644 reference/plot-6.png
 delete mode 100644 reference/plot-7.png
 delete mode 100644 reference/plot-8.png
 delete mode 100644 reference/plot-9.png

diff --git a/404.html b/404.html
index 7cd9b0344..a4e1e4f46 100644
--- a/404.html
+++ b/404.html
@@ -32,7 +32,7 @@
 
     <a class="navbar-brand me-2" href="https://msberends.github.io/AMR/index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/LICENSE-text.html b/LICENSE-text.html
index 855bad8a0..afc650d57 100644
--- a/LICENSE-text.html
+++ b/LICENSE-text.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/AMR.html b/articles/AMR.html
index 8c916f216..db4eb6624 100644
--- a/articles/AMR.html
+++ b/articles/AMR.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/AMR_for_Python.html b/articles/AMR_for_Python.html
index be1176ae4..6444322f7 100644
--- a/articles/AMR_for_Python.html
+++ b/articles/AMR_for_Python.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/AMR_with_tidymodels.html b/articles/AMR_with_tidymodels.html
index f2a22cf4a..4a0dda3db 100644
--- a/articles/AMR_with_tidymodels.html
+++ b/articles/AMR_with_tidymodels.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/EUCAST.html b/articles/EUCAST.html
index acccf972d..f08c9f892 100644
--- a/articles/EUCAST.html
+++ b/articles/EUCAST.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/MDR.html b/articles/MDR.html
index 1e28d0b68..e104af2ce 100644
--- a/articles/MDR.html
+++ b/articles/MDR.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/PCA.html b/articles/PCA.html
index 9256c16f6..b640e7955 100644
--- a/articles/PCA.html
+++ b/articles/PCA.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/WHONET.html b/articles/WHONET.html
index a9895b513..cf8b29469 100644
--- a/articles/WHONET.html
+++ b/articles/WHONET.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/datasets.html b/articles/datasets.html
index 36601c372..653053bff 100644
--- a/articles/datasets.html
+++ b/articles/datasets.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/index.html b/articles/index.html
index 2b6df3bab..2b2d65055 100644
--- a/articles/index.html
+++ b/articles/index.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/resistance_predict.html b/articles/resistance_predict.html
index 5964ab396..5bcf2de65 100644
--- a/articles/resistance_predict.html
+++ b/articles/resistance_predict.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/welcome_to_AMR.html b/articles/welcome_to_AMR.html
index ff93761fb..4fc093022 100644
--- a/articles/welcome_to_AMR.html
+++ b/articles/welcome_to_AMR.html
@@ -31,7 +31,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/authors.html b/authors.html
index 536991432..2a84a7c50 100644
--- a/authors.html
+++ b/authors.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/index.html b/index.html
index 06fd3db64..a6b169dd7 100644
--- a/index.html
+++ b/index.html
@@ -34,7 +34,7 @@
 
     <a class="navbar-brand me-2" href="index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/news/index.html b/news/index.html
index 748f4ac68..8368fcc01 100644
--- a/news/index.html
+++ b/news/index.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -48,18 +48,18 @@
     </div>
 
     <div class="section level2">
-<h2 class="pkg-version" data-toc-text="2.1.1.9153" id="amr-2119153">AMR 2.1.1.9153<a class="anchor" aria-label="anchor" href="#amr-2119153"></a></h2>
+<h2 class="pkg-version" data-toc-text="2.1.1.9154" id="amr-2119154">AMR 2.1.1.9154<a class="anchor" aria-label="anchor" href="#amr-2119154"></a></h2>
 <p><em>(this beta version will eventually become v3.0. We’re happy to reach a new major milestone soon, which will be all about the new One Health support! Install this beta using <a href="https://msberends.github.io/AMR/#latest-development-version">the instructions here</a>.)</em></p>
 <div class="section level5">
-<h5 id="a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9153">A New Milestone: AMR v3.0 with One Health Support (= Human + Veterinary + Environmental)<a class="anchor" aria-label="anchor" href="#a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9153"></a></h5>
+<h5 id="a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9154">A New Milestone: AMR v3.0 with One Health Support (= Human + Veterinary + Environmental)<a class="anchor" aria-label="anchor" href="#a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9154"></a></h5>
 <p>This package now supports not only tools for AMR data analysis in clinical settings, but also for veterinary and environmental microbiology. This was made possible through a collaboration with the <a href="https://www.upei.ca/avc" class="external-link">University of Prince Edward Island’s Atlantic Veterinary College</a>, Canada. To celebrate this great improvement of the package, we also updated the package logo to reflect this change.</p>
 </div>
 <div class="section level3">
-<h3 id="breaking-2-1-1-9153">Breaking<a class="anchor" aria-label="anchor" href="#breaking-2-1-1-9153"></a></h3>
+<h3 id="breaking-2-1-1-9154">Breaking<a class="anchor" aria-label="anchor" href="#breaking-2-1-1-9154"></a></h3>
 <ul><li>Removed all functions and references that used the deprecated <code>rsi</code> class, which were all replaced with their <code>sir</code> equivalents two years ago</li>
 </ul></div>
 <div class="section level3">
-<h3 id="new-2-1-1-9153">New<a class="anchor" aria-label="anchor" href="#new-2-1-1-9153"></a></h3>
+<h3 id="new-2-1-1-9154">New<a class="anchor" aria-label="anchor" href="#new-2-1-1-9154"></a></h3>
 <ul><li>
 <strong>One Health implementation</strong>
 <ul><li>Function <code><a href="../reference/as.sir.html">as.sir()</a></code> now has extensive support for veterinary breakpoints from CLSI. Use <code>breakpoint_type = "animal"</code> and set the <code>host</code> argument to a variable that contains animal species names.</li>
@@ -110,7 +110,7 @@
 </ul></li>
 </ul></div>
 <div class="section level3">
-<h3 id="changed-2-1-1-9153">Changed<a class="anchor" aria-label="anchor" href="#changed-2-1-1-9153"></a></h3>
+<h3 id="changed-2-1-1-9154">Changed<a class="anchor" aria-label="anchor" href="#changed-2-1-1-9154"></a></h3>
 <ul><li>SIR interpretation
 <ul><li>It is now possible to use column names for argument <code>ab</code>, <code>mo</code>, and <code>uti</code>: <code>as.sir(..., ab = "column1", mo = "column2", uti = "column3")</code>. This greatly improves the flexibility for users.</li>
 <li>Users can now set their own criteria (using regular expressions) as to what should be considered S, I, R, SDD, and NI.</li>
@@ -180,7 +180,7 @@
 <li>Added console colours support of <code>sir</code> class for Positron</li>
 </ul></div>
 <div class="section level3">
-<h3 id="other-2-1-1-9153">Other<a class="anchor" aria-label="anchor" href="#other-2-1-1-9153"></a></h3>
+<h3 id="other-2-1-1-9154">Other<a class="anchor" aria-label="anchor" href="#other-2-1-1-9154"></a></h3>
 <ul><li>Added Dr. Larisse Bolton as contributor for her fantastic implementation of WISCA in a mathematically solid way</li>
 <li>Added Matthew Saab, Dr. Jordan Stull, and Prof. Javier Sanchez as contributors for their tremendous input on veterinary breakpoints and interpretations</li>
 <li>Greatly improved <code>vctrs</code> integration, a Tidyverse package working in the background for many Tidyverse functions. For users, this means that functions such as <code>dplyr</code>’s <code><a href="https://dplyr.tidyverse.org/reference/bind_rows.html" class="external-link">bind_rows()</a></code>, <code><a href="https://dplyr.tidyverse.org/reference/rowwise.html" class="external-link">rowwise()</a></code> and <code><a href="https://dplyr.tidyverse.org/reference/c_across.html" class="external-link">c_across()</a></code> are now supported for e.g. columns of class <code>mic</code>. Despite this, this <code>AMR</code> package is still zero-dependent on any other package, including <code>dplyr</code> and <code>vctrs</code>.</li>
@@ -188,7 +188,7 @@
 <li>Stopped support for SAS (<code>.xpt</code>) files, since their file structure and extremely inefficient and requires more disk space than GitHub allows in a single commit.</li>
 </ul></div>
 <div class="section level3">
-<h3 id="older-versions-2-1-1-9153">Older Versions<a class="anchor" aria-label="anchor" href="#older-versions-2-1-1-9153"></a></h3>
+<h3 id="older-versions-2-1-1-9154">Older Versions<a class="anchor" aria-label="anchor" href="#older-versions-2-1-1-9154"></a></h3>
 <p>This changelog only contains changes from AMR v3.0 (February 2025) and later.</p>
 <ul><li>For prior v2 versions, please see <a href="https://github.com/msberends/AMR/blob/v2.1.1/NEWS.md" class="external-link">our v2 archive</a>.</li>
 <li>For prior v1 versions, please see <a href="https://github.com/msberends/AMR/blob/v1.8.2/NEWS.md" class="external-link">our v1 archive</a>.</li>
diff --git a/pkgdown.yml b/pkgdown.yml
index 39e8af51f..e4cb15baf 100644
--- a/pkgdown.yml
+++ b/pkgdown.yml
@@ -12,7 +12,7 @@ articles:
   resistance_predict: resistance_predict.html
   welcome_to_AMR: welcome_to_AMR.html
   WHONET: WHONET.html
-last_built: 2025-02-22T20:30Z
+last_built: 2025-02-22T21:11Z
 urls:
   reference: https://msberends.github.io/AMR/reference
   article: https://msberends.github.io/AMR/articles
diff --git a/reference/AMR-deprecated.html b/reference/AMR-deprecated.html
index 9d495b373..91d93f57b 100644
--- a/reference/AMR-deprecated.html
+++ b/reference/AMR-deprecated.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/AMR-options.html b/reference/AMR-options.html
index 72765ebbc..00003edea 100644
--- a/reference/AMR-options.html
+++ b/reference/AMR-options.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/AMR.html b/reference/AMR.html
index 957e81e6b..f425ffde6 100644
--- a/reference/AMR.html
+++ b/reference/AMR.html
@@ -21,7 +21,7 @@ The AMR package is available in English, Chinese, Czech, Danish, Dutch, Finnish,
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/WHOCC.html b/reference/WHOCC.html
index e508d372a..573599d8a 100644
--- a/reference/WHOCC.html
+++ b/reference/WHOCC.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/WHONET.html b/reference/WHONET.html
index a171e9bf0..d3a0ce4aa 100644
--- a/reference/WHONET.html
+++ b/reference/WHONET.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/ab_from_text.html b/reference/ab_from_text.html
index ea00f4c3d..80b2f42be 100644
--- a/reference/ab_from_text.html
+++ b/reference/ab_from_text.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/ab_property.html b/reference/ab_property.html
index 7b68e5894..3ac39c124 100644
--- a/reference/ab_property.html
+++ b/reference/ab_property.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/add_custom_antimicrobials.html b/reference/add_custom_antimicrobials.html
index 3d9f52bd6..e6db5a3a6 100644
--- a/reference/add_custom_antimicrobials.html
+++ b/reference/add_custom_antimicrobials.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/add_custom_microorganisms.html b/reference/add_custom_microorganisms.html
index 248edee60..c3802b8ba 100644
--- a/reference/add_custom_microorganisms.html
+++ b/reference/add_custom_microorganisms.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/age.html b/reference/age.html
index bd9e39371..49f06c318 100644
--- a/reference/age.html
+++ b/reference/age.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/age_groups.html b/reference/age_groups.html
index d4fb0de25..34014900c 100644
--- a/reference/age_groups.html
+++ b/reference/age_groups.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/antibiogram.html b/reference/antibiogram.html
index 8da1b7a0f..8788eed9d 100644
--- a/reference/antibiogram.html
+++ b/reference/antibiogram.html
@@ -9,7 +9,7 @@ Adhering to previously described approaches (see Source) and especially the Baye
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/antibiotics.html b/reference/antibiotics.html
index 0439a5dce..d19843d05 100644
--- a/reference/antibiotics.html
+++ b/reference/antibiotics.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/antimicrobial_selectors.html b/reference/antimicrobial_selectors.html
index 6d8bf7462..6f1d28b09 100644
--- a/reference/antimicrobial_selectors.html
+++ b/reference/antimicrobial_selectors.html
@@ -17,7 +17,7 @@ my_data_with_all_these_columns %&amp;gt;%
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.ab.html b/reference/as.ab.html
index c8f4754cf..0ea735718 100644
--- a/reference/as.ab.html
+++ b/reference/as.ab.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.av.html b/reference/as.av.html
index 2cf10d3b2..8207ab93e 100644
--- a/reference/as.av.html
+++ b/reference/as.av.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.disk.html b/reference/as.disk.html
index 95fb01e58..67234df50 100644
--- a/reference/as.disk.html
+++ b/reference/as.disk.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.mic.html b/reference/as.mic.html
index a0fa0797d..99d52ce30 100644
--- a/reference/as.mic.html
+++ b/reference/as.mic.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.mo.html b/reference/as.mo.html
index e5777d1b4..70ba0ee15 100644
--- a/reference/as.mo.html
+++ b/reference/as.mo.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/as.sir.html b/reference/as.sir.html
index 7d7553a38..1a330e00a 100644
--- a/reference/as.sir.html
+++ b/reference/as.sir.html
@@ -21,7 +21,7 @@ All breakpoints used for interpretation are available in our clinical_breakpoint
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -782,16 +782,16 @@ A microorganism is categorised as "Resistant" when there is a high likelihood of
 <span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #949494;"># A tibble: 57 × 16</span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>    datetime            index ab_given mo_given host_given ab   mo              </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span>  <span style="color: #BCBCBC;">*</span> <span style="color: #949494; font-style: italic;">&lt;dttm&gt;</span>              <span style="color: #949494; font-style: italic;">&lt;int&gt;</span> <span style="color: #949494; font-style: italic;">&lt;chr&gt;</span>    <span style="color: #949494; font-style: italic;">&lt;chr&gt;</span>    <span style="color: #949494; font-style: italic;">&lt;chr&gt;</span>      <span style="color: #949494; font-style: italic;">&lt;ab&gt;</span> <span style="color: #949494; font-style: italic;">&lt;mo&gt;</span>            </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 1</span> 2025-02-22 <span style="color: #949494;">20:31:37</span>     4 AMX      B_STRPT… human      AMX  <span style="color: #949494;">B_</span>STRPT<span style="color: #949494;">_</span>PNMN    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 2</span> 2025-02-22 <span style="color: #949494;">20:31:43</span>     4 genta    Escheri… human      GEN  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 3</span> 2025-02-22 <span style="color: #949494;">20:31:44</span>     4 genta    Escheri… human      GEN  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 4</span> 2025-02-22 <span style="color: #949494;">20:31:44</span>     4 genta    Escheri… cattle     GEN  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 5</span> 2025-02-22 <span style="color: #949494;">20:31:45</span>     4 genta    Escheri… cattle     GEN  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 6</span> 2025-02-22 <span style="color: #949494;">20:31:37</span>     3 AMX      B_STRPT… human      AMX  <span style="color: #949494;">B_</span>STRPT<span style="color: #949494;">_</span>PNMN    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 7</span> 2025-02-22 <span style="color: #949494;">20:31:43</span>     3 tobra    Escheri… human      TOB  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 8</span> 2025-02-22 <span style="color: #949494;">20:31:44</span>     3 tobra    Escheri… human      TOB  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 9</span> 2025-02-22 <span style="color: #949494;">20:31:44</span>     3 tobra    Escheri… horses     TOB  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;">10</span> 2025-02-22 <span style="color: #949494;">20:31:45</span>     3 tobra    Escheri… horses     TOB  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 1</span> 2025-02-22 <span style="color: #949494;">21:12:34</span>     4 AMX      B_STRPT… human      AMX  <span style="color: #949494;">B_</span>STRPT<span style="color: #949494;">_</span>PNMN    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 2</span> 2025-02-22 <span style="color: #949494;">21:12:41</span>     4 genta    Escheri… human      GEN  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 3</span> 2025-02-22 <span style="color: #949494;">21:12:41</span>     4 genta    Escheri… human      GEN  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 4</span> 2025-02-22 <span style="color: #949494;">21:12:42</span>     4 genta    Escheri… cattle     GEN  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 5</span> 2025-02-22 <span style="color: #949494;">21:12:42</span>     4 genta    Escheri… cattle     GEN  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 6</span> 2025-02-22 <span style="color: #949494;">21:12:34</span>     3 AMX      B_STRPT… human      AMX  <span style="color: #949494;">B_</span>STRPT<span style="color: #949494;">_</span>PNMN    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 7</span> 2025-02-22 <span style="color: #949494;">21:12:41</span>     3 tobra    Escheri… human      TOB  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 8</span> 2025-02-22 <span style="color: #949494;">21:12:41</span>     3 tobra    Escheri… human      TOB  <span style="color: #949494;">B_</span>[ORD]<span style="color: #949494;">_</span>ENTRBCTR</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;"> 9</span> 2025-02-22 <span style="color: #949494;">21:12:42</span>     3 tobra    Escheri… horses     TOB  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #BCBCBC;">10</span> 2025-02-22 <span style="color: #949494;">21:12:42</span>     3 tobra    Escheri… horses     TOB  <span style="color: #949494;">B_</span>ESCHR<span style="color: #949494;">_</span>COLI    </span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #949494;"># ℹ 47 more rows</span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #949494;"># ℹ 9 more variables: host &lt;chr&gt;, method &lt;chr&gt;, input &lt;chr&gt;, outcome &lt;sir&gt;,</span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> <span style="color: #949494;">#   notes &lt;chr&gt;, guideline &lt;chr&gt;, ref_table &lt;chr&gt;, uti &lt;lgl&gt;,</span></span>
diff --git a/reference/atc_online.html b/reference/atc_online.html
index eb5246469..6dca0afb9 100644
--- a/reference/atc_online.html
+++ b/reference/atc_online.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/av_from_text.html b/reference/av_from_text.html
index dc472e8ee..94596259a 100644
--- a/reference/av_from_text.html
+++ b/reference/av_from_text.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/av_property.html b/reference/av_property.html
index bbb6a4376..1439d31b1 100644
--- a/reference/av_property.html
+++ b/reference/av_property.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/availability.html b/reference/availability.html
index 56762a33e..34bd705ca 100644
--- a/reference/availability.html
+++ b/reference/availability.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/bug_drug_combinations.html b/reference/bug_drug_combinations.html
index bf04fef7f..3c1687e69 100644
--- a/reference/bug_drug_combinations.html
+++ b/reference/bug_drug_combinations.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/clinical_breakpoints.html b/reference/clinical_breakpoints.html
index f29f45e5a..d3e694839 100644
--- a/reference/clinical_breakpoints.html
+++ b/reference/clinical_breakpoints.html
@@ -21,7 +21,7 @@ Use as.sir() to transform MICs or disks measurements to SIR values."><meta prope
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/count.html b/reference/count.html
index 8ec724945..d3dec0d95 100644
--- a/reference/count.html
+++ b/reference/count.html
@@ -9,7 +9,7 @@ count_resistant() should be used to count resistant isolates, count_susceptible(
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/custom_eucast_rules.html b/reference/custom_eucast_rules.html
index 0ea642c1c..6f263631a 100644
--- a/reference/custom_eucast_rules.html
+++ b/reference/custom_eucast_rules.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/dosage.html b/reference/dosage.html
index 828348ccd..f12a24884 100644
--- a/reference/dosage.html
+++ b/reference/dosage.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/eucast_rules.html b/reference/eucast_rules.html
index d820483bb..5b8c80d26 100644
--- a/reference/eucast_rules.html
+++ b/reference/eucast_rules.html
@@ -9,7 +9,7 @@ To improve the interpretation of the antibiogram before EUCAST rules are applied
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/example_isolates.html b/reference/example_isolates.html
index dd3108240..d1f2051ad 100644
--- a/reference/example_isolates.html
+++ b/reference/example_isolates.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/example_isolates_unclean.html b/reference/example_isolates_unclean.html
index ace07a7c5..c283f8b96 100644
--- a/reference/example_isolates_unclean.html
+++ b/reference/example_isolates_unclean.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/export_ncbi_biosample.html b/reference/export_ncbi_biosample.html
index 74467b312..e77f3fa7c 100644
--- a/reference/export_ncbi_biosample.html
+++ b/reference/export_ncbi_biosample.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/first_isolate.html b/reference/first_isolate.html
index d923b55a3..bec841855 100644
--- a/reference/first_isolate.html
+++ b/reference/first_isolate.html
@@ -9,7 +9,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/g.test.html b/reference/g.test.html
index c0cbf6a8d..58e170b2a 100644
--- a/reference/g.test.html
+++ b/reference/g.test.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/get_episode.html b/reference/get_episode.html
index d37a137d4..1c1ab43c3 100644
--- a/reference/get_episode.html
+++ b/reference/get_episode.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/ggplot_pca.html b/reference/ggplot_pca.html
index 18c361d49..e580dfb10 100644
--- a/reference/ggplot_pca.html
+++ b/reference/ggplot_pca.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/ggplot_sir.html b/reference/ggplot_sir.html
index b01588779..6eb6b8207 100644
--- a/reference/ggplot_sir.html
+++ b/reference/ggplot_sir.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/guess_ab_col.html b/reference/guess_ab_col.html
index 4b1c77654..204c71519 100644
--- a/reference/guess_ab_col.html
+++ b/reference/guess_ab_col.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/index.html b/reference/index.html
index f965a7bc1..ba13fbb5e 100644
--- a/reference/index.html
+++ b/reference/index.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/intrinsic_resistant.html b/reference/intrinsic_resistant.html
index a55c127df..7b9d30b5e 100644
--- a/reference/intrinsic_resistant.html
+++ b/reference/intrinsic_resistant.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/italicise_taxonomy.html b/reference/italicise_taxonomy.html
index d95513ced..6e8a8778f 100644
--- a/reference/italicise_taxonomy.html
+++ b/reference/italicise_taxonomy.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/join.html b/reference/join.html
index c4e93ff2f..b6942e2d1 100644
--- a/reference/join.html
+++ b/reference/join.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/key_antimicrobials.html b/reference/key_antimicrobials.html
index a04bef142..c7cc4262a 100644
--- a/reference/key_antimicrobials.html
+++ b/reference/key_antimicrobials.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/kurtosis.html b/reference/kurtosis.html
index 3664dabba..b5ddaea9c 100644
--- a/reference/kurtosis.html
+++ b/reference/kurtosis.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/like.html b/reference/like.html
index 49018683c..6a37e9205 100644
--- a/reference/like.html
+++ b/reference/like.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/mdro.html b/reference/mdro.html
index 193d39bd4..7a3c0d08d 100644
--- a/reference/mdro.html
+++ b/reference/mdro.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/mean_amr_distance.html b/reference/mean_amr_distance.html
index 3a4420276..8db5474d3 100644
--- a/reference/mean_amr_distance.html
+++ b/reference/mean_amr_distance.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/microorganisms.codes.html b/reference/microorganisms.codes.html
index 01bcec8e1..63af07ede 100644
--- a/reference/microorganisms.codes.html
+++ b/reference/microorganisms.codes.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/microorganisms.groups.html b/reference/microorganisms.groups.html
index fc2caeea6..1aca98c50 100644
--- a/reference/microorganisms.groups.html
+++ b/reference/microorganisms.groups.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/microorganisms.html b/reference/microorganisms.html
index fb3cc5211..539467fe5 100644
--- a/reference/microorganisms.html
+++ b/reference/microorganisms.html
@@ -9,7 +9,7 @@ This data set is carefully crafted, yet made 100% reproducible from public and a
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/mo_matching_score.html b/reference/mo_matching_score.html
index 3b5bb2d46..f60ea9129 100644
--- a/reference/mo_matching_score.html
+++ b/reference/mo_matching_score.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/mo_property.html b/reference/mo_property.html
index e468a945f..1a68e885f 100644
--- a/reference/mo_property.html
+++ b/reference/mo_property.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/mo_source.html b/reference/mo_source.html
index feea25dba..3cee03bc3 100644
--- a/reference/mo_source.html
+++ b/reference/mo_source.html
@@ -9,7 +9,7 @@ This is the fastest way to have your organisation (or analysis) specific codes p
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/pca.html b/reference/pca.html
index badb73453..427f1af65 100644
--- a/reference/pca.html
+++ b/reference/pca.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/plot-1.png b/reference/plot-1.png
deleted file mode 100644
index c37288d46d14fdd673d0b982c26a3fb27e56939b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 26471
zcmeFZ2UOEpyDyAgN0CuyM5#Ii7)L3B5L#fAQ49#El+ZzuW`Ix(p#`ww3{8U+sa8rt
zuR?%;C>UB)2t5HLfCfltNgyG~{U^Tfd*<A`?!Dhy_xsNJ?pf!*7KH5oF3*1I?|Jst
zKQCLEN&X=7gM@^HB=q7RS0yBNgiA<lyS{rT&@z_(d<pp2dELV758xAcHcaucz{j4T
zi;kfZ67un)pDigf%U%)^$0VSCoV5+lS)fGZc%fY9gs?)xz0gO$4`2WNSk=y;cemqA
z_nrOe#~ohVgO4m1Y89EBzaoc}J5sL~`10ND1N8^Ie$re&bmiT(1JYqX5_kXlVRrlb
zBTp~xcK+-bDbTBrMfVRh*RB;eA{+Baozq6#D3bmJb7P>i5uAoXGdVwr$@{Ng|KD(+
zwT6Ed8I-h`hCB5aX_B{H;?m%4J8598&dP1%Kj91Ra7Em-xOopwT}(n;>4BxKqQ5RZ
zi4%2{@OZJkN%SdkXhsDtAyI)=XrKcf^jh|WZ;3lG;+_jbLs`%046<W29|nWjGRp-8
zkuVlxWEuyK$TftypT-YHkma%H>SbY5q%&(}WIaB4YTmKCg?@+bjE%P|w2O6Yu853b
z{ka3Ja24`8(ihoH-$sUG-Ak##`&c=y=m=Kp2wPv0#4)igFL#4A*Kp@N-*kkH1(3ek
zkH46j6S&*CU#1iqG;@7fvvfN$(=LM<<T@7Mj>J6};zDFw7*V=SDhT)7E3OkutDLe2
z%x)YOY?tqJw%O3n&^_Cb<+c~t;KFfh-`&6M54eU8x;vy-9?4YcXnw)1vW%9Dy(;hG
z88x%8?pGap*p=B&mX_C<4Zzly+sq+gs)!eTSe|`GnKomS{pG=rA%%<VxizW5Vgv-`
zzwY_=0zpc{Z++YQFnwGtA@8o_*~tr=@jLiJYtuMAX7=bPTyPnA1z9aQxc1i$&&}mK
zs~oj;nf>-AO?|cMjvtwlvL#47K;!av8c&ma=Z}*voQvN^UQNEX39G9-_KmfdD-1@O
zx0ifQa$#<%UJmTnyc?>7b{p$p5Q5IJ-0RnkqN^g8c9xwkpa^mTb@hz=-C2rYht5x9
zZO9H|dxS#}We)L-+4A5Rs#GBKOIl4)o`l%)(e0dD5a&JN(yps7kYy~xz1(K6+!({<
z(QyXH31FCqRHIMlVZ9P*gPf0=-75~|%WV29WJ~(_Yc$!h@l(CR{-sKmRA5y{xl>^a
zTC#d?3acS}(qQOle_*Utz08u}v&wwtM4jOzdf?7H_2Z9Njn#;~T!otsA(OtHGD`|r
zYXAD#O{-l^e_Jwzdw+2M!I*}rmtR$N?<w+Rw>FY+Ev4Bp?*wm^r(;*|EZ0}@2MpZ1
zUucG$9J4n*haNFtLHwW-CdGXLnTyfqSPhce4nqbUPjd0R3;YzGuTL>=<&itrk^ZRG
zsH@1uJ8+Ymxh>KfUpcco<`nmbn|#)Mxv@o3J4A}Wur3OuR`hgaODCd>xb^b`I^M0k
zc3_&5TbNc&y7j!~!vI@`UkX}C`{V@AkNY~19LXGe_cxR4YO>z)>OWYyjsBRvfY@!-
zR~F+%+=$19rVmuR9ogWHu&jI?CRl#+Ik81kJ8j%`42AdT?q}}lCG-&}cSeojx`R$=
zH|nVNF{zsH($RzrBPq|(UXovcv7S-$`>AhJi$}@11_$5dx=>1|eE6vWv8~Q5?Y%dh
z>*|(nAC3)huY5*=*-9a%yAzrs+t!)Rh0M$?OsTTa(%zj-hOE)?zV8xEar)^+lh3PV
zLO|PkJXX3!Z+N{}nIGd#WToGK%!Kb=NM^$<xN=y9stft5r}Tw)cGRk66Px@c`*)rC
zTKza>KIIzNq_B2lGH3*|T@k)JEJMGR`zNyRXuGSk%iIguNc5*Wt5S7>6%|<|1Jl^C
z?@PvxCXz#8hlq^7DVFn@X{Mo!x1mCUu-Rk5R8sM*8|}%1%Qs77hU&uWrm8eox(}~y
z5S7u?mS?2JjJ1f*cZ?Msf&?cI=fn;R)S@l_$(`s4%)F655otd~DRrG2QqvRO-`gh~
zIRH0^Xw%tmH>dFp49IZX{J@)$I8>oq_SgnsXZ|mfq6hX!c%)G)JcQM_6!1Mo!WA^C
z-K>EW=t-E)8UHf!2;uLLf_x0ePPhdvyrTC=x(I#CT`PnLfELCbK1}^vXon2XDscJp
zK?31MiH!HQ7~i(aUFH2KeCdil<*!s3?~bw=MQ34$zu}^p^n$QOhfOKrf&t>*2X__I
z8<jg5T3?Xq4Z-g!!&Yh=T$!~Jt-7m<HzYjjdCYk^lfr8qlhrbV!c88%UlP2Ndq<fc
zNT3a0xKnd}>Y5}NG@6*ATaw$D?$!3?^<MTw#Ve8%L8um`p4YGG909n~y<FoU;+wGm
zGT=mWYH7m>vfeff6+*p-5H0(b5aC`mInJpOaNp<{b>>f0lyvQt2`mab-wx-0&F?(V
zcyoI`@mfzDO4U7kOyoWNwYO+URE#yhN!s!IsuecxD*wExL)56u+$ks3ng^5DAsHI_
zb^a0YwWoO8^d0t!juo-hglT{rK<zujM(RFP$@T@*+nE$PS1$0D81iof-e@ehc1LH+
zY*DARa}f1~V{3fwWweFqud#ltRq`R(tzBcfr!usJ)o)^n4q)dwC;TU?7Tx-JzyycA
zx4Hpr!nYWdzgJXrBbmy2%LLM_ma;4Tb!WWGS;fko>43h4wFxLIhScJdP3&cE4nvpZ
zRw0ZsSTWFh`F?(vM%E6O_E5p+2g-6CPCGqSPwjmQk63E78H@~I3<9>|As=&PheW|7
z_JJ%2@{F7IKqm-1eN9qPHXp^ptfmOUH=NSZIWZ&LX1F8EjA>H$WgZd#kNpLu$m6bu
z3aYqEGCWcl!x(PjeRg85*65*-P5<Pq5#VLzG`Bt%c)cVyCLpjgN=>4o7l^^FPnt@(
zvkyjm>Jhtt$#dbmv^(W>N58+4A`B)rNC!SkU(Jxi`)}i;QoTs;34xi9>(c@pclY@?
zvC-kt_)u@!b9$$=9PhyhK^W7ve7-sszAk){?-sIYa;5}{{<1LMZW6jHeQo4Ky-Uc#
zJNmd~Vl8%PHdCy3hX71qa|`fH76{+W9qSZwv&<lEbMlUHe(b<B+^K1hAzizvG^Gl&
zb_iN%JX}8A=<GW7<#Obw-pV5rwQ7P52)@x{7kng9T5VV;FIgQvl=W$2iY{-_VT8Yi
zrqo8evuq?ys)tbhBclo2#Kp_)aI_l7<umiGrJE1Ay!SK}z9lX|Wpal^vZ|urvSs$*
z4>9SzKSmE?JW2k`ofWYyGw>#ceL`MBVa<lDYOXzhxH7HUhfl+J62`dQ@i`7?Vfc*y
z2QyqoX{E4+E_HaVp=hStH>ZI%jLVRVTEcjG@^CwNiqui~c;4rVg*NZSj-gra3DL43
zVA(J{v!taaX(ID$*<LmQm4w!C(dq;1_FQW6G&3HI#?$pBw=jPpRNR(9y_BvC(**cA
z90^Xw8Z=M6TvaS>326Gveqw=@JuHX!YWMFegymBnPs~W2*-n^nKo=c1qWB5D+5~vD
z;aH-<H2ONRGPc%3*o&KR@EJed9NZ>2By;X-`?XC0J%%jAGHUM6FATs#4FU2bB;p_o
z(lXv$`I4AoZWTRL#`}f5Tk0)$WH#N|d#<b&a=mTc+V!(f_1OFQXti?pkIa;ewT-`&
zhwzj;#;R)HR%?`pB3Cdm%)s5Tfz-!y{C5u|HXlp2jc#saH-8$p7lE0dor{?83>Uc4
zSevpZB7-jp-ad8_tufLXuBx-0lF-spUKFTn$e}%$`9a-%uCe<Lvotzip|PzT^>WNP
z-36`+_*MRQ{BkAUG~RPW9kG0>{ty<H9F&C8*>2grz}5_LH(l4g;GjGy@2{*k4uixJ
zS%YKt2bS#losxs~zB?J8)Txu<t7EZlA^r<-s92U%b_r9;X)~t!)FkE4q1iCP&{m1D
zcLyaUT9+9frq61=dNRSf+hY;}f2uuDUaBx0Wcuoap)I6^i@>ShNHH9!7=Qa)x|7~v
z(nPDiVkxz$Z8_%Id+<4z?gXqN_1#7%{lVc0|6sLptissJjOyX1Z{eZMBE!$#4FKQi
zk#5<+rp+;AAHFQ|db?!SpdpHyMuyo-+vks;L*qY5l2NvtGHT;}0n>MVo&BN~Hj*ie
zQ!ME#cB}p~ro4nvt@d*%vz?qhs6VGVx^B6<yqGbzwCf4W#RXEFD<$nd6l7BiAJk{U
zuo{T)@)0a$b9DTx$QMZNFKGm%jx?&B@*Lvi1aJa&H!}t*4%7T37550{=x9j`<;^QB
zAM(%HO@p#Np85ShjvGI_=FO0oG5!U8-n3$+?m>zZ0e8)tvi!ANwuL>WTT;AnpG+uW
zke71Y`+uBmsz0t5Y16(MGV|MCkYs*9JGZL-=I(<+yLQf$u_U;U>)UR|6BbP=ZDy1f
z^av+Xbnc-O{MjnLFgq!~^(;bJqysXr-G2iG)4&PL%<jCmyW|mpY>)ahWGif(in<la
zx=rU`=X4lJoptlI4XJ>O`CmM(%Qq*m06zDy);sgXAts(8M}kPS1~1$bCo&Qet%HC-
zMB`P=TSYMbKhLvZ$+LHrSj1TsTa~h5VB36bS@HhTitmY9a#eZc@2tCIcinq$66#R4
zCN7W)gKIl##mSrI>Q_3`NrA3DDtI30;_bRk#FT|1G`YOisI|GV@j9DXwW?0^cN4Y8
z$*&%woO#vU6d8$R(rb)E?HMg2f>2zcP7gJgDJ+pMmJxOHNHN4`Kd3et2-2?Y)qBpp
z{j#H>|Dld}y^05l3m1RojeE+?JBE7lhfgQ(5DjTHppHMj%&txeE*a}dVzE8t!~%*#
zM!!&czHkI=W@t3ncA!H%g^GV->*JZ(yTiR@%~3TQ%psjlJrSCg&<x&6UF3{5u$zHt
zqIzrHQHh>nc~>MO=mqn$c{T5YtxBj`;KBODtMZ}|5_@y6q8pXk7G9GfVEqmX&V*So
z&&)Hpo+NtNx_!|Zxh3k|eR95BEfay0wndZ*DjTtlWd@gXuTTKczy*_CQ|P=8^t3V9
zcV@3D*tt$Tc6{SJ5?G{A-+y1s&&%~CF*S{Ky*u!`aQQ`n&8y1dW_+5mjQ#Yx2S_BT
zoZ*TNM0P%4MV1#EyEK<aTC=z#gk7+!Cq#r)!1siYJS3Ip+2(4#bttq$l>t$mvU_@q
zL{vACjDpoWBab^TeWsJ_UAF_hfpAz^D&I87?l$E93}o;L8t}~ti?Ewf99<(tK)S)Z
z0A)#CLX%~&e5FK|9eIbO#bpTz|G{Frxtvg|lxUg~%(eT03z|jd&msigy*GDB#GMwA
z(2ey;IWstSW_#06${7WTie0&A_6UR#oZ^f;DjV|tXPDf6z`{IkZKZyTxNJ6P71RIQ
zUa<u1Em{pB<)v0RwG@z)XiMyhLjV!t$5{yoo4}L#D}W4vr}YoVyC`0+&Pw>X2+{Xv
z2bUQ4x(z~X&WOIneae&+)v_csuXrypN;tM)h3O-6$yauZ+t4Bn$fLJZv!!kGjXRe2
zdlPqwUL3PHvvAKO7d{becV6^eqEm4TlO5IV!gD)`6Kj<oV|RG{gyGyM$K4TN8bC&{
zSu36|;JEPa*!?Wtgk0?SjR2mjE4sG?&EGwptM&HC_m01};<aO<-c<(51)mbNc)UIJ
zN%RTO2;fWLiTm^5|EyM`{NHaLsFsL^{L@GM-#dl819Hoyx8KMcNvpeV<9h@Y@LLDb
z&MFE;s^nfGSp48gMbVZ;Z62fpOH~I5j+7UpCT|@$ScZ$M?%#XEdg@NQxS(OI%Euid
z+lhm8TMmi_{wLtm|6c@-s<*-R<Ip&Dz=r^}3H0_R!}%V1CE#CxVn%a^J`Y!D8|#j4
z<c1^BtiR}Ya!(u*b<Di84X5rQ6%om*hA+fAu@)S8ujoi}gjsIrIo87wwk#QJ3l6kM
z6ZLJqk=cgVZYiy8r(lIn5br_3hOzcIWHc44c(hpTKt0}WZyIB+(e*=9G;l4|++;ex
z8ciziD;7`|mgli}ol`><1nxLpC)CrZ`dWk7xILr}Ed>(F7e1_R_R$%Vx0$iHzEbMd
z+$+tu7&#gMT1DgpDEd3(O=-+_pf8E($Wxi5#JKAZ#oR;&F_w{%+7UR0A&+Ay8;A|7
zk)7>>M7oIjc}Rt%)-Iw54eZx&x!5@%gNi0Ptn!)lOC^h^2T5H6f}#q-Xg&QG-EWuM
z6}RiPTw3OHmkNIbhQ<YGy>Ion$<Q1Uyp!oH*QYNRrYGveGDe)sd{<+9>ynP%qDH){
ztm}S>ohY_)L>|w+Wy*4u?2ZeHUSFmI9{Szk=#oeWu&bM>PJiTa9E#^U^f)fay?vo8
z!YsZ3o?AWGV5o1p>d`(_8GEnN%6HwMKoiGIz}m1E$A((a{h9BgzJ7eZU*4dGa82H4
z3v+qe0#~qB9X_1|wx_%>&=7lDe_4?qM6fh`7PWCHa>&iYW)QdeoZ3D#sWPMPn#t+0
zj-cT7h0jt+XQ*{iE2;V8hw|6o*=1B7FOQ<kbDt}6pW9`$=|(U@^T$MiK=el1Jvvw;
zYNXrVRYbZRr=U)O{!_DG8Q%+F;_#Ie#p)X+8u@SnvirMV?c4#-z{_o5YuIw~qM|wU
zh-$V*p8y>OFE5Y6I||#~wCryKGfA|b*#r1!I8KT*T4>M28ap(XbF=ByGTzh1&DToD
zY<xNigveHZULHLb2F_0kwPTqXUWJH8R=`Cuh2j9gHr6QJ=DN-%Ph5sP=iXvTlK0|L
zvH31qvqQS$Yx_ifhueE5DMy(_46+XuDv&2PNLGj9+-&(g<!L#8hj>n!(vVo~E<HNh
zWH{GM=Hs-n&UT8pGMDM5{L&a)<W*!0(uu{OTjMfxde+3T?Z>6ER=eter&9INe~OqQ
zAz>koW)hdu{`=zjf8K!bAK~Z*fK!WBkWpi#)AZ)q$+GHL(vfSY;wWEd$r|=PrHzl&
z!mqGS;?%G2Xer>Pi_(RWS0{0W_OO!1A1|y<lkbY`ZqtjkM*oQ23uN?5N#s;flgkw}
zoK<Nj{g!<Qh*K%$3c4$qD7gruILenh^gn_55*SqKzHkD8d&f??@_eiVe-xYG@e_di
zJjmz$Kbhiu>|*Qb=OjF2Hw)?NqW<c|ZjJO`hGVw6>3+$^xtFq2GzaVL++CaD_bMgA
zUh?hc@}dH*V^&iC1;sw3Y+g)SH8wfV!{rIq3Q_D4pYq0Z$p#w6B68k&!YPUq6=pQ*
zj-(eChJ89OT?)Lze|?eANXO#3NWVyQmh<gM;T_tI<kG-ex|1$#D3(7IHD@^CDBv~u
z>;^(I9BU%$S7JsIunrhcz6nck9)H23Fk^7)o~C6@D_wRO6(`46^EeA-sOEXDDr0MT
zL@(Ulb?X?d_4c;Z2xgJN1`VKgtHi=X-uRp`g%Jke(xb8~pEI&*@I!Bi;@qJ1kN*fQ
z_#aNB{-;KDvm}mlpB$l-EH>#@gN`3=D8iawH#Py0$6hH(oQ1S*^(4J%4qz?NyXddU
z(t+(dVhi?dT4Du~&yx#h*=R3T2R$EKXy=oBOHKSzMq{$Mgxf|xOm4@TWEqBevqH(j
zV!<<m*mv_{yt&}$xvO*t7VCocYY&mMM(hrg;;qn;)#wQuiZ^ZkdmeUlOh1*aM1D_R
z#I9ncd*$(7EN#hdaMO#(i?y+`qM8nHzQ$O1cXbcgKGqFV$t|JFVx{`zWB*_kl1*R}
z?v$8aH^s|kHs!kQ<xG=Tad_SB&8x;ZE|I=MXOVZ1TXIkB7tP~g9%e8JsZ`Pqi$xP7
zq{;*(kv_<bfO0X|_P7)F+a#p$S{OrBW^laOgE1_NjfbhiIAwWY)B=^v!&ZkCh(l}K
zx8c4xjyL%b7G<aC>)_6LHBxD9XL6PD+<*{?L{@O;=|pla7S+>7F)<XM8^-B-?rP#;
zNw}ET*XkGuou0Rlg(R{P&Kby5(cx`+idY40&U&TN&>j()ni0D$$jl`^7-=N=XQml4
zv*<op_cB4?A<r1!GW5JBR=k->y)*W$JZnkt<!ny+gV#R`iW&p=dMX$Uz7WkJk?iBS
zqbXkAKeLo)LAkY)VegnF=B3B&%B@tB@|dPbJLl|D{09rO?|JCt8!ceE$;Tbvd)>7H
zpsjc!|6T{}a1P8mx3(OcM`()?)-|$9_^V`=5?L(4pmEugll(jMIqyZEz!wvX*!<ku
z3H#=2uOh$G5#_1&Y3XYNTR8p)0oGTmsD1Uj`b512=GDl76JihX1z}~27XrY@ga=3a
zc-Yty<GRrh(damecgfKvP9{7y<PhH&GGvT;<x~Ki*OAG)WCpba*>p%XI|^%7S|*M(
z9`#30PC`7((9<eh`7HRv^;b?|*{L`HSkIl(N5r!5WFaiKx??%YRs-?oHpO8AELy|k
zZO0Ek2>r^?D=OSvy`h%KS5A=98~Xpe3lNsAQKE2qj9N=;$}ehJDMH7(^2t|tEiGKZ
zC#;wODsaHSEJGLZ0WT(8XGakV#px997{~i2&Ov=6WKJI>CJ^Za6cdce#bi>iyl|P`
za{w*|penw=NOI^R(=~k1hnv^H#$CXA)kl1A;hFC3+W@lsZY-kQSf?I1WJSRUka}kD
zOWW^pdC*}Cli5?7KhP1lll`6yCP898U94TuMdia|Bir;-a^S*NOl98bApuqmCSFTU
zdVk`XJ`<R<ozRcHK0~&)H?eVz7-*=+0uc0h1KPxyKP{SJb$%anz5twVxDXS%$Wc2w
ze>ZfIr)Y5{`_>R0Cp#y#RVtGo^?A2vrH-nt`%uN14+Bk&kz<%?cNa)d{LQ2N6Sf+T
z$O96s1>*%%D6EaW>jxEAXTCSMIi<^Oxrpvw()dtqy_+rt1M+zMki!-PP%Bo#u6m~k
z+x_uPQzjR|x|^MAUoU^ta59D?>Bg7Q)DpBe>#<JXQ{n5SMrBWm=1|Mb>A~8g!Uk^E
z<4z+Wy&8VA%Z}uZJm9?cgr2jb)2c-9tZ}3-O&e=i8-7$;!h^HHA#+NBg&RHF4+cdM
zkCK@%9?bQ_;ql9q#>SiE`w|s~Nk^C5VCXBXy<~9jq6xgNbY=)Ete<&4JJ8&y1du2r
zaEhq`+qrAO^9Pn-tlA#Z&X(vWblq7qS?|Ls%iSN_WxSVLqV81=XOXZ5^Ci`dOK<nW
zT)H}3q84kLY04VKj*ENciUPyl?onO4r)qS~=LgxgZg;^-Sqqnr%`R^|Ea72ejH3%U
zgjf`Q%$uRzY@oQ_x=S4Qx?>9s+;W{3O=h0hL@szL;(_3{fQsdy@UvNPHh4f|a)*bd
z=gua(T{u<nsotg1YPO$Q?kbTElf5&^NsQ)GzM>36v>Os|8TR>l^qe^qXtEi_V(;*z
zc$PJ%f?xA+^ruQm$Y9e}{IBo4<Hx7=qrRT7j_AU64?jvlFL=gpqax+?f;_a`VPx5f
z^^SG`S!<r1%a4soiRkgPckOtvIi)C=nu<e_3>ItpRap2hI5!rytYJ3GE@MlpI(X1>
zFqzGu`;os@FIM}EO?6ts5%=(Rh=mWGWeC^i0#Z&f3d-VCv)@a?SqcH}ra>{oh!h7U
zk%elW9s7Byk9VUzWNQX#G^+PG{ZnWP2_CMVH(0UXR?@<DoG^7ERN7XfED$_2r%T=|
zlOA4ld>@r~RD4;ZRlVlW77bQ1eSS_M-=6Qn`t)F=&xG<LRp4Mc%zN{og3~YgeBH!O
z=sn13q_;_eOT}i*uBaTE^s1cj&5YoE4ve(98`W1=n-elHO+dVXj>He@i9P$UCxE0K
zW3N)`fP&$S3Kv~*&7SFs<pi3FOFymK1BUidtFd@H)K_hv=jy9Xizj?w4Ruo&ZkoW*
zfvgR3GZt?T&Xy-M9UnluBmG#Ulf^%YHa$%4*dI=K!+rgrBIG3dTx-<ub0ANY6ut83
z?{IlB6@?qBVFi5;FP2B(JDC(*eqPk3H95r&fitFRI*Us@aR-uqUW(zl@YH}f*!H+X
zjLbbGMLzjq0~@S7F2?S|VL-4zWx)xkSEt2Z-{CsonG|P7!HH*7L`zCoB>uc45V-Kx
z9(Rb{iN~Kn7?JL&)i~lOj)@ZKM@~+nomunM?AUTEc|UP(DRCHREajre7T7TuqMTv)
zWmr6!+)V4mX%{Wo`I?dGC|hQcu~Ty`m$tKWgZ`K<US5v@P;Fdi5S+xX2fXj+?`R}b
zdlop+5IqaeD!rGe(XWEV--e$7w7|_|#I7+;vg!}9{QeDw-7Ha-Ne&BSt<#T_(YWkh
zcD^yHFBhBYy~u&Z*A<Hkt8vqxhqo}F0;y6wVsd;JOe3>a(Wcp-<|JMr8Sbh6DQN7$
z(>JSA3_FXy*kI&L2|wrN#KT(gdpS<W58)i0o>qt_dp(Wp1FJ77bN)ETrN40GSpm?C
zECwhQssIy>i<Gu3z9whDG7!%uggJ1Nuvg#LVBZ6FQJxPsscpP=moEq`Lkli`pM7Bu
zARXzxS`7$&__?wu4hl9O-xM=y_z<u#(`cy0JHazY>)iK>Z&U{m9aLvWdEie>S_9kb
z@5{FUfo}FnXCQE#7&<qKko(8(A=dzT+l4Ng0Wr*Gc2q<0@9mfdb~UztPC+B9Eh>=Z
z8anrnGX`DJASH5aUM;DJ>PCULHSEHcmr_sHGry5L&9Gw4@~{@Gz*KU{%cy#KIqoFZ
zX0C7JB|VFNUcTYzSiZe5;GKb(hwOW&w^8SDdvS9;i}=MWb4DdfZGq%aTs#b%=<pr7
z=4pDjIwM_KxnzBE3QW;;I>%aPJD!7WK@WbHy`@2g<+LDxl{gY+-rm@xdZwfrUMe=k
zxC8#MRuPm^L`{BQFRE+!9|>xO=|69E=?or}<lxvzntfwtN5onrrXOvA{IYvG@xVcG
z)54x~Uopj*MSvC>7bo3+Uuc^TgETa5_)o+ZtIHSfuj{-`ZxPi5$C5Pxi>p90Y$=gI
zmehvF!bO2^p$3q(g>Hm^V@JZ~IiWpSHsmxV^4oRb*s|bzOas7=C#RL(D@vr?93dg(
z^AfFA0MeqA%ww4)3^zzqBpjJS=l)D?4=KM1)SQ3<3FD(?0_QLB8A1oriu|oGpX%t7
zE6#)_+wQuCjd<CiT#>-B*$=oM!g2fUCTyCkr!pRP+7W8sMmj8DbWJRx9w*?KU5NpJ
z1FG#aWH#k(^!SuDCQ5EB#hZ+$=B>Q~+_}WbYI=4Fzt@Gg{Ey1A#~pk5B{%Sp0F!1X
z)=SN^xc%YQIq>|ts^7%CQt>5kVcwk`%_u7oqwTmoPF}51bM*J2DehG-zDq>q8{ihA
z;kB{b<`!{*c95FJpg>_8J>$aPR0$mhcC|)n11Asm{MCQi2M+LuSb8CD%&y%1+IJ&J
zJu|qE8Y-LaKK$@5*$kHgi@(rlEFz^fPa6Qbf`0*B`-boSLZb8c??-0;V>x=j0qhuc
zxfb!N@^D1IZi=|NuXzG6N}${?=T~yR&hU4!U(NjKpCyk00NW}~XtuNW;+WZE!bZ9m
zIUZ|Dp0iECDj1;ps=epLY^o4hHUUPd<SJHoB8u`Zqr;OzV+#y$k!MXm@=D1hnBw0o
zR9VKVOuaPFt-1R68`qQLkcAbx1&&Z;RVsiXNB_kOU`FerzMR_ptFHC8jln&ArN@s5
z8vpJuge(_cEMVxsNr1b~zoGFh%pIYNceB5u_%6JJzw<no;5g<z_!;6=hW_&RWe|Y(
z@;}5^&>D)N{Jk9j^}<LbPI9f`JnK$FC*j|%fv^*vf#+XoCPdzXYv6;PajQo3efrqw
z8gB2uXxv{C+x0-MM8SoRG%bqirgLzzxO&@I8+k0_3Zy7|XE)e(4mOl%`#p8d)3k4O
zZciM?+i8?DAY-U=tozjgYD}Dh&iwZdDv7m!v|r1x<CH3lR;V*2;W?qEtS@v6S?{$U
zP+~dMdGs6wk|nI*Rvp?E0PCwa-C4`@L)F0bVBpc)Id)*F&v4+!uQAuN%f&vRVlJ`y
zA8KQG1Mj_*=L~6yyx5-A1kPSGK{Ij;?Nob5nWU=k;VN4ztToEJy_c-eaywMfrP*UJ
z<`S#w@jsH=vZRJT+5=?0zx2i9VQTuq_vEx{cBo9?Td_{H?gT<11D<(1s$&5#H&R)k
zqig*$5{rDknw^VX{kxpdm(8}0THrLw)hB)@2!YD)!r!IhZXzlLx9{ZfDK5N!7e4@b
zxknCEnp$wtk!Adie}xOVBYruCq}uY4FGmU!FLy4_3X9pKQIfOQafw!;7qhSZMgyI@
zOqdHC;XK-L?rosM_OtBXxQ7e4g*$hCH~{+XnZNq(b8#98asGFXEv25{zT<}3?>o-i
z@jrg#&x_|So(ntH;1bfn>0_q!^KVgzYi<-Wc{LJZ4YAP<<qi}!TG^puxdT8VCXrBB
zww;;wP1zS*c<Gt$U+CZF=LUZ2){3yfl;$G$h>xE<G?0>?mY8WQrNK=EDX!C%P$jsZ
zb^Uvg{E<#?<NXQZFiR#>iOy!x;|RM!6BnCP3A0~MC^%cTDrVzUyESrN1d62nY=&;W
z)S^m+8_btqT%kU>CW{e3!Vwr6=Qpo(9UuZmwx7B*y@?dmT{UeM4fx%Htn)|vd#Ky6
zqsbX!M(X$*>k&1m!c*FO8Udn4{8{Jy_vn9ofP5DW&C8$#LK&H~2hf(Mv=<;AtSkCA
zBJ{+eN+epB=stL)^99!rv$rF<XE}|<^jZq<5G~QV)5hlx3TD2-o`L3#Zj4=lcWr*j
znsCq1y422)nRYmnp`x$imH9xJQAK*GauZbmZfdWcmlA8AZ?(apCg``<k>(hf%0@4U
zkV~u3@ruczA+B~TK&odKHD1iW1~tbTdgbS5(#}E!<-}(lx6CmnP>5;dt*Be>)ejm0
z5*)0g%vZP|-WmB1nK77^24Q~I{_VqhbgQK>v??*wu)F2N^h%)lQ6JRHq*}ZZF*UVG
znP{T2F#?*ff@bDreO`_D<a&~LKxe4Iy_W^ZrNW4j9|HE5rDz;%YLz3WO;SvV=00d%
zmXHr3a^Ry@+2|8QWU7$~vD&~yF`srDYK71&yJDoReALxUiYl$5udhrL?GO!_IbltN
zFe69>?MH}aCfrX=Q<4Z@gMA*`uqWGich@@pay@DrbeHFc@llO1Ge=NmRW>3qFi?aQ
z2A8!rph$7RmSlYs3qXiC<@$*j-psSR`)?W~K3=HLm$kd?J@83M+mQkN>^+qY4b1tR
z3dk^>wg5u;{&oylUE&!U(3d<{$DzBZGx*ddkk{~o@`6OQwfi81WytS$gT*SSb_I5a
z^D>Qf7mf{msb@$e6((BXRf*-Q`X)q(&BU^}Q~G7sJ?{v(l8Mp+126^xrDoFJfby-O
zm{=p?tjb-n79DnW-9r^#TzTHd&3-A(e5HOdBm<NJefmf&>u2Q#@e<*_7;9zyU%U=h
z;#Ehkdh%u8-^8E_5{XgFuJ4MN(Ri?e`E8)KaFth!*$nOa<GQ=`3Z;*$DOGwR&nw6g
zlcFOp2Kg0lfUOzp5eJci7JZ_cq1c^D=wL%1id~b$Y*Cp)(=M{DI3P+W0o;zVvt)^w
zBBA*Ldax-4itvp`4IKGC&NCC##v5Eyv7OJDOxylFU{*Fh@O6Mc0cnZ{*$kg)5+8C)
zBwV~B&TO8&27Ny(b~69pHr(ufl0rC(fy{5BS9g02uEyGk&ru~Ru3gHWgy0IaTu-L%
z{LX<-M_!0I{_M5Nw>Dnw9z`*H{5?qpIzTO8|0*WtNcM3V@gM&Wmj2%p_aD>arbYjj
zk3SF|(V)e1dv803#V9z4^3YQ)JWzx`CaqxG(f;>P!LLNA?ETvxyotq_6{6>c&qMRF
zX@3=Ld#+0|UkSC`b^3Ncrm5=NF%Ox8O^IJ<pp=q^zWE^+;95n(xAV}zPq%(N#Er(F
z!1CMoEq?)l{R|%N;agxhfX>wF=@N^937(sQBH$RsGxEDHncvB{Vx&yurTUjWFY8op
zKw<jfMSCUU-q))^t&BHlauKx0mYrG{sp!dV`uw&2RI7!G$vS0XYW^loE@pj7^o~Oi
z@;7F|gr|C)>BKsa{`%zI(5@4OXXLv-xCvG=d^(fmp;fit9KI7=1!9&KmyH36!x^vz
zh*k^2k5SZBbArzGUrVIb^jpoVgg@5_jOC5Zy_rsi19l!gwb%WtO1PsIkAX)pVmNrP
ze8&g3)vTdghnha#RB;qXeBTq_15I2y=KFPq|7!St@8kc3AH`~xyIP0$7nL9or^IQ&
zv*5_3zM-VlO(Z5EwM1pZ32@o^w-OE1twX$g#3rk&xre&3aThdkY2_ZsEKCr#3eqqS
zxq>m%{YmUGK3+Z8w7Ogb(vV*gmO`m%G_~OGMf(F#l|HI5lxn-dZvqi=J|}RTpb8nZ
z&wWCLd>SQKT(F2ca1qo;h4g11S$y#%INanhW&mPBG{m?V65|cM6w*Iu4^@h(*g5+7
zmV|7p>2LF@stwn3wG1)iCd6)xks7f*b-{dP7!>6#Ke;>EDug+ZdaK26Zf;fv_=iGV
z!(+s4n1!KR$2>=$(oCT-!?%t}Xl~0+9akmlWbMCk2?sS!qm37A+frtA2gMGwaTKn_
zh2Wrhc{fm9_xw|=LhLXTUpLF4<>o67v*w!|MS(ObDiUJ?7a8Gs%ml(<w?yV`cdM1Q
zg1ea?3Ca!d{<9J=)i%SZq0$ne@4Xe$rEh+#h9^8%+|p=7KV2dZ`y{Uuxz^YmV6H(G
zmOfXPct%i4rM)|3-GeL(&3#~cv^QcdH>2O$1oG|OehG`OA3*x-Vb4&}$=NrD3n4RU
zu-VXjp34{3(ZlK6nG1^w)tfglHW)dK)*tOeXbeW{mZ=St8p}L|4EHj>gNm{QNZvL+
zt-$$Us<85CYKL^Is>ud15b*BlqrHJ))uKA$HIb#}g!X+?3BSF=>zDEoUW^xd{$Yxv
z<;wgwO5Ji4)XmmYEY;S7k0x`Q5%z;pN2i0MFh*yHK86I7lf?Xv(_*VN0PXc*5`LQ>
zQdq6TW5!PsVFm;f4OZe^aY#C~We;_PcioFWrxLzt2W4PrC6RBxhvNIdUeKjgg6>0u
zl@MY8CK|yF5=DbJugcfoBy{7Lsu8z*&~rbrHZ_TAsYagkhvK)dj{&U4L8*2&t{JZ`
z)ro_g$M<L%SNC6zL&T&1SNx{Par}O=1k~V7-#9n<OolF+S>l(rZePJK#1lG05$##x
z^TjwxGavxvJf`J<vf)MZA3?iR8eWL;tb~VY>ZI3Cv1CX3QWuDjLF4I=#V9~F9s2#+
z)&E6t|33y@{GY5+cSc^q5n6@7@KgP*stk!Kshc3;&3~Zdy@0n9*ki^Gh|Q^h3HR=L
z0)YUKA^HqaI(6Y`#4V9*e}`WLA_G6QNu3z1lr9|UTPcJB;MW|(IZqM?qGKJCxfy8y
z=w4}Cvu3zGroHGRvE;I8%}}a*rx+bS{9~%A=~rSoEJhfEDO7E6#B8%kZf}DoGd9Nu
zHh7k)n{MVzN?E*W#7K{l*mMu)0d4@uTB?5oyvA#3=1EM>;~P<B#<aIL=a@hQ!dh+2
z*$RX_s8y!r8!?s60F_jP?Vx);jf7cHs7{7Jtt?l%x+Y@aUE&^b>8XIn;@J)gqq`cd
z7H!1$K`XYDo{?|e8RUi8W-y;^6FF`5(lDppI@LeBs_Y6xov3X?lHO{jtZ$_3$j>?M
zl^u2y<0BvP;kZ8Sk!e=47ig>=L3*9xmIV?%q8&&7ML9_{<Ba&^%c0jkg(d{Z;8Dyz
zkrxL6X=0UiB_Fe?2Te8+tm~iGZC)jrnBc<}&AT$keb!E}--GInxnEyalZ<&iYIS}t
zk>6gAw4edcsz1ZH&l)!-!c!Ma%HQ0q(k51?M!zBn6(7;IiEvZ00ljGByv}`BeAvw9
zD#^O*$a3fYU%3S!RG!6U7WV?i?3EC{1WO(NlIfXDtP5C;^vkuHk>izu5Dc3IGBhvp
z>+|^8$-7c{&i?&HP09RNsKQv`uxaWhKZduM0e6{}5p5+cbOHa(q-XFkRNc+dkyu-f
zUTk=;Te{k3r~&NK3Nx>wkGHNws5C5Bb}D3-I@kA|tSNR0Au4Yyr^M!f2<fy%^k2=!
zL}I*rn8SX9ug`$dNhiyjv@%@795GGGQ4@X+tRfI|JXXjLr9I|T!_?CAfvYGDlXv!F
zXk*a8kA=fJsX6t2{RMI;7eYk2G>Ox~X<itqRy|=V%$+D4b_Du-4qf;JrIHJW-Blyr
zPQ@$cZIdzz3aqr8{_AAfhnWobJP_hiNJ#$m6pNL*^gK^tE=>jfSMDR)LS3$>=u(V5
z43>()y?J9@8H$mM?S;E(JJ<O?1sL;Y{Yyf%XJ;~KBMroa>&ZF%kntMi1xvleyNNHp
zI+RjS0+&M+`PY}tq|6=>zEa5Ga-!V2_wHI4lDyh-%-5g;njR}#;hv4zoY?wKkGS#@
z6u*DCzFy<^#MDz!m~nJ#3?ydD^_*K9(_47R)EbT)Z!L~LnuE!?DH__{&#OX0m`-s=
zR2|u%hVD2%cX7wV;LlzkGg`scRd8VBjg4VoKsv1nXvfTa-Y2MHp!fbb4eDk*(NN;)
zdx<8G-~t0T`dYy9Kzsbog`pOxx{GK|^@Vp6q+Qa-fTbuW7EAMeP62eO7w#J_A1iF0
z8-Di3lmM#kDC*fqn%sX&(OL9o*y|?5^=%*TqS%7d#BGN2^_RXWpDE5RSUYEKPO9Wt
zT^Z8TeGS?H9<rUdiP45*;e*Mf$ycqvaQ)1g_sLtlk}rKu4z!NqO+4Yx`v_}7x4vCX
z-3)3Kf|VPtje<*>?Y?mSc^9Bh3ge>rs%MX%j}(=Wv4`)X#E;6>sJb>cyye8u`#OD%
zyM!Z@oRH+&Mf9hc6U}A28mzkH_Xhx|jd^29Kg<@RG<(RZw9kB_f=s#%9qmWg&qYDY
zjw}i5@;m`K4(@CS+^I@E{K@)iU**|ktA_b9OM4r)&YiiQyIx=Npy;;uRjQrHA4*@A
z!@FU^#@`NB_sk7D9yKbw>`bZ*wPCa{d@Q==9mZU;XfF!x#(6E@TWf)Ags3RmUxQ=%
ze?vlw)4z(ml*Cu~1jz8SMNf<}@lVge;FnKM>*0|Lmw1s)V)UZ}A5)VHKwia=N;bw&
zn9Tr;@KQtb%B~aj?*pMn6IL@p^_EX*9*ci$;h3*1kK7l7so%+HE&jb@cpLBEzn>8H
z?YCIjd_pwsfG#jDh1?v~zxKUwuJpjyw`1^0Fp`-=D}a{eYk97;oX`%;%g{2Mc?vZi
z+GwbLd{In?nAZO^zYceq_it&%41<X;F-DriTth<sW113_ilL1k4#1aj_gjVJf=se$
zbx<qwkm)XAc{Z)-w5M}e_s3*8&ig@ISsWbioI3vN>j27eq8i3U<28l<%w|qCHToD4
z96wNtp?$TIueV%Tg$91!7)uBFsBT=vaFjRxz~m<$QDSAmvUBaC)@wls8Kq>BYQxx(
z8E}?YRMJS%&~r2<T9x=XjR3fyJKk@fegM3-<;o-oWgZbQnnjC(wqR&?Pn4gBo%wlk
z=Om=<OVRwO(fsse(+-bkc$JaIH;}C>n4a0aX^r)+Ftn7+)y|uk=?jT|FC89)(cWm(
z6^uIMjzI!0LdY|FPblu@MhqW823W6TehSNw>8<YV!$OR%B}T$hqstzk9qqf<WvZ`~
zo0j*HIn3b;N-`u%n-&+Du%0m0h?}>tSwA!P95&4znsrpzHK!)AjiUXMu)A4_)pB{D
zdC_3rxkKt&b<L^855ev(mbI0Q^@e@%4P;P;s{RXW*XihoU)4!R(L+x^H;pZnjNaXj
zK&LP3cq_TZ$EY`>M;sjy7I!$@6MLQoCy!{${&{O7!n)!}mibb0s9a9H#J*d;>=;jO
zEvVkSq1V%<OO{*<`AMPX*;pCAg{QhP1d7cIab8RU61?L2QpLC4UQb6KPq$=d-Sn&;
zW<;enX%aP*^^ISn1I(d}#Z;gIoJT0=<FJb7Z$ky@%3Dx4(9p%I(yJ3~x#8*#HWQX+
z=&hL!hXTUv7oTTMHqSgc9e7K`n@?q?&Z#-6RZ@g(T1kyYo3rI_jEgJlAhXw~p8c1O
zS{00{!)%h(SSxuRrw1Ye^2otK)UBs7WkC^VF=Jr)i=B#vsAXYfm+YS`qx!3i%_q{B
z^JDBL_RPtn!uv5JYBq-S&R~SSO+X2;uDOomLkLLu(z<r=AduGRB5&>MzcF^ydz^jQ
z5;%abdmi8*JEi08SY}EAnxVRd=sncC)f{;Dv(QL51_A3bj-LJWDTTAWH*=;oB1fb$
z4nLYHM}&06(J6Zup^_c_waQT`d-c9Lw_BdrR*&kY2Da5fNa&H>if=nAFD?51*x9TP
z2vTJ%Y-lFUpj_X}6;(cf+<1aKJ^V{e?zIj-se?_`^;g_HM`Pq0uS70S6s^|RDTiBM
z)SzBb2{&rEDv<I%BHbQgtr=V0qMH&42-5$^>ep`)(+0<grZzj<37SN>je3u3f+qZ%
zq>cBA6^2=`_QzKrYfOfH41swk08d1Zd`&NwuguIc=u7z65@t78o>Nt(0UMOcty16V
ze6LPBk^TE2;6H-UE$q3J4~k8vIZJkEB`MkuauP4c)K<mB^j2yrF!!XBiaOgO_G<b#
zV>ZuqWg6n<PXU?wS5Q1)Rx*UAZvKVImj&<4H_j^BXAiTDI9@N;s*aMiYG0m}<IJ8R
zcVzjGL66SYI8XTL44^L2lTrj(frc-S6hgiNNvG_LPC@U`^TNV+Xsa%@kFEjpZP_4q
zbB$S9sbA}riBHq2%zBN_aF_C)9To9Gq|pAmBbPqEN&|r{R{jEwrAF&|&GwXqVm7V1
z-t+9C)Thma+s>r^t18~baQ$LZSm0Gh#hIkHs3D`UtdW*^YFH|->Jy+X*H7ypbFzmS
zvro3S9ZKCan?smbu6)=-cbaFUu{Z;l75TMOqi6i0^z>g#Pb;5bM)h4>83b`Go<7>S
z({CR6bseNJUt*#;(>U&(a<K|u9plvHlhOVe&tAx;4W3A~;$NCWkS>ONlJk5QbtlO~
ztVEfQr?0AKO3%_ENDYgpE;Esb2Kk9ddXBe<P~*!dw7W86Pr~Vx*m3#nsk*ra4cXTj
zy+wc!CBM7@7Z&U4rYst(G;2%`;$vdfSvB?ORG0DMP#C5iUX&`xYEn<W4rusdo1yPW
z9t}3#T@CfWH<VfD7O(qyD3n2--y|eK!c(<wjmO?tkKnh1qym5KNa~2X^Ip3fz6ufv
z`Esf$#z@yoccE7L6Asg)H8c3$&t$mgWSety@3as1iQ{}#muK1OLATzpvDNX6q^0Bj
za@_;TJA=ow`)y^jJHr&6PY}W)T<OrYQqV@Hc{9PuG<V-14H=vX9c?e#LuIUFjlOh?
z=@!WJ&(2>)eMn2_mvzQtOZQM;tu@A4a`U?<E*`bX*;vZ+Sni>Z8ug02z@=}&j>4_n
zWWci@oi7Mzc}$H-936XO(9b!8xIRx6>D-9Py(IY4(ND*cqe21CwK~jrCKZZ1(HssO
zvkt&G7;`itg_^ed@gJ15V>Tl&V=DT;%s@!yb91LpH20~0payWZAa}LU()l_Cr9;nO
z+64aTW*L%U>@}J-iT(p4e{|6zyB;)B_*4`86X4BBC1t&-B5(e?<RBZ%ztxoq4q(r$
zv3&*<E@ap$Xj@9XY^(=-f+Jh%+O$(a*qMjZ)SQCR;ih^#=Ck3#1*n@#v5yY1x}yY#
zb0YC<I#xZ|fDHF*b+aqfh-N2Q?mw>HNc4Q_?l~GM?*v_Wx^_^LUvs)G$~^Kk^Wuv0
zi9@FY%tx5pFylJJq2Fd&o`i+@`ke#_Yd`f*!Yr}c*I^nZ<>ny?HLIpQyZXm6#*#0`
zEFQVN>(ob)esymi_dob=0)$rusCs?K4<uERK|?arl_yWM??fu{sO!hT)|U0TEPB3K
z)Ub)y2W)Wa*r_@ArLl5TDeFIZX3+g|m&*e7P(Aw-P(B#A)USB`*ESaQn?&uD<d`N^
z-wQzC^Gm+Fdia^Q*(*>a4nb?|o`DL^5G7?20;9_>Y6x}#WDypcN&VxWZ8&1e){pzk
z{5#4}5W>}o{tMAP1t3JJ5d-2km+>f|Zg)T-oVqw=nHKPL((6_?Tc)2q|MB@Zv|}*y
zxYt<vBswy3+1Tga%24SOg2#j4(>a{&tFWQkh*7|)d$kAMzV4=Oxib6YY<GDE%^I`$
zY?WY|YNRn^No|0}fTz*bPOO?W@5`m0A`2DZtBlR1UQVB!7pAV<IbGaXPocH=-vX#~
zxMV3mciz`sv9x5u``!xRDbW~r=B6MSlww|0_`u-^1@QW>p0bxr(vaHKh)dl;@{P!@
z3MJ^u^L;V-u8ZTrfip~kez`~It->y@&xK!O!@T&J1KOXYQ<w9e<<j-`_c`nC+vAqC
zlnc_h_>Ih4g;F1dj9t3gLg6Jm*<;i4m^P$EjnA=!)(yT{TfN`K^}O(KKI6CfSd()b
zXX*dm=}A9F2QaEz@&NesOS4-l!YYEld^S*U`1#@)vI15cB@F5qyS{vC8tEH)AvWJt
z4PTUh1!vlNX~he(Y?M6>Im;VBihxCEo!X4)sL2f7R1c_m4?RiXRaCMDt-k6mhFr~G
z^zWs5&gAMHE3kz26+da7euO?!zz86Px~DJdcKHDgSms?Y!b%*jDXc2<T#~}&Kq1Jz
z<O$lcCz0~1V9d&9@!G@!1kbUYt?2ct+*f|O`<8s8SpksZzH2apWWfrhr+-aF=00lX
zz^@z{lD!Q`7TjCg&?gGlz8~A53c1Zbl_lyg$L+PHG%jisXcWe#Oqdis!E3UbE&{MC
zVzFB*dN*zi2vsq19B&<Qs6rJDG-QR;e~1Kdt@rD3i|M=LzxZNBoT+=&vy)0;c%A-}
zq&F&%seW8;I$?!RrrhmM;9&PDfWvJtDX^~C$bTY-wNV$27{fe-zfQzOosE<pP0iX@
z^koTz|C}(N01Zs1##g~#H662?xNH>iJ6g36%pQi%H7z#I&lfHNK<wIK62B;b7(8(#
ztI?j^Oo@Me2@_HX0H*PcPv~|6CM3R74bv2Cnd{!gRos{Sf!b{1uIVw>0|UU<;l-)j
zP`8|wmYJ_l#w_sx!8Tnn#=jDQ`w!>{Tgfr%#nDXwl9ihVWZwt~u;wW$0|#yGpAG9d
zmLYghp%-(N8@ReZj0gb9b(IEP(ic5F#FoQ7=3R?_X>?P!y#_Dd9^sw$__)%tPrzP2
zkycfnzgjn6hESfjF1S+aIb<LttnTJDC(@16r)!iu(Vo6^C%d_@{NcG^{H}qBU~bB^
z;Es;h5Imcoe)quwxIY_}oqIki+&@2jSM5mszTuRlz@zS-x?E%t;8s8lJ{JhO$PAS3
z!EYg}c^<+4!iSoeC>mYkIy?jBvtIggS|xmbJdDCB-B<i_A|&I$w95sgLU%B5&Aa9F
z8DuFIj^`Wqu$-xRA=&OGaFah0tNqucLRYm&x`qktj~Z=wYKT`C?R4NS2I3!!CjoJ(
zM>fRqf}xd0;&GUF8xSn7%+*Gu1HrQUv$oQ)OH3Q+e#;dcRN)b=>_mM8P_u16p&gS)
z!32PT@J-RGRXS{A)7enA2pmQG*mQix!!g;a4KWzgtC%<+Wqk<I3=^$IJkMI(D(T0o
z{e}RLh`GjV{Hq~vDXRw8wM39Gjiz{2CAAs*RP**}w3gR!n){`dM^Bvbh$mWZ>k-hF
zEZSYD@go}M!F87~>LE>aAiemn=YP$C%PEV{FqMr_3?lV?wc2a>g-a^}nR)qvmaJiw
zie>)+_*-*KEOdl3rz6{S)!+=#?gWXcCX>A58Ch2cxTj~qT>Yl}EKH^06mj;nLTsT<
z%EVLJ?QLC%#Z%5Id&16H`ln7|1DHEQM`urszO69G9-cLiBrVRt#~BxbFJjD8BLXnx
zpp9!7d1#;2%54zIG9)_@ReD1(8m@3jX7vI(7mvc=!v@_4Zyb{xeG-xc9J@VDBMdFe
z@a}>N3u;PHW--8xfWw1+Q2U`%D7oonrt93Yl+5KApTeG_soM>otJD#L$aSc8eSd_r
zSCnE&CH3qg?#}=fW#|Bf;inUb8j*j@D)-~3TZl@ede!0oSUEa(EQfaI2|wm+d*zvL
zv)^i;%EAP0L-2*Ev`7Z{JVZh^3-Lk1_boHof$<DP(CYqOVwTnhRqqKOs!_Tcxsj)p
z`Y`US!V_9%9~Yjz@6|O5eDG(ez-K4RJ)M@D%OWJo?P&eO#b`@rA%VFW9gf*%_Ny?i
ziyLJ&-ck2N!5l2FROY8sX{ozEppmbTrgb3%`Qd&&sKM0yAj+bPw^Eurfp|)jvWb*>
z`^&KJ&^=42Tfd|5GKL90Vv%Z(*$ngH_d?k{zJ`TiF?o+@dD^2uL0t1bT`d|_R$>xz
zuQTt{3EtEbH^R$(pCnwnfy2HR0os0ZrQa+Y!q^I7Q}A3u(sdh5s@5nAvjLxu_}Gz|
z9h9+_S2zZivq&w=rmhB!=?HN=9kz3BV~hBf!tcFUMTxS<wEl0(`a2t}W`YTspAI#(
zdvmGj)K|xt<(qtasM}Lo$rEC7OB#&~jq$#i)9=692X)KPs{EV`nz*>K0NPb<)2WlX
z3B_orHfRwWpn~6sf0|y$2YQWJ%QvbLO;64ssy%pGW!Dsq=L7fhF{S-Ar8k0jPKfN2
zC!lR$75&21bwH!~L1vZ>z1s|8C(@cz)8VS;#-Gw4Xu2CTWFAybA0R-yN=8x&YCb7l
zcvwDBiFd_h6h`aUyj&G-!)#i--keez$Om=WDQMH>x6f}xN0R!VLfN_ht+{jmOETLU
zc%3}Qd~5QQ=ag2)%!C-DGo~Hyw;H>F7fRDIyf@?piIh<T0lSSIr%4*XOWx>Wc|$Ex
zQ31_#bR^Ts3#cg6cwqz)LlF^y^EmJOdH;g<hxhsUd7jT+`?J>Cd#(LlYd@`1_p7ue
z&5CsE{25p3MkHTL4&`6}r?JSXc^-3Y;%+LyO~Cmlzu%W0=W*<LtpwVo3}T{vZN1HV
zfPP>|+D1gYQIa>)5dvwSC`;v$uJFH$(U;vx9Gr@+$Au0z_l2{2&l8XGr18IXr`}f;
zwIHa`{;F4CUwA5>k5{d&)VXVFy9v_Waz(plB5&w#H2}Wl(-bT*a~=WCzD6q7w|@8)
ztf}7csoW=8_CVFawdWj@ZfA_0Jc_o61CqvRtA^K{bo}aT?}J&zNGccmdbK%TY0gCK
z-y`YU7RQNtK6&H{$d?)19(|i#MY0keF%x@;-q=rsIf=n&WQflP^NG1D6=QQ-{bY6D
z0FC#=hiSyvyS)Z7oLb3HPki)Ut&+c9AWh*_+jR{3U4M;ZhUF(m^wjl0d)G%V%g9&N
zM8Rk}yPWvyC_EG2^N>*H`mJ>|Jay&ZDsE^wW-=YA?(~@zd=rg;SqlD%6IRlgqwgW{
zqleCflIp}R<qKbOA%+0>UwF~hK*BXZc~L8hJYDHJudXKPJd8D-ok&uLs&T&zI6O{j
zbm`=L%{z2*aDKV)9n||=SO~${h6P(9fAzd#=m_XWR#gv%9CKF@nLi|-dv&-c<Gzom
z40St045(Mi9&TFW58;4rt(Yh)j(N3;A`kg~bg6*tvCbQ+NgnC#NIXtNYjn@H$=r|F
z%s23sby};Eg#f<uAN(M#zmt~xPlrnu@umMGFRJlmbjEzahbijNaK;FMTnOx|>@}=h
z6Xg*_bJB>3G6~LcD}PbQISpG9j6&kmzP28P<#~Wy%9DrdbA<0kGg0;Ih(}HX7>(xq
z9`2coe94tkqp<YD6g|pQr~F@ei|IN3ZJ<<vf63%4RpH{-KeEm)r5Eg=P5RCX`ts;^
z-{g#Yx~HVS6M1>|O;Hyl<8$q>#nPzYL}#9P6Ngqpr8bVysl>LAMQiIel&g<Elxo_4
zLYJ6etue8CF?LqooubHeF#dZ~u_zd#Sf@D8{!RV;LBR{xP?;5q+noo*EgY(sBdX=D
zY`MWDC#~K`UCWjf*x;|j<Ip^l#brBG*0r--+fcs2urwu?AIKwQ+x*z$NRi?unsw{W
zYCQ(n6pYu0NL~a@u>-UXN>|;;vI=fpP_a{eJ;1gsFNnlOB;Ff6De_vyO)ke!(?#lP
zFam*Z-{n7FD<K^+4@_l%C3PdZ^`f}N{po>~tiS=;WsmBlFIU$6_%7O_7OFVnuT#w{
zH|b&t5Y@R%dSo&VOrhu|Sm&)gqlc04i<@TD0V>=T0xT^5O5m9&2g3}_QF~tLI;ABm
zK1%~%b!LXGXZ?Gu$mu?bz0j02F=qt@aRoSGg#ON@j_|jdcVdvGP(%s0y#d6P)B9aJ
zJu%3_QzCee|I(u6IpG}LYzGG=ar@{(`(&F`O-^Ll>1kW-U{U9xnI{CsJqZeO^urp>
zr3iToT+7}CMJmGG`IgP{D#%b+h|=u|i|kW@;kaH7k?8L!&Vg(%$kx&YppHg#$m%tT
zE3zVBO6R_aKMLbm_cqfVv8Fb&kkaokW6n<oT-<BW^OfzvRoJn@@xXRS*#)if7if-A
zogAe4<Seqb>D09bjZ@TGqx~N4=#!Le(r&=3^uH&ViSy{9@{2h2C&ex%Ju7%t$$~HK
z(rrmeeZBy5#w^F1CSz@QAE9A1?)9@ohdMg%6EKl!w=>E%P04~g-2A60aZQxio#S&K
z^BOoqS*piv7F=u<-B|hsie0`UI6&VL&XH&ievw^X@bZs>qcHflY~&o{G&95d9y90D
zZ&2A~v|v!$h+$92S5bbl{pAyK0`IQk@+gvTBm&03V+=!XoWZcCTSA^2ICaa$aGM+Q
zWE!h2KlkbSMl$FLkyf9htlTiR`^nmRirntm+%T`ED>5{lo6>4Nr&L!3-TW+e=u`KW
znEf?SPy}@*1<D>v56vX#^*>FoozEWJZivoPpP;>Yz<P(>xn@kF8krmI;%-Q2bT!kw
zZI-+@9}_K0-VxYXTfRaD%`~iBB}ht!hEoq!gGTN(-H8OS7p?x9A72UNFm_MhAfxoR
zNPMrh2E*?@sMD`p@a?!C+XO|k%i(84wY%6$)GyLL8`4viLp$3A-$xcYT}^Eur?RT6
zaE<7abkE#lcJj2<{h+I1m{f6Z)%q`k89l@WZnOnRtZlXJv&|k!OIbIMY$%JaWEz((
zDa8Vfom}{6I(_S9#O-}IUvN1MQPy{qt)&bGuuS;+8}n4R&=2kG<T~}&p-Pt;9m(*p
z!I#N2=j1hBJAjGqN!%0+VDnfEN|*pq8O!NfV2D3?IoOyv^>>5`us-DrRxq+hoIN8*
zH9fEns$Vwdtw6FiPVn6dX%kIz8cfJYsIz?2)H%@vKWGG%da0<}2TDt~d45s7I?(Wp
z6%iVq9+tmY`ehdco&5X3+Wi$%>9yqxLfmF((Av6gh`t2%P2L!5>u(4g(7A<La3xA)
z5^0$j==C`d%`b4ej9mY9JF2bCB8BWcd%qK3wYdBWH~GzRIc(`KztQhgiR~-{b70V7
zh!ct{5)sZg!<~}H?lpxc;}9qkhMG{@HhrM?w5TBEcl1+CsXO;9b%v9t#uIP)R5tj?
z`v2G$D!8Zq)idL#OX)WF#CUyogueNA<nw?JfPqwKzDvEx)aonK{2yTdTU9-R7B8Nu
ztYI`=8D3;WeVHPm&xB=+Uk-1}wplN7J5$q>$c9-VSJBYHdB8~@Mn%jw={j0fg=&C>
z_#wQx4-cwH8q_g<(vIQVA%`x78wJ}%fvfGS<_!qyz3&COK!K~>sI+|1A#_WAk%_~W
zK$+M;6wyx(*Qb$<iC1y*uk2apX+MvacfuvXEXVtn85FCoVmmqF@^XjFZ4}*W3KEuq
zCCm+<gyez5X2%~P6m9v$9%(r+FjHH1Z&spT8alI>39*y{kT~-JA^93I98@91T34H`
zTI0kn$7?nxwjA+7b88w3D-(xj`N^#7sd03np3KT(tpy!?H3@+5DcBQbCz~#c0#6eI
zn%MFdZ|i&R4m?p8qezSyFoocg<Mq>zhzYlV(s3pu!jIJVyEg+JiX8;G9PIvlRKLNd
zgFm?WNBYIt-;Rg4Gg;&yYBbE9VjeL6iUZ~8eV*Q0+sg~gedh6M!ja_fT)$pI@#q}?
z)O8;`|G>|{8*oC5mg)BO<3(F)A)Pyp)9T>$u2Z;SPLt-rySprU^|Z>`=UZ(oE05KZ
zJ1Cw$vx*z!ueZ>)*T#?O`V{k|SX{(vdnt$Ec~Q{e13Tx(eY}iA@@p4is>ggIqBhZU
z6h>eCnE+`*c1u}FX9zIyGbj9;luSZb#3VG+yyNA+uWDPXJrk!8)Jgk3T`={jD5M=+
zyPEB7|E$Jt?i7|by_SN~w>>!JI7u1KBy|MV`BY647M;0gU2L`LfpvM0IXY>ye0}`0
zWYFU#fT>QHm@xml6r~guJsOzJQ*+Fr9OpjlyQ6!rD!=eEWF^3r9p&Kcfuif4qj!gn
zOoPL0Ly6rF8;XYy6nMgXJmUhU*OCd|>NX2c157HVP>FRf6hynetT1v`mTW$}En!f4
zc{KKDnae3}tYTx5KJzhT<e1%bnJI0oBnm4WM6J2iQx2(&;lFVIz@%Qt96I3H=7FDl
z8XgE+v7Oy$5j|Ahi*Gbh1`dm)S%vnERTq>et<oa!0OPge_;Pz0&lZVKDJo4yk0&Fr
zf<@04MyAXTN<&yrN{x+Oa@xTuX;v}7bMQ1-*ujKj^%`81RGMRDC~GKM$_#`%bdU}9
zsBf#lt>8W@i&L&+bzp~iro<7op<KiaR9xLiA?d>3iTmn`Uj%&PE}KcWgI!}p35gm4
zaFf9F<{bgS?dbPqgH6>Lv0k>Bn*E^O=rDmK|9#KnsS;I5H1-W97hSR(vnO*1O<}+L
zP}a{5P;Km%m7MCoFNXAf`ov#%K5##-MTRppuPEK8c&n7(V@nynaFwq`kMjf1FoCW9
zFMDGA0A%oSVhz-+YH{sWmV1lamCJ!v!!7b1u*IS%Y781aG=CUI#<o9k+-3^arc+HT
zTqC;X=qZlA!`e(CD6qA5@S@D9*tz)GTI&j-VX;`%seeCj0|tPI88I!2G6BXDPctkN
zB1UI^m0S==iU_5pGiG@bZ0XDfxIQSghVBsBMhmF@p4<VWN?`8!#<Cd?v@8P9QW+9v
zs8;EEX<VqylQBgnmpbI9g22eWeba&}DHELaK3r0w`r$*3C#5QPUwbtCwt-Ghc`hl{
zKIrruz#q*S1%nouppuyo=#>K*^&g9A0A#yNCE~k1;&(4UW%s2JpEi(ddY)P?9)lU!
zEHYr@1w-;%92?YG${+;)VY|Av+%;^i_?Xc_bdkCd7q4gTRMrBmG5X&pLP2}NXb0U-
zd3PYIq$jrV6$|uv;o$`2^D^_?ms1vYvkbQiOnXf*dCXLqd4a#m(W_p<f4Y<JKO{fF
zD{`unJXO4PtHxk?umst)9jpXo--p%EskwM6(W8s9oWw8BzOxfO3`_0sNmDM^E5kz!
zWAv!lXWi_g8S8l)A5(?q$Y7c;Bb@aP5qMsBlmeg6ahDk`7Q{0{qG^NDBbW>%)gh>r
z0)qvUl71qzy(Q3g4aE8Ox|>5VY*!`gfjVCrCWwyo>yLaEVIfQUa~C3e4V=AL2RF|P
zyAPlm<@{W*L;VI>nkH+Dnl;LKhWb#_gQ?OeomI0}J%S{x8qt^@P6pllFYg;$W3-WF
zGy8^$P2#<!<;1gw`qq=(_7Oi*P1e|7%$Ct-nw*oBhoRnYgcB3tf}nNW3HVdJ|KSlm
zlOAgkR$+GC#NpDiZby7eZN_u(YacIW%rBJSo|lozf$UR6pPKX+lh^b^Q2BVCQ+F5C
zs<uL!5z=l_B2>(7+WQ)TDzSVrxy@;!)w`x}AK+$Xe7f`ql=slp)$;983H;GPwdpo~
z#LrG#n7UATsbW&^>6>Az<rk(T!|U}sk?Y<JO%<~vqhGNt`NkgZmY0qjnf!B>gK+H<
z&(j9KVxXJ9bPbsA^yer7QiJ)XvmLP3NKyvB`v(WOl}1&+?v)+^uuV2LRV!k;gITuC
zhi?L`&aA9Lr~k?r&r$wJk19eS9-Wif93Bb1m*lkZBiQIMt%#%#Wvc#7e<YL8hYqWH
zp@Us`m`}db>3M6<Ss)8xOOhcQrMOxK1b>x!B{=)0_ALC%Bm{pSf`L#OgD2kC{B49~
zi;58X1@{O^XQ)R1aiId5Q&RwG^#dR^?eVz+w~`sJ!fs@6PZKE?UOazDN~s{4U4H-u
z@e_33Z~vUBYR}V}W_oPs-tl>$+|<{x&$oiRD;d$6Us57o`poviA>8@8y3F`q&ye;@
zkJvz}hU|<=H~^&VZa%q2=V39XXW?fmcLdjc21;<tZ@X~MV?67fF}H3*c?g#ac(jOd
z1L?rnW>xHO!Sx7-s_V7cSGMkbTVI|*a%=l|=|x%qLDwL5<5L2SGfFRRElYu${^kns
zddx{-ki(Od%U!Rw+>gf*csCf5>-NJyNV5^3A!tU@_{6&94$3>nx;nq%`--i=q62)C
z@F<Tyvsf_f|5eGzcgJ79X~XiJ<Kn5?_UEF!bv9ZK-Hj4xb<7PaYz5|Q!^f6~o*MYq
zuwTYQ-=G{t{~x?f%4p~Hjuq;`(b<0bciUfI1|0s{*GUdyeS;ybR+4JG`es4?8y&V5
zMw_huxY>yD=3m)SK;0Zw?rQaq*Z(*jxO5G1j8?J97LB|BkP2{Laa^49V|^}6RS3o+
zR1fIiEiTf{-iq^Pc0+SwfY<(?KR1mX-frN{LH>Qk*kqUde>L1CGsdO8IT)+mZ#Vx4
NI~8=Y?%SB#{{dAQ(og^Z

diff --git a/reference/plot-10.png b/reference/plot-10.png
deleted file mode 100644
index 8dec4586d25e254c5b7d69e5663b45871c57da56..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 27160
zcmeFa2UL{Vwl#{{hBnc(859&TE3_ye*?=NSQa~gth)5QY9Bo@g5l|2jiHd+^$&#@Z
z5DBH^j7ky;C~~Nx;LTn3IrrTA#vSk7f1G>Adv}~3=QJo)#rN%R@3rQdbFS6*loe%H
zE$3X$#>Tcv?zGH#HnznMY;244mo35Hv^;q}j4w;{&&tZ+EB=`|k?W2xKVLtsVZ+9@
z<u3XE0^jZtLpHXpY;rQkR2@PF+8m#5whfgYrnLIY|N85%+qNA&dP&WPjmxL|mu1I)
z-}?2ptuf_;%NPDCv!Gj}_O3VYZm$>1_!lmDEF<xY^0IB4U;Og0d~x<4Y+sjW@8mn=
zFPh`-J{;T}bj7bPs83O8B2Sm%zpJjMBW_OK)-}F>8P*?{Y9<`KigkJCWAYjov%dQz
zzEGEZ<MH^{;f3UzmybO*kgr>#e##?X+4k)C!g}fHJqrZLpO2pT*<k_s=KPtW73Ax~
z^ZzeCLLq3ZN?af=K0Y!hN35qJ$iKl*DNI6CV=|2*72IH0lX1Kte4K~*)m1mo!K${d
z?xd})?d)`^^iZmTaIkl%eY+^$+gTGnH|~`5dj2~*ZNtj7ixLZ?Y&VO1YTw1?k<Kw3
zEw81em7bn1bN1}nD``!6{JE~#2M!!?)ZsT(|1jl07b%@b(VP9YA@ce2ZT<cI-^Oc{
zXsHSs8XDR*l|rAZtABN;*M#d2wPsVZKh8au&eQu%dF_ir9lIt&xs5X=T$WviK{I~!
zR@y03Ud$QE#cbc%u|hkBnvBA|HPWu0lzCsLuD3-^P0iFiw>QkWgg<xSR}PL{p*PN_
zXJjM>axNH+YP`%h(~Ear6j@;Tvsm|I6~DC2bF;nD8*(>Cd}yXLCdS8m8Tg1sw@j+*
zIvUEz<c*YY$^Evnh;FcdDcez{WezGm{qe~@X2M)eX2PqFeDUCKNzcp_D_=l=q)j(}
zE-};Pf6%f$&(rfwm*3fX$4QIa{Hb>Fk_VgCaB$#~hCIZ*vYyg5C^uWT6^>`6gfMy8
zj?VA=vUvApy?J_rf1kgi*q#{2kLBf^-@F8TJUwr(*pTylH6HP}!9rb+Unkm~X~lQ~
zF`k}{u2XFWo!%l<j_r{?Vta!M#;ebc>2Q+Wu#3FZw#7VaM~zXa?FHYSHiu4c1N_<B
z)3b1HB6*}roYs{gjF)be{wXhi_``NzK9PcJg^r(_wQqbrDc6E$<1o`5*wCCA8@tC?
z#H{i4hbw}n>7ul}F+brgg~Dz1^-pGJHxy2v935j$wj|e3v*uf@O1Ie1bq)o|x7v5O
zTJ;1Om0Zv=Uw3QCs{2-*CB-A9JUnSel%doT7kt*p@H_U>nci@oTv~zigN+AyO-xMm
zYhRs9!4@qgFO_EYDRmZa(5skfzLA_isi|#Myw0le^4;Q{!DfE@EvMS-ebcJM_PNb7
zI!gU^oXtt3-rE)Cd6qw$ZxXwq@zSkTFOrh@GdtWENAQrp;f5bXM@PFP{BG;i7Zv!e
zrDM)c9-Cjd_4>Pwhp#EmmKV;R4RNFQ9W<&6xn^#Dz+n~kZAzP6%Z`@X?B1}JL5U#w
zovS5=i>`$2QHfqV(EPjqE+s|XUb6K(*^`Z@86ERdijotx8$~R%S8^WSo!M*>e}tNl
z_6>JQvlbP|A3v{cUA7~ooIm&d!)+(X&00-=y<3WN6*M+knCX_wKhvn|KK+hoXU&>j
ziWPmf;$qlL{(ZSaIq{`q$^5~UcFpWU!oqgql#c=X#4Bq;T?WLhT?=!XOGfi4+OIDi
z)5fi(P?K^`>FDS@;1t@sT72O3o5^O=$bteXUfv6dnsvrox^tD*nHDyL)34{cBzYY-
zm!`&tPuw-QY&lVCr7d&0#CoQA*F+3czaYV!XRK6u?qHP^V_^FIE-4N~gL67MTYRPG
zra8xYb9r7epH}mfM6>-CVc3_lZr_C~+j+S-nsRb-%y#-~n6?}=DB&~T&dWf&xRGei
z<2;b^vh85$<{HCbgd1_=L(Y=}UA>8kiJ`hNx$Z3%xr3bLVMna=M`sUFB>Z*WT)#Zo
z$umDK7MO+wyApUntNvC-Lq5*rj@+4^;NIEXiV#to&~h!@_N0IK?3Jo=Q)6QdgH8Cv
zCN0Z+A#uBbMvLC$`1ta<xv4@*pS?~CBU{)qf8v6jujJ(S%sVLtrB}lpdxAntzI%zZ
zJ8aj|)eXxVYP0XiubS+bsj~OW+%D@=;Wqqs>Ed|viNTplk&X(BoW5wPw&ez0hs)LB
z!NI|42rboOEwlp0kSKGtq+|G4O8(WfrU7h7KVgI+YO7LWWuD`hpP_unkreU}{mX8|
z<v)M^yv0#$$TC;_O1EjpBPv7fgX>iU&Pgg_fLNf~xpR;5JAK4cy}i9d-P(hva`xSB
zrgg_jPkcR|Kpxrc0i)X&*(D`c_k0WPn7x`>F*DYiJmTWMxzsStm8X22b-3yPQ{SFu
zY?+`-ot+tL$T63ue~&D4x!o)hJ@K@<gFH3mpB<cyXQxNnRv+$gpJxg!5uq-Z<If+j
zeyO8$>#obvJarvgHM{Sho{JtZ#mky9Gc&o9-X<N;Oy4c;F#Ls^$xmJG&|4MS5@~qy
z#+Ij*A);xe<=J2FaOE&vok!nJK9c3`TN~XoYFQN`+Lk;n^-Re0!l)A#rOCeCF{a#Z
z%ey}(*#&ObS~(YdyT8x6Jt?Vhj&5dZYHC;gQcg6Z!zkP}Cviu^l&D^dsc>**JFZbK
zxXgdol*yL*h6a15#GalW-2*W?+iu0U42e6fxzc459#nQERhGx~Mpa>HfVyr|pH;!a
z_Tz(7)^*tK+v-nl%L+d}k#;=c?(v`UvcFq)zA+OnTwqyHYE|4Wu>N?dW{gsJSYlF=
zuBod#ld9)GvpmC6zVl#Ip^w6XOIxk)-O&~M`o|i-zPsE`ugZmerRdiLB`sRtczQk+
zx3@SfB4Xd<hB#j~uIoHl7wNQj?_OTU<x3s0nb|hIJEdvUG=veY-qC{DZh_LJJTAX<
zmiqeQM<t`(<=g$sWN_zs^!|9Oii4$m=HafBO>dAcoK$Je^5IgFoSdAzy!v<;R^<n>
z-v7RM#f$80QNbYjD}q7S%*-BSb-0Nmp|8=K8x1rV{_<9bpPJF2tJP`Rr^oD9@4UT1
z*R!Og1Y!4cef|1<7ZN-J_1r|SRy^ehr`N{RYjk9`Sa2P>TH%R(B&(vLlB)P*zs9dq
ztu{&n={1USEj5bb2i6Pdww4NwK3}|I{rg?gGi6=z4%<&X@Z_?1v#s5+Hxwb$TYl%7
zed<XMuspj<e0jnvM2xFnS`?OasH?;%DinO|Kzd)g`Fatj-aD-V7f1PLhxa(T)Tyc4
z#-)8K7!634IdS5|04|%P&uiDNdE(;~6ck2!rRPP3+iujn91JoF+hi1Cb*j~wmVZuN
zeS=8b4VC@aj?K-@FA@_Ew7ZN|9vOMR#($Dsa$H$rveCdb>0n0-m7Kpm1oA|6p;cVM
z`{Ux`{7NkgnRdxf(xkd<IzDsFv>LTLQ4<f^^i*_>AP}u5k2bYUSXj7ZjgkaM_-xm%
z?Y!S+$0(E*+UEfM!6~tR0oOOm)E5^^Tig5DTV(VmvJgAnO0AifgZtTqy82&u@!!Yx
zY`hw5{tXG+n@*==@yl}NvfLS2LOpw~NS}UpFiP4N`|7CXe2}%V1HCWCzfgw9>HBk?
zOwqygnp6%Rx9&b)_gRDLxw2j1+L42)hUI>}vJW<<)E7SKwO&$;boVsf@%!_;)^nj^
zVWPiqH2Et@IffejbVGvc%fl0^%Vuor4Y6C~pUe!TQ6`;4k-u{BByBZg=7D#d^YcV)
zdam#|e7R{sd4SDV9(MiDpUbFjRILt8BZhW6MSiD$n`v^spPlnicCQ%ag9%=v7%C=Y
z?EZ02%#HaKlb_?%;!_c8ZAYKsu-e5qO}1DTc4f=?@9Z92`Oa3u)p@Quupqs@0nz*`
zr_s{~n-1;Fy;A1K-(n;fG_dXNX1{bv=6Fqm##wusO+i+JnQ)6S*PDrYE$93YK)xLs
zRX{imG0&DeUcFo}xuwaBR;60rAZ1e*rxse(EH?C}vp%N+VQB!nBBdlVIC)O(6;n5f
zuEN9rK4*9B!K{bOgkPx7FB%O`_F8ERx8$a1Z^Kz{6<GhVKZDz4_^m1Pg1jt=iG5G~
z?xgHtTX}e4yRM0ah2oMu-QC@$E0+Pw2u3^;vo*2n{um&~=&OlIPRVOZGulNVnZwkz
zZDQT952sLEOR_;B&aNZ-x>Meq0sE6vQc@bq9s6Tr#Ofw%Rg@;OGqqijNXn=_+>%bA
z<~EMf^CQ0QeCF7TsS^_uZS|i%eR_z@Rau@>QP-SQnoGB}4;0YT9o9^3s%jbg{y8$2
z`6-j0A;#m_^~g4p_Qu<Lv#52bd9HK#ij;i2S-O?p?@il%B?L99kfvI(WSb;NW~rRy
z|L~zfY7Bc&Aq=^YI%#igVNvC?SyYh5_!ib*<|d%djAfsaNus_`E^R%oM-@}gPP(hV
zIFISLIk0k4@=N8^k))@L;^Tc6O_Ms~@s5In3Z;+JACQP>g@}kdxW8!ik*&N^dG7|j
z1PU7aW=Mn}@r{EcoH<}b!qLIYfBxb5eRf@eoqem7te~Xf$H&J5Ler962Tel@p0x!`
zChyDC%C<HP*R!8zPANsZkNSs?(=CQYI$Uf7B^^h*6!rY~rEfcN=e?{n{j;a11JIuV
zc7dyA`+=-hKY6|DZx<f~Mz!eO1Sl6`RkEq9GbYR_PD!XsKdn;O3+Up2qYD9#whcNP
zNcJ90V+`dA+*!Rvb6E4jR6w%Jj=lT#H8e$ccCWs}&ch9`QYzMBiT9R`e=GK4+_3+C
z+$`BLZ{(C<{@W#7<4&SnmU&i!%%NP`WNh^Dj|~NpS)&=0*}7!6I^&>>j0~#wRb)Ds
zvEYt@V^Oys(U|T1bZ4HSS0b#SaeVK%@MrZBtJ@vX^b@PH#c<xOIrSz~Eg8q$rK4XO
znk{Erbk8G;U1}<uvZDq!s~6KNrb|g=KD$u5#a>%RMgp-pMZfsI;vfJ`{gs@ohVqIS
zZIj&s0>0-WPwns>8~Hluca~o+Iz}OQy_R`~8p$DH4&UxwMBuF|4??~YWZo8Vc}}s-
z(8@4TYqg<B9yD)GZ^Sv)M@ixH?AfzdNl8QFG}KcHJ$+Q$5>hY9^We~kQW{2_tOW9g
zkB6IlsOxX&9DFmBoHzP$f-~LzartE4=TlreYDmnCOh_;b9bCGKv$oF)<sj|8*Pmv|
zgAI}1o)*%4NoNt;1>0(CYq44P=rMfSKVqK(Gss1=i+zs>(ap^@y*s6LaMC%R`tO6S
z)W*_vcC0XKGTK!p85Z-98BxfiQhiu>vBypQwwQOX6&ZtBdV;Rh#H<pe++c|UR4CiO
zy-V{HZo1;{>3L+hXpw*K{A5;PF>o6f2M3B8MeEMnq)74P*uGrHDa%^2Gr4n<+RPCH
z3w)L@0^lwmiO~-iu}AZOr>#bcH<Y3_+04O{M2Ym367!LqY-~hU%9V)R7hG9Vqa+<F
zG1pisRo|PF+GM6}-)3u7CEA-X(X-dIre-pp>X4Seoa&IC32m`7`8?<D96E>`=GWJt
z=bo=5H2xk*kc(Y<_BvK|Um1lInMZJq^;)aq%eyQ5Mn~rv&g!M?^0zN|4hr$=<A9ZA
zHUti5cP=d%s+!#sjk18w%fqNZJnh^6r;o^Ltw4bBuG*xS``H~@Iw_3n&l^sr_+(4g
zg=gM#`ORnFfN&iTo@HZ-fsIA36rHsR&B-RbdB}nBC6b4EO;uf8R))`h!e{#WaP^#8
zzJaf#u&CE}Aan;onvQ*2fV}-I)u~k6d4r0*k<>aMIDh?-Z^Z{q>l2Q2Jv<@idQC6W
zOoJ@ML91`?H;P(cnH{n()H0Z>Tl!|^4PViYK<!`WM|kE_+NC3|3I-|O`ik^PK){_l
zcdnV3+(+ql4RLucBR!pOUTk$zLvK)EKtMofK|K$L72@oA+u;5_(`1S!!r>_yvx1qP
zvKy|huE~KU^zEpLl=U4_QMg^-;3})vWe}|4s&$dMvL?e=dUdx}%A~W-9y=cX@toZ~
zb1A#%HNubBORS>wP9!;{^62m|Bc!ppj*!iDw5Vk%@`6>@BiWt4Bmp22p-A?>Nx6J?
z%M%GbHMIa2$+*sm7B`%GKEjj4tkAU4#2$ZN8>`YuO3)`yo?Jx5(uz1_HCnby_T<Tv
zGBW%13M71veDj?C%I@B1++Me|c(CCUHIZO65E!XlBdEaEprp1ZTm(7x*Gx^lfR38j
zEj2fSg|I_plghuEd)W1eiHIwGT#?6aJFnFI3}uQ+&04R5%xXunv|#@AuYc+a#B^r3
zv{3Z5!kwIi)%7N~6lc;iKQ%T!t#P@%Y!fyJ)pE(7IUl0J1HL)^<`WrHUCX}QR$i{N
zaV^`Dg$pcsY-cZBM?{j9d5;|&K5K^WX|;2&B7N?AvKLlO%`s@QJN@MfmTcrWnS09E
z{erG8w?V0o_bqm*BNlJI-XUmpU(y?1$GH%js=ga{^uhhaOiWjw6`1Iouta%vh5n|~
z$7?CK!&WIU0Hr`s>q+6Mt6NZVX=*U5W2nY7bKkywUW1t}{EoBWNbKmJ9#RWwF1_N8
zu`%LfJ(XUTE}dLuos+?vv0L6_Cyv^!i9GK|8_Z}+${l(pKaXAeeZsnVk*{afb1SK?
zU-wRQuM7_x62G6AkdRR0U1;6rdqbjJOpmn~)dl>cb(HD??)s7=K4GU-qya2;m>a0D
zZQT7+o@@G4yYsG^!s%iG-f#0m7a5;@-3}vxw8y%ZQu8HIbB#2cChu_H6aL%f<14u(
zjwDa4qiSBgt*WNBt~cD>&02)&=~>F-o~IIgZ_TcqH6pFoPlW3xcY0so&-wnsf}(v9
zr^fm~flDK*pWK9~!r0_w+6~t~Nq8M8TExz0?>h(Hft!3*%13I$Xm)-kb>nbSyR>zN
z%gm@^18tDHDbLW*Ff^$%xV-6Od`DA(fOV7o+7{F(0rDLPS)!FS2OoS4)Td3+{nHb@
zP#>iy-qoI%0-!ba>8{>u&Wuq0Tf*lUp8O0f;va0BH%F(AeB77({vEsYW)2;_yXIa8
zGEYl6FcDR~Nxehaf*k4?8w(C5$f7Vzdk?3yn~PvRUHy|ZHs-Dx!gH{gkqPPziFX2l
zrs+Da6LtA+iAf<Yqh)2|CpYM*){4g@E9~^Y2fpP!DUx`<eLBJJE#UHX1>qs4I^5l@
z57fpd7l;9$hUq_P7U;HdHqQcqw5x{vFnQtcKF=xS>>w4pCdC^R4m#LmX9P*PI)&@m
z=m?E>kFL9Yef+cRu9}({C0b@-R%T|2`&_SkPreUpISzv#8BaT$bXR*E+{pW)P32Jv
z&Gci>efAh{3|?4tbo-aV$kf!-BXiT8{-t{K>OjA~q@;Ib$D`l@(-w5dsKi^O{i#I5
z$Utd9qrTSxzrJ1JE`kE(+B_g*)^$vmh**o&JWokE=*DR2*y$_YuDq95A1Qude8H4T
zI9)XlU^qW6zSG)mGA%qYZE&C|ZHzmU6lG=VQX39vu2qSaFPkx?u2{d{Gudq--Z3#|
zAHs&C^^($4T;^*FQEK|9X_k0%ZOCf1K3i?~@$sI;GiMy1u@(xYddVBBPTp_tsC^SX
z`(A9Jd=v$yrmHIDo>(o)7gX@&_Z4DouMBO;u8iT@l&Byzg=jH9$Jk}WMtG5tVga|5
z8U<sqk5|(Odied1+TwQF#4cm)4Y$7=8Edm=fA{Tt<2mlreMaZsuUq9UJX5}!EfmF8
zkcPQt9d7wq0sGYLayHfNk{Y>}S6ed@-_*ezvhM`FHaIRsof}9k=l24(DgG?WvjvCF
zG7tYMv(-v#G9%%$rz(_7oR(*^Un_GzFE1(;gFqd-zVr}09>s?D&IB)xRf_mpx-(i1
zy3frjQVDz<p;MDflTqqF6*T<F*>G>rVe;Kn!BqZJVROwqkW6K4*-=qDbRE9lJ~C0O
zVt}MJqqA*(-aZ@Lj9=O%8i?JEhpx&K>YDYTBJYr+e6re|_F)-)o=P}cn`Jg#EmK!q
z(G5n^*NZ=Ax1$8;MMWd-lQNraDh`$fDOXlj0`5O^u}Iz$ikiP{d=b0Er_LNPUDM%1
z;WP5Sy|go>5z{pHd1qA43q}y{b}mo_nv$5G9kT*TEs2c@dp2f0mO<N{3DBk};?9_%
zxN{Q(B!Swh%<dB*)j^VWM%-Gxx6%9C{rnE9QuQ1D<)10lZ@8`NI!dfK`nbyjQqw1H
z9p9otXOKYua=&ZljlOC#nH%)1Qf}vj3?>#g?Gyv_OYgWQDfR>@*o(PlHAoz&TXZ^w
zz*wgdat8&$!1PFG!&A#|@9)XA7==0RjBz^U%N<%VhTPD5n9_V$NGS1~4gdU5-dI%z
z|Co|HUBx%Bp}M--m_^=gv1{DcJ3~V1)hqz_t`S0DY3lLQjB;@)`?b{6>N-1&#=n1-
zbl&b10cKO&WqP<PBdLJipOI;9V)D@<yZeb;3&KG1C0oNZ$&N51N?2&RG+5vq^N#xZ
zdSgkE!-o?x%#&y~Qke}t?QV?0kV!cm4N#7Pv_F1c<8;{4UQuG?eTaJ@f=f<KXHu6l
zi==zydUdCN88>Ql(Rps{FO8T|)YLJW^-&KQ`PvhKQrfR<&wV$rC}3zjuRo#WbL%#k
zHY=2^CDylFqwnYI?f;{_R`Y|a0jKXCrWL_g0f!vME=>=AE%ScaoJbWj15WY5Cby+)
zFm!^(o@ZtXyMc&G9vMTn2OSN8q*{OymJl0zZ=TsPKX^lyYP(6w#eonap#}GIa`1dg
zam~$rPA?_FB66@xO<x*#{cd$zX8!DGdFM0Jq_XmiGwNhrivbDN-Q8k$v1%;DuGyH7
zE_n^Vay+WM+6ii2sp^iz_lR3!t{JwrLPA2JNu`WiJhNM`Qo?i|HdpWEX)%>GD0v`P
zwUrLavN<w&)LtTTTC0P8LAT#{N_u`wx@~go?X9Jyz=`W)LafhZmG66B5c%}u$B*B#
z3Yq5e!NQ3xnTL$39-#0Lan!~$9aM0){;W6Y%si&qa1tx1*>HlN-g1PWH{T%pd|-*2
z@&|)6?p!47?)*~eUuXG#Bv#4QBy>>e$k#s<5jg`yc~*;dJtU=Jr*j8GcL1E&LF=v$
zdHjXSpto}0pa|#$WbqBQ$l2o~+Vhm~^p67rWx<upKmBc6!FZHFG4?ch4W-Y?Udtle
zaCXqr-3u%n|J<1L{GpwSAsbKJUF|(OI=T^yPTr&_KA8RnNaPzrffsI0=*dC-_nw5z
z>4emC4?5D$5}z$x0FC5Q5N6J};3)bMx{m+Nm*4aL*k?NS$k&C>Z7R2;805*825lRK
z&9390QxJX#IIOL&F9B}c924&5RPjt`Uq6-FiKyjUDcq8<HG^m(WtE1fd^T@NyIR>l
zY2Seyqfa=<PLeYjv_idy^dC=)Y*#_%Sopj|3@1tw!Hy4)wvWvT)!?|2l`&|~b41x}
z^uAupa?84P>&R^!8U3)m^A48<k)TRIi#-_=T#-WC9ew}){WmB~4CZEMScMQyd*q83
zyeOtQQTAV-opPVIC$yZ8SpTbwsaNC#^-8HHCe_@VXcJ5SEj9Fme*KXeI2p&b;XoLZ
z1EHZI+~TDbzvv++R1Z@e!Zqi#w7ANj9PmPLv(0n|^wvk*g)X+~(A66`3pkh*62Q9#
z-V%<IFvz^>KONN+jb=M4CVFc_#QO~QBc`YyumKWVCJ9WEk$JF1T$ohW>(YhkIL|k7
z&N9B<<<Wg_!gl&l=~}ka0t*C=-fJ<A3H39zQVbS054G2KN2L9hQp)Y+>51(V+j91(
zvIKxPf8Ow)4s$cTC@zz~fD0%%nY7&Fu(kXGj~i=-7p`P$(6QH1RaJct3ZWA?aFC^=
zswhbcfi0Q4V&4$)@_GT(xYDV@d8PyOvKNSEC<Kv_)8d~T&{_|<%;)DdJKvL}bR`Vz
zQz?X_`)Y!0%*O`H*&a(Zd|B+_*3)R<>rH50(X0-frp~uZd2W;lv6)`SWmNaB)ZMrh
zTQPh0zJ1iO!pUIERgaq^3Ui15T+w-RDfjTP%^p%V<a({Umf)kHus*P<65U1aWx8B%
z?pLB(?Z5bXE3Qhv;^{#wf%E$XOE>=_yV**{rE5FS9=mJ^lRkUcl<5#+KT$}n1NMXi
z;7>N??1HV`n^o8%#<Q3FrCR^}oBtE}`oF4xs~V=PI30Y_YyG<1KPFy4^1ic*YyH_L
z`|ru`R2i`T5P2^0W?bLjya17>P>4Z-*{_q6{-N%n5{k=%$SVYYL*8AVyLt55yHz3&
zaP~i3O-peY`KBLuf?EyG7L0-Z%xL%L)>duup<mzKl{M_Xdt$KIYlEzC@B!<SU1w3^
zJOJZ%A>kqiNRMLvyAqtjW*1o{_513mvyn)kAJ#omg52}~EIq2;-FADrKI2ipKmy?9
z#hH(cii%>r1dKZ=Ayf2ErkvFfXdZjke*JH2`#-Uy{~f*rHI0FV%P)WZ^R^X2cWI}u
zH1~=P{11GwyHXl-or|$cH*%S$pGzrU1i6~+gO}fO2Y|j-WNj-zI(zyI)R1q8@GO>(
zRNgX%hH}hi^5P;kwexbl2^wi6CZ5yK*u>IDa4NmApNV2Y0x^Lh`hm;H>mQ2jSToFz
z&_$#X4i4$L5nlrUjWQbmn}bk#h@5+0jhnpmXnR0~LjBdsXX`jPWVs!$#9U`NHX>95
zWQu|-uNpw~;^46Qa&zhWY@*1sc}zs|@qbp<)8hf+<KQ^xH2&QII<$U$g2qV2A&LyL
z&;UTA75IFjnyYo4Vfi~oWA@|IymVyrD#N9v!6Aig6f#jI*l`53%&T%c^1DNBRLfDd
z>rotYlS|HZu^<<1>$dcFz~9|BD{s|34hYy$)80`X+3{p2Gc&y}D@v~LfH~dVxZ3h;
zu$$Ji@s?`gPjvIr!R(kb-&VG(#b1Dx0wU%&@U7=IyG?(!sy!!LOjK(Eo4?#*FA-|c
zUQ;+zX-TS;2avEGL(Ah(n~r8f|CLG{ILvvikao0_c|f7Pp}OZk|I+$crbr7wg@Xe~
z%C5t8iiq`oy#&C6+)J_U{`dx*2P+Hl%njc))wzHZWcK>)5*;a#8{a=Im<qggbDXhy
z?cARl&A5q|9Z!UrnTK(I3WWz0aDP;VLP?ZoD^ypl)FvfRmHE=<5xb)Bx8k6SuQ;I|
z(#Z^h>f8!{1Stn(0j*J-+cpiQg)_}uP<820O(A){vMi340S~bf0@4$H9nK*6=*I4c
zvXQ5}?|XVeg1DG-!}wABws5e>2EcRY@SJg~$x&~<{%%!dUY-P!?tb08H{@M^<=1Jf
zqaWjEWlq942L$k}&IsJEAs8VQ@;C?2jc{ZM+S!80d-ru#J08aMf{{hyGc12{5`pT`
zKHhMW_@NPh{`5(C;QXbd)Uw_nj_s$2_a@<D$_tNm6Fh^C3YL$1Q+!_ZIg|_{mp2*)
z;{jc1Ub>rX%80dO`BhZb5QfH+eZIOmz{fo*J~mcEz?QZB)}EL9DeuPymyv0O%%>wY
z4|E9f2RI?rMThuO*tS-l+4%*!xvRm4hucPAC3z2_=LL#dUS3U2O=9M;Voo&3${<~H
zk=Gx67ElqJP*+z6^rHSg-MB^{+%gg0gqlMjL8SnWY<Mn4Ug<HrFUUiLSKnVj_+B_f
zD=Qq`VlYV?EemF@64S^|`KP}S3`^>)Vtx+to~@Q=_I~m9B%Tg=h*i;wp_&75is7H-
zaaWSwvfNU?eOmT^V4wQ0=AZwLVEIq+f&J@7{;9UQ|Lw~em)f}arGmM+dCJ*~0(<vK
zw;qO9i#NCunQi%VXgoz`ltn02fS&+qcDX*SR=#lI$!v1RlP9C2?DO@yIy#~9Dx6{k
z;n2lORCc^V$<esuq0ttGIbt~s47?XkY*W2(f%ELKM-wmEp9BQpvu?WuFZ%f?`>IWX
zFB^9xs3Y<g^WR*!M?iq*ddt|+gYci|2OeF3IA0c#k-kYOAesnPu&UW@TI)HYY8;11
z2@9Kk;DSY(O41aRd5lPt)j#D)reFK~LO<}eF0Qon_2qZ^Wqv#6tabFE+9HpWTLkOD
ztA0XC?ZSnU;jixuno<ovH#ch%WVcGB^*4|l+oA6O&wC*zt0Hj`4@7r)psbCJO-lJT
zUxe~7mtZ1&RDC(H4a6&{`v_bE>9LN0$i~-~<)*XnTfpJ(130Wf9TtgL34_Odo#iX9
zNtV@}da#j$c+WueFT7Cep57~z|25Zx6~g{GKG6T|8}`5AiT&4F{O^y2YU^`7j%xy;
zg*f%cg5YA|6feEm7mYhuD5p!IY)Kv|=L=^?Nmal?bl?%lk$b4O!};EwyF6}3NELl#
zcKDr_4l$s!9X<B=N>MJ(=$rny_-`JIR(pcEecjkie#W*|Ms5kqKL;#X40j3#B0(|#
zQu3pSt!MbZxI+75Q7&&6Z;&Bjr<`>myPx{ylxf0%<pMs4Ww?g%IZ*dNCrlj=V10}#
zXCdR<!A$Ded$H#<X1+!fB?MsfL!I(6{0{s{XF%{nBq$nBPXyZgqoaNd#+LbTH!aO(
zh6%$eu=?|e))incEQ8m@RWr47?H4XwXx<h0)Tc8Bf%%Hr$-Y;wUVXBB{=unwY5B*G
z<1L#5YGdxf2PBLTtKXb%oUdC#M4yiNxhYf{1e%Ety@_lnSB!)~V{PtF%U=TRqSU@H
zAh_EDU=xA*ChCvAtQ{s$s!15<6uWUP<!l5SP`<0d>px~Tt3E?Uc>#^L(Xhy)zJ_uL
zgxNXNToEQ`WXnRnP5ik+fT)c;Ad{~W@c*}^{6dNq)WdI}4(PYOu{sFDnbZMO;&XPN
zpK?bva{1kzlLiI`;ZMTXlg|XyMb0TWe^GzQD$cbW9C+viUu$C_+AY-+HKBLyh(^&p
zb7uIc=;~E#$&zYm#dkc3z{|R*?R4`#;xK{NPZkh7<?J?JIO6@w1*bQwI+0H#94$!~
zEHM7&6+Q`t<_i?>f5+l7Mb7XgoPsv3?KaaLnq*(k1rAY#NIIsjZl7PAbk%gzB-8+|
z^YG!r(?uHk&Gs8n99>A#*61?%5CfJGiSAxW$*DyRzhbMb!8PJxPPF{00^8_Mc_=SQ
z+0pMHIezEPl4Iw7-m`Dt*@e39`vnf5?vN{LX>8nGBybS9nAq)Cty)#gzj#S{m`fO>
zIv=p(3&u8GD_WYTBsrlfBNIYgP7)2=ib)q!c!@;q={fzD-R&1Gvy@_|cy`u){^bns
z7xRWTy^GAn>&aj2-m!yYSl6~r&D3>>>|5f}CRYDIs8LP{tg{pG`lFZ5(dQX@{#nTl
zJ**J?-SUjoj~<8JYJY2G*n7`(IH^23*n*SThEmGOr?Y(!H~MdFlK59b_$MX=HH`-w
zTv7aWvb#^8$b;o7<=txW_mB6e=zoehR=k_F)>|`1xrc=hAAWBnH3mmTBZ!?&bXW|H
z&!UX{_I{1hz<3&taWPCaTL>v%@K>zVa4eL0N1}pFZxxZ8OCaty_ho~uPU09k#WGb)
z!q<d=UTH7N+B!S-(H77Mjc~;Y{@T9rRtjWQxOX84gX1G-lBkRv93!AVOW-{(bqQo~
zPRgmc!m3~qfuMHK1mcoYL<%8WI$n=yLik-ENxx61Q&Gpx{1xS|bW47LpsnuThsbfo
z34kyK(DMik{gHXI^SYA;JdBHiOq_`w9Y0dW<Al7Wc{ubsbF(OQjbvmt-}rPa)J<1Q
zD_EWa?&lw0hH_mYT0KcSNS@fbIXF@vLD-puS6=n+!%B7MZYH<u;TJIFVGmACDOwMG
z0axfxv@Gs<dJ^S8g|Xx86TTWe56gdBO?HPE#n5N+;mDCJVU*BjPz6RHCrW|V5YMob
zP7~q%W)sfP{`;E(+C=(-POwWJ7H+;4c@Y8{(Hv8X*K6@n;O*1XieyFw>T+`M=z2|g
zZuj}{Z_mMK_yQ0(46B=3%+f07^G`cEw!*$^s_i}nF5YKXm;)~_ZxZ}ZiXz&qXK<-)
z)NP67`@15V&M;qfCT6wYhc&1X@5x<h3HPeO<UkW>HJFQzKW6n%DW@*~b@-GFGEXrg
zNKTb75fq;5vdFuLu5RY9Kf16IcpRl~O~HQR4HgH{4gLk4*N&|1vLpvjWPU<#ejUO(
z$0O&9Ex~{NKrHYY>lLSC-8hCb!}Fm@g0N%#RdwX|1vHSl{(a_|P0^kKtiODDHazJq
z19k}+;o!mXw8`M8lQMbJ-@MN0w%gtbDZBCw+t{~Upip9_33e6wc^0iw<WU{tVUJv<
zz@Pjn08ZdeNzHZ*qtLoQ>mQf*_=Vn!ZQtH)ou;_Y*_goxh)7O;vQ>zFV+jq%CBI|`
zCydxU$R51BVF=~O?BA{13ArTUaGg-n&}^Cm1%G_pODVk<mu*=&Lfmb3x?n$RsXT1I
zs}!{az?ml5;l)Bk5i1<o9ZDZ;@~53-E%c|`ZdU$6tj}<QNIRT~r@`{?5MtE<K&KP*
zpho8%GaW~mrB<l@N?v+YZqMl=29~`T$9)4a;B8yaN+(CxZvB#&ob2n)Y;gC6C19s?
zFsw83Plx9lbEl@?^oz}R(#l>coDNQJyx+akHQjUT`rVJ$$9_ApAom7OcjgOTsf!oZ
zs$bN;q|RZbF%{#mh~t;{i=&4AwBGR1_1K<I)rN~T_wQTg`Fg9!ipJ`*&vq*GOf7A>
z_%f^LWz8W+sd=%okf51#e@j2d?usDE>e^b>!Ccn<+xR+;+kqlIKimFB9PDsK(4k!*
z|KcXwSU0}XLCxJiV_$M!syzcDD~CM`&e7%S7STXxx95^J0h101)@$cu`fu6?n`id5
z@duXI75vyzb4qwAF}Jo4;%KTedRQMYEv?3$2g${ZHff&Kc407>g&@Dot4a<-kIM`?
zXmHA}>#2cH*;#>nTqx_~x5^o0SPsMS|Ka_;H8xNpdtpGX6h}u3@x?ZQraJ^G@$s1`
zTU~d!!xJV?v)*PKi>tzCo>f;@r<HEmvPBfMsmL@v;7nBWcGktzM#Y7{V14J|9oa8C
zTorUErdi}?^6nx|ToEzh)D_6FKCQlJ4~z9?TYl!Rq`v#+4}Lu2rys43zsJ#o6SjcI
z?i(6%t7(`oQ*t|BY#e6^@-&0{a^&uSGje`7k(^4$-HoqZ+xVLvWl?dQZA)|WvolfW
zJB?pO-?`H;$E~$DAfZJp&7^JhCP6-&QlYRrCVK<R^IP8CSrz}4Z#)p-5MZsrX=U8C
z3EP(X`ktv+F_pam)vc`|kf$4*)^7YnA9aaFH{z|O?30_XEy9h!wkNHiuHFz_Fc;13
zfH12;<q*DaogISz4GmcoHr57}lt>?7+#H>dPzFXDPihGZ3)A#Hu`ZyzyZx7IZS{fY
zV{3bD8GTCbIpHzC-|T~Hqy3937Z+DeTO0GqYO~8VHJ?5W=Y(g4Ev*ZL<TE}#KCQFd
zBtgLRwKjju`Ph|DZF8NH`^IeJkI4F6Au3|8yNi;NQkeVPOdf>%{BUQAbGypZ!WRc1
zSJiJfTZ^l&ia2&F)vQVI`Kzh~vFevGF-Ao$lI&~24{yS0Z*FN>TESj)lek6q1e}Rg
z-CJ)Sx+k#Q*aR3`R#r@0+*ES)%4=M7rfuCh544D!#@|XS6_-GPu4m+>=jZDbB_Be5
z>Kho4cXoD8l)bqiQGj9u)cZIoDQUsj=I@Dlg0K2-5(i*M1atlViw|Lw;U|JNDPl<L
zn`Nvj>nyE2EadE;aUipim$Vp>rU4>ki^6Y^_Y-O5Ws6`pS?}QApxb@}iU*$@F?0}u
zK4bHb3QF08)4~4=c;D}Bb68`|++(aww>-M%rv1d&5HWNJh5U$B7ftX4-Fr29e{93X
z>q6{#@!|IN(2$e9^0D4lT>WNFhaQdfx;05WAVfrV8L|ndGH3f+Sg%OXs#|0Uf(OM6
z-6fxRVBkYboIJw%p>2Fm3?#d2ABgLr7h1jii4&EjtaGyaK^)Ip=g%iMMCIl>hcX_Q
zcN1gkCP}9oFkHl-NAOj%0qZkr^W&Db7tT=&lO`2dG4=J5s(Y-TC@U}gA)OvmW3F9g
zsuHVO{rU4FKA89wV^y0k+Ov|Fav*(TS$ovkz#bGz6{u6OKTFVr`x5RY0Q?{XP`B|&
z9z&2_Oi|Y7A8qT;DiVi-Bab;~N$G_ej>g7CeSLk7ZQX!VmFT&dA^Cx{z`A?0Zbmta
z-o17enuI=}abq-xq5yIoDzP(;BaCNy-95pUirAvqNMrYuTzKV1(cq9LUBY_rtFwq>
zUK<3eN~I=K+I@Xt44D10RQkiG=O^^C^st-eVEsL~LHqhKzb=9k01yQszMA;59-?w!
zoZJ07=mOl2>Q>IrxF)(A2g>b4tAZbn_w4Xp=>bWntlV7jKwT%}j2}z?N(;Y{930vt
zx`u+z$EH^?3vsVg7JANu8Fk!0Hj;&!<ZwG8VtxyjLW4F3m%t!rGD{*T%E8a<?Cd5j
z0?WH6*gp-&aEOVC#YwVK^U=0Ar!THJf;G9!(Q<)g`+2C$6BfQ)B1z4iKb~kJo+vyR
z;}(FC(z>*{_`;Ot{svZlzSM?PLfTuWEZn>I>6HP?<kY(olM-@;oA*Dwh-6uGAf-%o
z#;U~hyyKa>o|MxcR{<nSj6F%&J<o_K3>G<8rzvstJEZ|#t2+Nce>ruHJR(l8baVlN
z^-D7UBz>mDBA3YEfngX*q7B_(ExA~a9JPt<4*oB-`FqhbPzDdoA+%g%3=!9X036%u
z<d+^6o;+*HI-IYxn3`&&`X}AYWq>+k=!`OY`}<<VlqUF3QsG%NLZ=$jaf-DRw&szW
zg~cJ(5hGGn3Tt(hPcp6Vvp$WDu2;DEZ7$5J?SllGnlM-!ba<DV^zT#1DA2ObLPfHM
z{+MV#4?9t&{XEPtIsdrM)a1efdj^1MtJ4$##Lii07;ZN4Wd$4MEUY&4qMr6a^njV?
zGibp0`7J~Y{Kvi0(rx_*pc88@09jnxBtU(At~<aCZIW0k|MBBC@M}-Ua9LiQuM9jB
zC8W#BY#z4pPG8z^|4v!zP~X~_xj8vbS$ElabO)_i>(GAzDW0?j<}#-{eX(bVzcIfB
z)-C&gybyHHU87Alpi6sVXui_YEzO<*=U&|3qYctDNEi|T{D|?8bY!28Rjq$RgJbqF
zX@8azy!I(#Gam6j?()p^^e)^tl{t;yO#w0(<AL3t^nvtL1_u)8iDRKoY*FpZXmmj$
z_5S^aIT@X&*kP>-fj?1{G(hsPOAH^Cnut|Ow4JBHx+MC|bNj#BOX@R>!g<Clu(WHN
zsN;O=^Sj*7h<A^4{xeKh{@8SN0<yzNsSx{mjiebGI9#Ks*17>)PDh<Ddvz`{6-CWt
zSp1L0Euc0O<(@$dhMg)Hox2tf{(@z{&<EH4K=R`P{}E#R`a|UV&oRiqUch=3zyA-c
zgGs^!TU{zj;0n~TB*j?5s<B`D&2K<z<!~Xxf%}+{U>V#j6*nTTpLdvFH}44i4P9JP
z^QaUIQC$a#+FXSV@2QHS=#Fmd>?Lbh!9rC}=co4bv5!e+KEXp*CVK=voEV_G(9lpv
zm@uqiuIfRvPx(4^?MK^vj=ZSv{PE6{Iw7~RGBSi<%eE#T3_S;NFh+H66vQMn_*bbH
z((Cxj)YNqXhK7bXm3~C8Ew&sE4MtgQc4z%m_Xh0ONPSsS!#d-tr5?-jBqb&DTY!}Q
zhC!T`$b5mNPu8CgwXT9N{>7>EBQ$NX9ocD0t!`=xLeKjlxlwdvUCw88pgAOaD2H0_
z&;)bQXt#iScHEeG94Du)={!sY2(y5>bEIPsr&QDT*g9ZzzA~6I2o&y5IS_CL&`x8b
z5oi8oc6N42O;J&O11%XPhsF|i?pCE5<rUuE?oLh;H@+(uD%Az4(Q4A@oh!{;TxT3x
z@TmaC&a2ZG^wY()kkAWY^0*3jP7qAdc5rJ?==Nfu0R_$XxIQ>lgub}l;{qp|C>s%u
zs$#(uK=H+_iOQTrr7*MOgA)@B(`;?UZ0(^K<5xoXb-SqRO_GzPEi=4Usbo=h?hQCo
zB{5b(avE7~!U1WLw7*iB^f;JGF02hj=1-{K&Sv&&zu@qNy#XP%^%_VF&Vs1*T}Hh0
z23&vcqQg*nfPpqwPJw-a<pjkE$+lftg((a$Y*xNF>6Ldlwu*o+IbHETndoUt967A2
zg6jf{i^gWa`XPefNqaB=6INIw{o1+2(&zUf%L$@ff-CGWz()r6(k1Dun39F6dks;j
z7+YFW?yuWNciekPhdw$B#PYT8`*#*_R{SYJNw3}<9}nI(XmWDWE#z8FN?b$UCff}T
z>cwq9PU35J%754s6H@Ig)wBw%@iPGJ4Ba{433Y=Ch4#3>M3e=9@i8SepLq5L++TLl
z=mK+Fd`wKKfmW$R=0sV}?dFzGLf?B2n2FJImPjuBy*bT-QF<qkvhbqhkPnZv6lsAR
zqp~)pZM=ogmqgXQGwo9aftZwH2y9ddi{s8A);X|E<oEfagLiR>@ar*}8{X@XcSQ`<
z6?5cKB&W3Da|+;_nMigI^#zhNZA1GZrINU6Q%<pV@^=xO2=%N2e*yKq0p3Sg0r}{B
ztXEMka(WkhC$A<pvaW+m4F`@D(l2fOABZ~t`W0l=J^m*G{J)m^pS{#^DfNriDj0MY
zO_QJ|r^}d{93QWB;H003RptCah?LsY(W9g^vL>RQf6XY&2rM`BJy6?tC+RJ+VOPe4
zY1?u03ENegS<23?8n>d(%jGs6ox4lF_7}+l{VjfC*|<N5R94sq3108?6J0J=HPwPz
zi-ck#xp1X}D|4>7x|)<)ZqJr9{^F{W1&Imx69L_0)#nD=mfG6KknY@`MY}D_a~%hD
zU+?sXHzD|qjg3)qo)w;6)((ntj482hr96tfw@U`q`2!E3A|PS44U~w)IxZCeZ3J6D
z*;JToNiY^VeE20YGrZsB7<cm*JK^jWt9k}e^@)I9SXEV(TiEi8fdvxMV1UaJiBloR
zx`iye7+#P7uBV>Z=gI>GXr=drw*Y21*EFniV7!Wtk0*>tyslGyT^&kVWdkm{D@YeU
zIfUvxYwJ>Oj@n#{xM+I@8Q<cs<y0%&r9~D|=N6KC{ejwxdi9s0&Zk(2JQUGj1;E|^
zH-771=i#4y9{y7*@n1{*|AVFGG?hG>L52sx)NMY8`cND;?+TEA;SfKhdB6v$@CTYo
zvP6fYb+u|N$UFw{D@yQmxHCp&jlp`_3o++oR>4Ff3{`d4kF##PKQ>t9Fi<!j=`?b3
zavCLFxP)3HRGv24_Xy=irXLXKn8Ce<yeYz@f9G&Jyt&|?tCmU)-`tGJ8b4^Yk0?Gz
z(fK3ZZ7oVV4MAlpQ_r!8thg8a3oTKeLYegA<6$(GwqJJyK|wn7`3P|$(r-pSoQ)a9
zzmNvlNdgjI8G-yIt~(W;J?5JsX#R5#3v_)~BJdxh6VCWu>H6>pWUOhc5gFLxM~28z
z^AZ6Y(a&h+9SC0e@Zs8%YxI&4*ZCIpbgT$YU8Qq3e$w>Xwbypkr=^(%2~HAcqx?$O
zo0y$w_!8IQpQ!aJzaZ%tz&0gkkO$~&17L2K5{Esj0O6A}GBV1^&28ZP-KgSeO1+3d
zi%EQ<H98#fijBPpw$0#Pup}=lf6&~uxe!H1_|E6`VEV!fxp&_|WYzR#TcoR3SV(C1
z6L(hj{gn!Go2~{=WG;vhrsr6ccUwdGq#Hk@pWARVD?WaoT>XqkGWQK<C#QN&2gfne
z*^ZB4v!zh9cJDOZaKAKX@qMzq2A!t!57c_U>i7edw-4HGX=eJ^=WPdCzSwC9yWMy1
zvWX-VXelXHMYGOJ{f^yQvcqw(@@GQS6JbD&N`|!n!)I{c-D_fIHsC7vnpQqhoRbJZ
z?n+7(+;~k*jSBVLs~2?*iTGJ5ik0~AGc`Gh(xA)n)?`zs2{W~1$3v(>SEDG98wKN-
zWE&2r-!8{NF9RJ}8flRsYD$CC4Cs^DoiY2Lpm2)893r%sj4XOxMflS)Q`uB97>4K)
zxReT53e#@%ogXQv${RGdJd0dLvItc^Te}MUX&4^(4LiGfr_FIBMdvTw2FG~gnkupL
zdjkZ~kVP6@q0F%E3kmPb+gtv8_+^`ZJ9ar@<3W8HG?+ZXM`-#UU$<(D*kPn;IX}3?
zXCh5Pn;?|d)YS!mB(8UQ?5)qbw{QE4v7V>>@-R5il(}v8uJr6SIYFpf-oh-1e$+kA
ztB7=0wil*Lq2JWBwHGz?SwAwQiOZayY;=z~RXkmkL*S(3{Lp$xUxm3f1rlU(i(eNL
zeXFNJS=B&9VeQE;3Vvh^K-Bp?gjFPNtM;)*(wHcU_S!Uf{Ygt5QPoCXlFw$VJGa3y
z^-133DN|ObG{!vipgN*-2kOn`l#?Dtm%g)lai4Gec(J<3O%XY?ItQXpsu$f>mFUR}
zCjAgi&;wXqc0}*~*lnFC%c7NEJ<D01Cr8$m*!(L&LN#gnvJU2_<o=$`KL&3u^8PVj
z^WS**pMU&d@s1xZl(molzw`p{4DN_)*i(tJjl1y$#r2t}YL6Ka_O-x4N`CF7f%MC5
z^~pXyKJBHo23qcT!LPpN`9#xU>3-ZD^pyVI!8r4N*&hMIe_AJ-8~8FdEiG*-Ww5Vr
zWV(KjFO?p!H~zxH-?ow&UI1`Sn4V<=n@5Aw=1Yr+>mL2+r$N>cNl?>wk(mb>P2?*v
zdfM*_`dt7EV@G6)=pl&6L6XjPeqH2g5rvXd>8AvWNWq9VTFti0SwlH7#N;`luw^Fi
z{vY+k{|EwJosWVF=d|onXEw(30arrIeN3i+)pxRSU4pTpufHEM*hg@oz?QEUkVac0
z*nQ>v&~XP2sKJ5vk}DYmL|Pf;QZZFAnvf}L7Es5!{FQn@s-N*Agq1Ns^oWcy*WYv+
z(UUAdc=XtRx>4|u*mgfT6Xgfdpqq5V4i(PN%{I6*)gjUyg7i-MaWlBt`~O4+5k$9o
zMX0z4>9tA4{0`E9(plnN4v8ccEsu3hGM5$+8jVc7!_$PI{|qGX%lhpL+@sIOc0r7z
z5bP}hHCF?@Y?+usfL=!KAk4JNY_%@e^j(&Bvd&?3?xnZC<Btcj(cPbh?^(J~SDAkr
z{%nNN4&Qs-7bJuitr#zZ%kd#;xwM=nisY*3oAfjQc9K0Ur)Qx<KgQI9tAwgUSB;`*
z_hYDDobt=eJaNB-E7rgW9lw{kxw%+L_avkR!n~2(+~7I~A)}Ivq6MEXg*j>N!B9Qw
zuN{xzZ2xG}i1(9mQZwi-o%-vw81WYb&0UZn3Qn$WA(L|Tt)F%wYZE33Mhdb(-58!$
zLGXg6?_%3X1S+%@&3w8m^_Yx*M39BJd>UJw8)MLj*^TTKYqX^h8S2%<n<CAMFiCMT
zif6Ud^cU>PNbDZd>+mu3V*J`^q<%8E?n+Fk-5IzH36n&+i-~H+65L?Ep#V&nNb(Qw
z%n#^&t*+;4OC~HH78dULV~yemBGqBe4XIOLEpQ>zCEdcfg&&Z0Mt%t+!ziaZH|V*f
z+^IRrG=`H2cQbp3A|fP2?gMb8st|)qyW--LK4Nad&$1(b?!4Al9VH4XEDe)S4KYn?
zgTv)Dwm=npa*ASoFNwh^Kb!~chs4b{OK&Kwf|)Y}ox;*J(F*Z(Pm8q1uZF)XKv2#`
zR`BPOp8d+MH+44*S=JEtfO3p$P;a~ZRbh0esq*!~!NDL*b#$BU<bjlaX?r7F3VwN$
z)+z8XpisS}KxT)MS*WF(ZRAN<DqT<?aX$8Ce!irfAm(!l4IP4GlP{V1dCA3GOrz_{
zp(d3hn`%?f#r30*RP@`7W_zD+M*h9abazKUXO<ZMbRruaVTaJJ&Q}KB*hFI8>w^hy
zNv~0Yaw;xND8sZrYxI$I<7l2m2qM4o1s?mKfBZ~{;8B<%4AHY`*=U5mBso8#adA0I
z?0$^Ic$GfhM}~u-mnfM0-k`(A=QXvoOeLFxFT|=^qtT5|ZW5z1!QN7L4Ie)Sdaes|
z9VpQ6p21wSFW}4$R-CwsT-XE-x4|JF(zYa#HN$C^XuvKb4=XP!=w`q3f24yGk|HCw
zBLdnB;`}2GKF2<zTH#gXGG{QrG~AsY<Nh(XeIf}5Dm^Q!7Y$gIn2eAH0mY~{)IMqA
z)oZyHOo=&!&cap*U4$yHlALPgD}yE-f>Zk&=1tOScqBz`_r6A+Hi1&>&xdUL5vpSt
znW8e*OQE3XcM89oJh+bSb;{3quYZnL-HTIUL~gzd{;@vh0`d!c(YaQINeLK7klNnj
zaER8L->*}6Q-LH6!shaoZ73MNjdhyM=3#!91u6PLZda%(^_<75*H8z0W2BkJ`~}nB
zNd9(eGX#PxBk7l9j#aCw%$8BBK||XYGTI#BgAOILrbzfhc*?qhzDh@w0{(pDjAN?O
zSlCpaQ9f;3${vZkIv{OXP?`+ru*Ou%3Eg_A<$nbvQEVL>$cSo#ZeJ^Z#i>X~oFhH;
zMAJq`hku@3dOMn@LomP45N#S<E?Q)|3IZ+NRQ1w!h~s4>3X@=qqf+i8K6H`#$fHgI
zf+JaL{P~R(Gh~Sgm4q=k+h*=i1Tc+6Nn%6>9fKwcCwOq|ufnCg34;f)_=gbLTW{1T
z;PMm^Fa6;zAx*DM<QbB)5f<;Jg22qGAcz?;1WlHr=oz%Zv?oL8s~YGk83G;)fhi;#
zL&?&RqOOv38M@&o-NyHj+6uqB%9W$P#D?r6#EVU6DfHg8$OMV83q7Y5I4XirdR=-B
zg~v{uE;50?!5W4kBXZ7VrA8l-)Exnsb%6BElgvSE>6MrgMy7rMXp$*bGss&FWd1}p
zTFMUrD{A_#xP1uyIRnW3P3T9jX>SjMJF^GPm8_G3u@fxcK*3~2a-CC!kf!ZpD<l$9
z9bpnkO@5$|h=>vTGD)_g#beIQEuLBv)zf|si9eg^UvOU(1(!K^p*pTTZ`sJPtk*E=
zCPFmPG%=6H%3D^9EyD(mlWx%bmFfkf>!@p%zrg7tH{sU>n4w5y3=)+9M9J3Eo9ou=
zv7GC%B-RRyFnm)|b9@AIman3n5!s!}Yz-D&^{yn99)hV<UpS{&>4VP^D+NTIajxh0
z46C06%=AfU$eYTiRT7~c;{gwnF)4mmaB?cp<CqHlGhr$(H@;=9qB47)X#6)aYB&^F
zqwBGXq80bi#-AkdaOM24+b35ACeq&xLRz<h*@925V4g`M15dcjrM`HFVMaAya_C)q
zL@@wzKx}s+*n-pVhF5lc>E`c|sGBgXZu<;newIS=mmGlLC}xWtaUa>>{s{9`E_*k$
ztPOjY<Yw}Olz1pRajW?)5X*?WU2TY~edWy!!UwTFWOg9Q6fT&^Sb<F)L?Rlxrk$&f
zbUxtgb1khRv)S`$r*ig$L~wu4t?L}ZI0zd+^`xm)dpw#oJ18v*GpBcKymn!Wc~;W3
z-2T)$y~|4zh_Xax9`(R!;^+nyXFMq!>mi91z61=P)&5Hk4m6qOLK;oU2oA6H2gvu#
zJFY-UdRfx7enl&SU!xvIBJaUm3l}0@liS9{h~a4Ql`)`ou>29|TWy%WWvz1LnB*rB
zMClIvL=cnd29R#Ztkw5AjejIMWmG<!MWCfvm6pDjq2Rw8@R<BJ0=5KVzjdMP=<rlo
zKovBpeZ`lYU)Bs{R`1}`hMdJG2OUIQmiDI@q(BK2GOXYO+h$Z++95fXzWrCUq>t~d
zRp=HSAZj?_JF@$tgHti4rts^in(Pitq@6T2d))BqfZk2zUm$KywVj5M;1Nb@{08@>
z>k!Hra?VKfP?TnlYx@42cQU@T4O0udP4yVh-mF-hxp(=!6R(ZQF-O0<KN&Se&=r|%
zVu<Ynf`LS0&=g}Bn9V0Qn&Ru4P*U?}^dMLmnq<s`c>9RYX6p~zn(%8@%=i33F7goO
z$q}<7@z%`sC(rFAG9n2SkcGN1=Jt$VL;eD;+B300*d~%c4fh-X4@2fojbSWBH#*v>
zQw%D8D_p(p406yfxN>Po9uGy!`b^PMsZX@Aqxp`Q*zQc})YuzP?^xAFCc+@-43==N
za#(XAvG4hbyFEZH@e@aiOx7RJe2jD@h5Xg8IW+?h(NoOJFXJ)3w1r4inC!WYT(q;V
zPsigt04{>62<&MH^|@?IW*yIRcq;t9-QoEcBz6`;!m77io661{+m<-z=q~<ymW69C
zPg)h*mr5?ocp*uOoFIf=BLY&9cnKQ*0N_zSvs1Y!9Fn6H{EGSQ12&C`naz7KNQ{D}
zhY+IgVW<Z!4Rc(KaH2mPTe_v|(hA{CK0l0TyI*A#4zY+R4~1v{Jr$(#4vNH>JQQEz
zbnL~Zi<^LBZ6zBiKiJs%PeKkr&6ilp#&(Gf0gM#UM~?-dhKQ;^!K$b~+5hcg{~fvA
zUymaGDL&H(RzgRVcz%2&$IX4^ncD^%RT)fOSQJLr{e4NNm6RydBnFUN52n<)5j*}$
zFh<Cf26G+r40nyxD`lEA8XQ`E`Gv&z5}I}Z<N<-ntEFd$A8<>GK}T;Q)Kt-erCsn7
zQ$wdHXa?;jKoMIi40A1CLa8Sv0upyHJG%=rOe!!dnLaJOnC<8}(WSUETT|3~O^B?1
zeB`0n0GIcwU^{S5g<>R+1$1G6DgrGD<bxBX{dEUX-if2H38<svdJ~!G`8iT{ms}4e
zJe-6)?<rfbWnE-clCv{ItXB{1Wp`F?s%DI=5XgH-LYV;ej##n?5-e_8lG?hI&4d1B
z8QiZ+{=iGFCt8Wf5fd-;kj0K)Z*Ld{61DWkU+Ow`2SCmqRG?U4v`jC|Kbn@UOdG^4
z4E5@8>ICny$O~Jh$YU6StRxvyuOQM!04zL=F;j<sUbQNjsfi(ayLvGpka#um$+T`#
zu-e<(<LHhe9`%IMT#Wp#09-l(955u_sVGSX6f$xP;SeITqY1KvCkqhw3dl5Ek-8un
z^j*HK33q6-)pHpsCXrUoj|3kK==cDYry7(R!*ZqrgQ<wu(^S%WBW}2pm~(0q>y$oY
z4w)Yr1_s8-;hh^PD{P3Hz|Dm~OIX{6xhBcf-BdugosNU2TS*?Y4NRu3T}8l;4d#UJ
zkgF1-CJ@$uYu3{7+r3?*N%r~s<VJHlnEcpyO<$-B*_2@>V=WE%TW7MGt1(QreGsve
z^(QhZ6aOWU9dg{5X(81%56#G_a|n;&$r{BY&y^ZF0GvTX`}1Lk=!GY(U^~hryu!v`
zCh<fwbTgf%nXvN})H6}cv?Gh`lmj{ZDTV<E*6W;wWAGjp2q~RHY)Lp?<^0J0ppYt$
z4Ao&pMKl47_hquNz2YH+!J^~97n%ZEwyWFu5VpI-EMxFiIkJ{9jCrier`-Jh!|blM
zfV{mJj$3w4c1wWcSWh>p9?1yDN84quf;$#P+l8LJG?rQ#CMb^umrwXU9i%?W;4+!W
zDsb30m4Zsx5akVso(l4$gDne0F?Q00xUMj>ScsGXh0K;hHG?adlW-I<%ESB17Mg=B
zZIV2PCzi?D`~-mz+JF=l;3bccN$h@I82tGWpE&V^wusF`0So~f8`oRp>9$-~cz5%L
zJljo(ITrB0mN5d_JkE*1#%A`6xWF5ifgrZV6JgFz4(UZb+Wp|&u;n4s$cxL~uJ~=k
z!;jU+Psd-|rKx#ZGv8CyI3YddLFVrpE*bRlyp{_p`SZ`i3vORZ(c1s=k*$^ey^yxv
z*gIdvD?|s&fAi<dE-0uP85y}--_hRQgONRLzs-)03TC^xxXk8zW0R~UPO<2hHGT3)
zErT$5+Evr0#_05CU8VL;vXSRdXQt-n2KV;%md#!}{IS2^<Vj$ljg{3SoG-b7Exx7M
zg@sD-b<Zk9f-CogE@$(gj8CMAAZE^cor-*9krjOF)-67{<c71!4UE3ouV23gJBx^j
zXlQ9UPl=9=jWsnkPN!AkGJdiIhgh_BkVfkrA3uEP&><eptkhJ$$jHcws;UC<tT%6-
zf+*3eTEI4J`3tt;rH?TD2yR|&V{Pq+so}NM301WKr~I%B@&DJR0xdn5FabC~b#fN)
zNb&94r#~#10UWZ+$;(SiOjO(&r3)OYxOwm1xt~8P<>cfpWtiA~1P+_}T%Ha*5%K$8
zaB=U`Y7Sj52OHXiO}N2^IAPTQY)$e)=vsa!=ulTHYz&TO2CR6$F*MtA<rm>~ASZje
L`njxgN@xNAG6O^o

diff --git a/reference/plot-11.png b/reference/plot-11.png
deleted file mode 100644
index ed646966bc5c2e2a445843385614c4c049d9c0e7..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 33179
zcmeFacT|+;x;~6CiCc{lY$ysO5DP^RrPu6;pdg|ky+sg_E+D-ODauHVNUss3myuqF
zF;a#i5Tv&mr3wQKFu(xA%zXDVCVS`mt#iKL_s2eGowL?vt*txoGV{*!KF@t$_jO(O
z?US1~G!N|K+sDPlb>P}nbv-Vw&wRMJc3c1LbNJ3^-1BAl>vL;uO?CJizP2vcVc@U5
z53ZVcaB=a6p#SZPnqINx;`)l~n)=^w`=l(4`zBa;2C6N0i+kUTPW-to<MUgSBc?Zh
zJ}Z9Zv%klkiwXXl^snWIelv8-ySeQ#nWrrhn^95yUT^-;%U^!{?C+1h%n&ljh|Rjf
zQ|QEV;cl{cS+@#7ZS~FE*xfj#OY|9%s}DZSE-fXQfu9XpqNmYRX6vanv$A(h;y&Zt
zVw-)T=Pu6wyB-nDi~i47w?45%e+R4Y_2K;F(CK%aYp=%q3oek8qGH&L%&e@_rlzLW
z{rLt*Q|S}UVKiT|H#zWVs$6El4C5LzH-zrn`1n8%i6n6T{P~qfk^y)6WKH6lf7l~%
z9DSojuE9;t$XIRuJ@ZGqeH1gr{B_Iq_4R3*v%UQj@eWa0SFT{YeY;$(A3uI<{r&e}
zeg669q7A#1mB~N+@Iz=<e2D>#`68&Vk+IX|-EgV4t}YXgSLsrrkKG!TmX>asRby`*
zKYkpG?zF^jeSX~BHW2P;f0{^sNrOeeO0%!==8#R4!QHfl8pf}2yj<Vwe)aCNp~i9e
z2qo6dp;&D7&T2>I&2!<y8~wN-&r=V7|2q%QN@W%PJlv0$b_7@K#AUuSjjNw<E$scY
z-v3wIj<{(%p`A^lTCu0Xw-E=@OGe?7YsX8ZzGUX8uIFK~J_{wp)$fl7kCbBeaJ@4A
zwEjsAyoGQ_X=|x?q{2+Xt<f(I9#lNMYyPSN#kpc<f|r1maU4)rZ@R*3A-yzQ*J#NY
z8Ehf>Q~7y#sQzR|nw#Cqz$){T<Jv`ZYIp<}*Shwr{V(@0mP6E1bP0>ju3{9ah=_=l
z=iGtkgoFl)%w|KPU{CxS343DWCV|bQPBO^sAa(UqejlE%q|7SkED+PYM(nrNG}X3y
zwBpfuxQz|w{hr&iQ)Y0gcQ)0V^g@6E7oOdKWx`@N+jN5t>8ftsz2dkVt`)ufO6fCi
zlCK?0n72{QQaX-1X!$4K(YUxcpAXMNTAW5lN0Y0PRs6k<gf&}?)WiFCkB0Azto{D`
zO%|KHGb2Z1tF5*QYo}p0sRlSd0~1_@v3QGf)%Y>0-(cOuX-VH}ZxW=0k4cyxclGTs
zvz~5`%^Wb^+iWehR6QA4{qD#86Te^Pj!^KJQFnB7e3P!Gc2Ifwl~0Em>5$d-+xUq=
z0?TuF@VWoy`;f`f4vDa5HxxzE7-O?DG^MrQFHhFfy24JnRDUh&GA4h=0=7^0WJZ`o
zsBomb(8h=(?!K#Qlu`Y}(GsWO((1YF<czASszVkRiuVWxTu@zkqT2)~z-DQzM#`oo
zeDf!@&7b7yV;(keIHcs|jo0~L_gWK}8y)d^bUZciuxd7Zv=X8~vf912=+g#uk?+s}
znysNWFodj$>-$}+-#_+YjFJWR^GbxDxOA^c+j2Epo?1_vh$rk!=hA$af7`>ky`K&$
zo>Np*v}WyWxvjRHCF;nH-`-eN8}L&~(-Oy}GZ*Y=bD5Ef`GvFma|>nNT{*4|TQk?x
zg6vG<V#6=1sHhxPy~o&CJROyI^>j<5I~ms#SE3b_=r~v$5f!wS=?3qk>F@84R`Z79
z+>7doP_c}H0x=<>?d@%??cBGN+eIySKhuy9gHp--ik*ex^+JKlN3nf_gS>KM4=#(F
zRf=0~l*Tv1hp>j#S$Qm1|F!B?n{LZwl$zy0fao|oc7!mRYBPBiZKQyS!|BJpzevva
z8{*~bFrFNwH_-7u-TtJ*m^Dqzk;TQup0{rgsWFDTy?g_*221*?jMjH7cYizHno#aM
zT4~Gk^osfbiIjFUtXXPw2~Q{aj$&II*0LlchBu1qmfA<pThu=6lcoB0`>)E185O@_
zV;($sIpuq1dlb&ScOw0kCVwm$L&ekYQ3i@SELp4ZQA+7(jYsQ!Y2Wy{S`)T9T}yfE
z-+g9Fh;zi@y~ehREb3&Hc9t7E(#Wq~y^^3EKXvL&al`h!Q+nV=;_?dI^Y8(=v6};q
z{1oP?4r&m+ZsBF)!Q_{Z12=}x*3n7+9r<?@1^gFE3QJ0)FeG%rWF<QUvl&vq2ya?i
zWL9NZRGAt-OBBQ?`;%yCaxR3xce@2!2WMF<)`GLE|7i6hx60b@rIw9%Nwe_OP2y}%
z%DKE5J+kkS8!@k>(4_24IuSmZ%p8$gXe=JrM;aejd2gc;S#{XCR8+<*h7n}o@#YjC
zU%Tqgd+sPDTy}Q$Ld!|lH-3J8S}cqxb^U(6%u;MY+Q55<wJQ~I=?=wq{U1WwT=sOV
zltcxJr_TyUDPO$9K&J;zUh=XGe!a-Gv(lBUx>S#z@DO2i+-CBD+76S<W>$Ci4h<!Y
zKWIGgOrTK|FDX4dopnkHM~+fnZB3wb>SDHgcij5<y!FQ|`$vN$x%w{cq8?c!uvv$f
zzEEYZ_0}x}hx0}UtYr$Ou6B85?Iv{x%v>6V580kH$YM~_4;(n4Q@Tg9Du}u5H1j6$
zncw-z&*}?|aA|$@LK@s}it!sG*Scrh#c(Oh>o;b;pX*Yhitp9$*fOhd=x<6@)e$o+
zY)!}AJa-p1<W%mW#^Xngdye-@yZbgQly<aR7B{TM;_m+R(@(>0M1CVf2E+&J*QV`(
z>0%Z2_e|o(gBXjx$<1NBx_R7nwY6Gl1q1>{i*~}W^IcxF4e|CTB(2Cq2w6(-Njie~
z)3^I_v|Nlt-xUp18x_MjD(K<uHxJ6MNHfS3%;#CY(W3raUl&Hw=ePYW2k`Uf-xW0v
zCD@u&LJ%rVr$Q7Ht5G2xH^ec%Q#x4_WaZv=+C*EsBVnC6bylrS(5T2wu6fZwlq$p3
zw;xu~!f!5XtE)+_VI|_$XuzxzxouS3y!t}=5`}WIf~gHrvc41Tzgwd*Wgg6dss<U?
zGRU2!R=usMgr`bgu!>#q*3odhoE|*5z*uy}B*Td+e=)9Z=MqQxalb_Ctb*e};X-~v
zar;2AMe;hGZU-M$>{ytIuCDGz(_!7a1%^fKta%(uv<`x(M(Us|hB0bk*-hT*B;Vy$
zUjBBVxAz%({rmUNjaE9FZ9qu3?TDA`j#S<L;I^9-NFSFoYY6b&XcdmSi^t=A)1Ahu
zrRt{R%u073l>Z7-+|ttG#N1w=Y#_OF#|dC|R-#kKASK*2vh=mxn0lsr4<4OKiSZlQ
zooUPunrfn3%g^)jL`GlUX`CYcEF#ZXD5K4exe}OObB)Et#f4>MWmE4mMl{UQtY}Gt
zcj42u0oW+BE=gG5*tn<0T8Yx`+9!Tw-_{TI5GOa*aTpadLQtTrv~&#YMB6AemXFtf
ziCyd)@2O$AOp)2Z;;Gdn+blb6tA{;T^p}X=J^!WKuGUW0;F3@N)R0tca8<O;%<6Q9
z_K1Lc>(|5RI0V!z&ds$2v1sX0zS-We<F%svr}J|>tN8+N6;0cDe_9{A5MugyJ(<m-
zYg^J)84J(G18AMTY&tAIT2rr<6LvA>{_6!+Dl>kDovieBmc{v&77C8gFtvW_!K+=v
zQJ9^{XsHDhBM&M}{c<hh&U%BS^xOLsdIyB>g#k0y&f41AmVgL!Fh?!3)*$HOzg)i2
z1Ayxp^?HiW#FNZQh1nF3)zk=TAl2WOVFduDu6(@S*Q4#m8)GHq=>_vB<p<23sG!>$
zjw8OaRLDjGbbZFt^r7?zulL%jb^BPZW$;eZ>LCYaE2mqR7?;wN?7C7GFl;&|N10$y
zU}&qh`5X6{FKz$=(>nM{71lm#v>}CFW-NXW=iBM1bx@qB^gW7WH&oX9>Y1gs9cO}m
z=!INlw7RPqh0bWdNQg2!4U>^d#n!v}e7r2v$jl}0sI7Vk_oLIl+Cb#*LIA*l0S`D8
z?#p-7qg%y$%=C<3x)>e~xMArR=7H<|`f1&N-@_l%JzF{Il5D8c|Ai{V(YwQa@NB{p
z0|Nt3+Zm?%&LrA&dB8sQCn}L{ZS11mHE<jgGg2vP?Eetv;OtRQgATdF(d7+zOtjhV
z?sWg2j4P8H{YQ<=Z@twdGe~p|XCuzj8ebAP$zU+bd}%Y&izpAn@q5r$sU&%JGUATN
z!v2VI+IHa13NdQyhlQ)?5sg@8ZT#_wZecK(2H~Eeq3VtsX0Y+S{m|+-LDKhSXiL=G
zjo&IVuenHITL8*R=&q~#gtSGQ)!p=5NbQy{RL{Mz`h7u8j$J#ghOtzq?NA4Q=2)23
zRJx2OsZoFYY=4WhSr=ogz=|pAs(<cCDIA%-jvtnmBx=&$J=`uex3P%wo~10g43}oj
zSKKw3Tdn{8w;%U%7kV`qB0Q<!!g$&-1Zl3tS*eF%9x?4Xb>e*Q-g~+V(=h?6Hi2f(
z&Hu5s5IQ3gt4My1K^T93#?Lt0*$Y{EHY;~^?tNU<XSZCeZaXHMEjn7n7mNkv8oa8b
zopx)ft^fThq|N{m3)1e%3LX4s^>zME<`oVGr0k+1=hu$M=x~tB?7Gt)4BxP|x<Ayq
zUN`1<*74$`i`HmCAIXs`mp9_K7)Yk_GYK|;&Tzl8wg(9kHJHVbiV{!C03dr>GWpac
zyRPsnVaKkhH^J%kP9#UjkrH$zO-f&w+Qu)knFKcPqwnsF$pWAgufYd0Htf}?6f5?2
zz;+lMl^Dzc)=BcFhAQ4?WkUX4@a}9_W59J2&5}WG>gvktefm9vgK|Y%7M}eETWgPa
zEsvu+UY5Fa?+w<wev*G{v5wl{Zz>Zrxz&vcGWY0|*;+&Zw_#^mlDJ28`G+00OWk5+
zf{deKzvJ+*6?@%E)5C*S?Tq4TKy6vsrGrWfFWp873$~ISIdV=z7mBPN6t}EyLy<Jy
z>%Ukr7&A*l&_6wBJ})QRKwrkdVId(Oo(Oxxm2G&jyRx#<ZPw)6kqH|?#?UvSMUfDs
zlnUscJ}>gtt`t?)^xs~zwXQ|LC}6v=tSj`h{l~6vCIHI8wB}$|pQSg&oj1$ZGdsPw
zRN^*u-0S@#-j+9smx#{fkr6&Y2-Ig&Ttyk*Wuh3<p6^|nT^rfY*btqj^Km&`d!=*$
zvz>!U(j|M1I-jPau-=}-t+um_#mdv(ToM+V0ALUW05P^)(Dzrj)~`cu*AKLb<~)XE
zk)VbN3aUG_$CcOO;Vsob*48W$VeRUvXzdA9M7}uq!VqVfs>^P++G#GQ@C^X6tr~OJ
zPQkFY+?sean>^BJDfgY7BcRTSCXiVhRa-VFf)vf7EvAZ2WjwQ5jWwy(l_Ti$?g#t+
zo5lLNDl%OVrm<KkQP0gWbOY!0>tyZ==A}$;4R;55sRlxgqIvmpWA&oln}l2L{@;K9
zy+iXA&$z**rKJGBjNLUW<&~8)y^#NU^YfA*^jtT{YI&nXt-q8W)OeC~bHF3&3)zC_
z$0kW#^S1oOhE{zJ$w~(6FRx!8Z1r@WVvYH1jJgt{M*-#+-Ph@VSx0P9TB;UKS6yj3
z+~M~IYLT-}&)zot-qr3Ks8lecXh<C`h#|vC$Xd}2xnr&=;CojUqM#8GVpPM<hUcIK
zFOO1}kl&I=`bi!hqJY0s5F5iB#57YZ>9ue^pfa%mE`X<G@-PjO@$I^`y|Ml+<kn01
zqHcz*VwNkR)sVp_V)So3Ssbk*Hgc<-65RehWa6^m_D%JL58b`GYARjwli@y#j#f(Z
zCoF(QK#7OUYE;{4v>%MMH3``2u!=<0LJyg&mPfo%$?O^^G<l{=V6Ef2tvpJM#kahx
zyY5`Re0ggvGfGvli0{l~s8}%;ee57OzEf=PH&VnVD2Q3(hK*hA3XIH>6}Bps&-bRj
zbHA~ru+Y7(ltfAY-o5FX0)09Uy>6U0uwF=rgm4d)RKUq8C@CyWXk-3Ve20UaNyRC;
zmm$TodL-~xf%DSc{@#mH{BYBfn5~YTjb)Qx4%<69g`-l=*JtfRW>iZ9W63O*Pt4Xk
zG6G)iQ^)DZJB^IC$!pG5_g_Grkh4%(XTx7WUo>b(ST6dKPtJQ38OqDc)te$ypT55|
z137V3TO!4+<;0eWI(i=(kpv)MYpqrHRktFgFn`b@@<~@fem3Mj30bG1F9GLb58se~
zKJ{xLZKiW&S`woWsW_jTAm?hiL&WS5C8Ef_XUh-8e_H(^BWyyFy(;O=*yR8VIecf4
zN{xb@2w^^wL?V4yT#SH9WXBaq8urPW-StY{^-eA>5?iJi^`=oc5lIsI&jOdmO&n$7
z2C-OFJZ(KRGC~0bmRmJlVpi-;h4grdjiPADn}vGug@DI%#*eEm#~K#6xk^`9h7^qR
zR-6(%o0%5(WNygUg7V?RIYvph-$2>w+*5K)u3ep@Q_B2t+0|sB%xiA2U~S<&Xtib+
z4dywk73|H&ck<*U?5;LfGCUyQaR60LE1f-?`cN+j^ZaHj;kERtV*Q{MCQy|Lyi`fP
zOio5cg$zoRsMbgH6TsP%a^wJy6nGpxeSJrGc+4uDp2*?nbXMTVR!)^DQL7JH?f1<W
z05+w1b0Um3s(H#yJ%C<Ej(`IpUg<n~40hiw@ngcHE18d;+M3iuA(1lRaZr2uSIi5@
z#4n~Uhhed5j54(X3Z-g5FDriVj%47RHf;`)jE=9WFmBqc%fv8NcDU?F&LywDk*<#K
z09Bhr6=-~Uv%?KXQa-$$k+SZjJ0FrdSkd-rMI*U76uTEEX;wV$CcCk8{A7X5<E&~)
z>7yb=HYlyBeL}M5JFyIapB#b3LO@Vq6E1VBl)qrw=cFFvQIhCbUhCD>1k4~g`z4P|
zorb-NBWzn#w??Y5Hmg_L#WHhp&PvoRl<u(5**k4_4&mOl$)C7YApo)=O{VFl0}fjG
zNpR;Y(q@ZX3oV_!gQygUycS+p(*ZZytIWQH%^U5=VxqJ<Tp~$#o#Jl1-J2xiI3X;!
zbv4L;Q!^6ylL=sX3WlPNFJ2SH*B?~ceD@P=&Wbs21Xyf&j}cGSbGOqmhaIlZil?+i
z@>dK1t+d)`H61A|++A86V;bqderCNPx6d!mPc6I;{jexML@=njkO<XAw6;?CPIi-V
zzDk$Qm5P=YbUz(m#QH}}oJ|jKf4!AmE-7sQtJXhasiKdT?(<z#Y5Yb7t}_P_0vb(h
zX8r=o;~y3l#O>8qYw>E#Io-+kA%b0MotEYIZ#8#)e!?)!am~Q&`AzJ8iR!P8oi~kP
zle>d9%A#d8`8{u|22RHs-Jon~u)Zhl0hqhDTLNH_N81V2UYS+imzo25t^mNbiZ3zY
z+)DwmCO9=HyF+EOsEGJ-FtD^Jm|9hIx(k?R`&RLJv&tBAf_JUmGI}27#Hy6CSWU}G
z*V^_IN)C%MIA0yrKuS?n+<L-OK~Y1kA|l=|+7!T(iguCUnTzgh=czr5lfzpf$r^QV
z%K2tRYcNJ_C@*gSd4($^7Mj<%zn0Bg*sr7MKrybUH0u5t5^?_DUn%MAKPKJ^Oihob
zU<Bgywj}=+<)t#t+pzp-E0SGBU@jWt{XB!nC+t#gJxv-J9i`3PF6Ndy%AgW3s(6~w
z@9c$cmUub~NL_vRmd?8o%3)b4lw6CTt!e$B;vRJRw${)BC9&MMjYFHQ#Z5}TVP9D~
ztg9@=nvi4(0VO>F$*mhl_uiV0C!Vhw)9cWq=$(ecv`4__d(}Y3VwJAQ7spR1o`Up!
z>)erSi^0})uc@exi0No;iNp<$NTbGw5>~ek<fI4rsRp*4j=x^rz<uyw3z@Y@Hmr{{
zjq+drrnw=v#V^iUE!=KOb<l{g-Cz1+X_!RqxpdX3UoN-ol;k@^c=VG91x-WM59i(n
zY(Goinflr8EcX{*#K0MRDJgA{_i;BJ{&P3U0BLPTT8-P;sDqLuz^|M#Co3!4l9!Xl
z$HOyd>F*o>xvztI(syI>Vg2$W$u~Kel`yN@chl-FTUl8JObH8$P`(1Bezt+R=sH{F
z%}Saouw2iN-W`k59XjJWdBsm@Tk}Q)MVHZ7F(X4|Cj|hStGH#H6+tyl+(d6<P+<_F
z9EK)`(eLro@R-YrpRqGh2DUBIHIQPqm3S7hkp;!IqONs!+oJOWvXpIO%`HPZQ<dZp
zE6bx)>?;?GbBk2=n0a%lJH=3?@rL~Sp98jE@0S$cl8Q~W@STo5t#(x*P74m`2Ap<P
z_8B4&)03&x0J}DBVt9<L3Bpwdd-EJsminI^VPx=8OhTC#(f|=C&vYiolU3(lJZr(L
zen1s+(4{+fhw924)-X*|bgF}tY{QZ#E^N5WPO~ZUOTXb#+sVi~P-Iv-jg;HN8U8<4
zJv7|DZhgS43_CQ9g9@i_1L8k3cM<BWvjN5Q`dJlU1}szL$UPu&eUss3?SQOvHKD1_
z=Vq=F$03m%uIw%qH+%X0-&qN;S!xT3;aOExa_NH*KI6vSJI)US(2G6r#TTH|*a8hg
znV)ZW0TOtDo4w3Ut8{jD&fHjsfP4Eb0eJdhB9I9S5Hcj@q9Cg-Jk=CvGb}WInhG&;
zi*}%HrRAjd42?x6r~y|<EPej`xp+ruA%URA0DU4BPP|heeI0)31WhKM76eq3%Epk5
ziH(g7sS?mV@Z75LDnL`3z-}QG;~*>4c{zJE9+10B<o7l|#xo4HT)Y47?E9TbGN*BL
z0-Hh5boK`3V%m9wjA2`B*StMbr;LMA(Ukz~)80|;0|#R30R=2cjmFVbjU0UW=YWsu
zn5_!fTHR~a+hw`cZZs2RRp+(%LQpL!)oK>*Em39p)$ZYU@4k<-RT{{FGp}Y=`*1Fo
z(tBT2#Ve-H($MkEPCRBSK6)?!5U?HMWE)PcL3vCc_qJ+?r0y7rAM41sRqj7UZ#R&n
zJgC^tNE#h;D75RJ4{;x?azM#i%OL;`MoN)sh3IHQpg+7*+*1hj@=CzRYme{)sWOS^
zOSX9O4$Gpbr3G#6)3{kc1f39#)2A6?OeS*}!ApJ!5YPA;=Y6g2+v%GS3KW*IEAd)$
z_|Jp=zJ6B}tdlIeSlZ}5bSETYnALVReJqO!z`4V(*xjqx-_NfUDk6)tOdxf|j_fl5
zkil6-0VCaA)VnN~+PQ40*)1j&gLz9ERk(2S)G6`3wP&{dfs{*JtRe^5E<VtmdBq{E
z_dR<3!b(@h`yud&rCnWJJ2eu7omHvP4JcbB{d&T*4w~oZS-yTQF2)WlZdz4*3$gIl
zK!S7<z=r84i}L$DF9ejHKE1<M$HZF&`L-?g-1b|@^?YR}eyj{GEm;_JTWRKvA1hXW
zW!CA2GhZ(~XjCAXkMAs6^D`_8%C&u!Y-{56+h@Gj9ma-W*OwhQDSKU>tVPGf=>)Nh
z4Q*^nj89#v7XVbqTB5cKG|mGElm27Ds_|OWXu;keI&k#&;o(H~FMTgpigW6K1wfTo
z8#@&gr=tZqMMdXXzzzq-vn7v(NnN<`<?h|Pk#bTpTM#oaFtC}<-b$CBj?r&-AP*0p
zKrBF;(bwLUP%Uo)3u)%A^0v*v!68(TwH&h2Zq(rZx-#C<&prS+GhAzhmbVwwd`*bj
zl!y4y9`pvv$iqP8<@ucF;Q?F@Pw!N+oNFW&D_8&i!0GsN!zMsvx2^lI=h9hug@T5@
zXY$a>BJm9!7j^X=AXpRojJ7mcAAUy%H?;54f)#7p>Wq0ea8@V0$32Ze`*3=_Wni<A
zn+14cbEAsDKbwwW{p^eb3fc$gop`duk$s0*7jzZoZO01s<ZWEZ8Z5ibxFZyA;bl^2
zT#`}fn3fi_J-71iC+?Q98jrJ1W^=5lgpb=6Ub1>VzW^bm)}NgTK)CD7?=hx0TjfUU
znG^GC@FUMq!S?ws*GCIGKRIX$2Ch{vlD7w{;>J`#WfL>)-g`FXOI?-q+dGqLESpol
zQ%_G1CKlL1dbBc!+g}0bM+C`i-MIFlJz|LgLL-S4fXflIkTxK{p=;tj$0vC4AFiw*
zO2LiIq!Uq|Yyr99XxY8aP{4VBi4ZZ;jMjY6+(QSjA|YyKX2xPh3YMyRHqCz`&dlXW
zmUes#)WnA&6iT_0>Wxe9UJKanO^-aA<|8Cgzx+D%XE7V!fwHut>=i*akFN5HI(oWM
zs5G=({*hN*UthnxqJkqY!%Mor`UwGInh>+v(ozO752tk=Eh|g-2dH+!j*1yx-yE^p
ze9hYenS?f~9Vh+-g>cdJto%bAl&ic}0dA^H9D5sQ+y9HL39A}0Z(qA*0Si>UR7<gj
zAh{ChT`z38{dU(&>w-VM0Pz}d7eHH_y#f3c!pAn_wVsG?h*9Zw2<NZpgTet({u2-j
zpKu3$xLm?!F*MQdB*};&OuadaVa*x?f1CoxqQq^oWyehz_g>y&{o+#pPk*Mvn~{h>
zfDZ>c^-(tn(ao<P?We6TLLnJRVAK)9pg#YmxMn8l15`?mX%OqMZB%l<gA;s(>w?C%
zvN;(=)Kp!4ikPO1o9CK9J+WSyXkMAq#VA5P=0&pgTB)8c<Ga|n2}Q%s#r`xTvt`NL
z9X=9d${&39RKW6YD}d#dM@B}hVL^LFMigBbh=4#MwMJ8<;4!|f%h7%5h`jA4eCFPz
zFxzAV(g~l@r(AWV_)mi8-9DFmCE&8jl@BQs-7I7jjD;77W>3_Zy=p}c&ewHO)qZ;Y
znXYQh<9@DJADnPpzt(<Q&*fi@sD8rdlUuXtf)FQcD)E%!x(Oh!LAJ+Xn3ThopZqw|
zRw@p(7k3k?g|R~`GQ)%TB~O{+eOy|@mb<=P2N67N7h=cI@;eQ^{JI#l&0C{)O^zW&
zLfD$<*>tuxPQIV^o*N^t3TU1Pa^gt~3kxY}=_5dP1djkb2ONA69SHVj6gxp}2JoW@
zWBXUGn|mvbxNZuX|HR8R{|1mQ4~HZ3Vj!Tn(kRIK{lp;Gf+6Rcl;8Gue@Q1}Fko7^
z&T!o%R;7tl{2kADwMX?!#CD$&-+glseQ-Y~i(S8VArKgbN-eih=c;DdE0`9jwT>gX
z761W?&Xd4$hS@|aoFT1c%Qcs!trwcc?|*eEX7MOj-=19{rzJu-l?MrO0xm4@6ydsX
zW>kzTm%I44{h;A<{c>*bU+5J1znD>Odlc=z+PS?>$ywio19k<f*@xcVM!6aD_rU`K
zh>O?~aWd0be4sW(s-mLeil(Nfy1Em&K?Q_K2lJjG2uVnCM<nbA3Z=kP!3O03AehwE
z)q(W}n3C0UG{VK_#CW;A{uc)h^w*}`>V(sol$0c7y=|je(V3!f3GLv?mWT=oTXTMX
z{+@w>V>~?QF<-cFb27BW@Z1q=h}ZcPTWKLg3Eb;UI?BT{K0fYA2uy+m<}_M)-p}9P
z(V{)!F_D4Af;r{G{CvLKl}QSOJ@ia!Jspd*PYgo*&Ma{w*6dwZ&91Df(uDjn9+ihe
zBs>o+78QG7{}`ywD9L&`9(C1`T>kv|*8q9#$mOpUm6U8iZvzU!kfPTjqnv%u<k(4B
zr(92k$@vc-u-`rZGFjF+-&15#%l<k)iS1EurI2@uDkmqWX=i7ruD-Uu-p^!_QxrUa
z$oO@M*!SLsFjvd~$sv8!CiTzf4toQ>zRiFGhQ6L0TK)w6{f*YEKQ_ky#0wmXJ$LoL
zbQ!PRT=FKIM8*JXh`1|-b}C|)AlEg4#o{C?vK=i1(1-c<%EGN(u5q?*<DP@fNCu5&
z13+~~bh#UNT$<Jgtpqze{rftQ_k}C%`9|&w4~VfP0BP^LxI_}R3kacz+dz8;9!MHM
za|AgZEQC0(-n<~SiwfJ>W@~jCNG7m6eSJ*a_!qJfY@+wCAGwiyXZ!P&(WSwmqc%Zz
zo#C~4c--1Wa$<h2L*M4|&sPz8*I6oVSY!%6K}2dq0mr~Kq_Tn4mVW5zc?CAMy1KaW
z%dZ3=1{;@HUqPHBa#`X0`U@sNVhxY4pE@m5d<4DTukL)M;vEhaO%GKd>8}7=D>u`f
zE(q~GqC0TOBePNoFx3RdLqMyA#5S=Xa$+w1JiI)ii+UT1Qgu;_G_4|#GI@B~`jWI#
z8|Q0Ce$?*U+iu2qy4}yJMaL%ZPiK`E&iBq=eQvdG?|EEbOft*XAl*=NA2VOKpfSPJ
zSLA#_-kQC-tpooS>H2_x+vl!OguqD#<yMtm)6IcjIjLTz4<9~Mho9_#gDJcl!=kw{
z*9VEHij{}5ftYFpq+u?-qN+-_pe)3|TA(h>R7^5fF{Ixgzivss;Av<nT9K`1r>+Qp
z&b~Y{uk*a+sa~F3tfKJSJefcL49U9bx=$uDMhl<UcjD=dqCfvs5P#&SpX+9P-f>*`
z6B0Neu-J3Q!rYK^33S9&272U>B1=HI*&G>cIEM7(0H_!EJuHigib|kDkMJWy9u<vl
zQG2EuIK&^2VdR?!e^8l;T51Oi=&bPHbNa1M71hVY83da0i>wu>79F&4Kx-E{kb@je
z4!3n`i)-ieqen%=3q;Q6@Yt`}|14F(r)O%YKiB&=yJf?sh2b*V+-dE&@R6C^%*aq&
z;G*LS;*Vdw`WB$ctx-I%QIT@MN!h3lqPS{60c|Y<SOL(ib7ayT9RG-`**eJ?|3_Sf
z$g2qfI6{oi42s4#P}4)aiA(pijDm1!-Ib!i!voy9Ab{FDw=0<^Ej8p^NvD9(tdOv~
z4fTGTTJXrrJC`F0PLKfrWdfcD6~q|ARew(p50tt*JSAX}1HB{H?TQUlL}<nO*Dvyw
zO#akqQg-i(`ZpyX11bvd|L=e*ML2O=62(B!QbBh#MKREOY3#n!*b3v-e;7`OR+5i_
z&M#AZ-$}%YF8p>x&iVd*ETouWxuTa)LV-so9OR(>+L#v+r)HHD6%h!6HE}TRY0~4<
z523S9QH^eh5*eMHojp)*@F+F>W=UmW;B!8K2wy9x9}>mPu3PF(SHrYNIj5`&Q(9Xu
zxrA`%dLjFfE;swDlj!(;+rEJfu7b3&F>II43G%dDXN|O{(mQ#4|M<C_5J&jHVv}8J
z?fV#1pgBP@nbWr7gHY5ID}pWzwyRb-^Dc0lV#y0h$_-cz4sLZU`{Ii)n&$fpCK1e4
zS8wBf$%z&(oT3zBE5M+MxLJAj4s8+~Y!S$>6fcYF-N~akpDPFQk!z1Dw;o<O0_%gc
z8^F74KE!f!jvz7|frI97*s=4W9J3sIXLhib<BJUr9nSoOgvBCj8ltFj()=8Xv^npc
zo3Ap%_c4cbOaE8eQ2#IB;$Pu({r6kI2KkqGhX1{i|KUo?4AutsLD{O~F94zDXriu+
zj2zWi+SJSpe8GH1mny*?q&w(O02&{JzNC;G$}I@aZ4dDqbFH@J5W%aHdLjbSDHVbt
zG644=9FGqU_w~i-R+X2>j1|i@R`k;@@~vPfZ<in9C3o{=*<e)oYL*6R@`{CkFCnr=
z=Xx092S?y{#u6aMnO+p=OFmM`;K`cQe||HU!o1JO?g#ToLQ{e%Rb#^{)W$(uTl=Do
zgG!WsBwfsPDk15$acq0g{PMO{hbQ6dg4v1_@JSDO%_Q0l)c!n$%rX@cyS@@xXFSzQ
zH6;Ww%LUY+y#Fva_XN1ZeFnGr!3-|tzp*e>?Pg<<ei{LA2yu`uAk)HCkZf#_>Qv(~
z+XAI&CJu)im=4ba%N7bvJUl?wp~S*xBwhR7_F-?f4%BD2Def9Fz%A@NtDlSH@0W$J
zec#bBEaIfxKy8eh=fj6n=_D^t&t}M)DBXc11B^^TaeNP!-_8Y<;zB}@FDXF`X$)&;
z2wZ4u8yoPOQU{{wJUmcv_Vo7~Ky5GxnE_TDl^>?@7mMKzXY5k@7zrZ)#sCv|s%3Y<
zYoTW;glc3&XTfj9z4lTQDhlic0td9#w7}kds8aatx8Kkk6=`1^X?1JJ%oN`D=sV!8
z^By175zmtH^kH|e1WmtiuP2X<ok~qjg<K4-Ci;JtkvtFl7ux^-eU$&hqx@gdY4Pu+
z{%=}p#6;u*0F-S;h6QNIK;Ell^3J^?ZB7-Oo1h|#>M{xZ+FJx3u~;bW5tjvDH%<H(
zPV;ek5P;2_1dnMk&?@N5phWZYdT_CZA&asb@|Qq8YDRJA{4n1->I)w=T4J|4#2;kg
z>H$&upFi90gY>;7B!Yu{77k*9X*>t7-IVd~iz)>3Y0Fu)tu|B{P5%1R!KAm!$n3h=
z=^AutJ`nJB2|0+NlhcI7-Umtqyh--mIY~*$-r{@^Zo!1$Xx=&Wm=d1L4sr$<?W^VC
z;c?f*aMGbg)_fLH#m3ySzty=T84zb%`omqO<o-5~QjCnu6;6s#aNh!lM={0K{>961
zjTbVJ2OZIWDr!pPh!h$@vFNEnipnNA>bPA$e`Me3DvFM%&3x;~d?3&hk4NN+3L*H_
zVyOBTO4;Fgv9mseRXqigU`^`9Y{RZrW{hbO!~gax|K_7c>aMP?ND<Ue1KNFzC!MvH
zqlu^1bO*YO`AkOWr_o;mdjjsO)QteIkjThL)A*aPQkCCyjigE0Xh6G2=74QXR!xn<
zz3wy%%Egx0_KT)ero1>*gF+xaJ6e2aqGzHZ{B~G1>mo_>g;b=*!qBkfk$n{snNA{5
zvMPFN9M>=SI<};}!G1bRHhUnmhctVH#JJPa*M2G6(DM1stJZ};tX|LP!j2R(Aj!S`
z25DSwK%*1uNbAKz0J<G44x%UiZT}*V-SR;I?9IGZr;wBiRvZ^o+6c&8lilo{ZcV)F
z#=>w<Z||j|t#0UaK|Lej^E9R9RRWEUJZ7K^*0TKm4iAfU?FOFi2oD3m1EK^zvRhII
z2#}%RyHKfFO@?a3zrHv!Fc61fy43>Pg%AS#>c?c}c68sP9f~A)RXv%prAOj=uy1j7
z{S0}&5*sB`>g0!IA3+}2Cn(<nhlYb1rc}IObS21G%VvJr_EUJ2(5KZ#$hW^<6dtpo
zBmDNHKE;>(&=`0v=<h*wo~OvQlL}oY^~GEvdnK4EMA8dTSqP7QKIl6C(jZV!y$THf
zsK5E(z!eCrS*W%o{^7`=jWV#(6T!O=9;dO^DTBI>EZELS5(3t(H%Ygm0!&ZvE>N!y
zu0PI$R}@?Z&6ctqj03F=AWR0_axwQta#E7k=R0608?4UgIUtz{y)K@0yZCFK8R3}n
zs51auJh$ndUVsL6!7g*PaQ(3V(RbsLnLJ7d)#2Q(Hw5w>6rSJu%GIYnDwpc^&=?+e
zMi=b205u%R(fOl^r50BP?@GS0zP8r4{)5N9Q6{#0L+2zPpLI+4iE9hV*|OOOfz}I5
zU3a<52IQfR&JW!lEg~y@lG%Q3+J^w!Q;np1r_jOP{#r)YQ({oB2|ayf7ZsYUP^%X-
zKY7)-`n((gd{959;|e4<ayy11PeomkD%x7X4nsg%+&LVWWYwL%!3h(<v+yhj&0I=9
zL2E#T9JD#qBjXJ!|LRIaIlndU0F8tFvH?YvbC$`RT=OewVdW#w_bTmA;xAWag0Ftk
z7wpr#awH#{q7L-CT)Fag{~HLx{}n9ef8k9gIj+2jQr*6fx;>sil`e{8`IhL@+Gu}P
zoZ~cwer0l$+o!IsZc>*ur7I*ffye=KFx;G`OIkH3cY%2@nvMGQ#=r`{lEI75eLKzB
z0t+RgK~Ekd-xxS#`Jp)^61dUosRqEqi-(pp)Pa9LZ3zY~A_)Fj9Uo~-8w2HcOfdI6
z(-j1D{WLugni9Z#2KKjaKHsw^6I#m-ZnGE0{bNqqCaNvx<oHV@CSL#i(3jtw3sEor
z{L{-1BHBy;R=9BDvtxg6ZmiA7t*)InIWm<q(Q~EgZxa*K?;?cv?K>_U_L5Z1C;7vt
zzfT9h7pnf`kkH<m>t{SDWWkc;y>G6cNt}O(Pwk$?wLkP+e^ynHmcQ-gxjt3{lr2~?
zPrKBwv=n~4-TTIgd+I?O%)3E&_nG9-LP?lpgFJ!;AmTiLFo&rA#|=C@Mpl8I10NSf
zy70|z5;B%4jzcw-$^wr!A9R|9BRgzFcD9|B7den#bb}2(oc7vNXdQze$NIh+1$Xy6
zSrEK~&{v7<mz^xqz<^<D(9~1SQi(n2x?k_u3M?-Hg6N-vuxONn0oy8chwj>Q>gDjx
zw0n)>cz=Jarvkc^>wp+7n7-V|jk^JluwIclP7RV1654<#*g1pQp35odMdrl6-ZBt8
zd5VGXBO>^P_>fMEoK-iJ7R%4YY9)7C)s%kZsoMU$X@ikfakCtt7QKK>=kU<bn}b+A
z13^YBh>P`G-4JrHLk}fMU^D&t4`jn<IAgS2Szg`*q$+CZY%7y|+jj)nrqjLI<7h7}
zK_e!t!OpB2VVJv^vuiFiyurPT7+SmRh}mouoPZ?=gG}ht6J{yx;4=yZBO?A`<3dkR
z;gyfj>p;PoU9Rc~S|whxi&W~q6T)v;5YJi1`#y=}49@EYFACGde9&zk=i`I0ihS5I
z{+wj}DpCSgaEI0%1$-Yi!N<7neD}v4+*5Of6JWp1aOYVzEG9Vv^6@y$S@pSG|5tSm
z{rU52$U?4Y$bpW(#9bTziT>Lj!EpDQ`eCxDTHG<7l@PTZD84nie+dZ*nFInYbiZW%
z%k3)A?e1Ot<8a+Pb97k%dUN_m#lr7_YiKlo8V|jrsn<U~VWpT_a>m!z`d{vJSg|f5
zcteF&cKC{x$`kW}9-5`u@r?I?3`5_)e-GWvpos(H8@W9OY7z7#N(FAOB`A8%-)-mo
z($LpB+s6)c8;Bn(*hPiT7U{%h&WF4ZdF`|MOy9L0%`(exI9F!2D!&_tC;iX=ZjLJC
ztjx>xi-aGmtbwF;shN(eW^Ge!z=H*ja$RNWo?#8ztKHmv`J5LCeysN_*c(KW6*}m{
z=!!n;70#=xANYi1m*#usAoMdL8GPLp$+weWPWA=I`#t7I=&Xd!7_!FpKk$Z*ok(c6
zBSLpW-1^7Vw6J`U{j>sge)rD-{4%WO2v|3T;VeTZ8SX|mNPv`&pKn<Xz6qThg*SM^
znhSZk(M`M}-xXL+d-(QR|6GOR;5|G{2Y@fu8Q2ftYAd|KM!j`@AMbZZXJ!{Q&41|x
z;z+UB!a*~!KfM5#y5xn^{O0ottvK5t7<<kZ{|cEIK-A*}Hw*g&vMJPmxFQ*fq!_op
zTbj+~_PxMFhck8{^{Y^#Id9CT2XBGAr3h28>cF`LDh^#|b8FgN;h&dO-RE5VJ)?DJ
z9|_jO4u6&N4UMP1ul&KHWc%lTzdv?>^Mn6f-`fYVBxU-OZt&Z~u^5?|#Y~<gx4ukT
zIB@XbjrpxCJ(0&DAvQif@|mFy&>*i~Vw|$Zt{MO?DI?FmclL-#xMSY!P^gbueabp+
zPvfJo=i)n_jifo|>Di?bp;=1<P9?UD9%)+Oqz&>5PfZY2_^q*~uq=5-e=ZXVGE`O5
zTfS_XrJe0{;kCZP@G;%2w*{+~IX1luS?*4z2$z~mxNeZr!6o5Ycuoh!kAhwu9Ubs<
zM*C3pDS5tR-MPGB$Dh-ZG)?$(!j(zm^Rq6rOT-7r6bgO1vT7L3y*@P3)v#vkLwkCf
zt=-BVr_Z`UwbD;l>(*JTxua`IEQj*P?`HI%XZgiQo-J-+tt>rIS;hv3$HvT41)$Gt
zJyP|Nr}V^dkl<6Qb3=yvyA-;r;>CBJj09O3nbWuIv;>nP2K3S04VGj`2SUIuyphwV
zudg5*ft`xf(VsIA=}Yt$CgK1BKAG%$N^#hS7p<T%Z%4BA=JP}ra`fi%VkcwT(lTd6
z->HD(naIBh$Jqy(N|67ipTG|R0$QS@Jw;?nKn1;~LI9w?&0u7LsRBJ7kN}dBlWm|O
zvgY7Tz{dGMbc3XhdNDjf8-WD_;|3qp0LW_M7k??+y5u-8S3D*5XBCG%8w|lrn*lEY
zdW=WiZWQR2lxd2QM^PU(lobrnWRwbQYVR~Yz83hau_G{;HCFH#7vn_(r5`c%&CJY@
z#0Q<?^B|9lGF64hf#n%>bxKbZ%}h=6ce#=r7eT%Tt!-=&t%BquW{|omG5gVXU&F0N
zE6VaiD>c*0Kr}Ww_Yzej(hAgUX*s#spzAp|Qfr6Z9MX2@wXaxd%`$)ltXay-OQFVV
zWxeoofFcKUk#BC<P0QO}91Suv8<ka)gX0^iw3P6B0BU^TSSX;WsRd%)5hHWA<`>Zp
z=vV~v>x_dgJ80n@QGjbu<V;*Z*ZM$L%`X%jM;7Lq(isQsL&s`=Ei?&1+If7^1q()O
zcCp>Z(0nb=@zqJ5l5SPT?Hoz4>TMtWm<SST_tXnh>lnT1tn$cM&MAD6wX}=cYW15J
z_ea9+NrC?>rPYQyT)?y##dt5e=37%Ovou@E7u?*s_tT&WvKjT|`wZ5R448~od(Nx0
z!~=kUGziqB8M7YR_lQr<rX~Cd1qzm`?jsTaIe*+}opNA!8UF}xu4+Is4nmd^IIV88
zBFDl?dT%H-Tlz0@{MfM8b{3qE{B2_gz&7N}Eac_f)T?u2hdDw2pOZ#xgTfw{#B@x9
z69d{9pa?&AWZx(yi*ps9o(}GvMU{(%x!16*8-$}P5Fx<7hzb&T{0U0lE&&J#LcAR)
z<&Y}Dj}6L#y`dknq(h>v302b7v@RuxIzDB@cnJ$W)WVM-><U;Y35K?uWsojR8q2r?
zzvK3L?^&mfD%9zM5((Uh8yHB5sLA!XfIuWPtN8RLlc0qoC&gXsd^_hB?_Y&X0G7Zf
z$ZoVvt->!lb1ZEj|J4i!K&KVB&Q=6hVsQbdg@jOJh&<S9Ghs9TH3oAmU(hTfBm@o>
z#2?MjUXNq3@JTmU;|b<?KX1s64YXA1-9IRS#&%d&a2Oe<T7r!jf{WWM0sQ>=y+9AV
z`4edYJVbu$`8BW~;+wa?$J%`>os-jEZC>Q|f#SEI7XnzvXVNNW-+_MqS6By5sb3ZG
z?$hn(y9m(AJ--GnkiX(dG<cwY4w<IRQZQ581rY%foVA>U2Oh{6C0nMlpqq{l3ctw&
z8VQ=)SKu_b_RVs3G~lw`B(Oy(se)6<kR3SYMr?+LEVZV7BP;%nxbtCkOSNLC;ob)c
zm#9TIpN86E(b4$F3W)YwKTijH@&v%3d{WT+nPk_<WY-4YH(O-XFmgKHTrcb+vf*Gy
z4RyfXXbc5%`A5NCk5V9$nd9VUaQYuYI6Ov6rY-5?t~1op?>EBF8>}FRgnVtN8ygt|
z+o#<ZYAMA4B>mMPsGPUrxI=z@?i}B74`P}PxX|39EYLy(f6+hnN5<^4)G2;94jQ+o
zUwe#5V&}&vPzIofju7b|+Zk;fvlKQ$3SDfF=I{OWNu%!P)F3#!70p|^YRt9yHE`~g
z{RQ5$(EOMis0nj)VrLKk!<(Ml5M5gTvJ?CvHW}pP=p<J*Z$TI|O$DS^_t%#wHm#T+
zekAAYy8`{De|=CE-s2UDcnI|8yVStklnU}o*I!<k5o%PQp@;{o%lR%gxS<Vlte{l<
z>+P!mQY+|1{>4gpHhAx>Qc<(vFMq}vC^@R29DvXbF%jv_K<7xePp{_EI3Y%?xLQa^
z2(r;+glmA!U-u}i96C4!>>x13@*;aCGMJV?<AiaglZiQWysp5#384nS`2b{LMH5K=
zQBUL@+pqHe=flXk@?8MT#lT{*0O&7~;JQc!HtLi`;`Di~kEh%?(dsv|71TGg0(jKP
zOP4gO2p?;b8?$*i#dE-<LX?Ahf4$Q`wT~2vENEzo4T@=~i3=Hi0L+w(y3{oT>NCL`
zaZFQOX1MHKu2&%6A@K+-VihWuoGRk`keg*@(wqZz)9l|`I9iB*CXD|{f{8)C_bw9?
z=uAQ!2ikenP(h*5R#MI*<q4PUZk&QRiOjl#b?WF-1Yeg?e^m!hT7I#-H;qn5%nI`K
zbGijd^}cT4cLiWvc=-{!nrqk*Y#?AU+1a8^LXjGd3@_f^-qOIQ19v;-JDXYphIo$U
z874c0fxWkaG|dUIcQd|T`c27mzHbw1>O<&k%=P9#OI|pXae&mMK%m^ju&HclJ4Znm
z*hSCcW9TivI(dL+*`*DQZvZC~>{+kH5k(`;=KdvuU+5jbfI#t$s>;fNl0)e7;Dv=R
zcTt&ivLFu+>SPzu&+e<?cvOO&q?@7&BXXeA1k7w-0wD>7Rh2v&#alI)7o!>Iw7oXx
z1`Ze9{Iqub587D)+=2SN*$bHYzPwu!@f1)RL_Ias5m61mjKAQ76~g~mc!#NNtyL}1
zGNGgZH}kwPTu2Rsc<eIR^ORy|Ni4~5r>@oyDwKMIV{&!_A9dA+KVm3AMduFtNJ(B4
z&<K5X;ZhL`nb2HAA2ong;b2JxcXOJhqo|4u!i=Ht7cu*2q)u6^mTqU=(@7(MlOOVg
zqn-ng6aB%!SB(rE2mN#P^u8nA^yK7Z4}t~KIY8w;<f+l;J=)DS8T$s%7ZJIQVR}hu
zT3Xr=%m=C5OhLTrF-c41c^k=q_xmp?JQqI}mMoL6+~Ki1u=sY1%%kNO8E>!C#A7|B
zsguS};iJOaa13XHI$2#<>FxVX**`Il%okqG8n`|Grl6-qSGO|_6*&X?wEN{N+V<C>
zM1Ws-o-^=cS`yyRMSrf)*DTLmf(YF@)=*ue=kJ5YOHiR$4E_Q?ecm<Cd+)Ux2&WW_
zPq=RGK5$|=u49_h!VBD%?VDsdNkr&%N?aBMvzKoDj12C?dI`XO_+&}9oq=KpVVM^y
zf@h{TijZIckE75MuQ+Hmj06Y>)mSWQ_(fWJO-2cb2&fAf_$W=-$oP&Ey{)*z(B;-t
z>q(h6@A)7~`EaRQxontQWab*!_ty1i+BPChfxQm{9^D5#0#F%LgAh|-<!cH?Oyq!t
zSB=%m0&c>gc(4bn1I>8AnyWWGK5){VfEj_-Dt9s!xP5DA>f%9#B8c%vVcVC$ydOVM
z;q7~~4G63_dYWlX#8r~N0}1RA`a{&5@23(JWs4$UJ_XY6fIjkzk1mIT%irHWMk~v>
z?4Az4*hnczNb?4d5ex)GtY{=%*}jraV=lZH(g&2~f!T@I(k)71=y;8i;YKR)kDY_=
z-7t7PZsv-YndwSB3Bw7x?$KY5MCa+}wDjfa69WREOiS!bn7t>svVeDi!r3G6OO?{f
z`Y1<Rl~V!&U7o4J)4ta=i!*(T++NfV8Y{fihMloPpyxT!MAmyk>!ap8bYoV{9h<WK
z^nIyftN+;84h&{JpMqEH{_pkpdXM-Sz($XP{SMncYqIZ;+jd`2Tb&@*N7-N5=Q;D9
zneMZi82t48($^uU4k7a45dA(UD-Iny;PIJjjd>);bcvkhKv$g-WN#85acrsIUUtNa
z`Di|_L;g2_;kq9g73uoN)gQ4B<iqnqLM&i&<j`~ysaYmWuRHF)R5QCF%Ap&6y~hV3
z$6i~`s*5J6&WUVJ5s&}ogL-8E&{*I*P!Qr8D(0II_~;!|68fL$H~ueV>i<{yuK&Fb
zoc;QrlW6}9Pg84M;wzb~w^Jd7V1{}&-KPin539o_13{z+&N0X~TY`fC6L^vxx;#>a
ziRCxXO;rwPr8ZlaC|3H$4Z0NqE6~2qu;}SqTbPZ8913L-q-S!gYSOHybe5{=Q@WyN
zagK<S5il}O6gAuZi();$qI%AZx2|YA6Vi=@iPEt)y+<I(()C4rmTHq;`?JR;`;<@C
zq`NnyM?k*yPZfy(6@8T5IvL~FM*rLNO>b}di+E)60#WzGZGC@ewOE(9^E&%_;vxO)
z>v`TM%=*AS1q{==gw1zQEnG60brkt$0_6eo6I}mYHk3ZpLubTenfh9R<+nvomsmg_
zKeUa&D;$n{ouKhT-_9~ZHYjg_PID-i(t(*Z51ja(TIzQGq2Pyp<k)~S@_zWRukPo)
z+F3b3T4+FS!gnP6;^)HngjBQ!W!&}oEx1QSkRgYf#?gKaKr!W2Rm71BM-S!`O8^QL
zZFCr1>Sqr6F>>gGm<;q&UtNk6+>*lsS&*)QKLEn*Apjjf9q#zZ8>xyF+jE<D{WXhF
zV+*_nSjV%st5Dl9A~aFymES8;n90+<!NBg&d$3qYY|W4aGYA9%DGM3k^86mi$;%(7
z4B~JeAKOjf$&YO?Li;!jSfNE-B`ND0=P5=C2l&{SD#-Thm-YW)S_9JVV+Nd$5z)`M
zEjrzhgW6Kyv#F~WnpH_4nI2kfM2Q&4hHdL|6B=pMS-g<~85e<)8%TOf{tmY0XX_Sq
zq4^{^UF_YsmWfktV`F0t>78&I)9*KM*A;VTXj-yVVFm%kttcgC2H*9R>Rn^r<qR9?
zAv91_?a>wHj#~vv5pjn>1bwNI&1ql2csrDzMHbS9R>_*0k@ovW4QkA9%;0gYumQ#B
ztCn`8h*N|k8w`=-j@N1R#gWP+-RDMGLx7p4HY+MC&y=X~`wGaw+JFb#v+!O(tW7NV
z>+W{ZX$=-x?hR_Ir(~!$XfqVeR#G(wI7j(kIwAMpzAaeOfBT02`#S0V8_e#v^GKiF
zbzYzEFM06(HS(DqM#T_fNMP0nKa!JynDhH6vHp_yc%KiIz-FOwia^>{Z&v|zDTgMq
zB3*FuQu@|)(DMdLzC&j(ctQHjc|J5?2k`R^RBBy2n|cIsc{{*8h5`XBjwbN>{4w)3
z-#d|WViWBXg6Y8GA@bS>*kgXwLWHVJC@;soM$b#YWE5!OSpfxJ5e6yTqUv$L?!}%j
zcOk6%aaJ32e(%&1VKniiC(hqOU40BF&d~~4&TX06V}rfX2Zq>>#U?;sVJ_`d(B^ya
z34lw`tFHeublFYq11rLgK%iQI%@-JH5T`U#(ZGTzEEaqKK4Z4Xs|U?Ent?EURYSjb
z>e33hy4&0x@<2RO|3=@=R<V2C&;f9N2`iOXc(BL4#(dVHQXB^*W4mlRl7}v)(*I1z
zppXeq2VPGXqgKr44Ifts%3I`QxK73fdn}1&?s!Gib7t6h6qA<q@Oo@}LUuVb2a@qy
zKhvk*r3!a5=M5Y$f65`cv38fodafs41U?%SBWSL<|K=+nV9#w>$45~c7;1DrW3rfS
zD!M*p0c-_cJK^O=k`srr^zg6=B;Z1jhrFP#%`aXGm0tajNZT0lhA;F!W?x8Pa|`K^
z^D;6sGfB4BuR?ygIBEcVzzcmxxKV?7RJEM!jX`2q9;8Q8Gc$^X##;^lbbTMHnBJU0
z7riWAPgm!CfSPr0b6nn9A5?I)crW-2pmy-^Og6k1-aQV9PinU%m+T*P@3)P=>pM19
zM}I;Q;i?#b$DLO`N4QnPO}!6UEx&itLW21tZ~a$Lu_?{l^ycW|9=DC=g9HaZwFZSj
zuOqJla8s6^{aZHK)X~8q|M5DAn)|325Hp9??zFZ#pVbxh1Yg(8MnO`RUf%Z6gd>$%
zz&$`|S?Db?$>&>;q6|(vd-(X<_s@qfWP(-;1O<p5AgzuDSv}R$J%20=#7cKIiIO+m
zt6!FrGk|b~r>SnIhL%bNsGThI6Yc2$f>3N)Qomb!E2+K$im0Lh=6j@xC(uySvXE@i
zsYwgGKN648gx{G-@S2RZh{`3Z2KpiK4WzrX6)C2u?+X^r7;5a|WF*Y3i+u@2em`A-
z)7=IAL}-$in_75o6>tGy`p{Q*QoN8i_Q`2i4X90>H_*2#$jUHH&$IYmz+VMG<mrR)
z;y9~x%1>{htv^B5*_2bclY<hWh*wuX#K=KyLI}HnGY?^xw^U)-^`3fHBcK~lJ2e_Z
z2g?RiK~oDr6(p8}{~b1hht(|9&;{gRGf<9-8=pbO=dtHcKDO^s0(?3>Vm9o68Wq_h
z#6UoPH-@Y5&}j=!H*<WBC&}LcxC^8gHgv`KIZSVMZv}bjDUc;&wffDFJ~+kNbcK)+
z3I>lcdS81D`EC^xZq$k9K?{tZ*qa3J@p1B^jli<X*3>g-Nk9W&s%kmW$06_QPe?$a
zqfTWk7MP%bK3^E3bBKor=hG}!X=b7BlqU117l88{6@9Q1AvOVj1b&Q}B*!%RCM?PX
zjE{g3kM;8gxRK%tPKfH$g6$3MpYWn+-~bwp2NQKeLAL@sO*f(_<-jb(X?=#uyT67b
zk9_3mt~hN!uRud%-JtkxF)*??Y6K%A??n+dn<XQG<ZkrL=P=V>`5!pEtf7u(6*R!?
z2sA-pau5cjE5lr3)Qy2=T~0uYVLmHK1>OC(!uNB7<)CYNZEgkZTbD>%UETwAoF3EQ
zB!>k5k6yrq`~MAf&%f?1MEmQ4`(TAZHaC~BD@`$lzC>Uw!PGESI{;Z_2*e(Bbs(i?
zY6q~=BsK(Q{$BY0*dX6i^2!2;0$}J;)9i2r5(nP``)HnFId{u9R!b1r)d9^*ae9Ig
zS8cAtX7AIA@Wsgpu~kHc0KlsO4<szaUN`h}uvNtADaz{Ds>Pb>7Pb1P!da&~V}J=#
zD9t&@%{|*dI#S7o3JYDQw1K<&yvxs3|B<`xfuBFMfT^5rFqQ?#u@Kd@=V;Kq-fZYw
zWcwnbfhIgWXij4Q^z{{ZT$uw}<{4+~1jr*W^YBTk9w`-!)hL<6ZiLa+Xn1}RT!;n@
zO~MdQW7w07aVku?nB?~?J{O4Qv+wk(G4<8e5sQdMCk)nPwEW)N3kIJRbvS55BqRF8
z_kO7glX;0S^c=`yG;yXL2C$=7HiJQEoRJ<_EX)^Tx&fKjUz4#0vlTF*@ETG=LL9ds
zxTDQyL-`t|h^G#NNCx!Z!=|Dtk}1`l?=`eh@}trGFwzqyc%xBtIbBDysPhH5NN`YH
z5E4S<CYlp7Z+>fA0LE={Mzg|7z=$=}tqZqtxA{DW*XH{FkN9stHwPZu;!~#r7zcMz
z07j)igC@Aoj-Z*Vf%_e}kUWfL843x3Ath(XE<F(5zZMPMLM5A!kTpmG5bZfP1pvYa
zMySF!VOUBO1Z72|m<AX$)dG{CK$1q}J4|lMgpY-TodP+y7#LSDM*}xc=sGI%7-_1=
z;Q?|2^!CnM+_;SXc?I;ZAwVnnyC3#wK-5bW(;|l+6`KIX+eTe|p=LI1<uz}^OL$FJ
zuzsw-gfR(Di<1rVB6?hjxPz!<C{uw0fTU=KR(dzL@|S4!;b}vR0E(6ZD+=@Q2~Ef!
z$ipKUv=U}hHSUG>mPZ4*`p;A89>#GD{D1t*BUygDT$4y^K~rVW?9HZpg92-CzPo|V
zjL4&TiU-8(Ite6|FTnyMRb&Uc`@<014h|e7v2l9)(Us85VADW?VNHioOFJl`lBjbY
zy1>E!t)Xs<x@Ifm?q2!ua+s9@|E9vp3GjemVfsrGbjBWn2?$YWIvR2??(KgG&OLbu
ze_$;_tT&iKj*jlIhocLLB|x~NCR_NYOFj+^1+<B({8KPYYGow~6A)J6L{a2;{yG0v
z$EqYiPxGDjhXXbZnc<_A`B!j5JT2$<KbzYV?o!IQaps=?QpiaY0HK`mqp&ZVVG0d$
z>6E}AlNL~`SLLCf#2P>(&ihGCv_bcq3FGpoeY5~u&0j`pbXEE(%GV^*s-5{~3%d@i
zQ5TK^b5ldQ^6LK5KQZ^FZ&YWRijpeoFmZIGc$qDswgYJHr6qJA+YMN5r+honxUUtM
z^zJ#9gnVg;z|P%uoz(KRb#EjM=Is4F089V^9$TngLC=1t`<MTxwkr>(a$WniD^0uh
zu}Kj*J4y&!NakIkC`6IDK@uxg<}%f8*kvwMW(k=IA#;-?A=6r0rcSGcES3yQhUEO7
z)jr?({`ju%`mXC-Uza~z<X!Lkyw7vrzx%#__wU9h5$*j@g6B7H1_OwJgV-IkQ2PSS
z?WA{kx*g+ADr*Rpl?Wb$&Q#Xt@nNG_L9^g^_rS69|F}kaQ$p2@a_u$x)m}e#&bH(P
z>57)-oTX?T_lWH2jfoy96LmQS?>=@b+)te79LNt`z+ocVOWj^MQhi&y6pB;@oE15$
z#~eK{#G~CxqbYLoLtxyST}D&gzb)j6i0R6b!3-$F#S_X0X<s&ko<FH!kTcw*wea;{
z@kEe&lZr!yj5p+g0=;YIddSR+I<h^GlOX5>LIg>S)&p{(%!a#l!~hXq0%&?il8CMY
z8N30G?n~rv2(^i@;K;EBi4~8#n{a4Bm>8`ExgmDW^}FBypw>5Ih@!(Z5qY^T;$@DY
zr4|Vl^>5P*B6(nx5ymCLB3jyoAK3wuTX8P_DO4;}2XIC<Es?_`a$@JnEE5D%Zv>K*
zV?0l$lJwD7Ilamy+o{*qNxon08Cl(2hM3Za$M%j^bBj|NoO~Vpx+q%RgG6b7ECCU9
z(wuDxe+i9bQ6O;0=JF9GRKJC}zVQ42PBvEq0rX0<Q8)J!g&GRkb}hPx!zXTw>~<Q^
zyl^OdLcHqyhr&PmOF3lalXrZ4>kW#F8>&k)=v?yawL^wVP87{aOdy|e&}(9XnNu9j
z?Y_@{C`Umk?KLEV7D`Q<t58{JK$iv+IE2Flpc^q6J#}ZU{jkw(^1sM$K%YFBZR?#1
zp3og5NDGb=cE~sIsgAHeeUc<noW1?b&~!7#v^CL=D+4|GYoL+eNs|UaY66O&OcZL0
zyW83he<ONgdR2xp&iV49nY<7&?{KT!c%j=#GCSbOV9BGZBKt0PEE3Ui;ywvcuFMDr
zFQk+E6i_@aO^`lNN1J?;6Q<*>Sp3&U{Ma(J%cwQDNTGL^Q9oG@8Lh&XS4jX!=GZ{)
zn@*Wk#aW#V?EkOGGdqc&OpWk7=Az9VTmwxSE}pJ2ZT9bMur<r2LTFqm-@CA(3ZN0(
z3h-s)+a;2SdewbIErb`*nD_TrmNI=`ECneE`*o^bs&ibs%p~&|zgnmbO(mv{6pOle
zAO=A+fFKobIHk#*NK9*!IvucT=ytqR5d)c~qh3oh6Xq1l$C?r*Xu<2{Y=udQmv9Zc
zn4H2R+0M3jq{VU!A&gccf?YuD>#06%aVMW%UN#0GTks&i3^XH>Db?2I{QYB@RO1Ym
z&bjCA#sbH9LN<a50`P0;P?^~1U6t`~(HDfTyOjwOc#0<jUncQ0MUxJ4_({ygA=*}8
zRdnr?bJ*-OQ%!$u&eE0|QtHnP?<&!E6}evj&bahT&e`Vqb16V>^3a61aYH`*7U74%
zL4hA!ibs<ok7kTV6S4^!&<u6S@stm*I!{z$a;g5(z2Iin#W*Ai5>yd6b17AztMZ$T
zFZ*f31W81|EHf|(CB(|5Y|HD5b0XD5{e^J5AjuZc>mC4svX?~GDmNgPOo<_rZCoJq
z2GMC=UNQ+tp8Da~45xwQ0iNZX1~XBMC1Tjz`9lDbT)8ih;F8W6@WusfkUV+#R|Wgm
zQL!|&PNpVeEh4zV%M0ChtL^B-ASzj09yem?@~{w2rlYnt=w1ZZgzjCdl3LD}mu4Ai
zClgw1s7^Nu7>|5Rpd4y??&y`OBXhTiUI-&i&fn1{|Hlk^bh@TXmy>U(cS`6+$GE|%
zrU~cK`OjhRq3RY+UM}$$5Oc5Nj$R_&4ygNXq*lr!vPg_ZKZr2!6-QR-f`xD}`;60t
z>efLM^)HSdjNIT0tXc@pc?=iRZO0&eDo&^o)1)NG%S%L9E3rP|xZ_jVKP+rlxGNL$
zYG+OH0MD%V$jbA9Hw@Ii5%pYjKYrcp0-rtfhcXFBh$V#B6Jyf~kPHbid58{vs9$d%
zF<89KJU^#oD#jp0!mncA=9ir=?j|y5gqfyWW}O#MsD$F>G^pDa^HR^XF_`4&#oxz?
z{-VK<e_s7|lIP(qkla5Dd0k#29mr8ZeC*n0pj?ZQO8ms|RAQvoM$b+nIgGA2IY928
zoC|2%+nt0d_LZI=%5=Z&83yR$in4)bSqG*a#u-g(L!`U?5v`!z?#;#%-6b7ULZiAH
zm0OFIy40FWtG+x}y9~G|MkF1l`;}`1iGrcBZw?}05@Nl?keT%c2G!~+7G#xIl+b7N
zQx1%Er{8jRw0UDC=K@|Q8Hqv$auKO}2H0TSDOR8*2$N_P@C~Dz2+kl{Zzbpb67BnU
zBnG-nSoyl#yGT#G<h{I<K%<e8u9NUH9XF4oGtL$K;yNgiZUT^sd+Qd48zgK;g-l1n
zZG5>P_2gB*vbwr@^25yyOwH1QJW%Sm$OD~^62fl(16jh!jB3`9>$m3Xr}U&S!V4o}
zcjv(?-++1!$*(<z5-S4}XiNwwT&p?#eiwvvcUrLtBSpE#<8UQuefV6iWC15=`Dd%D
zaSN(ML{?3}eV{26aX3hxEnQt3y}MoaS!2u%7mjP<OxFPc6eTPyk2^amPmO|AEZ#J#
zE27-0*fNsP(qX4nnKDgu?1oCkk-`xDNpt!*svk|!4#x}Y?yE(Nxc;@PkMP;7pLiF1
z0QeJdm`D$RbL`rz@rA1y1xrbq5C|`aNUxEZ3YfG&%Q*VvbWB-@{bojG?qA#6`#3Q)
z*~j>z(VrR*aYuPDILP-lEUDW&bMM^d`8D>|^6h-{OJ%c}Y@8q0o(1Y?@0I&m3tC|{
ze_K|!t|mH8@br#*Yqy-S-QO5>q&4R^#|QkLLKPut{i(Y`Bk!oF7_MK!SJb{Wx^=Nc
zda$HjFLhVyFCn{(4W*=lqEEhGW_D<L{%N>KXqW78VSLX3Bj0AqT63=7U30*%cc3U~
zqCt+vT)k>lRD666IwaH5)1~F*TOL$U$vmsV!a_yA>2p;`;u=cli<MeUR7puoE1i1v
zx2oBpY+CcXQffZl<|klDnp!#%#DxL2#>UAIXvoC43L!PG90PmLr4F?%EoU?Nl!j(6
z+^s*=3c6F8ZIzEB4)W_)uMV9!k$hllKD9o&;&6BD>&`JR`s1hPhgEuw<FcEcCmzS(
z63Z%|;&QK%EW{^AfD@d$N^<p4g~PllU_LWJK|#w>mxQnL^ENitdnrxD84`|}--0R>
zPWIJ_bXu+tZ55cCp5CUeZjH2|l#EPAOUl0D3lpay2xNd}?Y1YsFDVAyRzU7<uB(ev
z-=(RpndyX^%PV|8oQ<p(*V3vnlXnbg6e)i8?3vG9NoncGBME;#^c0U(C}E9^<ZjaN
z8N{YNnxyF;m6Vh}IXTHg&&<j)*|X<zYg-!!%8M+_J;|I{YeNay3j$|OKCC&*+Pgi&
zI3g7@JuLSL=$q*42kGnUKSFFSkdWVI;p!o7#kwj3W%*xNdZnk?sds25^4_J7m2OM3
z9^Yc{*tvcjV`o}>Zu0UL2-ZO(-W_p&c$rlzR~kcp_al?JC9*yhgb%~~%JCp@lp-P`
ztgS5>WNFgfPG79;5iA&GxSY)m)^!fX3P?xRdl=bJUMv3C$jq>-POe@&QaT}jnSMnB
z(a|c%==E0l7TY%nkYO1SsW)%_O-Dz^@$J3cJB^-<zFJvR;qhs+Ub)|lTa~+xZMlbD
z1V2Cjsfp_NVw=p_%!FpGh033Q{y8cp1|tz~sNT(X-1TL&FaPP4<t*5+TyI6J%kv&B
z2Qmt8sJpuURAIHOoZL}7hqzv6XJ<qzxN{wr>HVjlgM!4*$0w#|S#a{&4@;If&l;|A
z?O*8O9nRz{S+#bp+3C{{@@%Vy?wPzx-*S>2=1P$|N_S0px#s0RLKr=KWkc%2?50Ew
z=}aGXUF_6!O`_kp-WCZ77G8g>GIdjZ;^{MFecW~bS|7vV%wS!o;L+}~6tsWL&$9AW
z&wsv`I>d0F9BlCcm8m`5C`py>PTNZp_J4zRzI*dtnm<==dehKwVDaL`v8tX~D01IN
zx|vcJCi1Sp?)n|MW3qZQ3Bf4<<fjtqs&3DyAW_XsPiItCS}$9+YzP}O*FH>GWwV};
zsz*lSoEy!H(yg(%T-IATehJTyZ|~~_3kmDdg&6Ole#q{rU?GbqB`a(8h22BRMT><`
z^Y}uCxrXbc$N`6izPRVTswc85E7kSfI*r2PMclnZtqzx!hiz1H`IW~(WHd1+t@$}S
z@AOsUy@8iwn}sUOv>C##GDG#kJ;z+Vb%*C3ul8k~q)>yUrKLHLtfUMyC7=d=xA?~;
z%hw+9@OXp+A!Kk`mGKLuX;(i+ioFd`6J2S$Kg(tKGPJnd`wuoo%kM2KE2BxMZ&LF*
z7@>$MGohkdwh(&B?U>$$$Cs0n!(RFXxb&8&acQ_d%E>wC=;#=cHTtZ~r^3YZxw%5K
zDJm5SNp*E~<W&vz)zrY(b)R|OY03DEc;gI6oA1g!vuCG9tXLDV6*sGVIE77Xr0P}o
zRLaiG%*1N=mK@v~X>$0m^EcU&bNvfdDzPRKXN}Fx&lvQV4ej30q)F>maxq}OR(1LK
zIGN*4``6J@npmBvb<wuAw$`(ESk+_jrZ}-_A-$fL9@qx^s5(p@RsX$ZR+D{C(z8IR
zq}J5b^mRi+8(vGX|D4zDyLS&{<w^5N7`DHEdd2x$?W2#=*K<@D(bOF3Z*wEoi8EHK
z6<>9AS)Dj>!t&wq-*-B{OTD&weHOK^8aDAQgr%)cJsS<=lfX3(zyuyfRiaqF(v2Ae
z{n6B}QYz*1`=_3oZ*fN&L4koS7)$y7;Rr&y8mO{e8K0QQW$o|qUez40k^)D?n*7w}
z)K^`l8zy|Axg5KRd}`sTz{WKWl;A7kI@;P9c6DLo88Ip3ctTgv{rh`-zpy(H7s}{;
zIT+g;9m|(fWca8l$nLOGx?xmQ)HX>;J&?J|29hmdf?$Gupn$N&v~SJY6)Vn}$BZyT
zqJ8(koLhKy_b~pTWvkSw4G#=YJXBMNYovXmf?0a+aJr$zK68uh0wdjJhmocT6^8)5
zt(%2PPoP4cPS5Yzv(oIQ8~&nV6epXrq0VDrB5I1>Y-{JvLY)LPZyI_AGGQ|eSFT(s
zq~expQ*X16;uIg(Mb4pYepXhN)Y0`nygWBQ>nWn?XD(qF{}y6;3Dq@>x^N1Rk7od_
z$iYYk-Lg%NV&KZAy?CL0?b<a*WX9HZvXP$?az2H9LnaVfYK3cF{8s77-XbTL&ofY@
zckp0NaR4P~Z|LRAKVW<GOz;I&Kue@CkI%s%MAd6JBPxnNj$U+K;lr5%bN-DRDgMVS
zEHY!aa@c&QtgYj9LpGf7?~7M8wzkeezO*bkKMO_`gZ0{(S`=50HyT^89Ya*vD_z~A
z9+WK4uySuscxZGW{(6svt7`$UND2(P&s~%u4q00ZuS{5=ckgU#q2Fgexjnzm+u=oZ
zcy?J<QhV&X4>+iU*h=VB%y7SXO<GB*DQu7&)1T}TPbg?<Y3b<dO6}NDs57-*K;Xg0
zkEiD0uE_T5$UcfMxlj=`GXyn`<2K9(XX$5mQjk2Fg<AO}wwbh?oRPh~eK|PN9Tncx
zHKLl<d#|r`o^)6m+8B43K7WVCe_aT6P~XrHZDp}ju=!uV`gyXuEw1paX}!9V&yuN=
zRP@2TZ()L6qwFEt*sJ=F(b0VLl1qw-?cEn9Y;0?r2;YT(^#O!ZLWPv{Nnd;R;BTtZ
z{Sm%(xta02RHGYlVoIeNzxBkrv6os#fGUHP@%a3H_ef90Q3<2OBj5)S;PCL_Lmtss
zMtyACX`wO8_D<_gSh)7m2d+9=nM%dA3nG$zmYsYK`zF`yN{qkG`<;=vXwmUQZ@#Og
zq9DSfbD7KXz_&jJE&bjGuxQZ+{*~NDF8uV@Z9o6}@5H~4VMcT5z5U-yB`ACEzV)3T
zfd71D^Z)I_Pj;hbDIqE<3NNuuLZYst76qeA%g?ZexSUG%{j4ngb?esU$&w_j`U<gs
ze{3aAK$f-53jCb>yEMPyfBD32GcT_aKR>^Z3`PnvfBPd06V)@(Ya_jF+kO!>ul85B
z#EM-Rx~67ky+tm1dVwgfCx8BYF6Q9YYgexhB0JR%Mlh?V50Y3pa}D+Njt|dbZ-B8I
zdD6~q9cEr3Y+bf=>6x$7W3wNE<#L$w`9VReaI}nqLPJdt9lGh~@6SPrz=(J`jE;Q-
zw1H8AZB6^}xKA>oA*{S;kqi9L0c@TKapu4PW#QX=EGmx)4PhIV4dF7zr>4vd4a5HU
zQE+o`tqG+})~0-LdW_+Nj2Fe<YiihH84r(s1Z>Cy5QjU;@{a|~=oAzb01rAp{HpBI
zeUl)Mn)74#dDaSNz_antH){Gl287#aP1ctEJ9k%nUV*`Z*~(?odN9FsP!`e<3OJ!|
z<hj**OJ+vKeq&=}18(wI>BO_Ex{(zNa0_X#UcJJ%xXS14+>P;`_l~TMR8eORX1xxf
zKQk*Q=aFF~Tp-7O3`B2!5s}>1R+CzIJme$F-3P*?q@<z}5^~`u%_BS}2aS<5Jmz)x
z7aT!q73U7e|KPz8@??dDJMfBzrbfE(2d#-3zKR#dKl;*Swr@XzUQThz?b{E=#Ke>b
z%$Ebi=k_rX+CZVrze?;nmic`^G`CCMg~gTx{2084Kr0hj_ToEp;|+33zEfveb3GN_
z6ri8tJ8G^UuI|~hN8B|0MFIjJjU?rrI?q$}A}0Zb4@3a(DN=jY<7yh}BPED+&}#*U
zj0+1%G4<TB5ONSM?VBd(_#j)Kn1h^rG8{-XvLXVwN)!zcdJ&VT0Q_0JfNHQsWJS2f
zR5WT86#A@&&*V)EV;28}UFnrey0X4}KbxCQKmdpn;N|_nyERphBRjxkronXRwIcaz
zNB@QxCntZPF-`)gqk;woUk6AUTlBCAR{iu7{q=2W&*{%6aF{xaWpK^MKXH%q=M@0~
z0h|Cch2{C;F^&B^r3@BJT93Ri$^!CEh1Gd6{ZDOijn|Eh9q_WZZ{JSCDR3Hmo4g~g
z5nq-VpsF!PPVJl+_XLI?XWg*y-A8*j#NKyI(Loe*aBoCJMw;5%<^ijvrKLGSd$rJf
zyAL|u+jFgy$pD+MZ98yFKVIV6owq-r(6yMw0=7`}^+g$f708E0&6?TpU^W$A2IzUe
zD{hdaX>;Mig~nbYfnce{eU{%=5B647a5#9>QNb403beYw1;umbfzgNGj-XfMp67SF
z^k@+dY!!xDMWnuYa}YrmEss-FSojF&gop1r?vX}^uQZU}M(*m`_q*DIAXtpWg8YIL
z8bM|!nzc-gjep7XN$f?r|I^##y*}u_^_b{4V2xwyOk0L=3O`y>$O*vx&`yowyQs*>
zlvl6zv&QG=zEG2?gl==x=`T~4xEXSKV`F0w;jR{!+_FVSQc{v5n-m>w&Z6!M-(-5^
z$dM~^V-pi?F9P}Ej+*07H~hdghnEES*RM}UWQ!znQyvG-H<^U$a8_$%>BW?z_6`op
z+_t#xC0tv6@0vlPx&>e<A~F-8+&8C-DyypOcgC#~6wI8SKF?$_>rd^lppQW#5ZSTh
z`}a3HcRx0kmzN*vE-Qk~OSyNin!WSIqD6~W>uK*k@^^!n9Xc2`L4!SACoDYtjEYxs
z>CQ(}=3w0yt`#S5Xz715-T#rtz{UQ5T+x4YVa~;iywZ!51)L1XW%P6mwe$8Izx+=)
C0ob?z

diff --git a/reference/plot-12.png b/reference/plot-12.png
deleted file mode 100644
index 58ca2e11305c2794b9cfc5398123c8a06b317b8e..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 80659
zcmce;2Q=1w95;Mf$u2vIl8{X@LrI~mLPln?_ujLRgpj?-ULkuYB(g`@BQASn@Av!f
zzVG)u=e*~Ap7TEEJ<omICx`3u|NnpE`~9rn6{sjLMTAd-k3yk{WTYjZqEI-_C=}L<
z^XK3vqfx0F@XxsykEJBxU-<9fevKFWgJ&(RZihl$2|)hG2%XzBK%uUpWF+q?J14A+
zyS%5}aXH&m==NAC7$1D3q3(?>L0ZCo;r+|4TRtW4<Ot7Qs!8}VSLyzdmaN7+_k7r@
z?wbxO${1_G5G~p{MILFNpX9Ag#0;1(Mb2BdpOeH>r>)KyOuAKSx3j$-`h%~ouh7x3
zXf4U5AvfF9kyx~Jbn?Ja{A7FUHhV`2_TT$Np*9k|M=}2W9+l~j&Gqm1`{a1q|GnfU
zY3;w?PAM@u|NH93Ea?r*e}3$(5H^ML?<<lSnh5^=R_uW<)xY1mzy06*!+vALsIt6@
zis97@Q`dWqm4kzW!#~P0rOPtS+Df9B`iF)x?_RyyQDT3*Z&T5X!5!#)r|>23O_@g-
zPrn<Ma!O(A@@~!lh|r^EjeNAUylf{zi;9qDBz?q48vb#5T5rx)DEhi@^5=>Qg@ZEX
z{{zqSzdKOPCF~umR~6%Js}e@9HA-Rncz>Dx#MsU!M-lWO`Z^|etTZD-GdJUABFX+e
zZfa!~@^D7@O>E{OGD>8@4{yjs8Qe_nuAwFJZhtLt4d>N8nWw3WiaE-*LJa#d2O?gq
z61S*vxgQ=pCd!r+l2M*8Jj1B%7P7ol>b-}9>0?uY;Y{Dw<}EH58j9u97;}beNQS4~
z8Rz4JQE}fpR5Ja^ZGQeo;U&NFd1K#oT*Qcxef~TiZj!xYMudLD=if>B{*w3qF)5v8
z-Xx5qcsEG#WYX<8oT^f3R##V*JUOD*{*SJ+|IK;Hc|VGg;^^0`hl~RDt0tb&6>;wd
zZbOW90_=&xRoz0I<0fuGw|E``XL@v*b&#HV)STmme>a8q(vzkj_5R7pb;t7?VbY9g
z$ijsIqPcX*Mu-6wVPa|d?COR02I@4_)Yiztee*E=@4|)kugQu2yZ%Q;SY=>o)6mk|
z3(=q=0^4Taw!8@cXfIsgF1Btfxe@Wl=~VgsR7*aOTtBjfGM<`r@*j~*alKP-JW3E)
z;~B<0)m4ug&-D5C772p#{~h@MUv~Ai3ao4+EG(>%D!VLg=V_ePk+R1pCnu(smP1Q}
zc@B#i(WWhLDah(~V~vt}Eh@$xzwe#xRx}jMbwnrs5t0g}8=MUlFT5h+xUkThzO=BQ
zsGtx!RHReX)zx*=gY(~kw0nhJ^w{6ue|t03>-b|by@;di^9swsp@QcQO*mA>v#r7S
z=Uy?3c9*cnX1aX$YA(rF65HKqCRuJ1Ki!N;UY~5xArakvPv^RDP4v&V>-c^@AN>25
zix}L{q|Mv|{r%}dT=i;?qd8R`KMsP0{pX-I$?Md`V&t<-x9DC1o%_Z;ExRAL$?Cf&
zPgf_GPS(A=2G35TIG%4eVv(?r<1Ys39H(`OpSicZ;DL>K<&LvrsOaubdoj9yuH-Fr
z*LU~^RJZDB|I6}+E_^g9@T4z_V=U_r3@s*V)#*<bq}mE*B0t_uIpqIdS6Amc<%7#8
zGkJEpH#j(GAWlgfQ1-H&RLgC(Or_!9BJN$K(jSy>F<Gy5c949QGcloEaX)9IW@3j^
zCcS{%KOn$pyRO3g7lX{mC%xvF-~SD#SI#2-gfeT!(OQkd-ndh8bxlo~Uh_FE=V?Dv
zb8~x<&oYsZZn%li1b6*B(um4nPIiA9{@L^7@RFO(e-l!aNmjc&Tr_Nwm6*w#<Sc<q
z#}%isYcd*Trjdn}Gx!BsHP4FG@@Mvr=D56aybhwha)d*i>bDZB59Z_ZTz}*h)w9Yh
zt*t$~LzleaqHt&PwdZD#j`7*)iK&T6Z>~z-cG6z06gt5!gs@p~@!-{~SKrZF;U5bN
zIaep@o<A=)@9jzyiIin3GyFlL<#{ltQEhLcviWbnigCN0I7NhwsZsA<6~(1f*zK`q
zH&SL6P3LuNgNzne!*K~?PF9vww3cJ+vqDWFz2s5ej*^#ghJ!ZbK`psS$=&EOlYS<d
zKC1c+KP54z?j+IO-QIL5hwX(YqB{dh8g;H^^}A(7H%SBExtumOdNbzK)oCe+9eiA?
zUgkV1Hs{YPcwS+;JX8Q9|4M8#P$A*Zq~}Hx?#_r=QSi}ExS1VzKJ<s8fHQ+q^(%tb
zDKqGX*(>Xcg0ysdJm~!(Nd_7kn!e51Ha6EuUGD6Dd69w{LE<aI&>QXgx@R{ltE;Os
z2S#bKrRE3%{q?d_Q=8_DQF{iwm#&Dq4v1;5tgRL9%jY*WHN72ud&xibR+nJ<Ol1hI
zP?|F9jb`o@p+aTW_LBUFq<Cq@&$>;%Zzqp_2u$Yht5uYhsaRN8P&5ck`9ID1ct3#n
zX8ezkBj+;x-@yVC+55D%6SXcdN0+ur1LsF1Qd24>T=QlGDLFKhZEX5`dftaz5#gVn
z4V0A7($boxB*(C1>vqZFji0S7EmeY_hQk0e-8qYVXeWNOyP78$V%k=05@%{*QLuk9
zZ3Xk@fbS=PjPJ7|ZK!cYi6Tx*#cg!`yQ`yR9!HMQ*y{G`)&~!FmdorH`)ru=9&^<x
zM{Bz-=C`V|MsEMG@>GKQQ+@oW!(sQAQRq>vBtvS$QB6aCwZrm2UmrBl5A82M>$ee)
zI}B?bHq)P`Jh!S-xh!BwyH%~p$t(jPqF>8tJcF&M_9^}8M!;ZCPl_gImc?lK;#$r6
z(Dl$eX}8UWjG@kvwH`K{9W}Vl#~QVIbaE4VOkvT-%vfP~hIP!@Z!h$|P4+rm%u*1%
zMnpukxHGI1%~)1e20dE!cw*Yx`Aacx()9E+UPK$9o*+t&-2FP#yWIH@>Xc_E(<ENh
zc2BFEH)r0qwIXXEX%9*gEL3$H%^c<2nLSGji^0XLWbI8idU4M&vHApA-q_zYm`Qs>
z&#e-9ZAPFH40?`@j^;_ED)!o4zI<81b$z_X=`oFHvV-=|n~{%Ht^}xVc?=8>3oN=t
z6I@4eEv~N*Pt>_FdrX*z%-+44a(e11lkOxZaQn90zGgq4kpt~I>^rZ+EHB}73-M_!
zhlnYMw~`FPXUAI(rw3@Zi1r0PD(#+b&)r)mGgJ*dT%MZ*iEe9E)r*<&^LIEqN<PDW
zs@tClT3TG}%@RLZ#7nNR9IuY>jFM&Aq4L_K(kM5N72B)oUKD6J2qvZ%yJI-Rp(+#g
z=mt`j4u3O?n}|~ewb(;LVNH}@K0R6=uG^c)6IrjdJ12gvc5P$28M>$<Z>+MB5jsI0
zyWwQjY6nV>diZv1xBJ6&ud^fc%A_N_<|^RVY>gb{xX&-nd+)he_^c_o&yYcNSz21^
z(+?G$cN8p#@guF>NN8%hS8=}w-s|SES2G#O=&}5K($-)#y*Yq{ExMz<gdJu!Cp}%i
zOTTd}uLHYCiG}<rZ!E2yiUF$yb!JBi`?%w%g)1yWq)jaxE*B;L;ouq{AJ?d`Ohmd%
z-QiN6K9e>YT|4uZt<~eCg-R#?XuUzls{SAo-|x%o3Uc?;t>j^yvmKq#BQ-!=%=Ma2
z65;Bjs+R$!eXP)o&&+{Ns(O4{rD^$vExBunzMn5rnlaaP-E_C2t4qGDqGI$+hn;c<
zI^mN(V+XDyPQ}>P)`vKl7)MhVvyJ9%6MyU#+U~s7{lO%TqWfmA@#h(r?vni6`P|)I
z$FlNrQ`n><wJxQMi`VH7-nhED%DY-}Q|B+$<(Y($69+JRtV4w`v9?z9lepIr%_W{Q
zPR$+vD5EdtU?sn_OL#GX{la<W+YfV1vPc6Z<qsyh+5ow$w0&=RQ9M(5J?vZMx<y>(
zIK!ughGcz1XZfld8GIvMx@lX#e$lpd@kdo@u6;O@FPp9Gn|-UyYEa?1u)M6#xf3=&
zKYz2KoRh3J(_|@9RlC>gWZ3H$$2r30o*U7cX;2_MgqQP6TRq0jIZB?%uUU?k%Rw>j
zZw<bjHI4;!Bf}LI?I0}wd30BzK^mQL`^&zGF7y@|lQZ*C_}0jvm8|8<mtQA{&%L4;
z_ka51rC!|pYbr%}FqzT8TBnSPPU&<i#{`D3H@;0Nvi-{Z@hW1;Sy|MI8K$jQ6@wD$
z#^dATDd35P)`N}8=j)(^!<RX~S;c(-fMlm`@JGs+gCpvk_~XmGqOV`S-q{wTl%xE8
zO>TT(fNy%-+%h<fj6QF^?lFwz>}tF0_5>8qSlNdyFP=0vH-Fah$@={HGYc&-Tl|mQ
zs;Ywz%j`df1RW_kBOhhRoF&EzCp+5CTl2;ewk<lOS(4UfJe4V`+l}+w$<tXxs@7`z
z>X=>kAcd-C>R_H~a>+p%pGZ)?P(+Q{aN(e8{hvq~Dz`kzgQ=6ROeb@vX}H9+F&oe9
z(RlcE8cP8N>boz=@%`i<`3_a9v1i%~b99sdC8_+5hsu7t^*d)`z9;tg@wkn|+p76H
zBri65Hi12S=OzRaK*{8f=HuhzcD(Iq@h*_^o#blHV!A>{$u{hc<#y)e0WyU<dXvI{
zT6ncw<|2`uQdg~a!_YPCHju#Of{G=mdqZ)z&Fp7`1-hhd_h93@u2rEKO~(AP_|1Q&
zKt)gj<Q4O*K*OPLb}o0`pd-r0_o)=VpRZQ2VrbxrF<&m+u*r5V8xY{Tw-czgnb~L>
zza}(@kk-bSUAb}vu<J9{Na49J(mVOkvl0^%Edwy7$YG)-Qs2LANgEG}7I|T+BWJYk
zdC;M=1YIGD@kg?_j;fYJ<mpQH>B`64X99u|HF>H<LvjM+$~Lx6us=w|k3Vr%{U+s5
z`_w4k3QhlH&8eY(wPKRf<g>8L^cO$oy3@2E&OVR5@vPO+ie(>Mx}E^B?Q1QwHWc_c
zqDL#GhrOYmj57L^y7yP@CLTQv#hZwVqJ&>RNPl8^8w;iUBlaC*lZZptjIYu-4sPeU
zW^VG}hs%G|sP9??N;0rg29>ZsZg~-=A0e#Rk2e97&8)EMA(fVW82Z`%9qTFYB+o+=
znLcQz9|4@XenC&zv|Z7tbuK<Ri!xY*lAYsu*q<!FA6MEf_8jS#d&NL$<l^}ktCtJ=
z=YPbnw(uv2?G}eFCfIf5Y|o~g6C|tMiJ6jRiqJn)z(Oe!8S#dTq-2I5N<lwWP|HgB
ztzYI**j87v7t_0)8ji*-0~EvBU5}!y;<?NK1d@2|-#?ibXy~=;7R?&g@zVNb{tPgM
zFn~h)%@!)Nx$lu#>`gkA)~PTBVkgIF@!@id(a%<sVo<l1YPY)jamDwWc<>AcaE+|!
zaw$=$Y+?=?L;o!&DbOvvb8M3O(Y8XcGOMYH<|X8qo$K;zYMzIR9WH#(nH(I<$SbUV
z1amt0Ve&d%=<-rtk(j{__mWhVRGIRLX#)y1Py;0r-kz(*{<KMih`r<-s&TuxN8@CA
zfz113MZ-2xbIPCdC~sDH&C@Hlu%Lw#Hpgre3!qTkJVw1yrB|+-sy42oP?g<$az~7;
zq=BElFriS^a#e4mDTz=hSvf}1Cs!{3B+T^AlUHmaxEp1V1L*9)J?XRtg=$o3(nYFn
zlmRp%&O2BrZXC8}`HENXY8ll*kGq?yC=IO!mM#jF0t+QJvt8yU41D7q4V{04osD-P
zA;cB;QK)@s*2r)Jbs&}iy~YlFswMx`b(C95*a1;y^qatl_MyoNg<Q9@`xZYlnjHL@
z)2F1yTEh89M7pcBw7ROIqmvxo{&8e<R6R5eKY;iO0$*<gb1nevnz5D4R;|*M4}VF0
zj*3RH!uT5oiWue-pqjAzK6joI7K~>luv&mx_4Q&QC%GSFnO0YAcNn>cVatg+ruVX}
zUQ(>c2BvExzu)*WK~>{bQ;OM9_jGfB!^%${nWt`UB1b&d_;Xms&D_*%-Xz$s?zxGO
z<MUWz$UxcbTX-P}!b6hwhi_*3{G@>k1r-zFd>V|?W&4Wsz?4>3U-B*<dM+fGHBWz9
z-c0dCIQ8@{lXA1r#dDY#!@J4;{^vxR%;K1ki*(`Rwc<~CbbXDCx8s<ih?Q+EJT@~o
zSH45%B$TGKH4+p2F#T=YED-%COc5}LQ@hTcRRF2nYKn_{cWh=7%;4t#yUHRWTVd}@
zLno*p@bfi03N?w_b@cKQC*!ro2Plq@jakATwC+Ao_M$i&05_mBMHasKw2WwSd>l8C
z<nCLRPxcZ5Q>7Va4tS<p_JLGBeE5)22A70RSzkZB<6`Q>hoifXpMn<Tksyd3G3Q@)
zy|WR$ySsa{VW?&=8Rbooz*acF<wA8G1GWFa%d25`<yShuMKfdLZ>!cQ<!`?P_`{?t
zsRO-)ysWmT{Vsne$d^f{qV;w~1VGvZJZ*FnFfX!VkgzVGWHIn27OyuER79?0-zL=W
ztfG~NaymwgQTm3G=x$;MdJ8|GIiLT~>%W4^Kv9eGehbAyTRR1>aWtXL6%cXig{)Sk
zC#U1UrGwwS`&Po<#l~#}vgUIJYk3hd=ohV*tytoJuwFQC3~SLc;I;Y}_HzSOqHeoC
zg8~yBhSV(sX7R$LC81qH>w~g~<%w#%DMVJApedhhpw26&1bW$p^Umd2JjWan?LSAA
z*X0xSsF6vdql*Gb=F+9<y+EPa+S%|jrB)+qe-H#<ArdxEF~}07e??{MXPS%zz1_TV
z#(xO1V&+>%qJDupck;Dr92W;wbyTdZibkhX9>N6yS+qzWcgOZPoND|0dd5z<8sxn_
znW%PfJXFFya$$)3d!M)PysWlu%a+Bo^-lzaC|^UNSX*)W7wwqZWn%Qpd%T?H86#u>
zwA~;L<Z0_|G>56_`5wu(S|i3?4;>gdoyqob)JtazP-h2tjC`a0x2^1$YH~of)wK*`
zY|~r-s?8M9`|aC#KjQum!1-+F9<_36-e*fzvh!?%c69sp?J8;_LjpYQln|kaXiW>R
z=l_+Td0nK2oP$0CUr-6Eue@Ck`-xNgBX%m|!TnW+re%hX5<YJ3lwoZ*IT6Q|o+7Uk
zGpHMg1|TruJbP0|1b&GSIfNn<YSj!(P8QA28y0BSX#jJ}d6uunv)E7voa|&twV`iq
zye3jUYw*{voUJYq1%#qgva4}N&{NPo2=d<q3{jv__I;K6uM`nOY4q417W50o7&%5~
zt?lEj?&W2jvqKqiuE<AO?&H8m#ZEUvlef3@+d@nbK3?sR$5}arX)#u5Gq%(W+v&1p
z<vq9El}lS&Cr5i`AV3`c>h{b8B2!wuQqqxoEV*%+)wxZI0#^8%&F+(z^5;;xuCR9h
z0e+m(5eDj=MYrEz7X`sj03h2>`%xj9*GP?1QT<MVmE585{|A8yPyn;|K_IdNfw*#c
z9du|EzwiU#fyiG!^+Oj^1L*olJU6fqas`YfE;y7p6!fQEeo+waV`${V4M~FLp~H9-
ztI00fYe36@_B5sPN_wMu0YQ9K=p?hVGpL$uu3AtXEGwt|BRuJX9}a>91}iDvs$M-_
zz-mZJ{IEA7*O?v^6PZ3>%2|kZv)c`**5SwP;f0K7?Y<7KhI~Z0`_sXd{9Wo5iCUC_
z>V|)_p1eHVNLqP0KP;(sZbJDOLlo?ARA>Ib$m_ElRvTLx^OPeh#%0}gxP~#9j+TUk
zB-EwzZ?}=@>&tlD=5YdRg$0rMOQ~;+8OCLLU_RIZ0ZWOBIDjIwjUuQ;Ey3>Gdt`kZ
zLeE{j-t_!cf~dQKreFs*A!sl1;RvD(JvyY)cD@&G-z9|ZeENC=)97qm@y!cQ6lzKo
zF*;)Ep{hGLIGD7D-vCX+MufH#leG5s0_@;OMh7P+Hab{3`&nW;H$g&K_gHsx-5FFh
zh|6TW34C`tfPM%>s4{@6up&+6XIsahd<buK-a>>rU|uy>h*2o-kHszc|6B`(7dkGC
z%9AOT^!>b{7?3vrKgooH$i(&TDB(d6<>G(D7QyKBTWY+q14d$7m$V#z8Q^2fC!U>U
zc^xvZ{w&rXYnwQl*joZQDe(*Q-8$&qyN-gi&wwKa^`gqqz=?!%EV?D388e&D&MGG{
zdmfEWKBH~BWR#E^5)$$@NraE@BNVH|Z7WWV&(Z``oca5c*xcWLS#gmDIy)B1F-0J1
zY>x8NJ9P9Fx)|9~+E+?Z6Y~jX`o+|<ly?5uRz#p990ESefyRbNaKeC5ncde;Pmao*
z)+Zu7ljO6?jJoMEp5DAIAdviLav3Cs)cXOg9_H58)_vGKQ}!?Xh>u9PO97gf{4#&g
z%|RJN5!@t2=OPIU7qAOO+Y{DER=P=ly?2+ll0@AG1_vMPU?|=|YnoB}8=*DAs~h)g
zg$)&y*(q;B>CMb!It<r;Mo36#JY1*+LOMd|j}NxSiA^-MlgysD3Es-R7-eA6pDE`G
z_z_v~3SNi(WMX@P%t<}tPLl;dn8$59I6zCxfgJ}<$0^$DRO;-o=qz{gcwXQTbQ$OG
z6YQyq#$Q^he4D0XaCp7j6Dk`%ouO{!#bL=dbF<RC+8eVAMRYbRucJ|wg0IgP_siMM
zWIN|Z%Q-swKq>^Jth(8Nr@QjAls%}5(gD4{Ig6mD`}beLuLS0adBr!HV@_u2&$2y^
zc4fkc^3~IahMq-Vm*Duv^a2726lHhir*7ecvdsRtw=fl-*dK$o9-ri~tFC3;gj0Qb
zxXe6u6YQF``Xfu3K4kOqFt@e8eEL_3g?0B=`<-wrDy&(n#ycBd_uQ+=dIvt76Mpi!
zo$u-ysx0J)<wli`VXENd<fIZ=KKQPw$Ya$EsgC??>#Hlclf%PVSh&Q};h@%NZn~kU
zRanSZR^7$AHG`_{Hm2^mHFO)6);~A?MxtM?#y&C{KtfN~uXH))riX{eZn;TZ63ykV
zD*pH%uB`mdllDVWY0t*c-`mGQX)_tGb{MI&nZ3#OdBS~5PA1*MnK{?}a7+BmPNuJ4
zLHx%c(?HoPToTj0wQ<*jx#-oF^B9%Pv9cqhi}oWe(kPTCA206<D`rwWs_QvgU6W=R
z1*zVqb6YNMqgIjojpo;ymi4=SuOTXIWRbf;ADz@(o=g_&5vO?T;In}mfLQ=TpXTt*
zKVnBYf%oN>botUHy(CPOGL$HhX==iA%;vWP76W0^9eli%O5FMJj~YMWt75`qdeT}k
z1NA7&x}9h69CBtu`Msc;sy)Fv0_9pJJt&`ukPvn{5WL6XQS9ox1*i1hlR{M1w%Rw&
z)wc4<fnQnC976`WQusr36Jmn$p->GhI-jGzGI*A<TsY4{6a0uq)N`Iwu#?Yd6BxoZ
zBaFXSRhC~9JK63>lv7I(-QKD)TUc2Q!*;(WgIEsRAOQ6y*_Zyx|0@9`xvhmUdi=gl
zADc*m+J9}=%fjkBZGiYVb#4`ot0NhR^wk`zGRIF;toDvyV&ffX!2ry6c6a5SF|K&m
zBn*CJbT*3K+IvjFWf)hQo$AhmfeOm6VvYZ?RCKyjv~tF!^TZRjnZuxp>M!vI(8agU
zPAAVw&(0csuv|eHy&=O@w;f(sf4GEq7A&$>c|VAXi*Bnl-m+RI`Ro|&u+VeOWAU*L
z&7K3oQOAR*vQ@iAJ)}3jbUe-Z@Ugd?AyRu|D-W0rRF}}4Pq;{vs6*-U#%_;)k-IB;
zwMoQjX0!6}=;%i9=|qRqB<fZZ;FUAZ4gRhWP5xdEwyxylTnW&?l_qw1C(NE}XxOjU
z?JXfbm6C#J1|Uh6i1us{01zGzwN?y4l`8J|Df~iY(p?lj7)3WLSh)$)=c$5ocCh)6
z+pBbJMR=zGhM+^X_L9n*2L8(eJQleDljR$IS~JK-j*pn6bGDCw6;tx~H-t?&JYlEo
zPY)H-*jWZ6CTm+b%UUr1F=v%zw5DzQ=Ly_XK0(23*s<_aXjZJ#OjG8)cYUd}BRl3+
zPHu%svs_3yBIO=Vm<?$;L#^xilvT~;=YD*EWvJo8{`j+T4CrPVfNUOz&vz#)_*VE^
zQ=T74h3UN_=B@_f-gO%n{Ou_}`j7o4ZZ$)BP8eKls++)PJ%E~lTH3R*@o6+Q<>Buw
zm!lbi3j1y`w1k<TgwOheP8RgcFr#xG`Z!T2Vs1nH`>wa}Z`o&0>qK}!dBFFBsw}hd
zr4-mfVI$TW-nm!2auF?OhA{iMFV&;K-R=@QGOuX9N0cJkJQmb83xfho2Xxru2_Ai6
z5s_7qpT^@VqUbAxU5->6;=Cevudu4F()c%h6b!WW9p;k*lgiR70Q~_}h8{1f4PLOv
zgUgLSv3Hz$DQ6uYMJ!YW$C42T3_^q;EbQX^ccDIjlu+UyFkGxB;Yp8+j~^#)8iqpe
z5fa}qCuo>+jnUahSXq9<U*cMQC-X((fYs-!Z(A5dN2hKcxK9*vhPJ$5ypN*tlektx
z3VU_!eJlgIV9Y&BFZ0v;p0U{>m&O_p1c4>6cYyakMO*RE`ZwM{?2BUKy{mS+F)jNV
z8XDw)UGV*~C;ZSoNf;Z%Ac<hF4<P2Wt|NeJ0L?P#7BtcuL+3*7_z2HxDhEe&x0ikX
zTzLhPI~4>_tB5I_F^3nvR7y(q=Q;kqk6(~&B{68IX_7?>Yl#rxyQ5&+Ej;Z;t*)9o
z(i`$u{$j5}HFiU*Qou=9Yp@0`k=EW_6@+2|NQO67cAQB9ZWt6y7V@-JLQM+7a|YDs
z?3Ve2LY0<w+=x)#zkcUW-Dr9)B8Ul;c*xPpc6Y_x)bvm1WV;ZOHR%3|v)@$fvNcBB
z6v^WTTzvIcsTA<8tYpt>VPWC<JLgjJtE$vN9d;6Vr(~;*4Qf<Uf?0<s#*G}Xts;0Z
zE%ol8CeXD#mxCXQ3Ryy%1PTCE6et(7zT}Sd#)~0qw=@Bvq8@O6x{njtJ7~7fv;zt5
z8jMkw0%a=Py$lQ07l!~Y)u@nQS74wHe?PGfh2`VTtdBkR#Tj+nl4n=I2x3`mW)3J{
zFN#^3xyOj{;wH3)7@c{{KmZsg#+U^h4crvCp`S$4EiYy?o>sBiwyY+Kb+$EzBk~b1
z;eCJ)#>Ue3@873Qz|5L5_nOMkM`gk4avRP2?=ce0whn5I+E$_T>d^~`P+9udBslS_
z24Qn48o?GUI@m%#7Opa-L>)6iIggbEcOPONz`)?{Dw;95iMj{`w*M{&Me-+ZAk{d$
z<8nZGs>*LXN*8UF*}yz4H|}NN0MM3c0`#{MKxX6hbGWz$)c$F`*$iRl(It&|e-)ld
zQ(%c<*QFSPC$xtqCe|QOV2C+b6N3lBOWy)V5#y_|u=bY&U=u;R1BfwiBKr$ozOwTY
z#M73TbMjDTKpg^$w!#YC^^zA<2w>6PjEjib0xVgN8UedmgQFLelvpF7OtCDkz7|mg
zF?i+n^DoofPR^)VU1d}w<M%qz{>9Hms`qe~BK4@D^o38*<)Kj0C3z=|V4+r0%beIU
z4JLY#q|m@d(-vo<4633fLP{xyE~!)qZLe8TU1puz9&1E9JgzqxZ23N8dWbJQ$_Urm
zlxmRQd)=&8Glx&5$D^W|kHLiKdrZ?SSkgFLWApjCjoDB*-D+y5XizEjLJ@(=$5nZn
zEz7diScBF~&L(A;??UNnQQU!w3#&zsNNcS}mx~%W!AAjSD#WSw)mz5kq{=Nf6iP7c
z!k-Y>NRv3x9dp@@l7rb(IB>0sbF`wa%Aw+?m5wVvQ`4nFbJNrBi;9YBG<eoQ9b<D1
z1%T|aq?8OA8!{seC!-BNYzMr(1lgQ}TGE2TQYJv*t@(2!xjSOuSls9CU84jWn9GHj
zP5jO>Xu5y~V1>YpzsbGV_<+&5uKC~ypcUfP^U&;HoV^r7i>FOr%KjM9m_3hY=|c|=
z0EO@L+$Z@<7s0dS%{;VD+pC(3dcN|Ng5qtX+Gk^LRl!jb!BNt?fRrr*P%P0MCeF@k
zeG52H9=zR*d#A3jfA(KwT^I1z1Dm0SR?Y%eDn9nY%#snKvxj*{6bF(}fI`3B!sYn{
zo$OT)!V?rfV!AuoZFFw=X?rf<jBW!vT>j2hG=M;D^$?m3Ku8<-Xv`N~ek~|xEGKu=
z#1r`q8vlDO$sGxJj;<SN8Z9q+#c+cjrjIr8H}9^E!{TIa=C%hH4fs^14|b6PfRH7S
z$KNx0oE=T7s;Wkxs0q>zh2MB=zgjjde<+rt{HS8mvu0u|Y0voNaEH;^Y^+$njSnaS
zNN2G9fYsGp3Fmr0<)`|A&6%hf)KcdQTR2lhpcH|i+P6@35$~E4gnmG90Xu~!a`5hr
z1S*_LwOd>8d>&8YShc-oxzI|GLI`8D`1wgs*0w?exgs>R(|M~?f2?HT_Z@Q|<g*Z3
z@-*kSqhNg#gV~0!yO>pSe(MV!*B_X4U)^~`p~bp_rSaM8%rpFM3?-?=gi_(mZMg^$
zT1AR}zSV=s%{yDhsZT|+r*^J^<n=p3;ZZn*W5nn7TmV_^0G(e18SEpMN}@HBgAB1s
zmkUJ%nPa8jE}{1xkB)5jw{LHtFx-&g;Nakt!V<S*au({|^ateyL=?!W;0Gy2f)`kz
zRxFWe^1Xz;qB%#1naR1TW4J+j<FIFHD#fFx4}@?;3WVuKB0iw+B76j>zb0g55;sJF
zVgR>TNBnT`YQ{gz2*M_KJxJ01G8{V%S4e^Xdf1z+(}!IVFPgdUP{cDHmGk!ViBer&
zdUdO5bYS3j)k%-pCd%@SB*Sj`Fq(j?t4v!=-lySnd3kO|#ser+O}d{9cW;P$g}}ts
zT)}kj;yP;B!hCr3!iM`;*K5mE#SQn`7;0Gm09M{J0_*<kB7uW%ARhI@u*_lknZ<au
z%4L4@9%jg4*i6oXDc}QljS!!kZJS75p!?2NXJsg2HywihBs{m*x&jhNQdWI!qAKzI
z)b|?CpKHCFe>rWf1`2k@({{V!RIwGvIrS~vn@xKV)H;U0hVza<cSYtY+|^j6HB$;a
zG<wEu>xaPVD(E1rTs~yZ7M+-av0K!lG{LT(W1uM4?F>B+f4EM$=UN*RQ_)5IB+ORl
z18jCAmQelYJKGMBO*Pp00E<S-E#eU#hh&#DN)3aD3bjP$cZ%u{bIUEqbD=gq`|;z4
zr8{=z$FcSAK`J8JdtFZTO8$DzoH_o7@7rIlte?U(@7`}rPtY$NyLCCzZg-9_pqcX_
zChD#Sge7+!nIdd>NM)&v-iRs7cLJ{ZUwQ8Rp0$cB3zsMN{ei@!?bYwq^}p2zKZb0#
zQ3zT>+jhOLzz2l{z*gu}v93{Gj*dT??p?XeMJ@g#0?;R#xd8&wT)r$?MdHfCq+6&S
zWe_?jkvdgnGriKH!y0*qN=JweUKGdV@$SE|yCwA^E|aU(0zCF&HD#uV%?Gt*ov}{Y
zlrZ<ErlutoMU^veZIbO^XZ(!^fRY6)URU+r@lnHB4eSGl4vq@hnyM2JRAQCHCPqg^
z3ijGM?UGnvl&h%Q+Y`uSpxtY_d&k2*22C;*VtAmbw{K3D#XzIM_ruKN(C-2W0-7`w
z?CxL!BftFUSpsa#q#&XV%vKyQ1giLxyN_OsJ!H5DF9NJF+ZJj^6&wUIOxCdmaKqo`
zc|Aiz{A+dzZ24&5qu!_B0!A=GJ}Fsb@%CJ92!M6OcM9JI3tJ&8(Q)M6S0X1E9v!)l
zqF7VnVEI6X%ran_0LcgdY`GCUVKJlm^<s<#{2Ur|Rr#YRgI`o>2jdthZYKZ)<4H}i
z;=T>mt+V(437Uj&@A{fC+wRI=Dss(nAC2E?AhWtLRbDwVbL<wrf>P}GPWAaO9&`|)
z9f)09v*ggORRev<UW8o7raUyITM&h^Yx_l+QoDwG9FWeOj6|H)YPaMe?xqhq*q}_P
zIIMy2k6^q83rW|X?2a<KcY~C%2x65A2wDoBe<ET@XbNqU1fN{ON{PD7$M=(>Hb>dC
zbLJwdHgJ|4g&OPq`4chB!C-X0v*sum!|;9IeF^N*r=XIQTTK??(mCG+(>(m)Q4W&F
z`oFEwc-boEb$hq1FYyos08BbCGLj3;ACkk~>a*rbP?ELYVc*TD>_3k}IeneA0l`H9
z25#)Ztp4$wJ6LeWpbk|7S^ey?g~aj>KBC=M^y&MB&<V2=_8V!|f}N{rs3i$@N+oV;
zh)g`n>W4uOiuP^;fe;F7Y^KZ(#6RywLIM5x#gD*Y(qnHCEq+!7&SWNx0pj1G6P=R8
z_M1pdH)mQb1J0uw6K)aS2WbXSB{;N;_HiOqs8HyCu5BU{(~i_ALv;x9iY2SUmzi1K
z=yYQ<lrNCgiGVPGm}mldEXx3FXBs^1WS~*ikSTEhl~UoT4(R00LV9S|U1jhX^HNjq
z!4<+0Z*NIbEP1%z;|Tm=B;EuZy5jy9JZ-NN2e*q^@Dv&)x)KNO+RzLAB=pAoW-;Ei
zRDgGwEwnw2<9!$btqow~<RqB3QbJ@70NGlkryS-8Fv0<LQl!8ISQS~w5!wo(0f?R;
z%vIdSw8lql9Ytwi2H?kKJPl;MhRU|-dacJuiWHx(2Ttv1=&vt0A7XJqr@GYL_*<|a
z!y6w=t}DX!(f|XCt&YwaLP!E|6zqWKfJ?%prNd#ZLivI`ZKFB7EoXm=$tbNWwh>GI
zH*L4(c@&cr4i`$fQT@OuE)$@Z_V&&Wf?olIF0HK4_S>Q&Z|+UsCVVw|7bAsOv1!@k
zv+>=G??+N#ys}6x;l~iBW20Dgsr~EZRBp^m{548JhJ*Loi_q#IQ2|&A3}Mu_{C%nv
zlh}Ulc~{-{9X^sb??t`KQfAFjWRVUBC~S9A5d%W@iWo*XT#&2gH+qa=l!u|rOpgC#
z(?=*85@}BB+UijmfWkp`uDL?0DDb`FI6&gsVWFF{kh+mNB?k@nwUGkUp#Bh+H@;$1
zP!%F^Rba|6MJRKi#7NI$uekU|i8YShqzu~qnGUi#Ve==hdf_ueZ!&`Xs_TI6ban3{
zgUFUWP5+@xcj*254NAy7sz({+T}n6#(xXlqMJ%~G$m5`A?pxS!o*(MznF>1jLBfM+
zH~m#atX7!6R&W@z-{yszzs6XD+1xEyfKUkGwqF!WW0Ein+4$TRUp-=OUCR0yN-tgr
z#qV)AX&|<KzY)UWI^zYpA)>qi5raiguz^h>pb<@JZd>?{2H)>a#hG9p0hr{-)qgd3
zHPqEvPfLKfQ$sChKuy+&q-U($EX3@RvVp-Xw49SwEmr$n0gi;3?O9=Dw>>aV(3)k^
z(Wrug0yr5^gt*mp<3N1!Jih*-W2Xx}G95GIy(H4KT&7Pg3poVCABehftufV|VxWW_
ze8y~L)+*XwwXVF+nI^8bXE*l<*eHXN-jS3wLDq|qP@1ML29^TXu7LR0^PM<oj$j~)
zO8U*Qs4Wfi8mDz|*ud=Vf%IzYm=Kn7+8w7ye}0Uq7ib*F&WT)sM*|aQAv!dI@AmDR
z)P$^mBV;<OM_s$Ju!mHOMe>3J{Qcc2+zO1%%@h8Z9Wb@zr<&AP2r|=f?e^QWszPI|
zIs@{<j_>mew%`f@ccA1K0XErNOx{nlle!bAoa!i_rz(CT>#5R&$IGP*lpFx{17Ih!
zzo}e-JF*i+e2l0c{&45)Ofcp%<tm&nJ22(#sv6>x(`i|g-3WR2Zl85m<DM^7<HrQD
zVAZ1yS1s5HIm$11&UZp?#==T~kFSy%&pXGut600{d25HfxgWN2UU9LF26JnSi1G^N
z1jNEt)0m7pZti1ADcnuGdQ->)+E%LKn_Py@Tw4WUnWW`NB2=m0-$G9T3zvW!{8@rQ
zvF40hDe{V!Zh06bIz)|LOKBA<io$($4|p-0fKcQ>uLd)6u%VB&4oq5XP$*Lw*m<dn
z5T9oy$9K#lbfyRGL*6<Z&p8>0DR>c)?JpseByHR2Ya36&y2(Qo<Hn|Xs<l~%i}H2@
zp{6de8!fYOcpiD1phAM63uY~NM;f;}eIdRJ!nX0tmy%$=Ye2wc^jfOF5Y?QdfgnEy
zs?h^L@-#i->D>JR-5&(K@K(;3xSjrB;DF(!!2out4rt6eh@SMs^1Zwtp|>k}pd@^K
z=tf8WFf@o`vY_EM16&EJ16Zr3U{wEQS$3%-uvIV}d?y@HOIp0KpV=-#+#)V6PBR1=
z7HT5|EHst8XJG;x*ASsOE)$Vz(MfLd4QD6o4xqV%HpQV;t&GU|4!sYkf^QZ$AtD~w
zR6Ab?xDa1tX?<PQ&aM<m*CNSTh>XO{D!i)d86W59=Q~I{r&s~OV<_%ii1!B1NRq0<
z1J37lohf)6kEIZiwH}a19s0VR*2Z!WPj|v?eNg;tOFXZ%)UI<DTFl{oo49A+GUR|W
z{|!M`Px%sjwwZlssa3$k$cq{fLC-iq)(w>Uglyo-&<n^OgQE^D2YNcRfY0Yqxstxa
zS)d+mgWd29Wb`LeQUL(VWcvqz*#bm%G6C)XFu-dccO2qD`otfInvtUeob~%JkO2V3
z^N*N_<ou8*w<y>$t`NXR+{?;+Jh1-MK{;A$2IUWgN(y`%7HsE#L@W$Mzz69`BJ~0e
ztC-#X26Bi9Fma;i;i6>#76~=vvcbk;lp&yGS9Ei$fzvRjO>-BL4}Z3aWp;4Z$U>O6
zBwiP4H=s9@!&ZU%RG^)ZD20@_Zs#(fUWaBPfdyXzP6VKkBWfEW+<<euJy_(a5{@7b
zIM>rEg@w``f~efj|4fZrscP(h9gn$4RV4lZQb;a0G}+YDYf$&x?imkM0N%Ay7_cR6
z+y@CU1r`gyH<`iY+WKX!^>9Ad8}81|&aL`bs7ARiY%U8LRO49!-ZfU|la|=V^vd($
zA3##J0CCWvlB@U}qDj~~0Y*`&8*-WBgVO-Un%<L;E>{M3&R@^CH|+t4Wstm}(Nm}N
zU`D?wCE)_Camow+berKRph+-XPCWs+R8+J}p$4xtJ-^T1(FX$4Hl(f<A-H*;{_s7d
zLRHo`Azbw+9N-E#y9x(m;Es%**vQ+1Nni%y4waM9Y^yG@7!XE^)jC^@Tcdv%uD)%s
z7F=nko)!nr4$-l7h%`XVC=G{zvhqsbhkcVfUtjQmrUnlIJf5{)Q6M2`5E4N;!OwLW
zfOhXXGK6I>o>UKsJqLetYcH>?J75V6Y@JVd0nHKf*K5BE4TB84Yt{*Bh}a&C_QGSG
zL%<p(c~&(wH4Z@SMpr3WBmc*90$^LvpKa3rlLucxG!GYaqUvBn=;3_r3wpwHZDS8t
zDY~f10wo8j6umbAh~y}<f`@uDOxoFYvgmd4j5eHC0L}zUBq!mamW{S7k)K38O~u_C
zPytL`8w$(1hXw{dg7GzSrel^&GI*1#lDU1j?Q7Q;9~h0%Zj7^3Tq`1OIXhW1KhU9;
zz{b8}SJU?r*h~Gv53AgU2Ayaf&ngh!SALzRKw103@h-QTjEt5R=_M(!j^Q%WYcHOW
z2R+!?*+K9EY)4q~1GPx&Rl14h>UzGB=p{A4j80e_UeaomQ{FfnLymw*(R(QO8K><{
zHNC1{uRq$3NIsVA7CXvctDFsyu@Ws~e;m(m(FZ4NvXheNJRn9`xG-4{Bo~Q}haUYb
zG3Z#Zc#|v0CH>*i4TUxFt`>SoFab0P>P>k@3kmm$kd_xoV|#4V!<9a<QI9gVVRWU#
zp*$dKq=vS4G+`^oB<|hw0c@kE*+YBF8b{_`3>mUaN_H?xmuP7}S3us%Y7ZCfsl`I>
zek3$SjCQ>?$N(!65Dz7ncD~nfo7Vtj^ov02SpH>{Jotc0i*MQt)+?M@$}~#$ct)jZ
z?Niurdia(tPJX9898}ZiD#zJYAXZKA%k@X7D1eUKpFnjB=8wJO&+mF2yoOey&d;(;
z`c~2KP{Bm#I{+0$DK2xXfQry(JlO%+{6VdEo=Rr=L$T+<V9cq4HTYpzIE=BG+r377
z;JfB1<YWLuu_bq#!EOVv=_vBCnmZAji-X4Wy&m;pBd*w08KlB7yK4(o)mV)3BgqFO
zo7sPXw(JHht;k2gLYvrAg>V}>EHPa#i5o={@CYb3uneOO)E&dT?Fz8>?_a&}sew7M
zg)UnGV911H`MS5rcd8t@e%h@HAEzfx&(jl_sb+1MoS=K;6}iu`CA=P}MDmz@3ol)U
z)Qv!@MVXjqU=nbn%9_-xn*FxEL%IBi{{n_CA*Qf-wp`M^uOvtt1<AH*UU#b@F7$D`
zvt#nDc6r}ndZ<-!!gHm*7LuJ^m7$ZMi6I6da<pgMc{Z5Q$%nOAkj&7xl$na6_O69>
zQi{I>t*NRhEKtZLzh{I12c+<u=iW@=GCH3W6&F7NrZH_DvuotA_PJjk7{gPsWTVq<
z6xzv@qoO|N#yc8%XoJJuUZ-|i_Pr0{#J%bP4}9#s=QP!bf%Fq3y8~+G1Y~qWT?%eh
z)wu4Mj@$RkF*;#1;&e?U0GN5+>rHbHXMI3WVLq|p{**Fl;OnkCTe9%jT_*0id3g|p
zo4WQ-$y(*U`pboBaa9|=!4K8sWn3n|)kHo*J@*q{^5#6Zta`BN?B3Z;LMxa8rin)Q
zHUg2MCMIo&!VV$^0`6?+>u+RjK)c}&7%Wt3<QA%|`Il5U{>Se^5!2f<k{N`}S1)0s
zN*lQeq29S|KI$l`3>usjT-r{Bkm|`4u6PgR5uQVG<bF}ZX*KMwf!^NKTith*K&U>9
z5I=~JQD9~75H|3`4!o<IFUfEXk!UDY)MNP3z5$<&P^JM}Ki|In@?FRZCj)T>_9Dq@
z=WsYo{AMix7MMbQ8Ib`XfUP))*qE!Qwb@*AUl}t)JiTLNAVLjdwX$$xrtdJ8Y(L|r
zfJ@W)Vsnz&wq2WuR*%{PL|?;_p~zybIctNUru*Sk8xUZIst#Xc#2JO4`6A^D1VQkT
z==-e*W8VIeVD#6su8^XDJ?y~~*P;p8+2pNC!Q^hci5VMO2TYWAH56#TTQ`DtADqj~
z)ARtEv7n-g1LvPu;eZFQ;`OiCT(!aYU?nJ4fVLSzFV1Pg*EBjZvIJ)<Ya-cHA>7+%
zsEZHK1Ek+Nd29<MxJqd04J<tEbqF~Q1pcuI{cMc#t`It+S8581Rn&Hf6LT1}<Gi2=
z*b|b0Q)$3}VGdwg{u(?@jDG>%s3gYJOMu0Y(t>A#NG}c}iV6nTf%6$I3N^3ILf!|4
z))+As1(N6s1T7`(H;%Ur;DL9G&7*$<wRb&ipKaK(v4Gq#_xEo`PGji0gOC7+2g5%q
zvqCu2S_TvYz;gh}+Zf$21|T>IkeK{DHuaeU6GWK4g~4G+2}nLYfhgGMHH`YN0?dYT
zd^8S_M}1S;t$p=^?iq%AYqx$m=oX0L^szUBP~3}hN*NOl30oAcE^7zjP7Fbr0uK!4
z5ZK!A1GhwtE7y?&jF8Z0lmS<N2npd3owoz?LmIUCzJ&))4eMyo%BTqsPq!My{~COA
zM1$7Pr~&Tp(l;6%iy20=e+1|Y$DzT|!kx{Q>B_By9(QCDbO`#7*cBmL$#DA#7xZAu
zsb*d>PjpA{!I{QM0`(*){VS`WVN9N#B>FuLlm5DwjC%@vrZlst2%(?X=T4*h7QV0|
z=Xm!$I$LLLAl@v~2dHxJ@2od=+U=J8KyTI4S7ckwt(%){uLZfP#S}ej%ip@A>+D=<
zD9+ppB<Q8p@RzaQ$DEMHsXx0D^BkB0oN7kSOe2E-`HhPmC9UReHfO);+@|%cWB$tm
zR6y7Xq$$X5juYdG+5i4_5!6SpLGM90`L=^d4pp1^Z73ymiI9LC8(Vq7*SNB^q)DLs
z-KB})!uf%}Pur`$8Pkn6+qHY)-c?LXc+TT&Z?D;f!xT<zqp}rUeEA8RR~4tA>^*{|
z^~>B*L)D093swP~^TPzO4UA`$dy3`_Dty1z?b_F8VBuc!2Y+EHjr6iUY-@kL0p(3B
z?qOv0a^nJaK+?p;Q@&Inef~|gokOKyLi$z3rfZ^a1H2i)uqy17U>}bW<Jx_T9pGX<
z#E>N_y@pcGc4uhv!OdiVfNh2OjR=|ObPVqP5IWIZWVxh2ff5+X%BV{zly~QnS=lvO
z;yg8W@F8L|g)%L$OjQr8IC-Lc7H706QQ4$;+OmFjJk*GFEtA!c9FByUUZ}wB<TZ+9
zrl>7ziXBUTTj42q28M23o+1lNqP&@|uC8pTo-@JUTe%Y(Hr1jGF=NpCK>F6@|Dft4
zKzJWk$egkF$U_{fRU4)y$X`R<59-w|gco)uPLzQ)gNO={g3B6AkOISf=_~jmuq;7#
zg2Z`+s2m0;Gb*6Do;cl4y@31aZZIe?j$dxl1oJAOO5f!+jLaU3l)$L~UG}3_TVsgH
zH?_k8kCC_lEeP}nM$iN!vAMVdAaCa3b2Uu*{V|5AY?Z4jO@VpSHpDWBNJx-FKI!SN
z%J~KHWKu`&2l*e82elyDCI~NmJ<XM<SrVwiFmxbNER+pv_XGLbh8(O0C~e4r9wdXb
zdGtTEDUCMCbPI*Z9&Mk?PXcHV<bT(DcP^Xw#K~tQ3Zw?01^oiNylD`lm7zK5`t3%>
z3}>$41R$I%*`6vuR<TCO%RnGaf%U?DAS(V<T^kt_C~7smH#rUOCnGpzX9uFeACpz%
zD*vZ$<rfSkqSO}`tVw{Hmp^F|DNSg&Z}f+jB_0|zSm^7H)G_y<znhvjrpvE{5m~_@
zZY8Lt(8{Vez7UkXXu)3{uTh;m9I9aTJo#Uq(Pn>0BB0nLBHA@5ku&w+HZ1tRhlkl|
zFb7g>N&C17TlWdyw9WDo>dT(lLUNQ?BYPjpf3h%8N(VT4G;-#(zcKw*XB3QdM3)ZV
zhST9Xa8hMgA1tHCQS7}iBO-Q9U&ykj62aB@4EDWx)Pu7Z$}QZ49M8*TVWUPF@KAdA
zH(X*;f_!*HqBP@|2aT65oW}-*7Wx-#tx6(#6pAWrY6myu!%`1C6Qc}xTGSC^4=n7G
zte-YL7ln}V(9pai^+u}g19}yY)39zdaxem_8;Ha_sxu!FIIy9z1cePmIp49YsQ>no
z7i<&wcnB0~NeMKbzj`)%$-LeN>VZrDgWI@n%V!-_I*vl=0{ejzqgk4q0-0XC+;os=
zq8q){7R#O`2d6jISSMxR=G&Q_$th|b4B^BR(D`tPZ2rI7hYmtdAs2<irSlENUA%B+
zwBD7z!qRc<<F!1bwPfjGHMwT*6yEhPF6Pr~juPtkPo79Ue5l)#{wl!6TC>$a`Jo=Q
z=ywvT53jy$rr<hYcUQhVhx_nS&uli!!@F1UIImuF6E<7#7`L)$5O=lA5MGpb%~`65
zT1O`&C@qe9994O&Ki?SwEyIez$CwXau3xd>nI82ytC}nm2;5vH4TNLArM|=Wb{h<G
zlxKW3Bqb#=!89FMc0iF?&_J|o>dk_)?1q2m!U78_rLOKu!|@!K4MfN^LI+K6-#Oi%
zbw1u~vjIzbMhjxjGRer1u13&Y_~!H>lzi<JVx=R+l_sZ9Uh^#;Jh|hPmwtk}2oV@6
z`iqI0J+0<q$!$<;9PY6j7lkIwW_A15g<7yvZrQK=GN*LgTl32_$sRO?X{_y$NJY=_
zJeDL|X&Z6ALl<KMComORBd^Mkg8B+(!NU6mLAX=dt+yQ;8{BJkd)k^FJA=+}n)R(L
z(<|4y5IDk*sdF>}<*1ot7C?9tFjs+9yA$}lg|84w&2rDr$iM;7C6TSjXTEiUS}V$+
zTg#@DOOr_^96fVAErMVq1{z??)YMfJ3|l&gI@XesTSf}+!3paQ9i8{1Z01%h8ite6
zk^z#8ABNl5FUo0;jRm2{?I4Hy3RoeSF`Rc-p41Z3@%~_p)1$6jNDHzC1i=vhLn8=P
zw;62pnbm2%;mp$~EWAIhmj#}T{DB@0;l<vYkWDkIaoHod$iaRj_<1rAW=fKPEle6}
zDR<SfUj7u(rK44h{xU`9A7<LG*4@z}KO4Nnl~odz4@{hZ$<Kli(eoh`<O2DxqD+bM
z`Vc=II;-0~<;&~v32a@}4hnrE&uOd=BB@!~RB<s6AraB$4E2Gw-4*}&^5M1u_6}nc
zlY2}N_ZY&!jh3{WrB1o}M1`V+9oNz`!z8Xut7urr(&QQFedyy-n22{LL*){w+;e+B
zRlP0wX!@C)skpS=`z8l#%OZ`LK)5fL>wjIijG2Bd?HB$`j9kW37Rn%Qxd;N|CsX^=
zYL`RXilN3WpJ|6if3vtpkL9nIqQ)+vq9U2zE2DBuey7H9RiaK?ijlMlLP!^5zAGO2
z&im_Odya)WVJ|<$OMJ=@)(9bc3|X%ZlZ>aa+b1w!dM_KM-~c`LLkUSqIXDl|1e#&7
z)nvU)Ao(A0x&EHf(V**6Apn5X>;-?97-r-#%J|5vkp_OnwmPo`A4-Ap1`)vids{p~
z*3R~}tOiE-n>Q@7D-<=mLCKPmjPeD~?!#A^hQU)7-4%4w?&#P}=atBUPN54zuka==
zAP4p#jI;q6hZ(oYBf8u{R_{PM!G^=ZqKyJ84+0dj7Tyv@d{g3+1HtfK!LwNle!lCl
z4O&2qtGpW>3h;u>1(a&Fn-3)!W9rJzM@Te*){?&$i8rBJnyg^MPA2r~dV7{>8O_9I
zh?Clx)h^d~xy9wes(G5_Wr1;fGR~Tn5*cwAT^m^H)0y#>UtW=jy}_k--*}DoLW@iW
z!hz<FjD^7CEH(BSkUCJ_kS4UZw?7~9rdC!(WjX59wyKlPRt(Pl_Vo7psImJiv$j0q
zVPI%lkr8L|^94*G(a*O~P2A7tvO>h7qEcWHV!&Wz&3N$BUgh;}k%FqlXE-@<XQL6z
zd8N2*<13lM6vQQsVPoM?X?j=G&T**0DS0*?vV%wub{vZgRa@;U&GPbFvF{qCjq9Cx
z0^WRsPwnt8`|(kc`^sddUNufb`=oQR|2G#}8L<o&d^(rkVJ@Olys=+FK@C_gtm#V4
ze&gGbG-YrDBk^sWPM$|9Cj-7e95BC58hADSM@e<OF~+Y53E5+u2+?0zxvZ0D?-?~F
zTe^;(H*XA6ex1Y7^^KM<_?3haQ>#dI@c@K(YC5{s#zqu;T0`4JrzF9*Byw=;bQynB
zx|MW$UKYn8Agh(^(#PSt*vrn-^q4t)S$Jtj*gFu9jk!Rw=_xy9!ADE}c!N>8G@P#=
z6*sO-rup~=a?l*-Jk8Oyw<nLwyq4UZ4#5z6>6gu6l9GH_`BNFOB^5@xXIg1f0KA6y
z70lmIQo_`?VOt4Whu!wd9bN{-7L&V||HIZ*t7&Nld^{Bu8weLu`}^1n2g3~ZUOS*9
ztF`Cd#pW8lbpZ#QFlu^wTy7ioZ-%A5*YTrmtKPw$|IlHuy}iBROe-O~agLs>c3M}5
zI%)OUakD2HyDe*PBU_rBG}A;wMXEN@es_hU_V}o`9o}OWkEflc3qs$JjEQjKyv!md
zWEDTn^MdA$w|n75kpGk!lQSG{<eI_$2??f@TvmH6bPEl=kmp6oKa?SiX%}XExwo3z
zndI8|<nnw4Xo4BSq+!9mN#GL}Hno5Zflt_LCHTym$f+yz^hUk{uOX+<x5%fgGKaJ0
z4bd(^I0r8{Ni6u_Neau`1#aW06{_-8m7l+fWH(6tUVD&sVz00KVsj46R4RB@unCzr
z7h4JPt8Fl(lU06@bF&5kJSH=6bJK=hz2x=LulFYK;UHHqtKR9yeXIE?(K71%D}v>9
z%v7T2(d0Xd>wJmMcS15AN2isQQGm{6FDUsN&~oQ4p>KQ1J9}Xw-I=d_6Qlue-n<#R
zArpT4Ob@4M?~`w|X7k`|T}wSy6l1&Hx&j0a%i)Z<raMHh{d04<<;dQqZG272q!=14
zuj4mNG>l_{TsP#V_2QW1#ZT;euk~Vz*#3S~Hva3^Vn=uF$xn*lhaLwskiLTxBq-VK
z>D8qrQre5XUp8lBZs1=&D(0sP#kCv*@~51qVjKse%8y((5<TQT{77nBJX7G)l<d|+
z>saBWs57TnQOo_`xAmy6gWB}zQ0#k2r|Ibh9>V6OhRd+ikxM?%tM$xJ41~qBGG0UF
zuLouFCrzlN-c%dUV{-m`f<sBN?37Ua61TnnoICp=+b#f2Uozk>|1Kp~b)nnZNzS0@
zP<d`{(%())sLvlO3*{VED3si*k&gn211Z7XAbBVymAYJA))PJXQ?(WvNBxxwfCBpZ
z`aKR;fN-bQ+4+8HE7FgGy4yCQQ9!w11E+kxw)~-m^rsj)*<cax2^i)t$HkiHY9%F0
zjT9D2_KfyHDQAalAqU2t#-tviKY_O{UJM)lxgIwyVK(svUz>q}A?=be=NdLD^3elb
zM#*W2RbA`gwbzoW()7<XsW*?kCn@<`;B6Y*=rx+)k_X{!)=ZM*4B?DTd~(k=NH6}l
z2S(OzI%J4RrNSAZSws88h{?|Gp0o8$hrf<r4@Xg5$}V2r<8FFP&QPUEoN4ms(dCQ1
zw9?_PU-AY%%D~hb?ltDS5Zd-2-~ss~KM9OYXIhX<25X>ik-H`kUG$Uwzli$oc&ytu
z?At6OJL^G_P$@DBq0&v1S=o}6jEY3aR&vV<m63=_WrQSzkgQ}Ul#B?4jIy%c<J<3j
zKX0G@sOQOjf5&xQ=XoCIaU7?oXVDlNh2jgEuIVJ-`|mP1XM!bez708z|LEZBD{#Xa
zDILx<v+;LqZ7Vn#{@s_&7Tn)n1RMTHm`AV)GM0Rh)0yb6*n$Xn7#hwo*LrKnRTNwY
zNb^(uEuVtOEH*g2*xB2dQ5HJi^YaR$a5saljFo7U&kJVOE+qjs(|nWbj!9)F#R8Lf
z9D!emuzwQZPz-q(Mxi(eX?MIs#FCbg3BFpshxo^G{OXe2G$bUJW3Trp;AszZczn(L
zGi1h5C9W0QX>M+4Sk-o`Lk?NCm}i(kEo0v;868fzQVICO9RvlLAh?@jV`G&>kbrFs
z2Ot-2L6Mjz5fKrGx;*d@)zT_04eYSSh|?Jv;ENtfn91x=5_L0GO%1i?)qH%Smxqmg
zcc8u6MiNga6NBm9w9K8_L=K`ou>rSYL?=}z6H}m&uBK^d*b9kNccF9FquedZUTjbq
zzk@56Z5cX{b&#FykWdqu;=>u2DT`uyz@h&5grCV|(bP#-z_ej3K^8_?ojB#-OcA$8
z7P6`^Pxt1bqr0cFgag~PyG{m2QF0A(3OH{u5P8hbJ8vu2#w%=)lffPhCvzRN5;xaZ
zTGrpeDk1cD%fVFfeDcYG2GyP(LTq~iOLX`)7`G9Tk;2^J)=**!d?5$-xOANBJfqOf
zxs;TV&*O5ao3l)yk=*=kdmP)}20edjQZbBb^88n?VmIK4S?mb~30UY{inkbP0Fp^g
zKCzf5t|7)rEV#a4^$xiR_~a!hu%A5<q^ax^PH*IP;~H*3wXDioZ`uu+MmR&?EgT}p
zG_ozrufCruK{iVw!JMt?o5(Da(NygH3`Yk!eiBLDK*3-;V<4;w$I~rUQ%8%%W=Te?
zA<=&Q&U({v75VBkF+O<g8nL41yYr|>c~I8vs5l29D=84!h%@nVN=DP|r@B*uButpA
zD<OEDTL;`2)Yf`FF(b?-pMp-dN%*3^bT_B(MAd&RbqfQJt*q!=qy<!cd7;4JAY9u%
zI?1;Uwys+vd=QxTkyuf{{Aa2V#dGeTe|)0q2&>BVgDZoqV?Tc6o5{1Ww|&g)%q#>(
zMTc}z7cZrV10s8RIFz`O<^Cg<1lUE=2vCJR53H<M`=%A&Y>=X(;Vi4N;-#0<xrWU_
z8WL7ky3vX>bmsn)?}ml`fP=6?8_|kJ!CE$|GF64O<o(`UjF3_(gW?58H)sv{wyZDK
z35F*o=Y!apHCKB!{`APRFE(82`LCc${QUVdt;S^(*Ew-axiuQ!1eCp6KLb3!^faza
zfQO3~a+iWv#?LMIidza*tU03VXcqqBnGpq?)%o-1KVPvMK1iJLuTaV#@tN7~U^{Pz
zLeHb_nPg+y)ya#0|8`YGAs^K5o~js*J?>hw*%+P-y`KGkUH~^jYQEjIU*1PrY>$7#
zRUdxB#dqkxOAXuONhHG8$Qt|ftxXBs&n}^V%ecG>09uc~slMvkhWv9aDV%RC_r&Kr
zckMw3Y}l<NPW33V<r0gO+5D^J&6baa&@||DIut9aSAMjv?%U(;sL}7;4uc69j)s(f
zfmQHli$=nk19a868E94l!@{G0C<ZNCd^r={HA1Be!Z$mYI@0Fq#YKhMx;j0@)a?Oz
z;@aP4yarTDWjAMZ<w!8X#Opf3_uad9^~<m9?;M|VY#HQvOISIrz|$!&Hsx?`^YgCp
zr(NUe)vgwo-_UAL#FPCqkg=={u1i<_r*?3;r)9b@*KZSSREnYA&PU$_`ngj(H?QE)
zp%p%}Ng2A6MZC!{U)6(?zwHWo9CIImlryjTTOqgnVBJFToSE+x%cw|h-Djs>TqNug
z<Muqb&R$2ibiZ6Jue9{QE!)p)SG7Bshl@|YcHePjs&5l%Y;y8!V}77Xg+J)EH|{Vq
zemmzW_V&+f<W<cIL(Jpr)_8d@wrs$;hu2x@P@g+7DUZd%UH0~kS&o@=8tnyD+0@eP
z#DPtTW;r(qnIlQ;B2OOcJ7kCzf3qso9=5^ePdO0gOj}SXbi8cwSe`6O{o}cbg++-|
zPO-MH;ymnvP8z?SXzDREtO2Z7>7Z*jwR^a*TQdR$Hh5jgG~Nx(v<B>XyY{!CAJ-}G
z5_6UhT9{j@j?2mq+8k(S<;6G{$Rg~{Rb}Yj$^@}XA71HgK84q+Snc0DSJL9gg5Dk%
z@1~_rc+t(ENkm2JgFm1x&N!yO%j>KC>w;$>p{LUB4H0_eH{{E`IHyaaYsR*`FttJ2
z;H&gk!@IyAxJc9Y`d7qU@9eTh1~6NGUl1E6fLuxDNGmP<4-F_K<W1_KN2rnZ9eW%M
zsTcxw7)$6#mA?&meaf8=Q|yo4JqunQiPQGBZ5)MCZ#sYZ+k>K>X-*@c3<YjOc6aVI
z)dim#(G{OPG535+GN=>jN4AQvel9r2s@*}B<Nr=At@;Ty5?Le&^o2LkRA9y4z@K(i
zeQ9f}VtR?d&XO-BaIT``&N~}#wR(3C52^$cyPT5u(=#c)vs$Feo11qf&@%HHO>(d+
zBCT-V?y1uu_H|8n4UB@;&!2A+kb=!#oNCHwyEtExH=5yKXIBj0T-Jw*wH}7BLud`s
zkp2ZTXdXRk9vfv1f~l$a2F_tLq`vuAMGWC_y>-JX>}@MJlblP<au_47Qzo!U>@Kjf
zu*^2gA%zId3+}qtoi2R+V(H{Q-^J=JwIDc5=STaU$Dxm!HpjAzO8}{lh#O2WUS!hj
zy2A(+80}Fqne;${5epKVRaA?p)@2g81i6uVP37Yg69zV=`Km?}>kU-&+Fbv+U0|Fu
z-8n{wouvSB(@Z>+ye1%nk4ZPsHPK3Cd4^Gv(<E*do_=<UsNw6z$HorryQdBvT5lcQ
z>*qsv%`dVGY7EH7iNii>5fq(03Q1PSAD<`&KUY2U@UVBj33HZFPOxNSnb%KYfE1v`
zR6CzpHY<bA;buWbT$u}G16ZC&OPvZCJLD2IMP@0dc80)wLMh^q4{Q&}p=|t-8rK3`
z;(?{K39~J!0fIoQ1D{`q<XVCZ)57j+QIz__v7*9>+UXX1s->Cb_%}HZ64@9spO8vg
zDFX^pMEo#2o$k&%Y9$u*(rp<*43;8{_uzCkM7VE~Ll0Ad8t#5N)G>)tWAl6^3Z*Q|
z(9qCx<^6&k4sv;LlBj<XaeiMo=r3);>~xGs2J!2+)szad590BNY9Q-M(|7$|;^X5(
zrTsF&97adGDz241v&<cO_5@O2C=k9uVBzt7A+NCgZL8cSh+{wPBdS-M;CSS5+<0-w
z_y9xcZ3N47ZgE)6nM`hmk~<_b^VZo!#Q%KbiGBQ=f?{rQx}PQr8g5<&18G#fs9Tv3
zBd=#T-a5;NP;MI=&lX;j9bnPguzc0WCr0;)fdrR=f0%N!oQ^{hJ?Z(^6umxx3R{V+
zJ2W#hy*kbNg}4_)29=tW;4nO!bUxI$FSre$T;HK5qjn5n1X)%nT2_&K*SPH4<`w0|
zL(fjV?e6BSt*u2^z6r~(aF#HNF5rlS3<j%j#m5IZ!)vhUHX>m!36Iwy%6pXD=g%{P
zn#cKpEz2nD)2s5So$PU?YpwnLE*CBj8i-64w3K*VQb|slDlJdCw*u-LQSFM161ONv
zGe4|T5PSV?DE`YjsE)`{Nn#xI7k6(|U#ql>Ifz_;e&N14_gXFYCh{+vaC$c#^eV^L
z>W}^}m?w;?9t4?;myeH%geFK!%~7VI*0dv^t{5BhpfOnKP!?hvT=pE!(Vsz^@s{8$
zn<r~23$cDLme=9LyFyG*R~D*+Po#S2u=%`Qq<O-?;~~zni<VaLqC2LS)t42!)Br@3
z2xC+E_cn8JmJRbicRyBOeNao9xw<1YRpa5{kHvbG8XNY?eD#BiiV8OZm}Xy~5JqN6
zm7vM}vam2@R@V7o$^75^PrF|~w*@0j-*7WN4%ZFrAia8aJLF};l6G{tj!8lNQhVMN
z-*1?*r^{WJ=il@+lks#{%(2I8kvFxZoRYYoUOJE(ZlXOM>zq_{n4{^s-`kAw*LVi=
zUled=@ci8<47{iX)kMO_=Rw6S;w}ZoQ?H7OLa(fR)_X3m!!%iZ!P&VQae<12<Gv0w
zLE59kdfL<NY9?dcn-2F&-8_HUx}v_GHg_Gp^l!kGNP%<XT@Q0|bTkDbHMZ#wT~TGQ
zjM|xg3v2!H35GkzLRPazDpeV*S^HdV)=!HB5HIo0F;XD%*7u*7<xD{s^6+uJe36V3
zRVsgji))_xk-YAf=*g+wVb$2jn#d}Nau~$eTHx4YuD*5tBu;N<&ajhNo!J*+RvRe}
zS5U|2OoBI=??hn>Bj=$tGQN(2i~3<&98L_Z$S72RYH5M66(|*^2^_8KtHyc}wywwo
z^jxXA5@swBwz-VFR{isIlCHV=f6#g497sm}$J6`=1UNLPsB}8;G0%|EgYLkIQx5GG
zcLh0CG-$?v3JB6GUMD`4xZvr#2}}9wO<p9XCnL*yv^cQaRfV-U{^d=!b9;2!v%jgh
z4Q~t#JLWR~Z7u~N29OYi<r*z;78GD?g7oZySnpHBStnZT^kbzvE5#ch2Fj*sR&xPk
ziY{cscq+iUE%iAi?;J~e@?-<n7rJx+RrjK>#lp@0wp~OdY5Q}7A(aoaY$3F}yhg>j
z!-eA-w#GN+uRBo)`SQB^36R*lAax;9#w>YK*mf-r#1*egak`b3#9?>2ni1DQhnwo&
z<z?!p-~$BZla7eeW9SzR_aYY=c2lfE8QajE<<<CICJS{e#Xiup_$uphK8o5~K!PZ9
zV-8HLCOQ`;;h1p{Lmdmk6^ihD7@q9;hVPK0Io~#*g5XA~gs%`$I8@8&*vatu!Yp&M
zActZEw!X*$f(9^$xxwusQ_dGI5CKV07e6ffj(bxdE;86LiO3Yw9QoZ(r{3Q9Xuyh<
z_$jT}>v(k}eZxC-r$n|u9;W>Z@w+K_wYY)N(gm<%9%zzf@DnYD-a$Q7ZBM&g(`S2}
z%=IesA&aO^aLvT8<gWI1P%Vo|1aT?fcgd}7kPXYB{KT{CFwogxY;lRsa{X^tI7_I-
z$83=}(Xd@%oa`m_xc__fL`Y&x!Xkn|0=G5(OP(WK#p}o~m?s=Ya0E<k$;`rXllak&
z4%@<kfeREe`4?al+#Z?y@)o`)A1)As;FHx5mZw^s8eC-tr{H4qzZn7AFC+`(Y1rGS
z^NjaOSCvWw8NhR#T~eMpAuPg}fJ_vJjYE>XA#?1yGK^NA8fD!QWq=z3dvq6>#Wr~V
zBVgHY{ydf%gxOv7jZOcJ{1E^QgiU%<^=`{34wn&5k&uX0l9U!ly1}m$%F1=y`2^h0
z|Jq$yod_sX$*w&p8Q?5@9pW>|)33Rp`f%4wv0E2F>n6%PiKl4686V2a%X<j-jAP|i
zll|#czv0t3<iO(J5Exd<@fDD$UxkFsv(FB4=<-k%SFAsn9`GIEUE8eo^Ud-vApg~H
zLX;aFU3_o+CYWG8Ep>yO7^w$Y>}CIpj9Ywa%n^`PQ+aqdwbANyO#kW7+jd{~p}vF_
zRAuQ$W+kI~8UPsp07I}oKX6~>7m918nJ&3a&raP;t#f>5ldtt<<v^RST$iUSvYIMX
zIV4xA)LR8prKb!%CKohgxfWN`>n)!Fqy2(Y_N2npfg9*Me17A1fUt%3oxZk(yeFPh
zkR@gcArT0@xxq%%VgLsL{_YVvRPchRWklE?o!Iefm&HD5=|Hy<W-FW6{+^y40BziY
zGaNs?QmHyRXq($y5U{8IOcH}7;F{hlWSdVjZ-Q&~+a%wHu5bFr{tx3NAWBJ3=9~br
z@Ri3hU7L1y)*<bVKf|Nv?CfX{c>Y3a0HmJ}ZC){eG(Fwj357WlH^mReTUJ#x*)=gy
zUo_%xdpBbPPu@@K@q13+)OaW?E#e}}Y}>xCZnA;zMi7{t%&e@okY#Eg6R4YHv|L_`
zE9$@TMa^)Ux+=A$SmmPL4F*fMYZ;C~{6)Wa?EDrZW9N@&K!kix@M_9DvqAA&)!MQ}
zKp?s6_sQaJy?S|_Bl#v=yiFgS3X=@w_4mZ9@NH8>_u!6b*wm8Nw963@M8>oe&qxUg
zZ17(3=hk~`xs;K)bKU4k9ymMWF7CjR;aa*es@}X;m#gPsF49cmijp$;@N#(e1j|$%
z*Ujv7!X+VaMW|rSlLA?N2WZ5VKV|>Kx-K3`at55+bGAh*ouN!c(ic}!*d&xm?z8y^
z3FoYPy1P$5iV*<sD<`fNE84EW)W4Qn|MGJjkRY@PYzlj)=q26lGtCk6yPIpzBlpt>
z2UGn*zpU`H3))twM8q|^=H>o@g#@`zMpgCV>FZje92Ulb0zksEk0qJq*zaCcXL_jE
zWiz*DZ`FU2W%?S!EJC}FjNnX2oX-c=GH`9n%2ojI`v71C8n7>2lWfGM&2W5?Yrk>q
z^~L*S3`5Y>2vdG1rvnDC>H5^f#4zvnI-Ga#h7wS4C6SO{MdQTr|HK{Wn1rRO9xI9(
z+99b{u{b(Ec)yAJ={nRHB<<u~gb9Ev`5mL>3pw$u<d+;exu!MG9Qnm}P}R*dll$ax
z1BTtZZoch%I?Qi0bJMya?v`gOo2z77OR4s1^wJG0;Ko>O@6hVI1o6PXqj5W92nfdC
z;Kq;CiycmH^5XFsj&8wxuJfHHqa70|l;m?~w6B$WHt00%?8^YqW;t-0WMLH<)wLqP
zd!tIRSnpE5*6<#o-C@+t(#*T^VK9z*Y^Q=qK!I`0)6DmAQ6!{YF`m#lZA)#jb$7qr
zSvm~s?NEFaFi_{hi~hfQYiOkebC(p6AHRNnc~E5wOwj0<s}G63RXQTFf7(IoE;4&%
z$YIl~w}$pCpF)Wj6hnJWrRs}=+%0V}I*lV2Ul>ojO|W_Ep(`sE$1X@8QaMxZwfy%l
zjxSTiU+Fjn@ez2ac1lzvXsV+@xwtfH3vwY3nj6<Ud-{k{q!Njm^F5UhBQ0zDhj+fI
zB1%@g8iepy4qHVSOK{a#W2j(^X7&Kea|LTPoaLJG_06T^zPI7K|78`GJ#n|P)sicQ
zc6se6#==1IIwoB|ilYp@37nQhE4R-{d-x<CC!V~T0ar>lLS4+0X~}C60~1=BV%#HA
ziP|<`;)8`U(RhpuUD=&%)v-de$lY`lU&+*#ZRiTcH6Wu#*hzVNp4FhT$7gzW-|r>p
z0Ucaic#xMYUH7YK%nT3-h!Z93AKZC~5G6h^8$zZ!cSho-3V~zaRjG3B-@*xWn|*0B
zkI`=-F|RP{=cZNYa&)H$Q5=GLd{PUD;)g;BzHuSEET~5)og@p8b6Vq}!7t(y5>%=h
z|8)#R7#U6PQb|{hZ~m9jUc3J;Uz`){lPCRgn#0|j=wd~^m5fA`*y|=f+L>xda$n2L
zL|Tb3pG>Pa1iSZYAdc>Dy8z(G>2R_cOyeYWUM}rl-6Q0`XdTo>+i%|g*Qo4=jtlZP
z_9nH~QQ72cw@!YWQLqI55zaz`X&ci+S27X?ws)bByU8N_XXl-N@D>0KU@>|X;`6F^
z+8jQ1<a4xzDcC=W*L|{ANeM+Qf+o=?yQQpUXo%ft)!v1gbsAp%8}wA>xCj$9-DCRg
zE5zAKw8D9VGBTYWEJVJ39?lX}ZxAa=cl87HfWUQX@&8{{Bq55Uo()9#YW~I2UvHv(
zhsWs!_iVDeNL0^{p5;{5O`C-8H8DoimS#9v&m?t~(Kpy}<obLa2Uxnc5U_51$c}vZ
zWNsHtxr<$Jd~mH?)1he7Llaj?b63BbC4`6ByB#Yu+w`&cf}>;N?-rb)&j)Q}<o=tF
zqpg-+ToRt;BfbW7T{rJ`K_>Tz@=eu*r<}#5n$2EtXs}8mQoDcf1=<NbyuxkIR$JCr
zST9zrULkR&{@6P8AL4svCSIs`Ar!E(I-Kx$49svu=}iC-yDG%%ZMoTetnJ|ufR}jd
zY!7GA{7LDt*FuYz4xB`rk)6E<olpcpFJ%^W11~qN-}Yx5A~{<VV?-HH2fy6uM_6I~
zn9>}U|Hovw$+As4<?7EJyf>(@r+}bRSVEP7eF`YL(c&Qe+wr7(J-+_*|2`QZ$J)VR
zBPq|Rb0?CtD)yLuWSsaD&>zI=w!uZU-{dzX<w#nQ@9Ma~;F0L&FjreqS5sv;zLjPs
z4X7I>k?)?xVKBc{wX4kQ`2T8r2bNQY>V;<Nz-IQu+nzNI?A|45o1ajrDIvk7UWH7E
z96GoQYe*?*YHZA^klP!k^1+#zB(GlA;PTVq<u#iO<oH3-iBsJ+yrqfB3fEh5?Wnd$
zx+2;z8UX<Z1H<AJ#k{UnziH*{H%<@XiPR9^-e6LG+_DIHs<SRq?Hp0ArpR)oI0nlX
zacB7|^LM%!c`Y+dxo&miSwOCN*-CE(Tq7V1aoQa(b>9O57f&ZsCnJD=>L!xyefKt_
z>)8M_2AB$olOcj|)O01U+3Q8^=e$a;9KL=~Dt(a^J&oR)<LSf)R-Ep(oYCsLgt;Is
zf6Ed4yjI80u#&)4tP5CQ2q<MYv-(&5Dm1(viaJ1hl+(AU*X{T~)*Vr;x8LK>TwG1X
zY0iU{KmgICfftQ?s*%N@6g@M~Xes8^(7fhtwZ$?(Rwpimi^yT)DGn?+7Ydt@)f1%u
zbLHtvfcCO!r@+Xa0Ry6#-E3g&b#p~0burva$e{zTQFggsxk61k@eJl8!2_sUGIDZ|
z2Qp&Iq4lK!8Q9n@Uu5czO7kt~Ijq_0jbT#^D+}#C;s6ps08qL-v=7Qavqy|OyeGaI
z8G0nZ=|~gUjgPT~DZ>MCD|h%|Lqo(Xtz(C9gqVGS7KVD8FeCUQHz80WGAklQTgV#z
zFIJQQPO&LSB#Wy*9tUrml|oC#eIj+kD~IV+U=G7BxHNPWgG9CJM}M2ZUU2M|8vMI&
z@u8iyX@9Tl+pZ1t^wWz&(L*zJTk7C^Nx@KxJDfYW$THKQUlsba5uiwH>D<-(S!F&j
zo39}*me0>6-{c*U=ng<DP<x<%Nq7H|ZAy>PjT=(%UCbMk=Kg^0!px*gz<-20iDAR>
zVIEYyUU1AqRg394p)fh4ur$OM*YoRyZZ2E1cag0DK7gPH6GhTMjY94KyR04FEPV6-
zolEG)YpAAzSacYGINqx(L6jlN8$v%b0K-il!97g|$oA0@Wd{xmP#3;;ZNhXIi}2En
zgtAl2O9j1Fo2Ddr*}r{|51KLbZQf<{ASb7(sg<JJ(8nUC4UgRh>ZK^jMt~`3^u~(n
zyxw))J}NxB&E2GG+y!(l?q#3)bQx5E`|V!zOx#vk0SdS8V(&(EgK5HJu9~3j6k!5H
z@~Nd$m#ueHOBWY-urE*3N3Z%ZJKIE3C?n6?985g*AyDJJ!M<3-ejgx|+<jn~z+90T
zcqyYXoAek94P+gyScO+Kl_>+?FE#%M2uZ71Wx;~Pd9BbyQY5v7DT*?Zjuyg~e~rf~
z``qpeVx0~B1cl2qMh881NIi)Lo5W3ObpxWk2(pRpxpU&EWx!+lS8sT0;_qJ;7EM6n
z0Q_}0dteTM*O9}=?OLbv+Syz+yV&<CVPxum{d9Cv+6HgYz|K@4>xs<rTFig=n1^kL
zWD<w%5+okHK@;{!mq#yDGXkTRkqMbTI7A-4G~de}?a^0RV4P%!78ofInU8L^FWMcE
zWD%U=!4DbsXvG4KMrSTvqHL4w)}*OjF&+MJ*Re%CLdt`5v(NGvm%hacP=es7-TY-h
zIbI;`=>?C76!>o{$oGYH^oR&|w+!F&c)doY8_Y}!&6<?*(Ot&x;B85KK824Y8{d>v
zx+3h%25OLHZTef}`6Ermz2X||9xUuH+p2XKIHa#^Y9f=^iyf3;WqWv}$-6}uo5|6x
znjH8~;47_Fdf6J&nMG<4C|qGJXpMdNWF--XL(GI{%WOWp>-o_oaaYn#qVmZCZWIl-
zf9a&eg4FJ#BJBO|boy${L+YI;a-u;zl1coK?XP-YcY{(@<4t?UB#<Y^mY6xqxa9s9
z$;%RsHP0Lizk#}sh-FnO#~<(^{&{}&;DMpTjmZD@f7`M)xCNZX-}>M#P4<Yfq8p-8
zCr-dA6v_&4giUT#!Krz{`-f58!(%kEe{H^~7%XMX1oE}1H9kXd(FQya?Qci#KJ&)+
zM0yOjZb8L1>HLRtjmxbLa@iM7clmMd_l{qEB`hM9XZ-Z@g10Q__B8v0K6wbfkfgC{
zzZd+(GIzsORiG)1{2g*MD)vDoO$R9%-|ld4Dm2@zKLnG_=B0D!Zsps%5oV{AzzSSp
zJk<R2-3(;G%0GrMU+MRzm=SO{X%aMlq4Mufmj0s0!CXD9(+jLdzMs~m+>-TFaFN9x
zTNIr*#T=ZSQ9Py}5plbX8ZxB?+2>)d;k`_*y|I@8m;s8@?%?`Nv)`f446FImM8j9a
z9@f;AVeuSB{#hx2nb%&WOH3R-MjjSE-zMDy<l}$%2bVg#APbR$HbGr`0M04s9_&M*
zc*`+qaW=jpf^cG0fKqJwhh9KZ!`DD8V1=OwS^ca|FCQiLn7VGBeFlbxR-)g2ypBTY
zPU0+@!#+psCw?vQaFz+%>ivRIYESXxt1k(?n|Mv{mfRF&kJ*3}$$BbiKZ>T^&22!9
z9B!XH!Kiw=E6wrqBPV~6^gKY=s6nNfX+TjSKtxS)(AysOdI2T(jTMkrl1Nfz)6|5o
zkYmL{)|u6vI3Qo?r{P=al0>y|nfiwEe;6#6gC7850ZZXmhukI@GE@PUM<~-hFe%q2
z><v>jUIBsYl8wmqfB#h1ta=GmA9g)@afG%Qz$Fs_yp;RwCs`52{WJly^|8t7{C=eK
z$vgZYdq@MO7HvSm2*dCc-#BM9JUgYnN=F>powV;{W~LcaFp~8?549liMww%eIna$O
zvCw$fmhf^+MDStePD|ezb<!bc=Uv1xhh(Y?_*0=GtQ(!97AdRcPJJ~;$j#sh*g@nq
zQ@<@5=%++p5~uERPWQ0iKjITF>6Y%3L?qt8Ml0oXh@NN4?ax`B=H}){mOUR#JIp5v
z2?)@Vh{O%5#^8suVm>nrU+UXVO{6lN?&|qgcXWcM5Fg+sv~UA$Ks<FKVM8v4<p8aw
zq>^Y7-Ey#TcBaVwd_gLLWpLs0C3(hh<n_C%4k#(Tr=60UVCa+X8$9&He%%5^h}8VB
zGSRsiGKizpQ#HLwB7rALd-O=bS-t65a!|br(&xFWf<#9StB>G`>2uYXS>zaHr5R;0
zMAhFjsXG3|&6iGUX~Vx%i5Ac(4IZrgV0Nd|g!QB>K^=5vD+%o^<+7h|Spi;CsaY1z
zV5ziO%xe{y2+}L!XO1K>(E*!rD%`2SpR@_p755E4YZQ^<3jFCIxR%d81|55Wd_2Q7
zp8>AwhSiEEp!N8~rz@93yk<A(bYx$^kM$OSCJL*0Zm9RIDx4V!&YhhM`Z#Yz%$4rf
zr~Cl`o$*%n`5L}8rnKAP7NuSom#Fu7B?Q;PjnE3dGHT)&wi1ix*Nx|VeSO<|gNU1~
zWwczkaZ@!RPB)ZFnxerF?ItkUUM2NYrIWggT(-cO{2FTBdH>DW3#J?P8BPBUFAt6<
zpkBA|a(Ra3=@bfa<31c{Oa%=3@JME6g3I=Cd`@fij+Owf-WP2x)N?2aP(_8ah<_B-
zdU3J*qsLdX9Vune#0QyeMBE3V3|UyB_KTa73iFRxv#K0+A~Z@x{jhQrsu0z(3GUkF
z#eMv_0hhY_#x$_qvopU!<+`1BwKnAx#;3hmT><u!b+xF?!IfP&6^QZ5gG4UIo%0nL
z4`AGTviTgr97K`OZ~XLij0GTx`^S(8Ld|>wrcdvczvPZ?mEG=^a!*_A5)vj^VP3W-
z+T1a}IrqBH(%(!aOQ4^lun2R((j%KN`RZT$Iw?N?LUMWD1C-za;3uR&vz(DHUF@_4
zi3+b?zJxRrd2C2(eE~c-AAo-3!zBmV=X>u%`XlrgMZJu3Eh8Z~L_uQd`gjN7`T42t
zbnK(bh?XCevV`~VF6PN`mZgtofH6Rnwq74aySh1}hf{#`Ih)6d;`@Qr7v?x7xts}I
zwF5@5H19!Mb^fn`l=&-WB;=R(@~7L<)6+k2<VVBV*tjbq;35Qhi=v-(ERth`Gfq6C
zy(%=I5o{bQDt1MS_kVGuVbp^ByG@y%l5&rJGPyJ-=R)(r#)sl1H^Ti&FtsF0@K8Zc
z(3$mL^Tj$He<sIKz?+n~ZO8c!ox<e!R*?lF$p*?4Mz%3xusk^R-#L-`+{m|Kws-3T
zg4aq|{*FHku|D`FOm~`k#bQ4ihr_nW8D7ha+}HhmmnWL&q7l3S6hFX$F7XnG{H)+I
zby~X5bo<_KCnw2r=U~Y88Ln)|%|+4^opi5HhA;o`s1GXFtYX?+CKohC8xt-6MV3@$
zns<#NF*b3BWuB3;6~th$>VpFA6G(`)b!yk%s&PSB>kgj;M9kFwk9$ZUyCO*pKEcLa
zbBFXILPJCMDkyB!ip@iT)@7tBj{ln;QiVnaI$$w+hFON=pEl0-Zx@+m^-(kK^1$i}
zYAc4QjXI%NqJas7WiojUlWz!|2x(gLl=5}3gZUOQr!0&x<aXzcL*Ke4oQLpI;}sOF
zVUjA_s`SX)z_9c4S{18*V5rblO)I&o1%e3`u;3XWuEd<K&v-nX9_kg~?xYzn<<_wC
z#8u@X-eBefw;(;ykxJ*&H|)$*n~iO2Ga%jwLEk~v_^^#lP~(Ao+#d4FX;79A>rTJW
zQ*Qe(_BsljXA{ZVV-m#ebBcK#sh88iv=|!<%kqNaY5L<yJ>a|pMB4$44ac@YZ%dJw
zz;p?~aY^zB?QfdMqRlhA`GYf0?KB14&Y_cIB5BkA#*Y+;t%5Va{((`};VutQ3C`~W
zV=^fOQ^ZWebLap3*(wx`*0h8)nkQ+9p92M{s}l@oSlu?q#Pz{w4p!C|(PwMRhB+7C
zmgVj`kO8=^LgS(HqytM}c8`x*m%o8%1ZZ7AHNjJjZ4kY?=k!h`(b=aE$EUYBfU#8u
zRROmkOVe8<9oP{h5@0{5nd%P{X@+EDck!iddtQ%;WSbYU5K0JQu%(h@dL>wwks350
zxw%Wog@9Ry-i|P2LInk!FeqUc<e;;Vt=gr9kjf<xiO!iPXBcg|fv#d$21ld@do<?!
z5zUdb@v9ETqFo+>utTRO0S!mu)Zrv_^qOys+|#!&Sg2w)CSM3`T^db|2Dh>Q?>bt#
zz%#kdqk)jRS6P{vL|mSz7U;}^Ez%KM>y+M4{}ePSho)F3Car)UxVymy-OT5~shz^{
zjSNwm%*%Tfvb8QIC60`uL6?xhH#X$hdi}A^=9rB!Cj*#}X&L4i0fEyu^k|=aKP))U
zc`RjXtJ*EJ$y4Q**Dz9yA>w-d@`Oreb?>hJl!<)tAWGD^3K+YlRe-x)ucGY6!CvyT
zu(0#95HMq22Aw+paAiG-u;eO6%k;#8g~3gigX6IiPi9lkNxktzk1SD=JtiF0d*Rpf
zorn5t^rq>{Gh-dnkZ0p}R)qw+j#P}M2}2V|%-wyxPe4OX+F=y*m#6=L$U@!*sivN;
ziZzcF$di(1^NB1RHSRL?(P1H<vvh6e@lf%Sv5JxY@!l8=i@QtY??M0_J^=?H#xpAC
zuP1~-=kNK6NqgtbZ*Hf^Ovepw^KEgyTp|EdJ;=a-O76Soq4;{~^6S6NYrc>f#N}^c
zy06V~42`6v{4w>^9?NUV7+y$o@Z!0f43<d+C}wSN%$`*J^!00`iR7zF_L9&z?WA2=
zvA0@A_sYr9p6=SKsv701Dy1Yz7Ku3&*|tOkB@yoMC9$e6a`%3o7INO<U}u^#lm99a
z#<HK~x~E99JD)5Z#i1noedNO0;nXeRDM!&|mqsC-<qF@BjjI+-BPRd3V}~tio;kAx
z>=95$T01*y-rP{dpQQ1%R_Qv|csR=3QPx0*V}>fT@U$d1orXx1O4TkU2KBz$o%-t#
zB3Kbg<-grPv`WQyb07PLrFj_>42#k%bM=-I(j|g^?sqz~$$}WHO$6{`+y5=!aZXDM
z2joh0P7%7>vsIXP6yp)n-2a1iB@S=5Q&1Sq<QsL!kq-TX!`>)pl=h?x*Dot29W07P
zr|YlHL=)JKoPU=AH8Wej9s)0DgHvzwC<{>sH|ute+F@WskjSM}V-EEu!FW-vxY!%R
z7k;?&qbf$)2I7ytq2b~MQP9RSm!K$+T2gPVJKuU{OGwN4sKOXcT1a3EM>hELvg+2=
z3t<^v6k}MGo*y~#$NK`L2A?FXr~KAu+vyZjx`e<3h|xo8g5s7$@TcV74~~nYF^@!@
zO4^_X>@;B_CWy>(wDAO}hu7F3M74VO8LgJ?@CAw)4C*Zwn;3(iSbXCk8qYYZnmDEE
zy*XvcWYxBbLB<cc=TqM~h<R<-QhR&?l}-<`6lBVzgkG8$l#JX2I`EHwcUTqaC}$~|
z8gWU7c){1pJ$C`k>HxWoU^2kuEG@TnCh7k&u$UQRAgYGMm#dPjc5T`H-$jPoAE(ud
z?f%w$gg8#oQ1&bv68yA&@7m=ruPzYHwCvn|*jf~0&<(e;s!|M!4?#gX>HJ0dEap|P
z4)S-?VAi*Z*R#ZdCVME?0Of)PCI?1A`S@-emLJ`l#wO43d=0!G#V06O$T5S9Dm$h@
z9=26}^Y7oDk-yBTJ$)6bAK^#;^hfM%)~%Ar`^>*&|3P;>YU}Ad6x11xjAwKbiITb3
zIw`}Dw_Y)|2#ar+27M)pghH@cm3fX4XtIP#5MZ1Q>IPJv#IFtVu&g(lz8blUq0gz%
z>4JEiR^6L;i3k+;K>P#4Y918Z;H;@Yad$X#;DQ`K#3XZ9{XQG%6Tx`+3@$EyXznAX
zmAx<$4d^QE^LRhC4qp{3KLEwc_gO;g7XU61(Eu;Ws^`Vo^lD*zE8UJN^W3hS3ziY0
zxCq7+fJTVbY%qzTf<NvyC4m>0l&h98!=wANfB$ly=liJukRmQzIJh!hts8oTYa`&w
z#)zumfd9t@0ALFF4>=U-yWd>HjC~ovprkp7J9N7ys-^j)<9`Z6V!<e+4ERAFmOY6B
zY~gl)1yCjDWnySh#~>TtkM~We3m=Gi{e!A0E62M$2olbhB9}VMcLl8zx2-zp#o7n+
zy}r-Q-`$?9tij&r0(hhqo=O75H|pgZPt!ZjSsNNsX^ZAOG<ZIAT#g6Z2tHVO;d?n0
zG4Y&OXHZn|N2X_G9e3B1W0HEaC<0n>+PD&7!PU%15%{*Ki6L6~ug<x(+vHm0uS>EU
zwEio8ZUT@|x~&grlYob`tfutc%-PK4cyCy5$3>f1S@N{0p-Tm(4x(*l=3BX@)`%yF
z-N=mb`k_u$#5;)gK~hddf?Fn(*w9*{f~$OYnh8r(v3e#3<|aXK+#+#?Ur7E*F7cKj
zkzj4}!<9l&{(&K@?^;{gfNElh5r0zLeyN>zj&;BG*hL~DIbpp$fFuzqETj>jZ8@HO
z0Le|$S}FN=hxRlz`W`_&ddbJ2eR)3K8``>~hpA!WLUjF>#WnVl>KWQ8eEj+CUY<~1
zKdbw73iRX)7b<=l=DH~faMbttS5KL=9o0Q)*m;Zs1wqTx$B}VhU_YoGJ7u3FH!2#(
z{TYP#d<-ZxUANeGZ|?l@P&AL64<u?gJFdrSiY7CMX%>@3_CCqK1%>aPc)TB<@~dU9
z&U9Z=`^^h|_QRIrRiXq<(DTn_39Z<3b@xt|xl+}r*52=n)?bvk$w9zaQQ}^h1!|_e
zH~<MFG3aWM7;Mu%wf>gpjn##U@4H<t5m%cfo_4wF=P@_PR_}G*wTE?B6=xZ6F{g;;
z(ABBRl+fX`Fi$8L!qM(*Zg5+%ronbMU6Vz0;rv@_LqnAVzwRsB^WU&mNi{wr<$YX~
zBk1{f)R|<~zoRWob!N-`2?&}}IB|!(f)_Hgo(HE(K+U|EyZ%?I*JrAH5tcOy-=l`n
z=i`2}{B;T$w9@LQqBosCTs0pcuHzWaFw42g5=5P$FJbZpn1~6iXM@+)u9gcu-K%sb
zF~!R*R?C0yzI`J-4B37G>5iY1=cStbeaVLnvN}iPxF_n;@n~CKIk?UyIO?_M2glp1
z-9?k?rvqk7G*z}lkI%gxV?#|Gm1&d(LIT+Im)A_|S~!5!5JU-+_4Vn{gAv58CuX!0
zdW7>|UTuQU)mtoq2*MNj(}D`5oY=T{oY+<Hiv2WM55uURg-4m9ck@$5r0zAPqjLaz
z9H_XlsAuJugOAMrT-;h9?p=0OIK1rfl48DbK-@v7Rg!%;73=PNXUv6M6}(iH#j&FY
z8A-6tV(MFJ_S9WoMbk$d231%0s|n9Omd;5Ku>Zediv%4Imi9?k@*v8TwiUF;=3BSB
zQPAx@b4#fnw^-n@vJ#p4-}*rBy?Vjtr-fGQkE#YEpZP|Nmzwu^j>S?2F7x6$an(U1
zUh*55UxoLB?nUr_cHX%Ok<@mmzdy}Gd{U|THqww5C+l|WGQS@G70}3<Xh&VvHrM)D
z+`@$nKsQS?C&bPCTlYRcHN6jF;Q`DvQ$y1LvM5&=q5$eb(?AmL>h09jWcLr)tPp4s
z3?)=X1W~K~^Jhk_SoAHCXPB1krif~NTCUU*V`mQoj8DKPf)zJ{jeQI|!iv{ID&`gz
zhG6(g3IzTb?dj@G$ztde0~ilF524Kw_7{u+J(HkX;kaOZ5pV4ZXA-dEQ11Y@K$3$1
z-DVvE@3&UD$jw`kL&5t4UG<|NEGY}|2sQ^>h0`bP1`dMkG+>BMamwr#z@DaFxUU!L
zV+^O2`ngK@(Bbo_6Q1_k2OCF>3|}Xnktkrnw9;$g3^YkOKw30KsR2dSC$*q?nQfAf
zp^tEF1@GQIjAxhtlSHEP7i<a_93ET2@Uk0BdXX4Shl|yJ((!QwYLXuEM1m3H?peSq
zZ5hQ$uS($T*diksl=$LlO`X@_=F*Ymfz{us>*r7JD44RCJd?`w>*<aOJja|#wY06$
z9A+CK;~DSDJqjt(X8oX~I-xfk5HOge?&}uZ-@+cl>S~}pm2^26))y_{Ig_M!q&ctu
zwV0m2*#JVr3}GLDNv?tb=UY*fAW5Gq#=;p*rgWEi%fiF_SDb$fPsO4-NXfYg9ltwz
z??58E{XIApd$`%q;>bVog~mkv5L1nuj`GrQ9QYr&9)cw`c8kakDE8OUd<fsouY4Yj
zj75S2uSGLPoalh5C=j4?dG=naOE!{BZa6lh{W84j=V_62s}2vre;AbIyKO{xJ7XPA
z|L-2quS}6)Q25&Acm-r`LTH3~l%SHk_<Z?t3$&H4u23o8K_-aZcnJUXUKo>IUK3nj
zj9#bgFaM##qM`Gf!1WCBdc<qmBytM~>O@R{9247i0-mxR@wLx)>dZ9tyKg@lb5scD
zACYLe!u`?Y%+$}HZ;1zC5wJSv1vsD$q#w*>Q(R?X(Qs+=5QK#SNlf>z`Tw*6WZeRB
zuWN4R_nwdcj|4d5m3*fGFNRFjncOIoI;`S*R2en;3JwKoQrYAMmji<{7#CyMcM<N;
zhi$hlz|(Ks$ppCWzseCx2pyc-2P7E!AWp$#4V#t}6^zyf{f)p*N**N^SOs_ac)2{-
z5E&I^#&iK>9la$Lf_f<A_s0N!`Lc`mDwghrFh^U&wu}-gCrDm-Z>FPDPoY>?p`LrJ
zeyv<S{v?lhs`m&VcG1z4a0lUXG(HGGN5m~svQb(`Ryu_7Fh_)Jcpx7n+9ZfF^bR-x
z;zf;HaIPgepoJJGzw+nF!5Z{2v8FF!J!4Y(^Gefm_vIgQ$}N9$8D_>-1{&IC&Bh*c
zJh=N>S0wEeW=)(Ve#~RIgjh87a60cqxBgJa#@a;nN<zA)aQEH)V={@fB;@&HuT`I9
zi9>7n;??r<GVvk^5zRl51xCly&v>moI=Bu-L`%mj3c)s=d8X&*^l7iRA;Bx6j(G@#
z;a`_xaOW#UVW}j&@Ea%s^!ixNMLVi4-*s1-Jf%ra0VY1GEOg2cd;nyiu<ebzb7G#!
zh<tdf)`@}9x+Z061ARN0#LDO|iEi>$rBE;^qx-#xts3#>%^R)FLLEIX9pml^FA)rb
zK%@YZ3*F-0H>^K6wZ|WOJe|hsw-J6Yhmf;~YQe>rKxtgzMe+@gg7XcGNrpa|yWC$6
z$D>c@q=l-GUC{-790m6=Hqro6s>5meYyw!kmy<%AG8{z+$#JLzn}uQl7Dcsr-z+!r
z$_v&^#k`{LPs=Y)>(=<!_`d2?#&;?VI-?W|nBfTsLc<hfy)Bg((H02>6LLf6&!az}
zOoavn9wIq8yY-FddZL24W1kb)8@fo(S7C-gBd-y*Vx!8?HxhQMt2;X8cTdr@$??Os
zyf?PhnLSCVbJVT<DCrt=wkOVW=dA_Qz7stK8<2g*H(J9UG)g&_r!Fb3OeWe3GeV$S
zg*gbE9AwZZjrN}7RW$xxVp}&u>jD7;mQ4ucOyk4_Lw9k&QGKT%+{{U&qtQZ$bN@Ui
z;*N;B^!)tws}?=7yt9qm=lkERK4R!QX}I|LAdSEAvr||{!ufv7SHRvwB7DBLc;<G_
zi4i0T)=+@m8J)k`qiqEV7)%iUwNt?hcik()Sg<7O1^XK1#Kk<R8~alIj$gibgo~@w
z$)kA)JqFV+NDH3b68JvMFqgXaN1UKo1ZMr-!icjR3zD_6<dsE+BfY=>`Tiox$K;mj
zZ+n`ILDK(Nd|$7Wp^mfc%+h<~?01%)p8Ju`rCLKQI}xJ5grH0c1QpbAd3=6uj*3L|
zm_XzB*_oA(J3|$AV6B(z?uINbX$*f${UgGpc+}94i1j3~`6^Ji%(1coU8OmiSzXj}
zf5<-u_^lG>?Zo15j*>|TW!@suQ8}6_%RFi@FB;7wWf!y1eNbGR<FDTD)!)a|_ZqOF
zTv<hJ;c@}W&4aZtBpLjZyRi&?Su`y)9tOc|!WLx}Mg(B4I1}l4@(Rk8uAA2=+(vM+
zzWQr#RjKR5o!j$_$3>oC_AGn_4BdI7UWlhAD;XS^eA(d&-I3C)1)1rGo##N)4HfIn
zuV{05cy4c-QuNp1pXKoh?gFCTE$kLJ^l%LA=}+^Jguu&JAFNDF_ufnz2!*nhA=)_0
zP#FUZNP$queQfLZ`!i;+%e(B#q}kEbdlcQAl4)V-O9%GlI!Q4Dz8q=xKT+$)9rOF`
zt$_=#e1Md3<;IUhgF;ce5MA$$1iLH8+;gJAd$WBrCfqpzOvX^!uI`tPoDiA^hP`=3
zG(y37CwRbTJ^F{D4@O}cJALZXhjY4KkfJgZdRWMB_dBu|p=^gJkAKTj{ZSGLRhues
zhHvdz+tPoa%LfF8lwflDmBk*WC<y_$SQ;|4Ta1NJ?1@Iox$e3Xyc{V_w5%;#95W_T
z?`++j$Hh(5NG&tmn*x@L0>V*OAuCaJo{xLn)9~GGv3FE8lo-%JTq-E6azAE3&lniF
zht~SJds!QG4MSkQ;Q#00mO)}d?8^!tqW^AjC^C8P>FBeJgaPqxIT>Z;2yL-#3hAy;
zY<K)R5aw|8X8Hgq-Kp+r*C`_e9vx>J+}iK8z#8*g;9xLj8AN`{e4;BQ&9)D*exH!n
z0F?#i1`<gY1Y}9_&i~Fg<Q%MA>WOogF{;}4_r3mi{ZKp(eUNP-LnnrRhx!@h>M|;%
z02>Hayf9`~Pv0E#A|Tv2J=HLb8V<_}jjQQPmDSu8d=(%Vjr671N=~};MD&aFx#$p7
z@z6rMcX3qT`U9q&{qb1Q@!xN~@2yK8)*Du?+vW9pNjU}K41@W$6Z!Yw^`6p&0ST66
zmVy5Ah5NtUpYUvyWTR)7)~M>1DdwhqS`{^UizMuB_T*2@Ev@8_TFOJA^gPMWu6<<|
z%OAhmZfEo4nYn-i@3BkmFE{P8I$9QS<UAdpZQkCE<*h%wr@l^Z`FY=X>-}H(PyBip
zWQP)EpO~+Te(n3^zfe?B1A9oK4N+L3)Asb~)5;8;dwCdb8gZrolby3IO58q-W@TmN
zV*#U4XcWn@gBA8@XbRw=))7Xx9*U1RA3uJa*K-~dU?7(I9=Sn`;>O~lEHoOnM?|DO
z;>E+JiDgp5JxQvvD*m%|Hu;a9TW+_tu5{KeFE3xjTnVdu6KThG(Nl(oj;A?xJQbRz
zt^~>dIwJSC%c;u(n4l5_%A+kNFg?x8<X_-;NaRu>5@DJcOCkZII8x=%%<#t;oe|NE
z)YL&FG?c+g`uq2<hWp$JYtaK<RP+EWAMHLnU_BpLl;Cf4g6Du&WH?psUMOJPJuxx9
zZsSk=!_<~&$H1j8S@AMGX&ME~Al%j(#|o1x)N0hcf2dR+cx`5CYU(t^$4tK@lt_Bd
zV%ZEWipXwE-rbZgp%yFJ(K29o<G{qg<MYx3@_W9I@~`{XR&l!h&A3_#G#^Ikk!RLs
z<DI0E>QC`bpHTY#{X6Lft9x;W?B3_UAnX)rKeqR@`rFDtH$x+%8zz$US)HIt>IBHV
z)6URKjr#f1$wZR-_U+par&-QwE_pZ2=>(i&^@5IHMWojc2~8ee(Wz6XrXwU{)Zp<E
zILN9&U*;|!U3U!=Z1_}Dqc<Gje0Q;^+Nm&V<5kc~p5{$lIe>}SBm=IJO&QrA?jJB7
z=a4wMefxIiE1M|d!+&(<0vFIe&{M%o-uM~ufLU>;049-YrVur}eiA9QW&gK!*L2Z}
z{#*|R^|OEf{76vsd;RI`tuWL*=vZ}6VvsP)FJ4Y|YXwnF%Mv7+e+{+wljKD@Pd?t?
z?<+gEh@MRL=WuZux~OwolN4S^+QNx<T!!MhG<(+KwL1ez+P0SYl#x-?GJli<&@vGb
z6{`)oZ=MNF>n6!DjJBk1AQfBeDdf?!V`TOU6Vr8Zks`+Jc<_-IKb%v&2Cq{p5~0b!
z+2&wD<fH5c_O}JglB==~W*+^8UW=9r3tUU3QtGGeOm*qXxWhx@2RYd{Gkni3frfa~
zu7BV8p5=hooX#9eAH=g+i-<t1Bn7YOtw@Z73M=UN$m;4LpSC~MXrh`<)87arnWS%K
z_MTOv{rZ>5eVs~5{3(WF<pvjzyD2hQ8W^N~xd08@R+zMdm1$sVn%L2BJi6{=;Csr;
zp&2jj@|?};2ICx*Q`<KNKkG8JTI3CsFupo5uzSajy{lL6?dYWG)ARktxrlMX=T>UY
ztk7XX>KY>fmeQ7M!WlN_xW10EknsgmyhY4@AmfQ-8)Z2)Zhx;K5+hAq&3BfKF+vhF
zAF8UVvd?T`3HA7I`iiIL5SVTx>E5<MSA`dr8neUVF{TNA=8b{zA2jZsJ(*iTDk{9a
z<?M!BOX@y&{$<YXOTDoH*MfEz(P1P&FJgF?)#sOv-O$0Hc8qFoZ@<(}don}Ypj5P-
zlBhv+$PMF1T@{F}ZWyn1JCgVykK6Sd%Pge=ds(+)KGsDV%kM1ft*!dHsZvfZ&S5XP
z!)x$J&Q>0`w;S=`)-Spgc2n<#xzXM4$+{yv!9A}EUo(0+J3Alsb%3&bkPsQy=ANtD
z%+CI-VbIm8Gt<y4{_t&nbCh5*Le>~-P?h}JPdw@CMLiXJPXgkMS^D_#<AaP0jd2+V
z=FAN^CdgM$oH!9oSi9jz)pfxYhBZm0YwNuqKC@tp1rJhNUfIC@eH4i(YRW66Di3#U
zB`8|hiT#+QQ^=!5CiDKfi8%1Au`H}!;r8jA_tGCF&+q$>_HdwNqd?)I+fM*W0Ds&V
z9QdD0nZdXlewYUo$1h))Oz#fPT^;<r^m0#F{WLz&1(hW6iKoD9c-s&szE@hB6yF$f
z@bANe?XN{e-p9)P`PleDMxG&<RQeSI1wk*PGQHd5p?RCe^zKRV>l0Ue46=`clJ$&l
z2uF0rO2nhLpj%a!Tpop`pM6V8)AACpCJ4m3sj8c-ygrE2Sf;OBBGa(PG@DA^{pdUU
zKQJM_@$n_Y5jZf0;C*isingtd;6tQneJv`(@V%Q__R2K1CS6%W;QQ=@;s(Oo^=^?y
zR<n!E(0@=}nLa|QgljK|RF5yI+i|^+eDW%{>_=)oS9Qa#^sw4M=<G<;L5G}J&S9y-
z?OOhMeKJ_&&n-$~uWCQyuIP{ca<LSngFn1Xi>w}mD+wDZ;qR#E&PNV%Nmnf-$8U*h
zJ*(pmSFb)+?{IsH=le3{j~?Ijtjp0yEEf-lQJA-~ZG0rOFhB2P5>_&!VQ>o70^P-;
z|NJYq2t^+qcVi45XEj*6Js#EZ<6~ZISao2$TTNhB@og%X*-Sc-_pC(^$YlD;Ku;Tq
z(KY-yxyRs!!9<|=`^cVAk7y0P4>K<=yr(Zc6d69Q6!TXh-J1H({5(+tD?r^DT(uR0
z!wEV$ASh;|Ecwy1DM}uUa6cBq^NzHV-iW*V`}ZoreJn;hy4H7cusRfGv{5e`1Aabx
zB7LqoIoWBO$Xw6l6&(9kAg2$y^5&alJFu*XaU5U^bHi@3D(D$-R|+3#cM~bxaa4}0
z@k)wd+3j6druh-%!amjo7F7avok5h1>K?#3_yGQkgsc~NOyb;kV~9V|1s;-L@59?A
zL`O($35O3D_-ky!L=+*<jVcH2I^`7<C<IQok(N3^*;~d%=&<6<^6ImtyNKWka*naJ
zn-f{Fc6z;)Q(lTEggN-W4~*Z&PieQUbhg0BOk74FI)beqKcb67uvZBW8r>(H@F3`h
zb6#_bO1R*GD-WJM+mZF1F*p!I<_O%Xz`jXsD_dZV4Iz>Y;}p+Bu_SQd3bRzjip;Ol
zsW*G$e-joVa82=ib5QK;gWOzJ(0xEmme{ewy7IUp11&XHJ56O^k-h?CCzO{Wjcz*!
z3e9p9&-^$i$LzJknAxi-h)OC(shDo}0}RzV%%x5U7s2|x-uOLu>&B<gpVz|IlP;_C
zCXArL6}K<6NNowlV0l{1=1@~tr^`0lg?-)egKTTNO~_|rFRAkR>C*NqKixJseUmts
z@psi9L1f#>=E2GF@s2!>-!DugpV2s7LI10b46X082i4!*94(sO8QeIateSRSqF;)c
zAu>d;to9EyDlBkVS0ZxSOSz0s!s>QA1cB56jD?znY~Qu%w@|crr-~1xaE28NNSJPC
zS|faRo;`Un);f9XM%>>y-6yghm6($UI!)Ro^^w~G9hKN|6tP%|n|MY+old)xW|&C_
zymoj_AY<vXYf6Cf1;2CMKkkTfkhs;?*Y|O|5o}){`Zs;63C~A*L^NztF;?4fx$k|X
z`QAx^pN66wTID=tmM(1;7mtPvE-Ztvjb+Kk68i5bYj8)Jfh~<jIH(;>wL06`**B*&
zet-KRQI?7~_hHV%hcv9Ji&nb1r(f7Hynp|mL;_-K?Rtps_mZ#48(&4zz{h7N7vZfG
z+u??x2saJ+>Ik1G!bzre*k=PZ74gnUuHW7<te`Ob^oZqsGv+`b2s4tr{<IyHAOBYF
zHIpvZZ#?Do(QGSRd#Ff6PQf>(lPE(+a{bk-XjtJ1Kr3+23Og-&$pc_x?hJUBj4;zA
zZ)K~5=`jU_Dx0;HCBFLxK2S8UGt(dSyV(9n3&&vn<c8KXUp3S3kN-^cJ@&<v==4{w
zxbbdsW&4p(c$RFb?!4^9Hyvw<gy^Ja(G0CbVkE>CbFYk+ykWzB9gLOlFO!j@DQ=z?
zQuXzE7db*{zf}5lV~83u>SDL=jB98R43}%m<`z`e&&}WW>}$(rIm|Yly1tZ3ittp+
zO85#u@mb&k(COF3=^rP)d_F(@wD8d1<z*^gfp1Y_d~a%P_EOTHz7NvDPC_+EMPgN+
zKJ+;kwEgoxFa57kOw3M6fAdDjyy&7JyIaHvQOza{ZATX}E{X0u{g2?;8R>!T@^^0E
ze)>Y-Nr&U#1>5ITUZ$pec)j>dBi_EbqTl5wGHm;(6PH{|EHe5&Wp<Vj!$qs?3-CMN
zC;kNoODOR@nXk%hc%M9p-{s5qlRr-AoOtQiiygT|i8htyTN`>Jf2bb1Qw5EK_XdVo
zQEPOVH_~4-DZeT|0w&JSE^lQa7$Wr_I&y>t**@*zLlRuD<Exv&7Y$Ap_16-h5e8J1
z{Y<mQ*%NUVNr_qfAu&$Ikk-CF76i)n_7fMGC=@QpMfSX|guCeFWS^Xa%w9~Q-dZlN
zk4M}-SN?ml9-cZxuM=SxGXyD80ZWax{uhoMsg_?iLWY2X<l*z@EJSsU`QL4QeaO#$
z?u{>S|8YY5zLLksQPrgeJrMzxbL-c}PCegmj?E4_F0FKLO<KHc1{t{Bsqry~KW;{e
z4mj7h_%8~RtdKz~ZYaYvp^t)PW%<I1$mI6Vy}IkEFJ1u5>f^X1DMdT2Y;0!43-R#O
z_ytwtXxV|?w}z|CeSDS^&cyk0em?)I?6v1v(^}YePeVmvi~hFPSX!u?<qRyETU%QL
zImLA!6c)-~s4Tts_4Ug};8V@d=vCf(aGSWD+EDrHbbCC|6y&}+jazpzz8{cD<+Dah
zEQ7_%+i_naR*x($h)OCkCb7#jQ2gYMgI70sHNEVU8R$$K-c$>PTWxYKs0~!4XQkgq
zMEHIQA$x+Ow}X?@`{e-_tR;m@-)NEi&YH1AGJC~U*N0J_7C+rQ%^!*0v?#|=^B-}M
zp?iv##fW)S$m11DAP&-q=-kDP*{aN#XH&T68uKtXk=PuhK#VkTHmW1AU1d1gB>4rD
z6ALKl5Q!y}23=pERu7xqm9I?jk_KLVx!Cx_N22;`2osOZ?~$g=UbZphC8Qex^zT~`
zfH&lJg@cWHJ3ij|G@xt8LwwWRN4j_16zL0C_X9ivLjQ*(`Qzg-qs1@nrn0UjM{PD*
zG=BI=c`%Q4PGJqL_JSc7i3FqWhO0|2B!_-HNv$PZx!{(6zUPJd9oM4ra+S@>(H?R8
z-Y0*(c=YVq?e<McbkTJWJrq%(Qhgr_vKb@d(=)yRxRU?Q`^WufjG%I%$}ugV58gU}
zinJPg;N%9cuV23soe)q8;9#$D|FRDTju1p%^SrIe1^<r=P;{|;q{Ob+2PUQP+a@GI
zEegGeMAc|E1`=oPJBuEqc!+HT3_?rl_5Iys+jj`M7K)5<#5Vp|arv{R&E8QeMCtUS
z?xkag^yLw4%lnT{3bg4TLJp;;JVT2SBoPE<36#EVDY-d0HPx&Km47;K*Me)}!@lpc
z=tKVx9mcs;1<0%}n-Ka4?iE6plpqw~;R*hhzdYKke)8nW&#C{u{A2`G?w3iz|CO|r
zi?ND`k?b-4FwHi}F*1B!i|)%#u6<NL|D)-n&dSZLBYHMC${v)I@By26nw3?J?+QEc
zt9aX#prYyBj=rk-pN%3kK_miiO2fZ`g<3Vn03%T8pFSOlKM~GX36Op4nTpaRKm2c8
zBZ~ZpwIR@$jkf!IP5$^}f#Z;fDTJm-5I(OM|AfL8Ch4=+H{!vZ%~6Pwkux$d2m%sd
zjj^A674|5Kz+e6Tw59XoM=Q`36}wL2i{N!=1DH!u>Kf(c0>+&Yt{rU-To{1-J&aG}
z{xSNpn)Y(02xc!#?~}{EuDKpNt|`gAD=F>}*$LyM05raktRA0p0stVvADMCS*RS78
zu9)*nJe`?KtihCexaHqqV+=VFU2-xDQXJHn)<s{e^Ih#!zKbckF-SZQ>~bo6`@>xX
zLSOFinPW8Rdqp|IKKdyo>z-?)D@&L7wz|+Cx^<&#>DQ;vp0y#vPbliU%7lwV-V9P&
z$7_N2`f$U+@khlRz=DAA{>-Jx!))w{nvRY;knxfpR8*u=<=!Mh)&oR#5+M(n^*pY~
zsm_j?6norN>AqRKY?`u=-cNdGoGlW&cXt+xh^!i1Dc*IUKRiTKtJwVqN8sRcyfl!M
z?YUN^SLKA+4f2uEk?r)~=RLv~9GNnFG$D7YZ)>=O`Tt?;OTe+*+J9e>CKXYc$rKVH
zB%(4zkqjj=l_63Q87fm63>hLr2!&)8Qi%)+A*4tmV-k`%na^+8`<!!~;eXDzudeUf
z`>S5h`#kGe>%M=}?Oe!#V(jB9I!zSGNk<Qgck#0tj~`|?bBz^}t~nNzm@K%pIz5AV
zlyzN;JI~Zuv@Q?N4V~Nk)rHOX_J&1%l*D%kl*s5|H%Vwj3R^Pg*4}0wx+{XFHPh>r
zOnSRI{ngxYj`-5mblWG1&Tz7_9<G~7nio3DEkXWfdAC}uh+kE&{C(b6{(}Cg;sVZv
zuTy?p-Y(wt*iF1kN+D}bd9cfumM0wEa!8I6o;_p24&_l|Iu!cut(yWDFf0oLf(dlm
z32Sp&AnyPrqr~&*Xx4uK6wXoxV@qpm!Od!$&8J0Czcryl+4J`Wj}8|qd_58%0UOo>
zIhPIC?+hcA1ggZ>-zsBRuRJ(<^)+rb-9=y3N9@Kr0TV7%x8=hW(Hd|~Gx#=@Z7>mL
z?&Ggljb!hLUt{_1$HdH9?G4)g<NG2sOKmYuCO;@GkV<SIq|U1L+%=2cH8E_PqRL#`
z4EyGdanQdU&i(p`PKmST)>o?R#?R&F2P_r6zs*%*(7-l-9t(rpRlSWBok3OkT)lSJ
z*9y?Dwl?-YhlgHi3RFQrCyOfk)4*;3647qXdr+NE3evCs`J*(0-bgg^Ge3Q54JxsM
z%;WWSE(J#DFg_}WKW3G(T9W+!$SrlIn1QBja;~a+&c!-#s8}2lcT)RPK79GI+NT_a
zZn;5VYfGC3o`AP{)?X`BqLPbWE#3*NKG;~Vo9Ji$qOIGZ04YIuQ@H)ls2bbDd`phw
z@qwN$&ChCR%4Rh<MqyscJi@I*%G<)sSQ$G`6tO57w!NsPb5-Bi^s6h}vi0uJ8R<|N
z=O)$`ck_?Eox<!JV$Qv;>yO(@=bCEvp2^?h6*XeObyuZu2Kz+De7t#I+?*B-LqmV<
zp7{{IYQ3S9ugs&NoLCyVtBTIuR8hXl<KOeLjULcnIwJWm1b(Fw(=KjuRb;tj{Qfqd
zx?v>6%+*wmel>KdCS~T_S07x7t$zOeBwj?Fknv4doJT$?M@d-vf9AiBNoOeVZwdOd
zH||$f)`JaorAc<=wvBz~>mSJCJ*1~O{nmHq#}}xEM2ub@8<G$_a+S_?z#MB`FR93Y
zwtr+3OIv$^<LR7}u1}Xg0q^fi3>zm$ut?0c$Mo7BVN}^Bugd=UJ8#)C7@~X_>|B**
z9Z`z|t9<$GVmnA`tVu#JhcP;jV#hY&*(@aIDvN$B<)wOP*xs=!Z7%ZL78W(v4(#e`
z=9=h8y<zR`m*9I$kcNVNi1*@0{C345mO+P@(Hw}$YiG$3Om~}?bu#b}o3b#>-=Cgn
zVZs6TgQHJA_kq5xWF8H|f#-uD3>@7W@d_NvuN(Pmf=a)-nq@dI(5C_lHAg`Lm`7Q8
zZ<*J$AvQS@nbANxB;R-Yzm%oh$$x!!47dcbs*nI>X1WB_d<eIDg9gJa-_X%8S)>+_
zcLj7tJ}5?d;T7EpKzfU)kuN0Y0jC;Lb)Sra^XQoz_=FmEbZ#L9#}#JEhyPI(g>IYF
zX&*(G2>k`v*<p5>Wu_Ww1|+r&7a=SMpb_r|CUWU&CI4O$aaDmmZ)7+th^VFyaF{L3
zP4`xBkAFW9I*$UcYh>~FNY>2sSpCh(>1p%c%D|TSYe+joDu@?)d@~e<D9Bi)+#)Pl
z5(JYMIE@TYX^fY9x$t4#p%8)-#5chgRba>6F#p2{RRD0+r=qNUTTfRP@@`=~Thv3D
zwU5$At<BjHHHDA!=2IuJndnR@Pd@qK&Jz(6@<=Q2718&HP32~u*nt8r$93!sSrvO+
zezPNhvHFtf_SGxM|Bv(vR9Ne<H)5rtdfR^VK`C05OAx>j9`Pl{0ahB-w-o=8YDZqL
zK?_W=%DuXA8kVJU@B1gv`cg_q*Vxh)>1hBw^2U~kGg>1?I&f7L7MCC}pNlU}5tuJx
ztR(3rjMp0z1kVh7d`v{}MxtBMMmRZ1k+1-i7`tU;67%vThb%i$5V?L2IYjsojW~JP
z-dOqntp68o8}fVi7fV6Zq{tV3GH|>$#v8O7b)83l?5(dpfB47|-C&3Wq%v+=_wc`{
z!}oPAH10h|R-VTF(Z}!`{Q3QD9p%f@)8$eP*W}~$R{a7vwEfxh%NJh_2vF4OUh*dN
z5NQnRh>ekArmD!8#@gS!FUb031OU$14?$JE*bQ0rm3NEdI?&X6b(^mS6Ea9TN>Sri
zH&J#3e)%0*9INQbSyxxL=jx<j@F`!tH(oM<r0@S0XM66P>OwTMQq&ZTTH1TsDe#bN
zd8A~qufNZ@PX97MF7AWiP26iZmGE|;FkzuTv^$`_f8sG`-knJQTjQ^@ep!e)1A-)l
zKc=Duq3qm0<nYdmT!G}-`YxaVsrLkvJb~Om8H#y$(@WcQ)&xs`B$r8fn49bH{>Y9e
z8c{1KSTXH4o2BA#J1OOlF29-){C_m+Md?ng4T?+hiv#hC>Yx;|x}6Jikn}|<9r0Yc
zb{;+9+lk(9r6H<FJaR<@D;1sHDqs)0vwQ<8N2ZZm?3=I}xF>}B05lrW(U|=WokA~Y
zYs8@4Ep%x63_*4pw`w!#>az=LEz*Y+2n8S6vo}<YgU<DnTAYp6?U#w_v?+M534-So
z%*7BOz3iCATp#Y?mbiV_q^Dj&B?`r!D`u=~o0<j_VJa#N_XQ*#SIU3#z?h1jKW%mm
zS*{$q&IeHzm@F#2!s*)B(y|J`AznjNZsPs+ZOiDOhW)L|gn4-_nT7|}BF9PIK$^~l
zE7-A<_pBBM)YCSsx`!QEmA7yzv<-8*l-zh{&AwsP;lqb<2qH9+LKl2?SM3_5{1<`0
zdJ>IN7K+8^7IRE0%LNq`-#@ie`l^1>){n~0u=<Gn$3WR*w3JXj8`jaop*M+lg_-Wm
z)L`>{BmkwD_K&%a2HLq%qw51&NKZorQlu)Fs|rd(AA|?6X(=x32D<psqeS1--RXU6
zIcOMSj0u@;NZNX}m3~v;6O6z?Z-HF^-vcZo*VNE?>6n2LujrBR)ip;30i|%_X!*oW
zb6j7)7fsx#t2}<r#TPMch7ufw-!{Bx3j|R8WbwlI_v+-<MTa)EtM^8VO&zXTSY2gd
zsn}qmbEmv}7ok(!GrsK&pC3I^-^~$dQZJ7HWNZw*->dI|J9!(4IXvPe&mTSNj`rLk
zBvdxuSF5eB??a+g@wyOB8A0URYNc0V$HdA=eEmAsQA^zsnC5-Svm3m+DV7;tk(E#6
zPgiBR^Ehe7?%j;<^Xuiq1SI`OU6lt7f)&ISbc<K`=sy^#OyGC7jLTH&<G&lEV>I)A
z8yiFb4c<_+I6FizS<b6JA3bo1_}?4(xyNGeYhX;^m?{88sQYaX|2N}eY`lSrb9su?
zlzHU7$Se}RD4~Yu)$Q>K;cv5SpjGmNf&@B`{J}AU6~U+ez*OTE$>V%=gS)T)*ek|C
zWmU4)Iy&OM2461xhI)bANoDGR>(L`@U?uZBg7{@cRc&kBZ8NKkLUX`9>1Mu|Y4;7M
zvSZ!6c{9Z;D=SO9e@MV?lZ4}s_iHkD3jNn;z(<d7c#NMEI=_93)#tdH(0)Bu!F|@Z
z;C)S}(v>T+5OT>s%4ZM?kBof6*@?3QM``NPM%o*D9DmSxbx%JtaUPv~e=a}yVyH?I
z;uNf2dFlJbj$25>EckZuY3GPc^K<>c3yHZxx2VMWjsynTO9FIktX0@4=c>L_K5K!Q
zx??w>B)ks@Vnv+uE<|qG7XN@MmEr~Z;X*<!8^do=qnGqZ1~}38s|5`s72VSO<xDBt
z&t^DYQ<prkh3LwS5WUxxRc0T<1pSwL-iHJt`sQLfSO9fOZ(HZ$^AnF*bR2wi#pKFy
zDv1CB7mZwpj3Fl{Zg(`?XgV0F-4t1TOjM0^oI|&Zs%Qg=J}jjYs$j2qSn0pS8=^jZ
zMGrxMF9biGDf3<lIqx0e?u*Ez6@JD!ByeH3jv&+FeSci#Sx+3?-F|7%p@2f?se0+z
zevi=jbf_isaPZ`w0PWnGch>xOdx8BPcEiX-!ex^lCfTAihTTuK68nVx?4%80$p7gz
z^t+>X2jM2{r>80U^VRlCf#cb<DU3B{!;J#k3y7{fB~KLNNWDnw3#C-r=P){#jZol%
z_aKw4SbPc*(&-B64Mu8n1>AQcOjvt+fSt!g48@D^*1AuQ9?!oh8OkRlrC-<QaLI#y
zz`JwAr*i~vpA>b6ybwf}NMzQ|`sG~sng1szwMos&PMOLudDYV|ZAEHIwgv}I`tY(=
zVjCwZFlNAhaGBek0sHpMeEn)D4<kilv9#=qOBqoRFwrb$Q1lE@bzv9ay=d~M>&8Y>
zCZsNv0NNYe^b6|fRnWC^<t;*C;{4VcBMx$0MofTYxV&!xhrU%)Q`1~Qi))=vd1~z1
zjx`S^&p<h87-_}|PUN&KILB9C%8tx{2zYu@1o6HcMYIy`vzI{#F4ocAu2P+?^pc^*
z)a2z%P8d~4EdrPTTMbFKyi~>E^1V4<_OrppzLF<xts$1O!%_FrP1#n{j3#m|tS09;
z@UHrJMdXlqV?THwugC>+cPS=>|2+`%2>&#hgKU@JihC>o6ooye?lx-FgY-16uT3&*
zw^5TJ^`H5VZBQ*NVHn`Ty$htWyDcm%3`vLpPEDcS7Dz%i?a<F+pSS`l5~PWx@ezFG
zquEE&z?OW?w>@-ujqb&;376KgBNhr1{SB*j6-kyenuY~_U-QM{EbtQJT5z-IL6$3T
zmae~h-OWpngslQ862;bkPB#l}RuH@F3W}hZSQWG>IFiXff{G{7Hzx=JY21b)=E^bW
zQMBJY34DD<63(Mml>?W1KnYd)aCxLS36Xwc_gdPb*$S^n!HuZvOGjfvuDDq6@Ekr<
z8WRiU9>~l`csC(zj3XTb@=FCn_4kYm*v^6i6x@i(j!+yy<zN^IAI;u`_JO{X;4uKt
zs0u&^y?V9xCgIwNA|WB2Z`$_WbWE2=mdDe_Hw8C(bC>oxJbLy_Z%79<`=KvoF&uX)
zyn!uJ%G*xF1nyf+uI_3%Tjs#l;63D<=_S*-t~~JJ>84(^5}Nz0N{^`ajkyxh<ry4c
z%U6)JH@!J}FH3{?`q?+H&ypV-o^wOy8t6cB+JKQ0*@E2mxhuEmSPp~E1uI1;cv;6E
z0efuF-dNCLLAAeV7D9Er%gS(|)m=puF^0I&@|+F6o?B_k^o9gOI^N|zxSH|#l@)sb
z*Byo;TORDWv;yU&7=;vRf;)D6dMTXn9y0qR&UpL|<T=t)d(-D1@Jg%$GeHTq74p1#
z>X56^oH+S0P`4OHK8y|x-GnB(<|~OI8ss#5e!kVWA1!_=0`TltMLg5<Wmu0tncQ*R
zsqmXCMyOqa6$!fuV%ody7Jt|Qz{(d5v^E0B1770Rr-7S%>U*w?^Fg*^Hh5v@S+UU8
z)!t@<PFt82Xg}bRZwUHBh=gD5XP*=S54!L1x{it3MEE{h8lsKh+tm1qCLu%U#>4c^
z@?{kLr}Y;DWkX@*RU@UPUa8a_Fks=&K2UIH#ZGYpG!;&!F=$?Gj9D9B4h{`<Uq;T~
z6Rmh4HfDqSE}S^GvM5&s_2RqS_cLu_o7F?nkFB^<u>!Ii4E|8a?yJr{|L^sEMJYYj
z?ZdT#g4vU#fD7Eg!waQAvZ0;3f@ytr-<Gho0<3}O0F(e{019GR*E^OZx)sCQdE*#W
z8uf;(AEBW`p}~QqsNCF5;o;$ik;=5%z8i4a=Rm!@dc}&R%j^76Khw~u&X=Xe_|TzM
zU{}{s0E+eomg-?FSZGr(UVz=w(h09#eQjzp%wN$#dfQEYxa}gQ#YW+lI+y5G7yvL+
zyddHzqBbcts(PB(sMP;D7m}p3)BF~g&jI>kCwh5Q6m7#t_iaF<apH^-)*~`RQK`os
zr7a4}86S5b0YY~zWwSZ*&*VZ}jmb|mxO9wkLh}Xs41MDn{rJPs)tGiioy9m=;@Ei!
z2Rmc{>}Xj)T~kq{W7eC_f|sbiv?X+V7LX;8+WBwVm99#<<~q_^fctbdPDl9pYWP#x
zErzX%VxQUHN??P}2WlRGIzE1UGg1mrMU)$V+}Bsvj3S@;n3X!qd-c$_r9KjP_hyR4
zH&~X|UXTyD`cj&KA&QB*V_s<EUKW_m+3S0G6QT1cGut(_WNB&1dcs_z#w>_R%rcR_
z_ioB)7%waH=%!R3TSn~-z;pKjZ9ruTWZ0ox`4y&baIaorVNh&ED`pu?g<(I&7wqln
z(dd+p{+jz9_*u|V-(_@@r9UrgfFDPZ@~twh0}1SK2R=+sU$<4%sH>i4Afn=oI)`<%
z&FH*x4=_ZY!-uZi(ztS^;C49kR-AO35A9~wmFU_`bKO&hXY<(QHQF2SQhRmBh~l>F
zb1eJ@;y0Bo@|d%00T%<qV@cGUVd3Efb|!#`1Dzl0p~F{F6@#k-cv(gz+WB?xsR<$m
znV*~LpVan!e94+H`Iob9BY;L<Yk9nc_qJ<&ennTVL+e;m8e}aG))=)x?HFApykZ^h
z)5gdo1!{@y;?Pj&&!tO&HUA}iGTw3jd*Lsj>GR0qc?<rov9;9??*ta+83=cP`RS-i
z0cm#|IAV%f0>CntT^dCm0+PBHeO-Mnf0XOH7Di?1@ple%fKL3ls}CN_4N2F)LrfJ-
z3R2W2=9_4!V79^;Tc^CYTuPW4GATj`A3DGPe}heCZm*pYITc|^DCJV(V`EtI80%4)
zpYK<!BD38g6F=L&@u)$>*y|B@F(-jQxf26Ip7Hbhd0(ferFmENqNS?D4u^K}b~%t&
zW#wCs=?KP^3mmpGv8W7^qn4+eYJBRfl;k<N#-{m(#k)AFKyRQEf)a<shOTFmE*-d>
z0LUDCXE}Z#DgAd#NooGgg~Trqn$tCS5y&W7f~ddgbAT!pX=m#p=4eO>LV1a4b}LCQ
z1#@QJqOAXVIoE16#MsMo27k7^zk)nL-=->t^<g^z&-nYVRSSLFD{<p#+c5aY{X8LN
zf_7##UiuN1(iJ~xT2e0k`VRmS$AgjBC6+IZGCDue(5o9i%hY~R20aDXtqT`+7K2C+
zZhI{U0Rq>2*D8|eaU#<&e5&DC0BRsTd8ncQfCCG;(nSD9$~^(iHz-t&nVL4RHEW-!
zsjVFY?Y{|wQgka5*QlRDOQ;aP5$~Z<=X#A+_c}($*)lJg_0T$TWWqr3^2ZhIx)6r2
zmD+Fpr7Q|&3455+(C0@uu+v8}Cy}U2!6GIach}GSKmz%NIZd*+qE#i4fX7aqN_UdA
z;6}p=76}uIAUgry@d;ETsN1GjbUZpCEkhNk3yE%T;)@(2E%wxC@Nov)o?Ig&-lduN
z;>A`1A|BJ#l^kplF)3oE*iZg!TyyHVNbTP`I}d8g6{l};@>ezy`U$zDZ;v*K@b%w>
zdW)deQlJ120m-REJ!U`HB!r#b);OpeCc7dW!mdjT(~X}$CljVnSy|->%c#SKY!7={
zJJ$apXkcE<-fdaKL@9;5OW=Gf>gv>s4^aY=a-Bry<8zOJxn8+$iy7(GQMTXjRNUm(
zHL|ogl4IX}5LyI_<JZtAu;y`eqMOd;{eqrq(WE+rRaq6%+RhSXTHW_UnLwB7%AmLh
zy7WLS?#USs#gM&!7q>}D%*}oMrO2g8*XK*S@ABxl`#vz2tkpp=79aLTf9IcF#zKNv
zKBpD@>VT!Tyn7T?KASw$*r_eBx|7Ia%uFhM<uo)({8)*ibC4ix!+yr;Hd+=i$Wey@
zvQ_MwIaYP7L|`u~cFK-O2fHPY170G)dZA182S9nKR-cI`K79FTdtPcvir^qyC}2PW
zMXsbkC3qUp539a~e2x27pYqeQ!<5Zi`bC$kh}i}elY%4XKInoA9+Cn)klM5F@Vouv
z8p^K>j@9hf4^3KQ;Oy{B3EEnOG|Ogyl3l{+U4LTZBC?-R%%tX^JVni31eCl*<OPm7
z@`b>tHI)4@0MLZKmEga4AkRc$Z%VP&)7O8p;x{(Ix{ZP<f3bNIf+#y%i<FMOyAxk*
zFk!9S(6Kx8EPv?rLF=%yS$P9#)n{9)yO6Cz)|cl(i)ZXrc(1#W7WM|>IFH#+@jZxr
zELEH@S4@fvuIl~!+oj|V`kZ!5;$>-(i^jjw+$htIkJ;;#mb#%Rcg=`j03uO!GZU37
zFz1%ZX_Z}W9)x)7zy$*Fu*treV9Q@+#K+|*cQ+Q9D@Mg`ECg@V)Lc(YqM1iNGGV6)
zc)wWiuGdiKM%-T4I%lKw`EOPWhMR|W#ag)7E{hWwH~64sW{}`$xI2=2grJ{f#8c4h
zSK<ml^otOY3L@gGi6w&A36zjQ0psgPpZm=Ci{h!wXALecF8%p#`_|k#2r)+zLVlxJ
zr_Y?I;lA;X4q+8{_I9njEml4>IJkmhROBiyRqrp6_$wFrOv#-aQih>jO0C-9llarq
zlg_pJyoG`E!qwQUvH(F{;KJBXunk6cGHzp98+Gn=W=6*QnS6rc)sLw;`v+|4a=v_-
zLa{=}a=P|J&~daSTuw7U)>5m$Mr<AxI3(TfxwOzyBydIA4(B_dn3WW_rMb?k02WD!
zSrSHIeSe%!SW7ykYG?`&AD9@Vy@1!Y@i043fgk$Vo8ieP*)ygwSQ7VXR#3GgNde7~
zzR+*!yO(HITyLrq1SkP07Scc}3IcYYV@vroKxVJ5qL{~}9&>&WB;BviRAIZQ%jqSs
z6zKc@5wOKjJ{`*e=(PWt5zzf9e(~Ao$Mg(+Dpwo|pFBBKemLbm)cyQZt{JrJNK28q
zt4s6$1u<A=BgTdTiTt8=BnFkArtCL1{^rZVsra{SBVyT$wU_3qJm(Kl%D<IVdxEMA
zku6$UoC9f%{`1{6`E?9|!5l@mbWKd+W*6Wh8oRlB^8*m3Z0QD&wTFj;#AXN)H-2eP
z1#E%$Xf#8}UeGw^NPq;m2b(!7T6RCE;>}!(e_<y8XhQTXP~F&?{15TNi;KtITR~Ib
zz&;Ie?RA`*LQ>f>oWZ>q5>P=@V-)5~)X&HGHbyWn(no3K-s?P*Q`$daW3-+@`*+3(
z+d&W1r%~sAvkvIm+1U*$Yueb@oS6Rh@I3&0WV6aina5bmL)w7>GN^22$UY4kgbubG
zD1o>@QHY}2#wntlT+UL|3%<v-$8S5CF!8IsfD!m8`<NTfMPyHRYn#FopsdY45#%u=
z(uR&0@USaD5Hg1Y6Pkvivu99^_yTG=J8r%yq*~}uzF@zr?o}u*T>&X=TfG7=sW^}W
zw4pc5R@t2fI)?b}grXw$>-!19{yi*l|DL!<yHM-U2daX_dy3>@(&8TrV%6)waQP!q
zclfG~Zwk4}JZkk2!Y`7%K$tGbK}Zpb_Lw-qf4}SvDyr}4&-;b*qlY+TF&WaEB)emL
z;z7v2E*A7{ar^BUUk-VQL)IPN_bR9)>8i=2jrviYF5)Yp4c16VH+et(O@>4O<T!EB
zJEDvfUV)<)kPF2w>x3GL8=Re?e3<BTQl#iZ*hk&lx7zDjZ%Fv|cLu1r8>LI={a+Lo
zmIoqYzXREaq%^~{L}94KBr`dg7PqqYnyhk?VvU!~6&uMwt=22mV^U%3W9lkzY~^7*
z<~-u?=VG^@qL_-dMd^`VoqE&$KcAUFW-ga$JdQovw`t!-A#*k><n$$FrHigM`xvyx
zLiyH!>Z07%8owm3)GaujgIK0=F4oB08<ADkl_2I*o}wAB{0Y#$iiBBF_M`a{(2C2r
zq@o;E`07SvFReuR2FzIMv4r6T$~&wn6mh$hg}E(P{Wv)-woL~$ey=vCCt-8B6xaAk
zBo$@E$}w2%MxQV)+8jkB5PI#XnwF2a@=3Tzo<r9Vj4O~a;ML*sVPmjVV;L1Lm*C+&
z_FMT9t=$h?5#*Kx7285>(%bDz>lkXu0_@(H_BvV)79cOG==YcV6syvhmFR=mih+l0
zrv3MCz@{S#l7b+u=*O@L`r!b{JR*E`o{z5vT+dw=)aWQx^4O$EJq1Wf9<ATbzPX#H
zABo#5-zYc<A$h0X%nSbEU5|NckYGej@lAO5(1))2T>cr9y2}NcjZvVupqNb0RV|c(
z%r#y?2dFF$&M2Io2owj`_~pDbj>g;L0;LTX7P|3lXlV|peUmj{QVC4omf-gW_tYF}
zJuD{d$^<VlI7WN9y(L)#$1CJAaJQEZJ44oGNmhBOJ@v{%@htJV>*TiFu{pVhpQr><
zGcuZ9(KPcFx={VT94d<*4(DN_Ht*f388?_(_VTHFw+tPBB%II=YtFcrU|_Jq>}s@V
zWReNamhLgvXQJvZVXMrKKKV4+MkOpFHSGtv5lKG(a{jrV?n>>w;5G4HHe0UsHbbGY
z?`C&<r1+}Wm(kal{G)t4vQl#Q?j)x*Y1;L!JsdpMxZf`cOd?ol=Xl)r#wy_rXs^-F
zq_VN4<@Cy89`B8W-3K2XOS|#?&8>C!r7uKRZXO7hxC@}*pG%-Nxl{2bhWXwPa@_Uv
zh&|q(k2{I`sTP`Kv&wEat5oyVn1<ulM5^$n`{`$XCK{aNbbuagTC{ogp>;&A$3_YT
z(af>JRCasVu|RH*pNoNg0X8H1XEn#e=61DjI9B{pOT)_IPlHOw+q+E`FWMLaRi*Y_
z0&=67-|>TSbX@SdAa%{ZqG#8zGM*P~jO8G@R8#R#5f2S#4vVHB(?L(q5@CHWti32>
z2Mt8k$PYZB#)_)hzlXiDR;+{$EQ{h|uj1!dh`{(L+vj)SwhH5uj0`#ye-tkmhQlHv
z&f%lfp=O2D90Fwfg{c-IuQtXT5OUGjJyB}kdsk5_@|j-1=db+{@j#T$Snd`c9@eed
z^B-<^j%a;*Xk%}Gx9xn*Nk~;<+0i>rn4xj1d{opJ9pUI|!%Q#_pl_FyZ%C6x$gCY<
zi5CDs4k_+ZFV5}fE}f}Eh29H*a-7!VYcJ{A|M?5ffFA7oYBqzsG(LVjymE60gP;~I
zIr&bmAx9tlZ)L;I<riNBzii!UbAtc|3?Vgp5)R0rNLi){Xs|;<F?Hoz$o=<Xs2fCg
zIu#f_*ro<?1yxDsMq<QbnOq{T{N@nS#=Zn|1n0#q*g|3gW2h1iLWgt#piwspJBk;I
zaJ;G~w6wGa#S1a@&-b0LtA&=bp0e<c9oVv|&Hnzl4q|HU4UVF(Yayiv;iay=rpN?5
z^W7kf8Cixjw_D-I4VZ#k1o-ytqR*!Rjc-QxCWj!vyDsPt8Ud`JX#WGo&6MC@k|gc}
zI=!xUA-tp%hE5OR_c)evi(~`~L{2fO5ZH}0M}(urR5XZ&PeY>dx9N$(Ett6inO)td
zrsbRlnL6AiG|8VWGouGn2ypWy(iT7r$$X!hpbzo5br2}GHj`AwTjo%ib->x>T_0vZ
zTY`bp_%rRLQxTQTot^Q9hujsoIwe_I+1Zs535)#=-&|HPrV;?oZ6Gidg3~oj`%}Tx
z!NVC`Zj~qA&M!<)Tf}00;X(zDzr8(_Omy8lckcXB;;Bf+N$!cQUt|0S2IM<nH>aXK
z6>+ggWA~yWccmf<RsqfE-90lHP(tk9(JesryWzMXv9x4lYI3r`@2*WzcO|+2R02?=
zv%?0GmS<ZX%8v>aR6aSzVfP^a>rn7a^oigl#oH2(iV6(5>46`w^(iNIN($xZlY5Ui
z0)=EH*-GzgY9?{PbrUV{cN%a%ZBZ{7z@)3o2uBgbm^r7*X@S;(zBj{OOx_9SUtg|6
zJ-x=*8OT{N@(l@{1UA_l4_DCtrRx|EX?{pgi9Jvd2H*IAJxT)qv4mrM_8UXU;?o;4
zay9qtYJxNpG}=AphEdWh+kgT=e@4apd-exMY5rq*zcjj0->e9_byTC@3#oS9-5$*>
zZ^%&vdYs#z2T_dv$m;_Ra)6!&`#hPo=-qry#nDj`p=aAKx0(XXAz*UXtD&qL<=tM~
ztAPrWF0mLPKyS|EJOoPxu#$Cb?|-1^AjKj*b;mCp%sdieU;v=b`|GCgcfrG>-d2O9
zNr+lLAss~77M!0YTeZ9G#RPy#5MGEvBC@ltbo7pOfuEqIU|8}=nOf()v8geF{Dqq&
zoIYvFJRW3YW_~xAZ-rs=I7&td=CKUzDuqSb?|E?by_bvIvKF;|9WRdr3)qDH++Hk_
zR%SI}LWYsLqzp(79SOF$$2h<SoJW8bbS8Qt?88*to9~-kz`z%Z7h)6!ty+=m*<2t8
zdNac${nv?Fr>v)~m^_w%ZPXJSanQepv~g%a<f<E#ot>~7vuX9%0KK#no>}Z#Ul?$p
zKmm$LVfMJ+hkIdyqj+Ig6Sr#$k#F+^rB)ZYvsMDr-ImEHbdZRhkjGAX!U>N170erA
zzg={J>i{nWOuBGT3^t{j)3&H4(m989Glwu3FL3_wdoT%$rn2FYj%R^_`Zd#9<EmcD
z1v+!eIe3C?!@gC$56X_1gAQ~`>^gY!7IK-JdT5hCR36N2J)dlf!2{5*;02?oU8_DJ
zK@(VUhF(RyccVsr!~z(`E^x5Y#OuibQ_J6am*RV@FYczoIS1aKngYN0^-Cp-DBcF{
z_wr)rxfk7Udl`nJ)T$6Q<?_S6<4@c+t^?n|R+GX|gQ{!tZU4mBYqcxdCN<|@v}wLx
z|M1O@x8eJ%y<|evpwNpmdu+?(u$a|49Y}Ry#R=M?9}z{TZLm?c4As4=-eoEpSvxs&
zA=x$TuBt<C$nW!B5|@t$$=rWuc_(2{&kSGzzyVfHHuw0{iAxIVTH+xFx??qOn0F*@
z8!bSz$|nS9Jne>!8or(&iTTr8#R6~3oRs9mPa^@6M81m+iZ{B~x(A$5phcvT!-?P4
zVN~kUR9=R)ad;!3vWlFYMQ=~W;1P1v#B;%kDC&EiH`o3jl86A0+ZaQ58l6I-f{a~<
zJchytL?pV@>=*ggdFS2gH{H7+7V<}x`2K<Hr`<yXNTRHto2s_NyR#2anwyx8N!VOO
zf?)Jjw*9yCQ#N!YxyskqUL0V(MYqn9R4NI%S8NjCpqA<Df4*h=k25*<Q$M%4?S05H
znhX+`ih>%*+(R=r@sm}arBp*QeeH9@Oq4&Yi+}8vwrt!tNsmtb24;?Npb=!7kb<ii
zM~W+ofgB8ll%M>d#A3SE2|GiQyGO=Mk!qs+k@sGa0is*(bax^LAm`$@y%~o2HF|7H
z`9j$NW+|#yOTA>s1W_DW7>@d8yTX`XLeb<YmUn>WE>a7M-CFGmuCrHYa(y9Ckd%t`
zbUSwJAOh1n70jd2cR$Cs(YY#u)eQvdN-38-SrvVK*&=aY)-m~F#Lm?#6egk?%y5m6
zNWHyJWhc?T5B6fn`XQTpOI@WcsKi-MkV-t__m2V0kS4?n6i&S@@5gNzJs#h%3ix4l
z-9xn{94|L>b0uK44YJ;~I<MU=R`rQf=@BmT(il}Z>B_rNuqq)@A0B__x)$Q4qu3*s
zbHJ*#^zVdMc=vBql0+r-TRHH+OrgapG3~LZhM6ayno;qo_w@6ItoggJs8`z7(o*FW
zspbPE3Nx*?AqdZI`=^XyYTKDd2Und6{fK@Y^F3%OzdByeMfXE$V{)Z$Go=)<3YYWF
z1FUtt0_(Qo-~^EZZQzvmC5R-R6`k0}V<sMz8$}X&aKIfiF$t=vIZVnUsLDRTwlNA8
zuDedA6QO%|Kld@c207Jq@-{+kiuy0#8xuU6j?^UWMUgE>203*N11OJz3C51h2Q)V(
z*M_CgflF>4MgEXEh&||82=49VG$ob252x?vEMt;jk@o7B*)Cs#wR1sV>02re<N(g-
z<q0L&-C#C%8`O<w9sByfm@3Y`lnb(0Wycf3{6aDsOtlcx+DCwb;jL`%Zl+q6RIW2~
zZHxQ$<8cp31%g_&mmpHtj8W}sdxRPnAuuA!?>79S?EitO!n`9@q6J%ouW_OC>6R>G
z^Uub_vg|%`jnH0;3;&1V!l)7jiQM2%-!m0A9C>(ZZaX+j7Sc{_qhB4s{0Tm}v`Z?x
zL!f!av5GyW$YexnjzmI*?ei%QdDUnj&xxw8>19IC+yB2Zfhwjp7ps4)oNJMuj3fZB
zb7^t@3fOW~w`9*C!m0}|zD1n7$8|@f)@H<`_D17rm_I5Al-Z56ftFWHJUkNgmZhhB
z3+r_MfFD>^!eVZak4SB5&U%ZBsAk@{AE0Ocjpv>EeNeWD&NUx2fD%+k|Avi(joF^a
zGV+6GZ;M4M1SA!}l%K-y6ktkC*ZGoLHa;@9e5MfJa6QiJ+UChJcp~~I9+i|RN+n-E
zV0sB9l3J#ZLcHDES=hOi3y>-@dHtF+{#AKpDZ~2NB)ZH0Xg39POifJ(b2>-B?x4zd
z)TS*m4-|a-#(Whi-nFTo-T=FK<jA1_o4b5|-yR4&VQ)a#Mc2g4?yPri*eQ?3$)^&R
z(`jy%uz27rXK<q9kYjg$OhgXKDwWH|Ie-ME=kgtRHT1aYT&K`CCnP1&pzQ&)2K0pZ
zm~&ISFt1kE@#&GnhsR%e(gyxUg-=FgCg9WIEGqn#<+O%o`6K(mQ&wcY=<vrXBu4IE
z7CgfAebtOmOX#3g*w0OJ9e!)tRR0Z2+_NVR)7p9f&V$s8JINbOX;0+m5tgVH74ULA
z;I|g6<!s;7)qK(G0%{ISt9H44yd~4D8_d1>Wq;hYwXfW~=7fnv6Mg^?QBWjgImt$+
zrKIc!8w;ToN&uu8Ck>9r9(|Hl%iHyLtoI*#2$>|FFY5EIX}ObOsb{IxfW;5`Zm1l9
z_%AyILdtNSd^Fj^e)PgLKD_Xqo<+_{gd}cEDQ?H<kLH2!W8lM4NTqph<ekCZZVFK|
zHVyCj;=m+u=s4FYWR_YUuW3J@`eFCQ@w~{C(SLrX<tj0yCppFi4l9>=0}z_4u#VtU
zkzj?o5Tr^+lb@Or5pL=3a~@%ZKpKS!#Z@O-Ri~J1&x9_VJK_`wkYVKz-mvTic-Gj*
za!7*FGQ~FEP%}R(O?~E?j`NvKmT6FmH1Dsz_+o<85%ss8pA=CUn^C+nZUK`jxPXlc
z@!5hNR*AYM;=KMmOIB+u77LHj7Kuit|McW4xc_d!7YGYV#K^U`s`9YWA6aRhe8j5#
z;tS%UP-QW~Kt$cl^usmWyjyEJb&G3xbZ&HxoGm?aMVpfXyXz;9#o*+kA|oV!DAmZO
z<ce%<p1Ukhe#Y;Z<9d2e*KA+KD0}ukm6#^Yi#FrQ_&{~?ZM@4Ke&+wa6L2H=N-$~|
z-L;PNWWXY8V#clbiDmGl{-Jv#ng(InXK{qmBQwP$sYr0+bEL`IC@xih32W=L-}8Pt
z^<d7DC@e5u-9fLv*D(sog{WD)DFzE2jadxu^x^}7&dZj-a=1Cbq5g?Bl0|?`BJYLF
zh{Cr>e)!AdE?+g%(7A0-QI4rHRV0+e$fq8Pu=H5GchMm81Rl9EPZ@rU_>yRt%rDf4
zDhq}0s+4~pg1gACw6`@lXZ=k<CHWMAyko_YGsT>wf`(Q6bl&>uZ8+e6r_+0dOt{<|
z+4xZg?+@dh_yA=ad!Nu0*PI*h38=9l0>VkwL!R^O9%bGYCLPDBnUTXfD86Kc;WB%C
zNnpa3J4CCo%$z4Blt&=vH$fX0oAUBqu6&PKnD4X1Umw0-Rfi2b(3hjAf}b_82dBeN
zjPndQA2{k8?wj(C827;@<>%)Ux6~{4Z)b2cUxF|J7*y%Nrze+iI04$_3HbF#m-d=o
z(U<k-<mp`R+;MoVw>WImGupIOo0ZP#T>yz4p`&H_9i-G+R@<<GX95&CFa@O7FsVSh
zfKD*8mPhzp%?3@{A{x4q?Rc}$6;qqsfH#g4PYM5#eS61>8v1yB6Kfz8+P9fU={Ii;
zvgQ`%M8E`YHRbl%chDYuJOQWx05#la5MJYDzFayoT|NRvlh(2fEC?~N(Q0CsuiapB
z0dMkQC6+;nu5H4iM0CTa4%{WLLr@IDnh48_hc3S#no0nnLD>bb6N;$79SLr@TI@+#
zj<i3V3t4_`Vqk$->pJLlFxjQ)44gSuDL8>iU=t>s;wuy3E~#1(5Qc(=u^44YBftnE
zo+sjDJI}dOHxF)u4~8Dz^Xk9XT%4VIp#SA3igncCIJ_R^<S-EDN0R!z@!0~rZ`O>P
z#Q?C`dyOMc*Piep<s1ok5$;rOZUlPNu$1F6eL$zlWwEq#Uqi55o?LD2x^;J+9ymt7
zUim$hxOV9$rh*C{BZr&}J%{h($6UgD4~Ot>UO4W{+RLzGvzV;o_G*5oYu7_{HaI#7
zY)k3SOc*<}DQK1c{_i&NF8x=MpA>p}o=+aW^{q83)+g?_M@d=+YwPfchgI^m<hf7r
zBmL`Q>Szrjut#A!bFk+P;J)AvC|)RVKq?F&q>mv)3QrsfrBJo0Xx~?Ycg^(<@zKj;
zBW2vuE(;|g+#9!HQ13?UWYm~maa7PkC7WFrTj$@(<4ai_Pv|iz4Z&Gmrb|<%+vn%-
z!DGnup(EU$JshdE`k6M0cWX4iGcttaTbC$3lc7spT<uMPejQ8>F7EK-uIag{4R@@#
z!w;wSmsiAXB}Lo}SX`vzdW|U*85w0i&xn1!;~<dMf95biLgtN9?ai=rB6i1(iu1eB
z>yk)s6nJ^b=8(e^W)=l#B?_TSkgJmMt}08z#3Mt9MyPxVtkI*yHA|26?NGo@JR10p
z;qz|q-yt6fW)ci5!})rZWL(o$(AX`kqgg>Bj}VtwMw5L2ZmlWtYl?ssn}eLGlm%RE
zgB=@^$>3Y?K#&`B<A}WlQ3&_ye{rI<Fr5RT%(pw0S-FTjVqb(^bV05uhzOST@eTTu
zogD1!Z%NoBrIcWtJ-l2qBo+si527W&A{38@XP+#-x|IS_f&fLR{)>=+`OgKuS!-h6
zC_tUBeo$h|@U^XUv`zsx0p=OL{19Dyzia1yZ6?yoQM_=|Or{)xcF2tsF;bd1uTAax
zbNSCm=SB)qTs5L&Ys|scXU)!7RN3QP*c^jdbfKT-)(du;P&3Ehs;EI4Qo`4QT<^$p
zch+>QKul{$wFQN=WA_&85d}y^Az{$p0*rkV1phW?d#ohx6{U(wkOB#QFQ4AbS?&F>
z-dBfdB}Ux!`=;n&{>AI4O{-~K>ByY~GoS<%*wJy~riEBDZGYERGIM}peeq(P!}Uxw
zrxdc$4O#s<SXo)wq%#8-$OELQkSr4@Z&qSOfk`%l`xa7p?TSSKtnTI<?-u5H4@GQT
z6mI0s`e5|Wz42d<G(a+(f*9xle-^KR>$-RGUHy!i_r|Wirc}<|Q48H|nfpA#d@Ep#
zwPx};^CSoY$?MZi@sB^*0KQ=+zAZ9<jUcQmtG;5C5f<R~g;o+10kwGMFV~@4ZNs1C
zXcGJjkc>~59-KmEyWV-UDIe!seXt_55M+Qf+ji6A$c#LQFOB0;rM_@5@Q<u4`-V^a
zPZ1!Kt`H|>dqUk-kDldvYmh<Oz7xkoTfs|sNG~#E53YMRO(zuA{wYqdHrHsfV!u?C
zC|~E6Ge}M)SBkL>w@CCJ{1=FWCPLz|@AtHNNVE&4AHzSCEL1f(Fc6ii_*A5pAg47!
zFHA=7+<pI@tUF>}BPKG~)`*8wE-9_6$tZ1mxZ3ztdCXBrf8RUy@X@0KBfHX6uZh|K
z^_ZFwQxAPr8jYo@t9!4Wf#G;uRFJ{PLz!#kVDd>4?cwxOJ~u8?9sS_wc15-&U;Q;J
zzbnp<t2b7Ih)@viOGuA4-x{iz6KTmp-r9bJX<0z(*d-w5frncuv)vuJwmZgU{8Eao
z?L?w-4hE*gjjB>>U%(+SrFwV|`)66!G#Y~l<V&KUyou=rugkY_Ec`*X9pNfvI$Uq8
zlVKrQSY$hgPj850ZHVPl?-yRgV))W+W<0Fqo@r`Jw|!WcVCs)Ax2)53(Bx}r%*v6$
zFw0mLp>~zEA7FO-4%3;n$o`5`_xuc@dQMKxICU1|%(VAAx@YPB+c)`n%R~ocwEI`1
zsY%VQOHjFUQda#H8yj1%hni)hfU2_g?&gB8h;;tCsbbhUX0X5@T+sjT!ieWavm&W7
z9Q;l}YMv*)W$UIg<+h687tY+0b+yOtOn9hh-$ll#y@1GZQ{ixl=CkGs`g<@GVawh4
z!#EjtS#E46kbEh6c=DP&ldIL!ZFzP_$0Dy)NHJ{72eZNdt_6^jlUui#i3Bhf<&*Kx
znwykWpJ<NO<i0=bD+Eq5^^^OK8Ci|b4#W8^D^fJ_iv<0PQXc18+qp#zPn0}*L<K^K
z1KTr^@y5j*x0!A^FI_v$^z@hQT~j_3o)E*D#DpJ+=u2B#KkQLMBX&H7Pd#Gce9@-t
zO=}?<I>+-JUlV%<2GUSb$kbP|XFx#*vnf?kN!Qo#XFKQ}FNyO^RCBd>bKB4E(GKs^
zXT<vu(ea2rDz46oRImrj_-GCsYx_FZ^X(O%y3V~&`R9Sxk-F%<tLzlOUeXCt%rJUE
ze?dO{Flj7TQIL=D{9<KXJzn1R6!<efBq=tCJ^0+(YLF-JOIc5DMqo07hLmSlBeXWy
zJaN01zstZ{-mSrRvMhbh29jhz&FmN*4AI-*<X!B+;OW<4gdfU+gSNIq$J>zJ!VKp_
zB_M;7V>z{#Ak^cbU)>e1Z&F4!_xzF!JZh+f0x?<72Z{Oaza2L(6skHwYxq8H9~-Q3
z-1-&*<+Wldt1+|!-R6{L3_Aj#qwkCC<}5y|JZpC42Pi*!3<C&6^G}Aw;#wFzIxwXG
zwj|Gw7c32664s3i2$KZFTuFe;HJ<Azq%=R@QDm?%4Vcp$nXZh_K^(cxk5!TUSt9iV
zV_tY}{@U+?RRCgQChbmXibmc5Tm;alWvE$>Js}s(j8x94)C#sb|KT0bv`jwM+~Ajx
z(vc)E^ugqaq=7XZ;y?#(tm)u^h?pJHy9!M3BnCyyctLW~8Q5e<UsroLG)(f3UR_`Q
zl0viaIIKZLusBB4gG5_EjI<>E2_6H_KW9B}UqL<iS2il}&%V(ax4o(d^%RCR*P=?1
zX+A520_J$|b7S_oQA5Uiw<^o01IC$DPB&_-7cV=N1os1+X(T&F#V+q3bwIJO`89b?
z<hecl*%==Ig$TwfBA7QKw^Ceg*pGjQYxPeYmz-LvzN=lAXg4Y{`kou9S!<HZIXqWC
zbvgix%eD~o2hc@oZKfOusr_HDKTaKE47Fzcs_;0Yi$8JV>g&N+J;b{DCQD08-^Mq`
z;ArB&0P{O?*H;9ArDt+Kxc-eZh)EC}oYk3c&)S*wRB8A6;n*`6Y5eow9IP(HY97L@
zgpTeesSEj>r*_jD1Pl73!9>H4vvb7WR|F&R@JaW_Rdn1MGJN9OW-+X$ard+1LD2P9
zsrd>kX)~d&J286n23`r<PaRE|YK7ZpVf<D3fQPnCZUh9TyfJm-?aliMlk)gTB+@vZ
z%j9wo_dj<IQBG69mq>n;+0{3x!7~JZ9h5Bc;{3USfEy3cW1P$m*CFPu_Tn<Ik3iH8
zI(JvWS#9VjulnqlOuFEt_PMM=d<r9Q8h2yh`@3lTj5L?k7WV!LW|*723>c0IJSqbn
zjUz4TlMM<`_0aofX_9!Gwrj=Rtm-|0bRUGHJa$pjLtfjd5qfB&)Xq0P<XXP4*Z0@a
z-gv;N3Wx;mZw%<QqYBXH&NM1mE-Xy4^)EDFuEqVNY)OnqLHvD`hQu1Mq<X8Kt#mlw
zc<fY0)b{Ax-!AIU+0ouO5Sxg?Ch<cD{dXn@^Km@FUOnN*=6t(ZhOZ^KXJe_WalxCr
zZ^Y93#dFL!(7TC*98`0N;qL*=^|S-SQbMS#xAvQJ?KS@>u)fvj#iI4r5;Bq6SIraO
z_2LV}@8_r}1!5}Qar0!4e1t}q{H+k9c&#y!ScxR1H1n_&&f~AVTly!&wT;Zb+AOu6
z|2!HW^Yytdg!Q8hhhn-eoIh_7Y$35Pkt;7r@q(vb!nsSB&)Da1moIhDv$oG+EN?6r
zGlkQ@?_BjcCO*%@mzUEvoV-_cd%4h9(u>3Tqsxc4Q`gj0)ZEArN6@3dWP~K^@S=HM
zE8W6AoGf=y%>`_KJ?~LM&hYq<V1u-cMGUI>cNO(fv=ta-rS%^Lvlt)fY0)mJmv*hZ
zyqx?jlyP`E!osmfV+NUH1EdR7lU7|=X~(}Q4}U;obP#Ts)Ym3rrp4!f3NsArRbxiX
z&k?HOf+*xaOge58-ZJed9O?XnT00-H?Mh54pitA|3k5c<(RG)WmXT?2Jw;w9U*EKc
zAJ>?CJhX@XcFWMMmZ8`Wx!17=={$8HN=gqfC4(H^{J6!+77PdpMi(N23Kc<JQy9H7
zdA~~vnJToel^&{pPfw!~eTt={j3E{bAu59Y-Y{x#Qbz0LMt-ud51Fyq8i9FQDY@ed
zWFXL{C~eV5w3VpDVa1cUyyUoJU}#7QE*mrwKi)^(I3}3nqNW^Crlp>+i31Z{5!x8h
zahg8-JZQI*kRp2dZLPGztiDkf4BETub6`so<}RTZrJ}&)<V|$Z{K4?_>JL2irPI=8
z%2xr6-hbxh7R)2X{OBMfQ6Knta1#eOHXS#>rVjpxLV?Qze0+;6k1%X@rGUn~#;^Kq
z(jK@EqM(rg0_b~QmvalZu2LAKnz|$%(<~nPwWiO36oW@PE}{)4Lu)}LkM-OU>p|t3
za`3W+6>MF4j72cF#5H=TDc>9V`E&o?*dN}}ThK)Al8{(oHZLH>SoGtGF;l3_r2wqz
z)Zk&~!k`d(gN<Y+*cRmJG*z8Q>xvbESOZ%(mD$Fl(NTnfLg$Gn5~KX+6mK}~x#{ui
zgB(*D!-a$TG5dKYJ{C7<{ABZazHc)=Cm*8_v>AZNr~eUsJV;DIV<?ZfC`3j_vJ&?m
zvJk~gPu9SL+DzC%NVL+(-L&5rM|}V3yW^=SnGmE;AtnkWpg-BMIogKwv!HIg;pRc^
zVXDiX<n61yJGRj?3><9pY(AWKH$!7iA#W+`k~QOmTc_{2JBQe{8DI!S@9<9zhPG>j
zPyb|b)fNh1f_#TiVJuXVt<G#3Ai%z1h+TWCE>`YE#;|z~GqKy4%}Wx8+h*Syqw(#(
zy}3v4!>X3nc(k$Q=(p|U+ZO$n*nqV);DhrLgAqt=PJH0v5)$5k8%R1L?wV~Ymr-B9
zzE2l6tCV4h1`$eo54cl=K0(y?6!FaDb|B(#$B=APBNl<EX)zrI=HvRC1HYB0j-FZx
z?8PbI?H<^tXJ0-zYzBy0I9R%@iY8m?>8fmTyevuBPyc7jFFG2Gb{}+EH_<mD_rU7G
zE}^r(qAu?6=OMZP!~(2Y@&KOy0FeP;(AbNyjXC(wZ+@1||KxT^qxH4lxoZ>le7&c$
z{K`&Zj`UD<NrQ8t`x{M{G&@ot_{4{OLMu&JMmSc<32fJ?Ks4a4NW6TCq^y4fG^Mx-
zA6$Tw6e!I)w5)Jf_QD`kMqK11@%-<E;*uk!bfd>8W<P~j+T1&XjQK{C^q0ThSu<QM
zjM|_%f_wG3Q`~VrG4Xurg9TgDT>|-+ir09TzAd^1qc>hHO#gxA;B|Qp9p|M_B+?g}
z2rvc2dH1Gw40b4NDm{&dxNgDC0MR!LcD8$+W!n~E3t2@$oIi(5yFE3Untvkb&MHi>
zIEm+gczpf6-@@uEP4;uv(08k!&51A;RbWyG&&L|8JZ3gFB5l@54a2cvaCFhgNSmp{
zRZ8|xr&QK_PW7O%Epn2UV;_+`k!}^FJ4MRb1VJ*9fQkYO-0Q)9SRjZnlhTE;BU!Cq
z`Yy(pZuUT?$Gytu7)3+rijCNT*J5gegLKTT1i`(Bl=X_B7qH(01ywC$9}k(?V1p<v
zk{ASvO8yYf#{<RQTg^6Sx8~p#z@jL9I~_0c{T$XNY4V%XUi|s<N8NLKJe@l_f^*Wi
z1<2@J-odcOB?q4WGqai$j(@J4)GQv?)@F)38A@g_zK@aF2=(9t(c^pv=)uEL;_lNk
zoE@hh2wazGa7Z*#mrJtDRs^P1AcyD-uKFC@dnR4ZeN?!_3-J<^UM?NoZ`aQu!;=EF
z&nqkQNVb-K#ic8V&Ls)G;7e~_E#<$bPG(BDqS%J9KK1#zhtoDsp5s{M;N6yB_|CI(
z{OwI&xL{`Q%Utn0*U&qLNLUPhx^z%VEJ_Qpomhjn`Hw^(d{lkJK7|I)Z{Nk2-LyuS
z%}%Zdh_OgmBUK5@Iygcp9a<4bD~G05Fe^4j@}>SvUELR8e9?&TSHT{Wyz&c^pI$=q
zy#0Lt^jyw`Hfi#HemXHTj1sp-;Wcta-(pltg?Ag9ws5!(<1c7lv5{jx<NUhr?2}K`
zEfIqU!Op_sR^HvEAiKU;!D1yOK+-U<3EUI^u8^I1(-HH++%}*a&5HgxF}DB&NXp{v
zLjK3Utj7wTwHPTkquawIkmgh0X517ASio4c^81rTCd0cwYd8Zd1YB!b3O8Ls<cG({
z!4711Km>+&035z|!)XK&0#Cj_d6t)#H~Ci~BOPsXRkg*(um)|h89BErZWmHHQOV<&
zJw19qR!y4;uPWs1r2uC!OJU@0$_kKm;B|ngL2=mLbb8n&lY8}Txr=$NBE%a;A&HIP
z<x(A$mGdp1nMVvZgp(^i^P<f*v@5^${QUX(Wjsb}L8$4h=BbM&hfQl8Dn(_oh|PlR
z+<qAEL#!Vp4FRVwws~@d62~?PJV$s-s`qP0ngjR8Wn=QIo;=s#3sL*Y3~gPrq<W}5
zwk=AppP+zi-M09ym6djrM_nWz2=6F4wrh0pycD*^G{u<*^jp)jIF>P?XQst&S86&j
zGMiF)FY9r;l;GK44$o0WyN*|Hr=k!e780bI;t>SmHs8y2NIIs1G%^@YV2)-=cQFt`
zH;oxd3ampSe;pGY_OPQ|gRc_oebG9@Z8Pp$n6q_ch8C49PRLtr|M<r`+%hUVcnyY3
zv0Z$$eb^(a+Jf$d>4uFPwf=f0R%q#o{+`ei9hTavEoH-4<e#9o({2+KmQ$@?+S}!O
zFVq=W8El*R-d|c;8W#9yZg!SJfq*RoyTkdI_|MWvPFYiJoCy>D3W~|q0e(S$iDAc_
zXfi2+H1k=&?F66zwAT5$xwL@!?iK4c^UcxfD$S4F>(=E9lVL~hHfNnAv-Q{jH;`ag
z<OfiYj{0EI1!9l26k?xR-WY!D&|J$eFrjEJwD29^>a`+z2JHi1*`ChV@7gHfI`uen
zzZI#AQ#|k`-tTOt#p=Y%jNCjb3Xn+8<_5qXah{7;C?oo>KcB%3ubZaGdoIdjAKD%`
zj7FM@0<GH9gW<p_X-}5f9-U$PMPV$u`l0y8*J*E5iGz2fBj3{4UD#D77YLy$J{;y&
zXuVElvg1j-g-r*QE;5<Z)VFHMy8Ycn(hFB{xjpSMZ@AnB<$L~xa*~~iS}^6f>c55h
zsMyjts&(L}_22n%Gjiomj7HxB41ti~+P*U@Z32b-6XHhp#vc77`V(!t1udd=Dxg#d
zICZgL<<Td)BlaTK`T2>FjEqJ_wB;=lnS)74f3E=C{av!OsQx0H$g(w4dQzM4%;r8Q
zZS2MbyX;E?ANl8pI~RwON*uAlMHq!^C{JyN7IEM*q;EMq@hP{TbuP%KN=2d#o0!mB
zl^~e5Z~nd!21=*dp${6NN7MQLYsu5zFCX%3$Y#az?ZvTDneDk?OH_|h8VL|Mg2k$e
zN0QDNl(do_N0s%sm{iK(vm=4!m-nI--t=0DCztL5na%s-&G#&Ym!+&Yfg799PwQZy
z28O^tMK_-C!T<?TFgu!t=hS}t49h-JIm@92I5FPI$72qx@>~h82_0kemTRM%j&GS8
z2|eMF%y(Apegmm_1GxI{<qIdxLb(|XG7XUhM74-)g{wZw7D7cm+&mJgXxtQjFhxb_
zGO1j6fjs@m4L)?<m=aP0o(m${*7FfEJ9qvD(TsQTL9E3=l#yHRq~<lCJ=kIp(LBK=
znx>+6ZTYi=1YCEJBpAH1-rQtfC$?Hztox=IYK+m<D+a)z#+UG?n}4WA?CExrc=yBm
znm}l6U(;Le^t`}Bb2IOqRh9FrNROe8Gipklu}Bhu-Axtl_cIdD)Ws{BqHWUfQ>S()
zFv9FtN+qNB`PS2%c!0q(#jK>Dp6wYvnjj(M|3*kq&xJ|lz}{HeU2zY;tB=38*`fAq
zD<@hQy=z}b`rHyC-yfLs$n|b<Mg9MA>kf>?M$S3e&#Nsq!A%3vMUxyK&v6_&TDfit
z@%4?&G0#2Io@X!DNpXLYrR<8L*X?i(4BUCycd`PrmRV}VR=}7FEja2F&C)+~;fa~t
zr<Yq=)30y2cDUB@->Xguj_Q(<Mj6}8iK~@N%Jgp%_~biy-8p`8Zeg$L88_T0T%scG
zt{DClLmfMyjJjjR**2JmD{$|CXk^T;@EA3%nKXzvPnMk$K-HK6M(Jd#vYyhfGFy|G
zX<d4nKV4GX&#k#L0qa87MA?7j$U2<n^$by5dD?nX-S#I(=+m-5&cW|Up~Tc#`Sly^
z1Y+R+%YM&a!w5urL$w1e1imV`X$Lq$9~Bsh68UQBU%wsGb=6}|+jcOhFhG8#Bpi-p
zktzs;AG)ZiGzw^dg$h0;!Kb1PF*!Iihk^%>|CmcE>Z@osu?@f=ITWMk_b2*JN;LH5
z_T6t*Qugm+&M+ixeAH}S4wJ^HPBRd0m(|C&uBK_C8Tt8B5b@RZ3@k0-E^e7mm-OyH
zj!Z}cOexl8nlX$}KNS^!^vTTCqt~<78TfIph8hACZ>nC+uDOywg(tL`%HL{U=7;_O
zlX}lfuhV8aooA3Z@h~d5RYM@rLlxgkS5bdoPEEw}&<1ytFv%U5Lz5QmCR-c@s()$>
zQ_6Sn*FKCr6It6QlC#vb*0gpUzX8@tL!i9Hajl#2yK|^#C8Fri_JC%g3o-jB%7hSH
z+y;^1m7}AhPc73n)w#XLd^QXj)MkQORb#<uO0Vv@0L%%9;aox4qM%&|L`N4te!{&H
zog53ACmhzO3V<1-FnGI?lOY^=9T=E0r8r#2!ILZSqAfG)FL2$~Q(1?<^xhZv1YjLn
zXgs(-=cLxauZ#}(CkJgff~I(w=>|Dyo7bGU(?S-V<@k&n>N3pS@l*CTJ*O-H>2|W{
z-?ErJdaLhge}0tt6uXMX=!xntUzkDjA=2xXL7r8PZ~!IXg*q=!!t7(s(ed_|(xh+J
zMO(U?btMI>P{V&~s9pwcGkPhK=7HsfA{86WaR*E9$TT2lxO%5#CTwyn7)hL8hVK~#
z;AD1Bl*S;bI0km!tA+qn^xau1(a0n2?`iC&tW8g)Pa5#^q}>AB7JQ*9<s3PCVlKW~
zLm4FzPUt#JDk`_A<eV{&TC@DsT;=a0C(KQu6O0%GgF*10o4PR+zKKQww)6t)z-Anw
z^2xrw!akkGdYWIitOwMUs&}}`Is-^$!dgrVZIq+IF+lzG4!=O+`#Mx=%h>#h(G%K>
z`T1?b_Iq3h4l3(yDCunYqkS9`$eW{H1Hy~g$2CbnFeJ&Q4ZA$}HlYxQ#C9M;tJBwK
zg0cXZhA)pdN`QNZ29rwuh^SbXuWyYD(ON?q-Y7?>JA|oc%VxV${gT<^^VRY5e)S9k
zy};d%w+LVT+wC*6&xG1&?E{?GDF??l7kP8<LVNS8{FL#p5^{cO0B>6eK6gW(Q-zg-
zg8<2+2n|02Ru;_!NJVOlSp1yp^2MPtOV0?!>R)tZ8&$OF5rT6sem*Iq9|h)P!L?X#
zAW`*&_&@rI)MCd+A*U_Br7jy!6Fn5p14ySv>@|Egk6|aRZ0hL9a#<ZwPGiS#^ueC%
z2x{9O%R@^cb_7`Q?UA8Oh-AquVHARgEAa6o?|&*GEZu{VMN2sR4Omrb@VuxY{@fDy
zNHsTNsP!+-LHFs*%4I#+r7sDmSz&{Q5%8u+Eg88A>8%g6QX91PO)(+Ugm^L#Ht^(!
zxxPeJq^So9Dpu0hZwZK7#V{bo8BEPwa-clz07woN6+Hty{fup~N7E}*EsSuI#nkb{
z9@Tm|ps9@YI6{<_097G-+e<<<{>e!DHvs{vj7A{yi;IiVLB^>s!7#<S_N;+{juS3=
z!5LY4ZFltOgS}?EQq1|3KUY{VX<OP{Gt3i?%fj$&Yz#;W3kKKpy8MyEgMHVw5D5j8
zevbtGJGwv6;jnuElRFxhKLO*s%6cFT*Wm?iP?!^;m>~m7f1$m9q8k+!%`TIbr_ytF
zu(U90$y4DqCv%-LW_cx20D4(Stf9a@@{8JlAE+@NJxs?x)OTq5FC6p}PKSI8)WHFq
z1?Qzro_|nAHa2ov$c#aVL&-%5`D1$`*&%@~Nwmqv)M{<2p}yyCoIOm&r>1SS)s^za
z%vLE1iPTPS{Sxj!M3kjqq%L*3Pm6C;wlJIdLq|z>NgoEC20dUY_P&ls$2g!T-GWhe
zjqUC4aZIiSE{@jq*A6Y+8qd#%L*EK5(pI#J{dlBVpSLL{<<HA^hEy|hW!O6yf2-O+
zzJt+u@e0OyIXf8i0p-Ejypx!HF>&~~dX8=H2aj#GGj7E~=6M)PE3(8IcJNdb@9Gt%
zg`2Dyix7>NZM+$qJf^zBK_`4~hkXf~mMZ!W0`#pdv8|@VPP;6daJ0+YzPir!{W(U=
zh^Pu{XcT4-n*MaeKjGw`*G(o|KnH4zH|Sj>wLClmMe!oBphH&f>fp@?yJ;e8uFsQ2
zU`fmh|5EIJO?&L96@nCWpL_kps8Qr^=CcG}uz+>q4Oj#G{V~jZ{xZ2eIy}KW7W+J+
zJe>k6_n<(;TSEzm*8m1!_ojOD@e{lLTXSz7&2`(hkA6{_q=ZnUWF8_3sZckSu}sNa
z88SqfB~v32GG~s=V<dB_RElIu%3LCpF(kt8xO6}Jeb-)l|JL4X|FK`odX}f|zCOcs
zo!5CD=W!e-tM-NSxPF|(@jJ6P4_Mq9PAqoztLy(_uX0lg$X8R72Bq)K@rPqA79L*h
z&fE~C2_)Xv`D$Q`!c8svNmRSd?vf7MPi<GMG4*)uYj1gG<;MD=2GxB87{GuK!ou<R
zU1xHHpGi>0LQox?5yX8a#m^FUS>=`zSzf##qVl?C!!I;5nDN$iAeW#C%v|1Q@<!!V
zvS&}x7#~}(<f<J?p#;-p+(WhUD<hBAqa74puZt0dv;W}tJj_K4Z$PSQ+2ORav|!MU
z?;%=20+)?MV4`dE`=d2zov@0dP{3F{3cI0wT&9hq5Ao<-mINYPS_eUTYs;-IlKKEv
z!q{t@i#<EDyxU_J3x$7{@>I8Ue}rvkT2FfY?4Kg3w;)F10Ysn`zTEm1o;S9h-=Gp|
zh9_W=7Iz~ix>sboNtZijR|<IZdDmciQ}}y>)VPAWPctYPdc@F!s@Zlq=SO#l_q49J
z&iFTwTFe74Xrn1%@$gWjxpO<Ku2(|BOn@k$+}vXsVk7ng4rR0DI?*#2j~h5S{h`<9
zkoHoM)PSPKSG=V3uJpc%(xG2gdt|GyKTlWk3&g9l2%k-!L;pC&-dp76jH5toz5a~z
zy=W^CHztZncRIh4ot5=&VU^tzce~HeZ|vO)WN!(sge;WtevDPfvYpi7fGG}^2`vSv
zP7m}(>-@cpB^FaR1A&lSkH1^)EGhEBWCl4IaudT{x_Ykj8Pe@*yPafm5u>z^|Al><
zhlsi4>?Kw0otOcD?n!g&)g!!hh3(dF;T<=j&-j`}CZhRC=-Z=Aio{j}WLq>A;e`>|
zy+e<>ZtXgtP{V8HP)zYH#6vPeO@X{QGA7XIRw=hq`EZp``lbgL-o0IU?a=h@v5y)=
zP$_)s0R{D*;?bkN-Jefgt+&vU^9JH^-1ydNiWmkM5{?;)e(Swk7+Q}kUe=2h<~D6S
z`F--$-F;sJ8qD+(?wD-1K^F^?!<sC^zbWoT`AUV715_4@4@gTU(?5}jU|i!i3MRa@
z<67Po6W@Krcj+PT<FdcG;c5CSd+F71SM(a%Yr`CamJ+q1Qdxo(UB(xi6tjf$+p=XX
zm_-?=d}{hMIDa}`$tc}3b>Tl;fa0QW@^Rf4gxj-h66#{ARauhw9Gf<;i)^k9udT*0
zf|(8g?g-9;nsr*m5)KELaC>-H6w&TAwzs>t7861Y6!{ZSi=OmA%^t5TyW>sf?tVvU
z_F=&Xq1+RB%ptcPsL?POp>4^v@U=r5&oA%t88N)pQrHHP0SVZcI8D|P=S)Iy+JNz|
z2eyf*#&&-5I<fx;iun`#*8u{Xg6}>!I(o#__1E=8S0|_5#rd&BCj-669ZJpDc{?Ju
zU~OeeBF!!r{DH0ix0j{C19Lkx8sS+Cz8kn9$uhm*H{CEtc?$kEC)O1Bde`+B6~Z9*
z$DtC?eWsCa_0^}&+^RPdKdpFhRaX=Xnaqg&Sm7#4Y?RUe!<<u?1Tzv)4^u~sKhPVz
z``CIrpth~C@!rfaxNplG{YN39xv}x-4MwZ0%Wx*kUN~^?IU{^*<BI?EJ*E4(N+k}$
zH4MX9Pam*<0(azig101VBIMJrY6*JuT(|rfXx`{#SEG*h@3z1kL^v&b>b%N(QLJl?
zDU6sgQMb+a)xLOjXwtaDl$@JZot%eqs{t(mY>O3#o&tsh=&yL`z_v&3Qt|4J1Mlck
zH3g=33q3;$cKc{Y@)fB{8!}E)Fa{YX1m;5J7;>Q;%Kd1=la3LK#NF#E3gx6wz}~mD
zPa3fWflA@kJzK3>$G&zHg&F?TSA0l2MT6NYdwuln!s0()vYx(JxT=f}Fam5r!K!@O
zyUM^MxMFsw^Q9lK6%Sq09o@GZmeOYD(IO;GYTPp0S#keKc6#6PJI8p#q8q(~{O|Lx
zMKR(t@rbTeY#lyn!qPxuyrR-&R+*83?KS+9aN<F-FfYT{N`)Fx1Z^i(UYh*2b&48O
z^fecohl=LPC*1Ja;%OMS6;-!VDfHvDlAFl(xx|qiVJ`P7MGbVp{sUo1j8NO`iVDzb
zT+PA3QTh8M?a3G5BeI#0C9&hwCaodx0e4GErk#|ur;_7ZunN;?Us?RR{BEbn29{Ig
zwZ*WVMAbLW<*S_3WwPQ>@8X_YIMt@@rP&8;6r5lB&-X>N-aP5C{&nf_M8!;PEbsJY
zPBsu;H?DXV-g33xK^3nLYWixR)U0&R$9{cB(+{@xtay!RmmoEqrcWQpJD^_g(5bnA
z7IPNwrdxv?)NFpmnl+y<UR){zP<)6hgd!%(9M|IW{aP~ru~bbQ<S<%m-p_8NH*uHD
zLYgauaa7WYV>On3Y2O^mcn{>2^tk{nxVl}QNfFffAr5-Bx8QQx&-~B>bqySe)~-dP
zH5QlDfrx=GjPEP*PbC@xUOfh-1HS=^AB4b1oqGdZG<KuSun=o~lT@CkgwxRW-~L-R
z@~K<@%3q1MpfY)n3uT>fGpw=XA0s2)acfv`#MQbk{;4&t5Y!aJ<l!O*p(nIv!tFY_
ziW)f;p2fIiKpA(;%SO-MpNp!ufT{pq_--G=ZDrp&7+Ri7{d8Wk<$&mxD3dTy9&g{i
z4Lhx2b`4u4kj&m$JnXE><bHPg;;RmEbm7C^m`%|q?SjcXdO+w?sPJ9OOIhFwH;<lg
zV1|;SwdRVvPyX?E2x^9zir}qR8}R0=o;|qVTuApY7fnrYW9uHhNWKHbNL~g~b7NFs
zklR^RKN&aqu-RwA?4Di%OJK_`nOId`5m&FAg}eyvTk}#8*RjyAAxkCgW1orIe6*yo
z+jkP=%j7T`0J7%JFTPJ_x%MMD=~q8(6~{Jz0Nm!lj**j`dy+JJCdVMIh|C@>q8_Q=
ztig223+G%V*N&_mBe%YRQWg|)C&V(nV6UVkicq6BB+GdeK!e&%!xm-!UcI|-%y_k%
zG!{?`Sy!+I*Hp6w@}{VbTN}_YXgH(sqx}yevURGi@aEXYNFPh_-`zTw0I<|p4k`mn
z^vsF31BpW6QIu%f38ovlj>F+IH%p6WZ@m4-Afg_2crMsiSvz9u(~8ZJ@0*&MqW3ht
zD_S*+?+D@^-*e|zpD$*NDM`lmt8K+z;jp?;>0qJe*<9Sv_{0P~khR7MTNzvFQTy&q
z%YAAVOB)jemJSTTP}qG;NK&`zka*x9<yVA|fSrvVyB_WQ_y;+Fu>hoU)tJ!APj+RE
ze4+Tk&suonYN*mM*?UK{Dpzbhz|tX(&8<jIZXyBuK<E4xN2ZhwItVPsxw(K>5JU*%
z6L?6#va}6gnab&v7Zyxc=KtRP%*HldJsuu(SPf|C7vPi!|K*mo>c;D9BJQ@YEb@Ep
zR>WKhW)7G)8RVdG))y3-zhzxB;*uTUQsDk%5f}DEl(a#$y~H;_jI+&GbzENs{V$f%
zxA_MZ9#!kc9b0w2Mc!;1s#54Fp1NI)(hBv%Ie{~a(#NvD@+nW&`a>JvTP-E?7cQNP
z*I-&nrsJWZ0x3rK=M&5b{$Oa~cuUx7aEvt8Y=p`(CEe<Bj@`e1pXketJ*RcBDejRY
zOY;0vD1fqzKYrgmqgsqm_YxG^8$Ev@S@8NJedNdyc!Cql`%c)+o+@{;(rfTdgRl3$
z<dJ#{&;kC^w*4$#kNtw{43S|oDHa>;;KB*y?C%B^es;V73KcXOk|skl*r7m${2E25
zxk<W)Id&@eZU$pw8MN?re%s)*uDQF>;%dnZ4&>*^<Ck?jsRNXY@ct(vMPGW3=R1QB
zi1*P>pdTJ)d%uZ)^zpCm>{nY*n86m`9(g~RT|~Od$NjQ$3JNqHk8fpT!`bxuDXfz)
z_v>AQ+B(ER;Axn|q1BMF9faZKGr4s^bzI#JSq<1KxJ06LZ1BxEkLU<^Ryg>Lkqz3m
z!={Afy{6j=2qbaxfu<DgPBQis2M}V~_;~dO;{HYi2Vkk&$jqpgFt4B@c6>FmGrNWL
zi^%9kwEI4t-+ZBRf|d-WC4=<fytWk7;4!dgk5`waJN^?Sis%yjjn-U(G(s-9_p7Qn
zv6T>E(a?!&ycT^-rQ*Va4P);oAJ>`d4G<b!$Q3Igx8L!`mxT{TlYvv1e*7WVz!z0N
z%eLTESGN=C2odXJN*e?q$TUn^3eJ_=im|>xtK%a06+1WG%2?mnWhVv6^R;}>5gUpR
zjC~8v=|muG7+xNE1dc(h{Yq)lRF~9<8Y@tiU`ubwH15R4NzhdzSlBZjgfDAOY-<{9
zn*R;&e0;J_wC(&4>sQY3Ps;L!Iwqn>BqB&l7m3JDV9EhUk*pP4-m|jGcKaETH6=S@
z1CF>8p|tze0-E-ms7>*~nTo|lk%zTdO4UQs^?quT|M=v?f^4>^;zZGGWe|qC({sDc
z6Hf@<bdfuA^_O+KssD&N$#z5%)6vttDE*P4_`|F}%N-Ne`(<U9YHv7Vp-_mkD|+g!
zx4*x*#Ry2g#yG#EcWh>AsuzN~u*Dq6odQ@ycMV+#dt_BbnH1tQ(BKWb(lTUbzMU<I
zIq1_LZHLp9!R5>edm_>^GuNP<N)hX>x>U|>fs(ZR3qW)5RyK3sw1^(>LgO>N)0gp<
zLem;6B<s!k)Yz-mHyU6>J2wVTW#Bt$&rGjpELv6BaM1gj2O-aGb<wsf=@4(INB}WF
zelcYqig#7nH8D63l{A}#GNSglJo<taOURv66z`u>-g`#I#xTgPcJoFE-e#dbkRD-G
zxYWMwpi4zpq1N6WuX$UnYBG8Q;Mxx(g_~z?m`e7P^Y{Erh*%>L4?!SaHLnCX=?3G!
zqC^Z%P8Ec|F-zk(ZUdii#u+`|>fQZ4FvcZQW&O9beV-<C5*2IgVT}cbjM}UeFxP>B
zf#jO)g@QsJZJXHQC4YT!k`Zz2G~OpvBZxa0#~c0g#x3dO%Jyrzk^RYxmFTJmmLlQ7
zD6!kX{b4-422<6d7nL?@C?atMghNfa>FXPPfGcY!;;xZ2VByKxl;kdWfPg0nhv1a%
zq+Vcw+`w4rD9(>IG+=leArbmz(7}I-Fm?qa&W41l4}W``N8f+k<EN(r?|mbNvT%V`
z{PLkv;1_+tazPkV%}92!$zs3BX8OfoLJL2Q;a*q@4<9-{=U`7BvfNr&S9t!-!Z!d=
z!%#3j_X;%Qw~(_y6ZSpQh^JExQGzv)m5jKs==57Z{G{AoqMR9rMlzzI^*X=vlKVeb
z=+=$>VmVmcWhW|tM`Zzy(ex6>GL~{lm4W$6%^Hn`mOLA7|K!vl1DvmUa;H<j87cMW
z{CtL5@po85=G<HAK4#aavu~^lS}Xd4jNXtNz?6w`B55!EJrGM^twS5HCnr=RAQfDd
z<3wB9nw^U>Ts!r2cD?@}+4Ic)$(Y?`vap1tr0y}T*?Ef|lb@~APiN`2Jd`uZI$*2m
zDCnOYUUY$ZkA*R)C0%xjcY0fAB0iwC%@b&({5EWQ$2;8`MvXH^7DnHAcVio$oCz7S
z3K&7-<V>(Z^D4$lolUnCD7Qr^yLC7_JQN!Mt=UMrq<*R8b2agpP=xG$Euhh_&(#%V
z;M`{&+JUF?_%(fNx<}}%zKL<i))nZ29oV)?tlPUJj!_igcA%3?ARL=fAfsOZ)xhI_
z3UlD*fIUjhYI5}E!IIn^V1Gjhgbo`5>&mu`T*;FYSr0*n$>G(TG&!a#gRv)+-be$x
zOrU^%DzTl9htgsnz{D?9@J?nqQVk|WK-yVJAoEV!3uK3AEwG`F4776s2<AoWP^Dv`
zD^Ovp?-z1OLy0oP7onU4T@p(9o*J2!lVadqa_mgfM*|WCelq{&^P)-$5#+V~Z%ceT
z0J^=Li=R2i!OuI3;dHZeZB_5S+Qp;U;r1=wl^D+@`04xc`&&ET!?Ll`7jEUm%22Pb
zt;#n$5A37O&~ZJX_F`8Kb6}z&Gh}yJ!TS{TP1@T`=AYatvpUCJ$h<*gdiVdiQ@(H-
zhB*J#DHj9_%J|U;8BGiV-T`<*F>_u{>lg^D5PdwgiTI5BZgcm0(<Q^@pIrVRvN$>H
zt*x8esoal1UvlUrq(_(<Q3jZDYoqHIIHHBgO^|M7+lJolZ4DN#vz|@^0Qe~V29T6T
z-#}UFr_)ImG~Li`g-rkhfvp5=aCMb}SeWlmCBiR!_LfpQY76O<LV1O^>15(08zJTZ
z%_^c)6f$*=SP6v5*;B~8;@uazD+RpOBUtEt?J-jx<8v(~WXyxj#nCMQp)q+;u44~t
z;xe=I@PB%e*Q%f88=x8`uMEd0I;f%CNNOi}<LQxg#(XrL&YMP#x2&0iVm|m#vfvly
za}4A`U^Xvd^1(Xx+4#5^d}s>n`;HKQ8*`rCTDFQ#Y|7{d&-={e(%Wl~H*m7CT}kF=
z-L~x&1Zz|D0a@tTB8#H*2?ab9jV2nHWjh@2V2DSYgAlr2yTg1*W<N4&@v@X_f4*;c
zu$0GK)@Ap|9K6th-N&>Qfl>%;x<#yfU>IW1)IlJ93@ga5{kW&A7?}C04cml&kDVAn
zCp42e9vKhhRRiIkj9?W3_`=%C>E0V6zx^JsTi=8<@^E<HlTz%Ly<o&)BD3N8-c($J
zKLM58y^+_ctZ~cVy5ZcUP8>=@WH(skgzWwxLmBiBfI_R3Lf^~Un|Z+30M(@i6OI~w
z|L0p_c8et6cFnz9?$q|CC!^<^P9JQ~8u<(64b+^c*5GJ9=`fH@|GE=d3!@J&&dOFn
zxU0JzMuRAYW+d6T-S~m_*5BPVPiM6~^F+=bBt74_9+uaJh<vDBF)RIEao={xK~%JT
z0ZO-8zdI!?*?-tmLAwe$^W0s&{iK!w5!Ag}C(>M<A%FMDvCx*GxQk&o2s(&dnVk73
z)>`(ntDw}*phw#J=;&R{^EqzJB~EW^69GsW*rS5NhtznkcarNL@zja42TkPN`B7zK
z^xO1Ar1{B-m^(AC*jO~*vw+gjnhDYX;t0j?eBf(P(l!9=Xg7)WB=lzM1|-t)>aeR5
ze{;$dWeC2L`6t<5KgYF$#SA8A>vg$k-(vCDHOPSmxw5pdIMK#R=2stlqoVK%ART;n
zK03a;_qTlu?kbBsH<$Q>(_+#|wwqcL;Nh@;MYD#f>HY2<Czz2jJ~|kciHw^k%M=xW
zp+~^$r>6wW46!Icl)1ir^<6n{S(uPa(G&8uUq&9J3;>MWroqnD34}NJ{uR{^n{6_3
zFBXAT;XbNhak|)4vPvu=nIFy?VqHIX+BAE$$*|Lpwkgn7n?@pwIp>9RQj4?1KRLrQ
zB3_*@z`zL&Vx9Hb@fl3*qOuiT?-qPtpMLA~HceCDQQ}<)0v#P6F6+r^`p%@T4|#TX
zs3P)pPLY(B5u``K9e|+qBB^eRirEf)vtB{k?O$0|?-aaQyzla^s68W#P}h}~Qldkn
zgFZp5Gi|?pqgbosPMf-4zgE2xtu5+#tf0L#i}kRC18NSX6G*an5;~mw+3B_9+Vlm^
zE$~>dWmr$TI{5h~@z-9jvFP*$z)Y-_=xCcdJLzD!2I@DnX_~3zmcXoFnkzn;30w7I
z(W{R%AMN<}w)H-jOCIm=ZiCosu51fc)D2#)tPqqQ;GN^-!eDM!X$1+~NNZn<UCF~J
zR?=&2;x1Zaj8nyw8{lJ<Uj{D}PU#O;Up?{R{W`9!1T5sFsJq)&qrN6z+TI}EVZ1gn
zQKCC!uOr%n3|hk7OwmAzIFF?#V0)Q%7ZYYdU@}G-Xmn7NqocmS07eEw>r*t%=A?ln
zp>?gMxV%XzeGxOyPM`Qj-u-R9*!p1(YKD;h2iI@E_Q4qpaS&NfOoE3GY&*&uMkV&-
z>OL0ThZt9L2c_Xf>E31PO&&BDAd&b?2%JMhmwZ%4D5#8xB^V6@qb|8`+1tb3km=sw
zh7)6uWqj+_u);Fo9@O(AVys_fRw*!Jm4uMl=Xh!eiO=jx`e-<3fjDX~6t!0Tcx+>Y
zsB^9%;f8P>J_S~872P|Un3q|(kPVw&h65jCIcAEYu%+DYdEKWPSgXPtMxkghp=0^3
z>BK>{ty`-B`BU9abR;N<9*D|pejsOnFQK4uo!=isj2zlrM^JOLp4dumB`vNXzbWYa
z->-{@j#*Ko)JHcS`cblF&}~u_U<g@r>9O(NXR*h)@Al?i#Y>j8pbjoV_@0xFP#$7k
z6S9wplu*VH*X#TzbAq@wuvc%v1-=(HHzJp<kMA>WFf+h01sMRDn?*>$WQ-FzkYI?;
zNW9gfrVTUZxI+Bgzd2f3`mO1fkvec7hu0t1cjjTibeRse8$pCOtwFaQ+t+yZy}+T3
zcRl#H#CMEba(vPpE~qkJJ}Gkp7e=(k$^5u+9V*63MKEhuzzmD(bz#n82kCy3LLAzL
z-~*GZJMCQSYZ8WvNgexSBJhIY^jN!nh8G4v5o>6OZ7^Cks_xn6-Xz0dgktnf=a$Yl
z$P;AATt#9F_!C$M#9bAE9jie+`s?|awUDq3moxrui{U^1U)cET@28U2nhJc90I3w_
zzt|%M=5?EiT#azg$*exodx5mz<HK{u>uYwMGeG~HjAQ}<a~54mKa3@Q$M4nb3#Cu`
zZ<A>GlsT?CarEc#n@M<VO<v<^biXNMxecfnTs)SZ%%AfmSFHHly&<$KG{C*yPX;QX
zzRiL4F&a6d2-;Xzf9Ed}0|H`{TUAw+hQ{E?D!JK1KRy=Fa4rlU-3`&N*Y7PSxHoBV
z+89tyctVYXBDxwLNTv5SQ&FIuj2UbI>=%KYk)f5gLqdB)$DT%~qOw)nycHtMU$3|z
z9{oZ^^zn7#%a%9X1HkA1qJd5!lM&HSi?I*cjhtjUMXAD%r4`$i7PISIS+)6!bSeoB
zcLY;x%nr7{xL^Dwyf)qQV03Tl8%rU#Xas-ZYHl}0jTg&S?C{1~_FAmr^kzTzE-rJp
zhaK@SJ3B4wiNSwsmZQsY`yYdsu=OYO0PVHl{Yjm}?zR?^`^n(b?i*FhS0|1@3B|Tk
zVd)j#)PswY^Xu3GxRsVQY)WG%2G=@wHR(s169qSh*e~4zNi)Ob2^mxbYpMrGc$G>n
zlc7)$epG#}a)oPS@6tlmQu^W&OwK77^n9t7b^>rP#RoPvQBx(zuJ8=(kuw1W!)nKx
z2R1R|#v@KaHvkprdC34A7ho(o$-r&^xzr|r&S($C+ynS}z&DR#TNo!ou@d|(`w_te
zNFd<~($Mk301C2HiVs32fvmpy@#l|j0{^xbe@aB|Hxifrd?WNdALuq95j}(16)j$8
zfP3AqSW{+b_k8Ia%hNY}DtuvyXYc%5q~D0#<FdC;aM_O<Gj`54CwdP4!}+{MZgIkY
zNi8M=NH@qp+bF&0D;>cRIQI}G8M5%r2S30g$bYemi1(P*GEY~4?-N+C5l|LUZX&t6
z4#}n%Yn)5$=g*&$v!f8Jkc0-z8Bt2%@88ri15FDgOIlMO70^*y9@RpA|MP-%x4myR
z#7gB4PJq!5bz{I>7gU(Q1hO-F>iZp5USP6A362a8*d*!v<K_brLfL|A?7;|4_F`d`
z+wkC%f2^AX3_3zWLXy33+5yYBKPc^YTlKQ+;Dno-<1OKA5O2`TecJl15l|N(LB+tG
zb>cWq0~RarE*0bZsQX6f(LDD#?m4Wr4AmVy=z8}p%EZqXnOrZ)NlTxaI$V$9hCpF6
zzls8CpnghD<kLI;&p+4P<9X^Fn&$nDzg6hQjC$x^W><IZG|n`B=Zey|OQ3CSld-py
zgv9u%{GHBWy#PgO$o9ezuagT7#M=}DQem%W<OJRcJg?~q_Gw-dzBCThS8Fk7MnR7t
z6PW-9>PwX;U?b>KyR8aYEk51#C5+c3V1)-+jPYZGyoCqO((EgLg$PA74`AAO*<eA-
zkdwx;GsUR_YRpN81)G6JUqJFHr%v_shbe74w*_anPPQ<tNJ=TVrEg$BMm!?3G4%Wc
zC)3cGOw6B5*5#U+Ur-DdS$+7K>5q1sd>zgW^z>)(x)g&s0*|Q7b(W2mfA?S9rIM<N
zKY^>lpHQBC>_N$4hZHQrs+wX%$D6#)m3ECLSFxDs?Ha1cfnUoYv9LzFqX1a%bkEv_
zTpWTd2VEp%MP%$SwfKE0o-=VqMJAyl8@$8D3THI-A38Q3vL^ras7tV5A+q1;Goe><
z<UkL{)~#9>t}-{nz0UM;(EFEqUFh<wc))9SrC5UW9JpG+2-1c+P1-iU!@3V=$Jq$o
z2zcD0IZE26x4|$o#dr!ro6mH=mjLNRY)kp;DX5+aI0<8d)2C0%LjvpPC(Of6Z;Em1
zvt46ex=uVoY7PAk%ti~UA0iSzcu#&+TBdudzl6tX60~4X-6BRHyIuwANf259-)CgB
zSk`oVtYj4PO6Osv`|#4p$D?%|c2Mbm;{R}P`&MgcMw|y1W>21GC*JXFbJP=wp$1$A
z9!B=0>yR~IB!WqeU%<TYlwZBIBkLf^#0#o(&@kN}uMXfI-zhCa$uXnHXDZXVRv%W%
zuR(?rknllk*hVe9t+`qJ%(E8<V(U*f%R1NZZBw1yC6W@}r>hZuwdO8H+<{;Ym9Wv`
z*_z@{=XpFw;dbDx5j2EO`3mYQ$zCPz8)&(9V#bg4)_iS2od>6UUa!U4$yYum!ye@C
zdE7WNyx_{j&!2y$&gn(2E8YOiQBhZl_S&=LzFpIC*E3E5VZ15b^ZZ=g3kqe+Q6+^#
z+VQGM1wY%0p3eTtV`y;wRv)W-k*f9&-G_pLg5;W%FNiK|7Yy*!kfjht^s<zFdHMz6
zZ=L?L&X(A3lsT9yxzY+gdab{hd>JpRR<@W>b5CfSaOHxqXJMFnH}9LJfEGiBN7!^z
zMPa^EW(SMrP(hbn$=ph8VU50ZS5IhmroMoqrb_uf&DxI&mn36tYxOoy7glYSd4*nc
zOmDrJ_z9*|>%{%3Qy=mN&e4Q~SOu?B3N0NA>sOmz(90bwjP^J_x8Mcp3^MC!E8d-m
z4R7-Qe0=%k+)ABmkfliXJZ^TXDVUKKo{lR*kO;8)ED`3MpHV){^-O3rHBojFV172B
zYIa?mtmhWd%5iFAx6dm*<~w?5DLL54v68mWK^7Jp5loNJHxpr}&lZX|sdTe%UU}jF
z5Jf1@87xN(EZ4tZuch(%**vSM3Cls$b4X?OIm;8DrTMC*YBFi4a<UfE>f90vg;)gG
z1L^q!qpQTJAjQ<pdW<Wm$nV_KaNrtvAw*`v=#`+HN=EO{5C=m8f`7A+`eFHt?Wj@y
z6Z4V|vb#xMLab5<9N33XBziWILG~s2Zp(6~12T*T0}T5Pxw@9fh*<nTAbG&m<=-#I
zjp2vm!@g^SZ3PhiCb;=u;E|!+y3|c<oZ*pR)12rGss>`_Op5JJFA%i8RP-$DB8Ul@
z^T^Br(gwu9vI|4UvzJ6c){!^pXtBWUohq^GO6g;{cF--83FW<!0`v&@9)@?38sK%M
zk!Sm1AV=z_^*H6>fK5Syp(tXhtT2bd%6daLifuyoup(Wl^{{z;wNfpU8*IEi-NmE&
z5==W;>CXPkrchh#Mzv2Idr3x)`9HvAj=)M>(yr65<3iQrlp{S6?6}VKVU3&xSDJyv
z<*JyxQ{k6)m@Ccba(YuWQLDXngGAg+7;XUK+rW8e1FJrhy<dC$gY$-E=kUMJEN06x
zxI(Hie8FOf&_n-zg&aZ&^7qg2zh)<plK<zg7}%KpZ~xiC(#)u=wErBXZ}y>USzSMM
z$Q{;<HItgu_1CZ8I+4<!8pxl-C8jMxZOtqYnZ&Up{?PjB_wP>q&`v*nUO>d@x|uu^
zlL3|53*OeXho7A|y?rNx2-{PZ?Hi9dTq^5YeD;-YN4~q+!|{c;Vn58Iy7;6%b<{h`
zd9CtI?p;aw`{nEZU_T|#=%D-5gZ<R(FGgH^BOe>*8}91%rY_%KEq2)ZXUqmCZ?*hK
zND%z}dUdJpRrKkvlU8Z5$!_=-R?}Codwm{XXY1#z&o}7eo7{d18qKXN8ZxDHX|?<e
zZ}Ye}Z{wBQU$6aiA6<1X{+)b_jZmAu6VF;>!@Y2RbUEsCP2tFqER=zvDR5`TXyA^x
z)$*6!e~pSn)$!C7y49~@-!3SqjI%T}1qOaAsi|TasVt8h8phF8-?54@w4quBHs|F0
zLTC!ckhOmJ&>pv1eG@0=(xWAj3q1ecq|ZMZ<J5nD{vY?F;ztfjTb|!)ugXSyyORAy
zKS1UCO~roGar5A!VP5_#7gPxE44zWaRj(Ftd`I17(qbc0)s`&87HO{M;`{FnQk=G3
z;OWiU=Q<kK#n8JQZGSjQgfgvOf$#I`EAAcKy?ZwVD}}-;@Ng~JPw(S`r;c$H%AuGI
ze;*!F%MUf*_%XLqnpIX|Ve^0lrM>Ux@2@|9b1Px~aaMkQ$v%d-l!?7d-|zLl)M;sK
zJngzd`dEv=7S)!o-&S2@eW!PC{-@i4q<%OiU0NsYW3HEEwMF{emC*)?`nt${7n*ct
z@9h2eYA^DyHU9to$Bfk#G`;QStT)L617-qe>ChC+xApR(tbzqol4AsU1Ax*Fx3&Mh
zC#c`=aO*|B+=pfC>3OhC|KoEwF%|A#L1EgwS&)zKXy_FD3sljw&8WXcDLRKSpmPZF
z%xr9#fWLc2r77(bXrq4rbS^Q1Tg15Xl-%z($Cuwu&(h*AEd6wtq+5e(?NUS47zCxR
z6{hHfrkqQ-bi}pO-nahN#mA@pY5%`IxavaTy3vi4ivw@jV6v$cx`~l7B0Sv8k&7}S
zARwS&WE4+V-I$?q|3?tbJ&r`rndrq|RZB{tv~;ZZFx~kh*_|+i-C0-ob=o~5B;<@M
zJ7omJmD(^%n@0&{_epF`a;f+JE5w}re{t31wNQMxA7ciBeu-T?w-vf5!#!11c%l*!
zv9a07$;m=bMkm}bB{W6j{t+vTqlAfB)cPSx5#PqumLM9oKRg!<_f}^}dtTPm*vyW9
zo_n^z>t1*`6Wh%*uByG|t<BDoTwI3%`+xAEqLFYKJQ2vgPZK9fpW({#=ntb|u`257
zA8-oJ?>ct<JHtv*N{3!#YJIGf5G$Q>Xz_)fJOI@lackwZFksRtNOH*?gXQ3-UIbTu
zVqc}BuOG914OLNEM#eD=v#(Tbsr>OB{wVM9h^FDsW&Qfj4tK~=nORuQX({*4eZciV
z9|J&5#(6j#WYJV8KC?`!yx@1|m3rtJ?Wb@idB<FY@3ucObk`Zl1;L6=6|ZlLxS1ca
zmaMB+so`Xx3)=XoB_&1=(}2mmsYcTl31?VJKmw_b(lo(q;k$_q9DU%HBAK?L(Gpp>
zJG9~4N_u*FF&Hd1H#3s$=eIgKEfWHoLAj;h!IP7dQ_t6l4#K9tP)h7#yp4POBVm>0
z5TOZrM>BNy+Q!Ct=;T;q+VUKRmf1M!Q`GiW9j7hd<Ty_KpbO+jZY7XjnlUVXM~lQX
zmU`QOHbuYrPt^^WobqsU^Y7npfWyVw(y}o<(aZX{G$jfW&x;w${$Y<mGM)b3b{v|I
zn`d5+xCrgoaR{l(NsOa961>gU@hn9x|6$-alsGxB0=K4DAvwaE2;~+rC{{%q|A7#%
z&t@^twLcV|t$TVBHHABJGc(qk@RL0(m5a2ja37n`JJgC>q%Yp%&P+>tg}^6V*Gx(d
zrDpiYMkFM356s~5VldfKSAk}HFR1WD1?V|9r6I=p9y0VFK`SVOy!>udAFX?ycp}6y
z@~Yl}1zYd4YdGOjfsqWN@OG)8X*`r>b@Rw5FE4NVR$3p~r9Y0I@bw5y0XongM=M_n
z;nYEJXhKu)zG@4F5t1q8G>DYN*6f)b($=rV=+vUc6+@@sQ}`)zoqWxy$W+*BE=&Y_
zYj59p!rEGpvw4LR&C%d(5!{%lBOoKwj+=Qduy?NxISI6ljBFjZP~=^o`sPqQsE^)r
z4o<bai4EzXp7=gVh>p(GHNK<>I`;=7*RGd_N93gnk%9IP3@FbHSDSDP;IIIW=!eNj
zZ}+WpqJelMikpXzuOn`SeCf*<FRn*|YETy|EictHC0oYx>C>kzQg((cyw9hvGgMN}
zYZ)2}E0tpucE5w?v;)KwQUbEF#&|v+rx+X8gIK)s+Ord7LPjyJuJ|08%QHEFZzvMh
zaqB9O8c6a{ZM0)xFbQlS=g{soJ_B3E!VKoB!v#xYSn*;vN16{|Qdk7H+Fh~Fh0Zt6
zSk6ri=|6UVMD{9d%0DtN$ghJzUIu*rM5aC#=A0<2z`bEd9|xsC6K~B{bq&cfLY?60
z6&prbT>=?ZV<E|I*7&5w^zhLb-6D6Paq?sun$&`lk}yhD(#_%A?~=ov`QX6`tSlpr
z4e~-U$E!}W-ULA9HCxR&3=@h7s~C46V&|1t-D`qWMR+K?%#F%$KPHy^Q-=UlOg{f^
z8lB786`~Y_^Bu)%p_rGI#JMzEqnmT#&zMl8xfR^dw(PamjvAn8yhMyjxreZ_*3w>c
zsOXa9qs}wAM!P<g8)L+>%ggn>s(J<#D9`)HKUarxCpOfSt*6|1vD%m-33RFz8@i#f
zFLHp-TBhV1ntUUEYkIeyR+oQj9QgB-N;5=Fw(2z74OLsTy}X`qPn>x{@!5D}oO;j!
zwqRINyu8++P~>*sJ*lhv{mJ~OmX%c|QV6E4TRS3lQ#my+F3j^NA@jgykxZw`yMWgK
zdkg(Svs;KF;W05;2-zn~^0TvBs+JZm4A8GA-(6Q&gt3sGaLGV&5(BS@t`G{PYsF4C
zDyK52%auYc-rj8*$e=;TX1D39W1-mH-p)#rlz2!pr@$e%0c;nV0>)GngQC3Doyf>E
z$hovar)K+tef^V8i}4R>>XGLOB;Eic2JR3)gZ+KfmRGD|WV{zEWoLV7$X<p)M+Eo5
z-(SOqA>yhZ!X?}|vsc||4aKN;YN#b<&$%G(A&A&R8ot>$&Nnw}lCv8LRc%8<x&-xk
zvQ4m%m72{Rze{db47|C4GcP^C-T$}i6}#t!!puxegQyW7wwI4^N?6`;aBv8}bEomf
z5LWk)j7`h{^EcbGXOjybK8%Qq>+ENvD$02@|2495ZaCYajMe7WxG4@#qpZ@>Q=t+t
zb8)FNQraTplmQ249X-8Fn7#~=0!>GK^}U1jDqT3C`LW+|zZIHVT9<ax4j#d24)a?H
zuRm@hon=~>DEa~3x=uJ8yD$n*`S@{N3yU-iJ9$;M&WSOwO%I~Z(o*U5tYY##zu&n^
zID9=eJ=T-x$UOLJc^lloHu(G1d*b%GRsa6{-|LD0R$BZ||MB@-ydD-78~CaF2M4zi
z2{B^vFqHIAJ%w|_!}<*zkMXx6qxVx%6tK4O&8HrBhuag!H(A+u$Rx&NWeY8K@g;gB
zLtHhCPLPh1Q_)#dQ=HJLu(KZq$H;cT80`PSyMg-*_E@}SMV|>FIKBEehH6H{$LCVf
z(DKX3=tH}5(#|d~D=Vx0z-ByLeU>w><QezsYhuhTwEG4i2~_;}jxNs4ZnwVm=4NJ=
zKqe6ubv->FBU#n&3^v!p|Bd3tDPCC#Ym;=j$6rnbs<ylb+LHv_bf(Z24$nHLPGvlQ
zewe7Az$<M$_^Y$!%Wkz6I*x}2(ejC}tpYm_@o}us7OsW4kZ{=Gd~YvvOvfGXcRG(t
z^7y+H<nQ7bvT8{)(ioN0(bP=F#6bRytDW4c5LwHUJ0o^6rw4Ak^fg4Nwrc=%GiT%>
z*1(%MWL>0T7dYwfP?Cvx5x9>*<UdAt*WzB=<)5v?FSxE`Z(p!>?OJXn9RA+GZ=7J4
zjl05BV2%TBC*XqL1R`V+I6w4buE4cW17b(&Dk|PkZ3#Emn}aC=j}oWH_!VF*TiX|h
z(i1<hn0O#6WFX7OoQR__lD7Jm3z+ko55V+#|K(9XLOwxQjXKB)(O-LoYd-P>ZX>}&
zPE3B>RkwmdxtRMmt$hyh7*A3^BDgiM-15&iw<4*4z8xDW4X~pw+)Br$o^0TdNrAzm
z$z%8F&@-%2xpi;etPxr2<6LSaLp4C8gteCdGSN40-n18*2MoN67gEBz(em<IZ2PmT
zDj_s9v@T%-8p;eik67ZA!fuV?Mm))=tgKXVq@qxyIgkrdC>v=}q9tjRA#0<Hh^$A3
zl5709Tx<T_yLTTzY%ws@H8RRgPUcc-F17E&u;K8{C41=pLEGj|o%4{cPt~7((fHbN
zwF7^~tCK>?>-PpZ*=e6T^--q)pPjdciR^fJR)|Bzi~hi>zo^XC!J+Vvm$&aPAoQ-T
zgQetP>F7KTV}}fiEDWU09cPWzV2gP=_WK3suo@?QIlbH-_2FX-wU8%KC_ZA+7=dY7
z=DoCNYiGlH6OZk~XHPy}-s196GBi$vkB_gUA|GUqMhD=XzI!|vLFsdh{SF2JEyyrR
zbBOR2T$?DAr+cqmWx)rprl~8GdF1)?l2ZpvkoRv**9p{;@X(59Tj&TGd$gy4M5-r;
z|E{Y;9v&W+zhNgaV&^^BnDl`2*T(|W5xXi}O{0Uh_Iva1F_AeV%%#}rgO)E`5;q`=
zo*L9pYQA^qnoOk$KIz0VnsCM*Y0QmCsq-Hj2Z}kp{}f=`K6mI$YmovoD6QTW#DYZ>
zPb#5UMR~0C-~(;dmqmv$dmh60+n6ub`uR3!J{KGpAeBB`r<Q&PA1y!s?-Lqqy6<}=
zh+RMli@FUoE`!c@67?=}bh&$wUc3tC6jf?3F!l8h2r%fHnW*8^B@v74SpR97RrGWs
z7d{oHzVyW2B=%$RwWZk45s{I_rFk#`hVjw*g9&IMrO&L#ntptLxm0^@hE0^zfBQC_
z?Z*c@BsAM3)RuRNw0$xT9{bjySmcPEorY-7syC+|p1}iz?5Nnm(UFJjP`vM6=6N=8
zvtyP|dXRKkqR_znxzJtA1?XAKH3fU6VDguXOdoQSY0$oIBm`qCqF@}p+%unm<bzKs
zM(oVt3eFRFS3Y%r0h==p^CKlfd-iCBVpv!@8fe)_Z^h~U0jvn??^c}l_rdA=j))t<
zl(6C8);$)tG3=7f{m)0!U;$)ls2Sj6h!-2Bwv~A13G(s|_%8sV8<eX09hjnWV+uJw
zA~;XA*jLn)p`ppBq`F{NhOIVs@rh!<=FGf2b(Ak_4{`#}!=JXnE{MlyZhIMG$;5YA
z0;9EwEW;VwRdLkfK04wow&M&YhJwO%p(&dKKIHF|lQSW^tfTW4Mp~lM#DjCm?!YW)
zX^H-SRCv|nGBCYH2{R+AKLKG*fhTrR`4eQmu&GYOytlT`o`A1<H~sxRx8ViS?fzYa
zsCY=yboT67;4~PylZt>p(o<z3n<zT<>@1?nnSFWCpx{6uH6mYv+hH~~HpDj)Dc`<*
z|Fc0ZU$p9vCT`E#Q4X(AQBjf9gA(>%6rhU)xzz?!w@G9oPP9;cISIy|VRipeP@|1N
zj~e*9LC;(ytb>o-W=(<|UL6~z5pTEZP00o<t1j`JS4lK@K9q6=q>7rFn%qjrv|8cL
zVyn*Ng_xRzf4D<QP*hY`Q!^9=KBA4<?d7#tegdt*ZSI3fjeQVzcd!uDBk1Vg?MtHm
z6x{0v@vE}(@&JRO=7Ot{HjWGIw4%$u?39`@)s_k6HR?k|Nxg5p4FvvcTu$Br)a={0
z&(`jnOT{#-b~6Js5s`ZJUm{JWQ0@hwA{|7~R900j+K<zS!XC7o^2oEQ#mndb`e=fu
zUmk)f$36!IobSrVjx{2DIHOxRglXQIW@af}8u-AA>&1W#{h-H>%U^>B(k%b|k@7z+
bU7C~QNLNnH+GKl*T%nSpx<cOJ(^vl&Z7%6z

diff --git a/reference/plot-13.png b/reference/plot-13.png
deleted file mode 100644
index 156ca234eb8639263e61b15b37bdd05944839e64..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 68687
zcmeGEc{G*#8$OOdn?xB?nTL{82qBrtn5iUX$doc?$hgg0LJ|%#&xB;kJhP>eBij~|
znPi@}%=7o&o%iRj@A|Iq`u+D?zx6xIIjqBehS&4D@B6y1>$;y%?c1u<Cs|G+2tuuP
zOGyVo$j}HvYEE$+{?AZcCIP-3H`h>Ag75Ipj$*w(d^_QE%fJOe*h7f_q9UhOEfC}!
zqNa3H7oD;=V)GB%hVRkpoi5+S#eCKi=TE$#qu{1{uK$%Ox%HE304=Wo?<cjO_ww|4
zi{zyi^-tT}AL#_C{o;;Qe{ZgHMbGI&_%+n&#RPGR^HkBs!Xf{V(bLyG=wDvM;bcaR
z$oJCSHEz|5)Qj}AHJIL!wzPU^T2xo<Eqk!E()EGAgZ}SzA_&$)!GZCAKO+guBqsm+
znQ-O*3x3RhIYf4@=5~tsg!>6X6k57iljnlkzcXv_C7^fAOfoPq@MMre8NbG*J87=<
zY)H>7o*4DKp6Itm0(s`8v(Cy$u8628cdO8oq}<!7l2h&yNN$vR&Kva{ZneqDN#0m>
zIE-APh!yU2b3D!e+pSv?vh3+YUkqKus&YkrwJ90!M^;3rS`SSl46u7CGlSNm^X3;3
zmd~{egm{sz%8kcN);C(q_&b)k{?Gb$W(P9bx4xRUR*~k=yK^U!J74?q<@Bz4qf?kU
zDk`eWXHK@4@juu+?sG2c*5%aO8qqvFJi7c0#>Es^*W={m7EHmKJ-uJQz8soD5>$9&
zb%Yb&aEC`m78nV@A@-kGTU&n_)4+yRWlFGhIxP#vS(F|jvaQ!Bw<%`~vHs$gs1>O-
z@u?c+yboVl{&!8=XHbIw{~r-#Tldov37t{5b?*}W_=O3l?TY?IIZU(DH<D4tlLuAR
zfi73aX@ejAERld!oOHQ5Y_1<?QEPnce`}ZF=?c3;MMY&;03}Nw_W8~2yo+j|&F=;&
zv`oE1H1kSI?qq7;inFkBb?sepMl>&QMDoR|t406WX;9CJ?l_A|z+tf%?sicvi*keb
z0+dkd6XAR%uJdY12?^S@M((v-+iC|MhnjS77Td}s;vUrOkzTsKM>{UaR*u<{;2fSb
zqTCL&FxX5uPTY^%MwHe6zwwlb4+ig<ot-^c>yqcb_%qYE-reTO6D40?U&C_CaEIC{
z24m;;iw^M=|2>Ng)UgI*%coB@Jx1L6r4IL2C)+l@iFC~-TBUUdQ?pihu2}}PK4Rqk
z-=o7{CK(lY%oHr<F#L|yq2?yvgF<b8|7L~EfBu~+t8yGGCg(uQ3pQ<MHSLY8%;Th1
zT<eYg_t5W|9e0dxe<HlQXeC!b%_;`*cWc5)KlssQj!3)7FX6kt7@IGvdHs%3)h*~V
zDrWfaV_01Z^pM%YkD}cPTK{*aFWs&&R`2;yY|N(yi43xmfF9|8e`v=0+S)o0QdtOx
z*DYF^^?w)yO$sA@eqX4YPO^K`{$yCd-Z0l@uw-(zCM@=Umy|$*a%3HL?a`_me@?l+
z-g9Sxc-z%Y8&}o7=fs=5r}f`%HLP|>r4LnV9olQZ<Co>Xo9Ay^7ttZuY~;V8=ik@6
z)o{4G5dHRZ7$aYzrr3cM$$y8QWg{6CqX<%{+x`{iw)&ZIVbpu2uh_VLeUkG?|6#R5
zUqML|ui9YEvuwXD4}ZM^y^;swY-!HkaTcA{91fL+_VWi*&IYv71Pu4wz4YDQij||?
z{^sI%6R){K)B2t1<c&-Ie=cbq>?BlpteB)qd;Tdjg~O(DMSY<eA|<D>Mx$#ow|MAJ
z6d9D~POzAIFJ0Ht)D&GGt#z51e@-cou<0B8=xZ7O0W7+8O~h13yq&f6yUeYK@Nk>0
z<q>Mfy+0j-jx}Ki_)JAfpKT|_m2j2p&C6#h`d5w)S8g<<o4_VAZ1y`ah%7nUjy%%N
z*R{2lTIerv+}8isvfOLETb0$TgxhzVloQ|JxyItPvOH41vA@=JV|`Tn@>{v>Ksh+A
zRl4s3$EF=i#TDZkr_9l#!>t<)ODq0Id(@HShV|}6KZ+jq<Gt5*8V=Xwj>;z_iM#iz
zVFg_0@=*01xucyf*3nI4kC7ZF@j}C*+*BzK+ZkLcwg1Xi^O0|m!eMIhbxFzhGTxgh
zYYV4OpUykjliMe>@-l=`+Yj-zji$+XuMfL)U2j;D+i#V7y4~0%vo-uiPA-p(l4178
zLq~h`klc0)t*DsT54?9t&99Tjo>Q^ZjnxB>SEiF4e|XNM!qr)MEnfc>#At$trwg~4
zBH^;0N$WdXXxi~a7^-XDLA8cpqi(+L`%HUp<Dv&Wd2$Cf_Jb93g`IqU+czYWIS*!4
zCEezAVGSC+DvN8UlY&Pd=IgGfuK46lNRIA}ZrRR#&!KMI8gB{Qox9^dNT`xq4ZR~e
z>{Q{q=ia#grvt7C5+>))S5=nc$@WX>y`_)a;7b0D<*R%)`6ame0Ix`UeScIdS<GhQ
zW6<trLBGuAo#bYl?Y;fk%n*+6{s;a3m9EK*#;#FtdKTsE?sC6_@EfD}xT_CYMn^Bn
z?Wi#acjLX&{SSA7<t96mZ`|SB#*)Lyx>Y&vizVC7Hro9t?B3m2VD)k=)$eVkl^b?q
zzcGr=j^j7@j`yZz6VFILShRvo1qlHvwB2kYF-4?WX<3<>$yblK57``DP48NmkvsBi
zWme8UT)%UqyW%~zXIQ7z%r$Obev3Qny4?@!&ZO(S@g2m(mWKXy!hWRgK5Y5>;%q-U
zZ;TZtJJhK`HWFVDEPLost6V+otnqL`RO{Z>h(|h98sbB_Rq&uFY18|5nsmNm2u<pS
z-GvJms7*5_s>=onX*i_OxRMZs%wYfGT$}gWmqpi2T7{B!x8RJ_j`fB`ViijVds_=T
zGwJ=$$tYQhTgwV_=N&36^L^#)dPF6}B!AC#XH%0gK;hbgO(j8kFflO^9Qdo!uK#^W
zNlE5boO+JSA&pG3KK~Cw)zuDL^iZbuyNM<Bn8z<!M%=lgtZj$dBG~Qb72QOJ9Hy<<
zLzM=rmPSWS_EbJ@^qZDgrFrH~(220o>`9Kt<{rxUOgD*3OBWgm^yA%OZ@vmX>I8%~
z6d}_i$~s#0h{;}R^Vcf6r){*>K_K0x@BMvh$04EDRqWoK4pK4gDIz}N_hi!@ZHMml
z`Ao}>EnbiBFEVX5S)1vSo9S;PIQA}<y8rpz9_%<QXxh|ZKXYI?FYEmK*1e(Ic|G%U
zb2hV$oxR4qu}QlIk*{BiHY}~-q)M*#-`sHdlcgM1VcYj!o;}@41R|%m>HEt&et%--
zS61N2$p^k8{r&w;SgC8*x_g`5MD}JvPhU!^Lv_zqU5L7MPcG6)eE5gU<Ln#WfK^+E
z#I)AfuWxN^*o@S<?e@b?4PLNQD%Gdje>Mb_i00L2^Sa&n0=JDmJ-+lVJ6ONn?nuA4
zd1+qicVsr+6gT`)_nqQ1okJVPS<D5HbKnL2cQovWYwr5*xA_m_x3tb;b`D72J;XIA
z{+tb6GGaM6v^-*K6*6pE8maeKa6H;{bQ|;CYKMheR4hO6tTG>3Dn!kQ^{M%fHhAU*
z)#D6)Z>-o)XH9%tNuT%aeA-`D7+if->%hYuE?>*BpK>UU<;dWMx5&H1bhq*$cZVBi
zPTH*g`aD?eI3_VB7jp^S@bvvyK(5WdW&C?bH(atig{a2h8TA_ZH#0`a$Q>3gREpLu
zS#CDw`~Aq&zD$_$weqO9Fvw$Xjovy6x2#`s=b+tKvH7HlQx12iPaKn(aQb;#Zo$%z
zquW|n&)6;YZM(yY!}nKXcdbTKWQLmWmhUy$BvJSG;l{{JcBt^*sv7^)V|U8k{far+
z*q*8}c~4)zz{AWScUJFBhehL`fBC$Ag<c}28nu{_OQw5+{gQ?u76x8~7V-`Mqb>iz
z503(6ArV*e>q|ZRrKC~Sqea6eZWhwe5ijW9_SpAkmD|Dt@1vvNm*n!t_8P}r0Go}p
z$sIleByMMBX7+CzLr6%-#-Lq^S~O1X$froZH0$2QERChHW>fF2<yg@j{mEaS!@71x
zBmMU$-I0ALpyK!&+Ra3KF?P5##;s<$^RM1|3#}i1<T-{Jy}#PF(O2SkKYCr&e_ge1
zGD6Cz_?<6r?5jq<w)}f}>##q1r(R(aJ^!>{idE0i6^LK@aO2(KPWq97-2MdZMKymY
z$o;)P!^*EfuBme?c-v0XsiL}_QU7$HOu5hYKCEj$)SSZVEn+F$G}KGq`DW$1)_J4L
zE7{~)GOLy=!Ce~?y<JITF8sk!l@*>TICHCMtI_LdZ>%JiPcLm_Z?N)flE|=evtOf_
z<0#uKZG@k;|LZOxDu!B8$xk~J=<G;7lJcnbCyV{z5$6!3<XQ1n%Zl)}fmrf==w$dA
z?iD?le73>^k7&;=yWFVT*Ss|0ecyYyP5zFP`#o13&X3!R^oUkkb#WZ5=gIXQ?ROph
zw4c{A@ywBhyR?H;+Jj57ZR4`TIe8Fc>my?mHuRfCrD3P+%`cPBKDuWpQlPtSaCssW
z%E)%C$WI&B08MPzmoHx`oF)|1Vp8XP1RQPy^n;{h#_*5Q!e{eqQ)`}TF*6Y=j3TNo
zfZDKS{7b|#y*JQB#8yL$a*QVEOo#d<F~~Gc`KEqFCQmxc_<h%s$9l`VWDEF;YZOd(
zzsHM*ddlbM+=ZquUpLFUTBDD#But6R_<CBa{RUYtRqOc2Gwi1#+s=lo+|Sn)y>{*9
zB~m0-{o<6f1YE#XqZ+))cKbqbNJwn^6Bd)#X9!#~)UVWYz7w+UB|MGL2aq&mft^gz
z3f~weEt-ksBTwJ?i|imHyy5Olmou#%^Q#}-DE82*EqjtD-Z}!MXO_=&=j24ODmMTs
zqgtIx!|;F!V{3Esk7?}#{r#%gp_{B?PqhW&@8#=qJyC3X+hY?R5%I66*4Z#+heed}
z^7iT^3o~<+s3$vqXlTJ%xo2T+uA7ur`kdOgl8_P!($W01=VU@6R>vExX#qGwV_pXd
z)$L%T->&63nM1rw5)`bt0ju;_ZemF8{%n(V$8Gs7JRJKTeNjX5KFgsT$=34k#WU}^
z%vG$ExK7;}#wdEs!AWLk-@R>Tl|I%W@1c>`<Bm=4n(;kFknB!D<Xqe{P7_FqZg?yk
z<>*{d&c2(L@bm(l9lcdQw&Z?-I?<s@kkM9t=0)oxc(}107VyP@Mo4g;&menWD>&=Y
zIQ_HqF(f*5AQEzPT)&V~JB|L%@49jf&y2&wFDqnhKp%cxTzuA@IRF;Pf(XgmIQ@s0
zgpm|w;lv6ncg8T~FQN!aOjPu3hd{k%zAmr!NIm3gSTJF&d(a-l8##v#T|_l$o0|2*
zdY&Y+bRlK`O@+29tC2v^H@_|c;EqW-Py3h!W2V7PpCsi`{C%!MA#>a)0xmZ(_lnIQ
z4;Iu}JX5RC;sVkmy2RRq43>;ucTD2BQ%XX(Dv3s6FQmAj!#-z0TY&v0?s&WOqy#?Y
ziLi|aJ;O<|2h8M>Ph4Gf(~tINVl{a-zFvr|F{VMj-EuNiL9jC^A`*$s<k-lR8+p$~
zy+~${IjoZ`!(F(pkPxl|Cq#G&(9~L#*XGYPD*<8DcDTr!tUyL<n(QPXlQd5eNH8J2
zq^CN{ba`WqijN~n_h@qzGU+1YNwr?JO>Pd`7NZbEUW5WJQgL<^HLeHw2fm`ugsT{s
zQCmkpV%iJ`iYCjjXd37n1S_cjgxV?5!oY_jxahD&fFZ2b=ma8~XCyFG!rX;?(}Xpb
z8rSO46=!p==SIW}bn@S_OSxZWqk&W`&!<=HTF;$P09Pd-<^21limK|_$hJ!BZZ3wL
zIp@a4M#Dz0>zuQa<JbUFKt<}@^oBJ~=jB|?;!+~tz)lh=N3WV!YgpgP*Zm$qO76PS
zxMd5y1eCgTc!IMZ42vw&_fo@@Nu9`1g5zayZouq5vL)It^RJ9V_-42eW42?8TaA)|
z->H@bdz69_T=NJpD|k=<klnEAX(F&v%2-O*%w2sRHBoglL*NTI<a)=(y6%0mm&a1p
z6gK@rUq17z)F&fwzRiT2{x0{&Il6=*Yti?*yg54gpLVEbxp4oG%Wpa1PFx!Q?CQQW
znDbPOPpxjbZUL6mxcD-9TZyvfC7BK4(QA8s?`L)oodbaLH<0$=2~|8D5H=;xR#a4E
zVPSEur$X4x4B`%wK96ITfigQt!Z7wclg&3Q=!4{k!7w_VqWq35n?1PP@C%?J3kOG>
zs08n)E|ef<`8Sm(;z~Qy$^PZrBS~B`GR*Xpmc#_B{6Z5Wm2mcCiPN$_?tmU?_%_h8
z;6uIL_a&UwZ6b(q9#)>X<|9GW8G($QPNT=hznFPDC6hVb^*#Ph>FY7q=yLD02(_B_
zo%*p_QxXJ&tA)<ib-*J2N^+;LUfoULU6|o~#~PMe?o!meq%py=;~qz17#*CK{f{*u
zNKN4F^=S!3NfHyynQzPu)-n+a|6NteO_q^0r<gSlxeGvsIfb~C+A%aaP@x2qCWcjO
zGQV@M1z|KadvtSjqPQWI2ykO)-gA0KCzKx7lOQ#j@jHR|wJd5KL!m6zpsr9^5~?`E
z3Q+<ODx0Z%Uqq19&r~L8GB|chn1un*fH5th<=0xmHc*}Bat<a?XW|5mc{|F0DH;(V
z=a`Xpq$5<pHX00$=*@@hboF7CW{yrMjY;EL={tub&)?1x)U>n(y1X{a!?nbuFm#bj
z9V#c(l^mVSK*r)%N?gKJig6ZNLoKZ9W8pUs>ocV~*AuSBs(V2xzhCsAufIPpM<c!}
zvu+^tvpJyG{+>H>MZSBRwZ=3;c2Qp)v!k5E*?zj{@Ew_ul9IA8F<B36(mh>b>!$0S
ztTZ^*V4a+qdA?eMAgxe8=InkASkU4zf2vmeXruR1|H$E5LTLW&b!fWolJlW*8K9U6
z#OG0?LxO`zMjip@3e=c2tj_i9HIG`OAJ9E79j)})VRqJL7MAN*;<&GxHVseNVg{Za
zZvbL9z&JbLlIjG*j8*xA;|(%0GHEh*+*x3Kc;Y*1jY)hyF+c6Kt%M6mxzq?P!-iaB
zq8VClpNl^9osj?pSi5MaOK?rCkpLbgc!8HrbjsONWyW!42UgH?lLd1he~KG-sSZ-T
z*~~BUjTN$$htRN3e@naoSb6tYPJCC4NOw<<$mGh!apj>iVanASm`L4%`+Jo(Nyg1b
zK>o>98S^vLAI_kT88o_Frw`R+CNP94cO7piX+Erb+fMrxdbfmbv2A>42HdrXtbd5Y
z`NZ6O+sgR?JA}mKSg42-<eWTTu@li5hmGUD-PRRXvsW!$iZ{3*%~B-nrgrW=9=aDK
zi|pG2xN(RVvg@<$Yq~~ew)t`5gJ#c@^DG&PDJ`dGgb_a_Lj>VC8}^x7%YDtcnjNE;
zm}?`agA2c$aUK5lZBL{ivQ!SV3iJSabFL897{TrqN`%90I@^2$%ZN058hCUIOIe{J
zGakzmd2Qy19Pc#-<OKkZ^~M$7t|cABOr_=|ITGnyJlgR~5ovgZVYHT+DgBfH0*DZM
z@kbOk8@D)4h6p+b-*IIoDVp$6O8NW^61Xcd)n}(%VAeP0zz`#6ngM<*9hU&-M{--K
zprTy1A{PV}TJGF9nPK25YNVmquxOC-rb_nRFQO=c3*%%)|8!a|oqi;o`1{Gra@gmP
ztaL<K?sG*s0qHN!Cy5UD!OFy>Etwkkh$*l0_ZCa55QNp)P(msFte9iNYH$Rb1eY3-
z_uAdt12%ouAte;WNcA_I)S?9%*LC6x0`!=JGnK;_YK>LK*<`$Q_qJB_E}!}SJxAlZ
z<EV7C28mDS&bBy85YIB_YB=XQxFnqGy7=?j!G`HkCB#6?9oadM7bIW1$c|q<)_|b|
z?E1F-v-RvKnIse?VEKniP~T`c36NQtf_?Y%bglr+i?{i9QdI<Uv8b>xSF4byxcIxY
zG>&SGlWZg=8Lj*sWqOM4IDIkzAC%e-ACPj!u+O6YhdydAq~&isGil?AQDfvojUOj5
z5xtPyX}dDo7|o-4ik?u#0oVBm7jYAgvW8A!Gl}W6Bm!rHxEx}%0cqt?|2FrN`8AbP
z#Q3X(NB_bkH&Ou~N@_`>UQ&gElpr#0ar0ybWVVUxm4;eCOGbYpxrq3*da`S5$?o&B
z(bOA17nCT;PMu(P5|dUQ@O!HvAJR9oqV7r{amZ*z5X^wf@djyIO1b)(H19W>JiuGc
zyI*PgMKxE`^u^V>%tnrQi%<~R8!+`XH5f+1%L|Cy$&l9feYWu5*VY(ExLjZhf)vh6
zb|)*5|IN#@<My<E@8v_Fq9Z93Zm8|3M@+!ZMe|q@{?t?X0V}omA-_p3`x>z<Kuu`7
zb9Pq9l0dSY^^o2~MNa~e%a{A^z!Xd*-TRPEK8CQ(h0@Y%jC`w{R<~5i1#?@u?EWqQ
z6Fbz7OAew4p&%;i)Lx~?WzzkRwPV78)e<;N^@D<1{S*5bn$gV*#`6k)X?<Pq&BbmY
z#@^1P`<v|em;k}s6W_FcvATm&x}yx9Ie&-w=Le!7O%)gx0gz_6W`mMMPguB$BLBSg
zkrnJHJ99GV%`I-}z50pX{Ck^rBqrOZP%Gc4X=rRA{J#mRD5l&%eZ4sth^eNuA8Rru
zcCMqVW1fXsB0y{LsBLDQ!Mbj)j5(V6n79vEq#gV-!$^cUWlut>TFve<p-CbOQQ|#O
z7gXPL&0JI?!ceR9U9YKFJ=b5KPF$ov0UTcYk~1b3AnD$9l%or1T|~Ld0(%KoUCz<u
zN=6YR6CzJ;S6r(?{QTzfNP(lzB&$PwpBsfo_vM);ouUWO#TXZUIXZs)xTuiOxH~hZ
zb3pAjmQwyt7x_eM2<-x4qSdg;r`Af*+Kd-RV`8N$iTMFB4aA3NUF9jO2?1BQr1^jO
zN-B<yy0cY{3y))6Q!Zf*i&U_b$6kw!Z#e%W61&U9+#@<SjWWzQyj<{zsp+5**}mBr
z*k8u~pMLV(ZezaBA)XEm5McZnE`n)>%NKcmK2g&as0zU0iJD)QoGpLfXx<4QiE1H}
z7V_S}#0y-?@YU?8kV31fsOVp&!jurvY#XF30qKD8xuvE1j*cZm7T?sx*&Y-<fPS@{
z@ieqJaTWs?SSZi|n^#~n-4i8Bk+m#H0(;sP3W;=)O;W>BV#8`qkyS*v911{O%C4e~
z?D~uQ?%k2eu^*~};){Gc$=;Aqy5ch<HQ4CAWzg*2cVBM~=-MSGWkR_=zh!fT5?7P2
zW-WV4!Cr9dBR?_ftU$V(G3`&JrNTr&NlNe1NHV%Fm8glcW|jZv)8s01YI*9+L83$U
z>6%2+X3CIODCYPGcBwa=mVMB!4NMl@Pf4%EoV&$MA0X$zs*DX|9x8ntQuTDe`A5m;
z$!9=d+Oro^_)RZrU!w}7-KD{ig>4TYBYYf@-jd@*$#W}LtQdvpjz!&iqSE#;5mlns
zwRfz6f~$nXvFPJH6h%^EnqA=6<aAJ(#jIxU=Ia_29z$@rfm1+E7#AKR$j#KIIY=Q0
zp2F~=(Bt&O^*lEstGQu!aB#rF&i?nlb@HDy7e6Ril%v7jPE*LlcIDpOP!_>tFtM|5
zJSBImG^{kWDmMUbH@Et(X*(OmtHiYO@`1vG!H#|<R|1zcf5)H2)gQdLE+cum-)lSa
z?A35^kwX&_lW#wP^OT$Fzmp=~ovBWD3`eHS-ywOGf7TtBQD#{5ryNzMTWPrQ6uVFe
zXrXg<>&&B5&%cOvGP6_S5FR!f@6Cp;Yhma9%>Q@zB4Tw~I|L3@+B^06c+y<0*h|<C
zG?LGLy#Rd=l%(m-<hI8i*0=Z>dXyQB-)dj}WT~4AYxAEMV7QRsn$|tYvXzEQ70hjj
z<(Gld4rMVRskQ0O9$?>fiyrv8jB>KDd~XRPzkcmn=0%OT%dzUU#`GAPpY<LkBQg@A
zqM!J6vC7aBlLKS0S6UB14N96RQ}9iS0do)mQb05WLIr|%)?EU#a*dD8n!Q!%3{8md
z>1Flr#ESw=;Z~AQ!y63aEDdhWztZp`)LkVg1^=u?F>#ophED^agGL|=!|=?V;B15%
znIWdB1<<FUrqt_GV{T9EN2;(%xoeWsa_EY(^nsd?*V1wmsu?#(Y`~Ilgx>M_czCpd
zUyX0x&v9rzaNle8pB72)&eckmG%@`L!fjN0MeGB=29P#-#yS*f=B4jIRbMy|f_@G(
zgI6lquQ^HT@a-{|milXraG%u?*YAn!4cGzk^e1V6i97m^F`^vfCHu1|a7Vb4>qRsB
ziS&^DpB>kZRsY=I!m?lVpulFR%0BunR2E~mzWV}tqArjn9pVw}MIg`rB51e{;}<k0
zq91D=#kZI%7*LLk&>DIH4>ux^%?H2<9k})}%Sb?ElBxKa2|=>n48h)kPC+vD0T3p&
zMh|L@a+5jJ?q5FRN*KRsA<Vq7u~BOr0qG>`)Uk%tmP;p~n`8eDa+MGjfPgwi#%#hX
z=h~;u)6W?(=^N|oKS%0ojUt%C2!&6@3IfS#)&CNS@dc|liu1cEX0~aHJ%&OU6yPY3
zc|1@a67{cAP^@?x+(D$n4a6^w$g&ZY#0MmrowfdwTMtE{AqvmP%!Ix>fQD<FtRVoy
zXmS%vdG4~8m)Bzdw~3&5iLE;`eqqzkw8B+-Tm4&wnACo#5cgA`6+zga(TRACCjpn<
zyh$#&4?LLn5miVlup|O{za630p4wP2uJafQdvjv;91swW=f;Ed=m;bfuYfGj<-KU*
ze2Q=8n=+Q~WYF_Q0r=R2%K*PJ*C}t`#9b?6Z2p6Sfp7<-g(}9zZut=Xzhyr$w>_0l
z)E;!)938b>8(VAZm#PmfuW&>Tm1yKdhs9<LNo>hM7nZRHe_uW7r2%QR{fQ1VE)Ma%
z6-`aDAoA`dF%np&SE>LO5`{_jbrP$r#7vi46d0OoREiq$0&ewG+qz=OeSbNnu)8Fq
z1EDI}Knhg;S{2V^F`A-6*UpmFJ*+f@R^}djNv3XEVC!xBc&jy#TqW#~m#E+)UU;O6
z3J5YQ8T`olaDT@k9`zID2u(XFJ$`C-mVD6h>C<n1Ban8@FUoto{A^CF$m%)tZAQt&
z`a&ztIQ_WkEFBd{xyIBtPq={WETfZ~m!}E>w?q7~ZIx2}3RCha50deVaTdVfswJhc
zcaYf0?_G>hGe%DYImCCLC1Pyy6p<o$yk<gFih8ELl|WyaMJxa%aX~fWR?_|+oGTpc
zC)OpQp(nfpu1lUk9it{%KQNS3t=iy4fsHUMkYkaO`KFxp6m~rY<p}V)l>S@-fLlmy
zAQ?jaJ?HfIOta%;<K(~^{}o&T#;I*Q<~eh+G=Ud?9Ki8kg3AJtfA}KTpAx}pz(>%r
zkC@zN9zoTtG<2)w!g2ySKA1&IbI2`y5%@n%0{M0ZIaV+wGOp({rD(49<)0;q&Cnwx
z0YL(n8TW`r5HCHmKk`I)4s13EHj;QLR!rRpwj+hT0avVq%uF{he<ZO(kPHLO6>@}=
zBBT{EMl=tjDGDiZcIW)wLMh4<(|#*S+VVL)At^T>a+hu8dt#32K;cy!n-8jpv<N?g
z+vl@-AGEp0Ng^aO6ybS{`mAZG#W1dhF$xl$812KgF4uu$tnQ#lfz<@W0g&mc{s(zt
zTM&(m0k;c)<~@)+@f~ISVG_TvqPx4hz5V?&?#x&374#7$T5x(0pP%<qB|PJdaA2m~
z0EMZ=+;|TU!U$RvSqMH-_m$D2qKDNuTh>q|1&}18xc~CcK<^sfH8f+BM&tb&AGgs!
z&Sxw*(M04Mfwgt2r|M^y^E2ApT#Dfa9uyvsG5|T}+7!r9`MM`HRk<)<fC2@`2Ooq>
z2U-ybhw2UzJP&To2MFKspttS5EoJ<G1ROR!5NNt{o14d$=nXB=X#Gg(KVJ*0u36#O
zq{d}@wh4vRj86oc_sdNDb}V2;Dj?#M9h(N0ga_Y8g}v0}wP>EsC21&z1KhY^7#ti7
zP#A6zIIb~mY{y3(ilisP%<mQ^MEYziuQZSL_gkeuY=ef#3YT*G4cWJZM@#{-mN`9)
z3ikc(%v~?)h>|5p_>idejGhGn$}~K*daIL<<*W|pf=(S;bRg^!=LF1>r{Za6DFk{%
zN!?zjX<Z2GK1KL=HH4NEvhxGs#9lmJTd{l|_Msz=GmPQtcM$m&kH)^Pu=a=!8PMaO
z-IVJWqbG1kI4ea1YC3_Q@Qkw{+f9fa^eC1o4`lquk9G55Twhh5xNW6jM#Rv-w-O#W
zJ+L%zdgd9mc~Z`8j6=;tFtdFghj@`;r8-z2<TY$bS%}Ke+|bxz6?eW){;_$c)GJnP
z=TL%@b3myPTbZ|GJE>zfv*rTv9XmYZsV8LVxU3<VGrB6nk2Sb$jCo*KfmZD6>B$Bz
zW$CoZO<11xE@VlbToCug)4sp>nek1je%$m76mL<R&#%sUb$O5WtZxBx4{%c}8c=x#
zNVrus!e+K0tu4_@Ut*&HU<XhK>}VgD&Pe9^DP}g&lE~xV{w2}x^PUY44+lO)e(de@
z8=y-xZVuW-^EgTxfq6hc?~Eo+8`uE14d3(hp#*~pQt?(2_Fq)tES|1x@>acVqL80~
zy>`AndmL);a)ugD%&;y(r@tX!(oqI!Wg32?f?IyZaA=zjSl6_Rj-Zgl)^(pV6>DCf
z3PsA50?1V@h%G;vYu6^sFU~UH4!+eMq2%%6j7^U;Nr%UWS5h7;;tj!`b>5`b$g)`W
zHs$UKv{WEq89E>#ll85sQ>TZ!Qua8JW8os&tnBP=rSz_RGk!qfn{3c^mGMu{PgY?1
zYnM+9h7dnA^X=NQHOVN{I>~VG1&(mVwEyhKK7RaIR3alA_KYtsylv9;R28!edg3>m
zZ(C_UGk@7}R|@E5Z$Oprfz)&>raur&OHk{8_YO#Dvu6s1Bp?dz4Pb78;MV7#b~l)%
z$0eV{Jv(n52?Yz*8B#)oi%?bz83cdG3x)H>PAz0WQKb;@70?*iLNNp}p3&5&^w(+)
z@P8_QNf2$Ig?VD0)w8(<7ztdm`O|>CU?;!Dd*-Ahnvn6f-Rw89->=Qu)UULS&Fa3)
z_bz*qsS9sF$tckKF^F<RfRtlmaeT_F`Y$*jlG5NV%lp#Sb{#m0P#S`J4Lokr0SU4d
zC_ZrU(L6INWn-w34v9#quOnK(Gqw!<ZR6#wx?6@ki47MiuD0`-LQz}SnbcowUHG1!
zN%M=Tj}77g4Quwi)!=Y4Gor-HLQ(heG5Y?y$B^Ilr0b8tN;^PIKuV*ZXM<8lENG|_
zmlj#%kp5(lD99<m@|{JMfKm(ExP?JnR`<IHMTS%EY8b0tPSfjDfo*yL46LlI<t6ER
zySp`sw66O6439-Eh+F}N7Qjdg9tcpP&q>GQ^+Txt02sib7ptxeUqs(u8PeYuNv;Nu
zD`hBvK}72w)H3kcytYE{{}ouMuH$oY4q2{}_Fa!UX4d#!NoUu0>fsLz9i5!CiGo><
z&Ro*kRm`%AI)cb*)1nnK=@|17pU*q9wIaPFb&1ddmfsB61eEbgY1WRE&gbuP0rsaK
zsYmwP22z2A(*!*QK`Q9gP{_8t%(w-a4&rw@?DOJszCxzc;8*4hR$%ucuZAjdeaiv%
ztsLqy!eM3b!v?r!tmAxS#=|&eM!`B0+4O2ggSXzqW0AnI&?)6Z#LesnoZWaee}@iN
z)IYtcs1lnb$T|R8GQxorfjR^mVwP8vm0a?jVbOb?{C5uuu0-7uB5+a@3DU-~j7I*B
z@1H)s(8)i;-Xk9d>)H$~A=uFtS2aeS^cNdTx-TYxO%()30jM89oZ-!-+!ygbe<pC-
z^yF&!Op{5zxIjIdLBwptlFK~@#;krk{-2JsEwn%|{C%j*#*cu_^4eRfwn)ka(^TH!
zp&uNI<UAN-!6hrI1>P?(doc%3xEo*!>Q?hpoW!5XB|@eqvUTx%BzRjK>kXh`0NV>Z
zAaw;TA`-&BFFR89J!rN!0qWNgK2o~%Xx*6GT_{OkRh|G6Wqmx5_V!NN-b%A61lj8A
zsz?`?!{qw~_km?|62sA^0jP$^3<LEWf*6DzFlhoWp-PJng4lMQ#h+J;Np~S`ll-oc
zcpgT*i{ncIsu65=0Xp)=eh1h99PO|`zMYfB(jy{GEW1_Pl`7`H)~9Dm^v}J&l!jIq
z+x~LHa$P^h!k|Z%{(<Q>st90c{c5Q+b|Hl^<1eg#o{<Q)P;UCU`L-xVP#|w{E6fPs
zKd0Qdo!E8aSPXD?{P(sbp#22v^m&^<<!4cjGjs&6-tFxg5a;=OL^nz!eKkFXpM^jk
zfnq{*gv=5BQGv9w@4yY!wcS+%texvZQFYY^Tg|qCBB&+kJ3KL^rl$#6u|!k?jfp;g
zE?jdRpP+ko_q(Bsh?4;b+`@^vnc8^RAH1<An_2O~;Q0gm4)p!7rt4i0>)-(HoHNa}
zB8#efrCzU9We&F|f$<xxu!#Z5N&qkk(6B_e+Tq@c9vGpj?1zA6EoDAp=I7}3d~J<1
zT%`~F$(Xa_lQwzg>LY_V-&Sl{VKJ2lDkDO`q@n}do|Rzj0Nb03q}Tepp($4oMC;=%
z3?Lsu&6;(aEcCn)MezIw^2lOo8@C#)LYq`mntIM6ZU{6N*9x(3G?2!(uH~YN#I9X?
z6#3BEJ(N`b27rtD-Yvk3|90v!f*w>aBA4DJa#$J$+Q8me^707ou~Hy8h~$tW0ix@9
zi1iz}gcEIDL`NcCRc^QYc3g;o)A0t^<0PXHFQU@Y{Aw($tUX}-g{{8=&~ve$U-WxI
z^U+>&zap*7Z+B+-9zbnF7tuh?4(2q;!O}pTh%cCP$k$aLc=`ZrUAnvwoi)bvl9*Mh
zWv=m!jF?Mz)T2SiaKXiG22nDqM*~SGH7PeUmzVy7NI5|}`M$Z?6nvNY4h^d+#y}NP
zck+8J*Uq?Z6DG?2?05c58CKf%y|!CRdeC|N0yJ`<z6(^~T9~mCff8o9DSO>dCDErt
zl``7+B)Ter@2V&XL7D++2gvs0)@>@3xH7@251q$fc<nWKU0CRJkit#a3EzL$y5~j)
zlkX(Nf~BvBgP<OhX^QtRPWY{YWnFi%7ZH!~(mn_@lldf*y!-mAb1irv#j;QTWOZzK
z)GP~m1t_fCql2-dzNG7R>r;aMo`Y?MeuO~I6c?pW_SGY~8w>q6jAQiR@?A(w(8^FA
zp|csXsOZyV?Em^i_<!Af)<wg<mZ6@W)3ccPHk(tTrTYB%nZbeAtP3kE)#-8VmgOKY
z*xomo)xkLs65Ur=f=|spelS1$JkJU@!Oq6E_%)r<0}{~B5~;I<FO#ePVb6^W^&Bpn
zdF^iI*`(Y7U2L*?&bpQVrWJTf8$ose^FEY6aJybSXqDR!B%<ZNJ`v!CtPbO`22iEI
zX&%%IxNddz89;KtYs)7ee*xmgw>0?{H^sF8f%|gly^dV&^<G%klR?m@fN+Q!r$VPb
zmLHFRqSRb0SQj)7+pwMjE%T0p<i`vlEqpD)vwTF_YN1douZM^Pvnd4|w6{jROCYjx
zvsEK)!HzLAj9aQ6D+YhhCYWxUe0FR+9QPgG`-fRVn@}(E50D2$Rz67K>Vdp|zAgy>
z>dvPOuCq%^eUMUoNnc|aMf(2TsfTJAusmKQXGnQWxz?zvtFpDVb+GxbnOdM`oTM{R
zq-|>6(ou}P!Daf}U7$;NVot2DTz9<<63$g{`D{FEOe`FMR9fM_WC-$hoD4PPN&%<b
zgj3Dqqh)AiwA+f-rf-!KZ_iHw;(JzN>|Qe^L#UFawCU_qq8S5rfE_eGUSKrIqe71z
zEPLFkEdrBuMWH}-<sT(9bar44@K{`B+8)EDO2u|so()xUU8?M_dNKwy#IRQSM<Q<p
zl-63y>GB`{Vyd+-AM^5eO|t_X)6y+~vgTrZ#~d(U(L7soyR}suvf#+*C?k1Blct4b
z4f~AFLS4Pb&#-ix;3_FoU^jJ+E&1M@Zp>Y7`bU|+1b{<WIsoIp+oH|cnOZRl3N&f1
zY~Vb2LuOJ+9;O>q`t1CPJK6!|5Wl%Jv=nx)g5i>+XaR@U2MV<rI{wW2^OUb|mD8dF
zzldr<@0Qau=ix)K4OU_oUx)(Nd||U`4*&gUz8eLmGE2=8^(25lB8y0_PD}R-r=gs{
z=9VPq-}Gl^zmVbVmv2#L&_t^{@Pg3e(s0Ta0<qXIDMNI!a;Lf|eSEq&RNYlfirJg@
zP*}Jii)^d<F3JP3VctIhBtFqa08ju-xt>0!10L|kF1mOMCsMgyk{v%jy+X1a);39G
zQid#CuNMnQ1J&g_tH>G6?F%+o@3wx*pyz2mf7H~zyY(CQf0x^X=Eoh~VB7aT&Vd#C
ztyc=oynx#70tYVGmKrtUP5mS?BNWBis%D<dI7<L&HrP=cx%Pqr?**Eu!*fs$@kQHj
zC+>pvHW>(uF8@Op88Ak>ljFgRxgI3wa}!J^VCI%vcs&fhQx&2ks%d{puzAuB^61#%
zX7vW}?Wt8x`e65k$XI9eUuBGb%U9B%1I?jr&lrt-Hbf%(bD=a8ArCE$L#K&|D++*&
zo9eO}51Jhygy(bLH3a*B%dBu1k+hT|oi%(*#z7R_OZ*QOh<3D$5KcdRS}<(en3~qP
z<{$nMJo-~OW|x#}TRCVz{_bI1P8)}}W~9vKPl=<&`vnhdfB-JvqeQGv%w#Eou14`{
zcX(v?>G$1m{&UtTYaZ0CW`_cJo(ZRfWn;uR9Mj+WqtGCmI4uw<yZeq9-}mxv&wZp6
z>?PWXw?N3XGWM~T1R8ud-hYRY*bNe&HS9E)mU>4lx@8>#J=8abGx`ejL}pGCD2Rj~
z^y}b1f;t5E-?%G5Ckmh$0Bba^E(3gg@_Ya+h5@}TH|;r$_->6*`*#77cn2waHO18Q
zgINHP=N<9eTPnJIN}d(69L0MtEFV<YmZ{-tu<#(rU=TC(437W_KmV86{S<-g9o!q3
zfdIEs0z-vn0)I&c^id~+cm$Q#cBCVVFZ{#b0ZyU+d|D3ilQzH?b6A0~D5>@3afBd5
zVzMHn)kBkwd^7Q-S%DO#*~7iBaE@rH_**icot|VPdZxkq06mS()W5NY71mK;>9LN2
zg#eMt&fHvqsGY%6g8v&pwCE(*BoUD<L<(0F1!HN#_x+u<*xPwv*y{yq&t1ZJ4y-Ja
ze)~1BQbdjH){A?W@VpzC>O&R^t~!rAy(1aW32McTlYA1@0x#&8pB4!2wCOr1#>K%(
zjqkrh3<SWF2Dnm}cgmgF*dBT#XgH(af;<TtV4cw^I<`bj71c-%=#sJ&Ic9Z$AKYAc
zDh9M7Vt6%C^Ist<gSgKP2N;mIf)pkyx*s}`%lT%sg8l%gnj6TjdLwGg-1^qmP|41C
zsEUwc*)wQ?h`DCIui(10^djijMs%S{i!_HIw`^RaI|mJQC-0Q=PQ|JVMKD}PBg}qy
zZ4Ov%z_37_@ohb#G*1Vv12qUPn1+JG=i?1tST4c+5GO()da`u#f4fVhJvbU~d#aOB
z|EyTK92}jRJ+AdsqF9ONHefjXtd6JI@W0nl#$+)g4Ba}}1_oq20Nunfs8}d#riHJ=
z!=K$d-UEFKFlr#H9Am6Q@Zg~kwd+cg87#91^Miy3<OTTHa-rLN+{`mh%`ww_W#LLq
z7`T>hin3Iit}(HU+rte}4)@t!_dd@w{xl_$(Mp)vo0}FH9|i{xaP53Oq99Xy4O;P`
zY1tcERIm%|L@9rsy@4Jm2=Z?e%GoJ<AtE0TB;_Gs*^<b%wl>do4#4*6;OC0wsaNzg
z{4zv=4lHUyn}A9P3~FuCGW2IPV4fl{T4~<^5#0PDL2~90cnu`HK<+AZ#Q^-kDQ7Z0
zEH}SL-M|-h>*pQOzE=5Tiut-_s<(4{Y<}oEF~j*uU~x>nyuHt!)JhXBAK-UEKf8^?
z^bhP%BMC`sxez9~AA5uk&6$EnK0hpaV1N;mXX^UTi~#Ml|A`=lf1e0P^XS>*{*F%y
z!dh-B2<Gd)O2o*=gCr}_c0?r}C^<dy7%1(-)`Va)XZI(k33ZW=u45?S2uE`OkL0wU
zl*2|P(BNoOz!EspE<`or(j^1%z%^6sN=&;o1k?P&!X#Umi!g*B5DxF6+uwGa%{&EZ
z2hb`w?H}E?!-%);{!N+z&{))PF@(WkfP!Qo!%7elw2=S^^Kh`A<ptvrFyF%qrXJuA
zm#4{65}GM9{EQiCLYQZIKuCKFrqZq@;qhmn2ARBwbSd)Z2OgGax2vbB01basq3l@A
zxkSx*nH5g}g4jtKcFDJo+alDWi|gij8KRsFcw)}~{QM@Nd5ER`$@er*>h16~Oa937
zV5Idi47PiWl>vSC@q&5jPn$beT46gN<9BhZ?5pVE&+7Zc7$rifA&Yb4Tv0EHMqq&G
z&=+?3Op9C{SO%@#u#8;}7<m|Y%B%QmCPY~lj<zi9k5gv2rDWfMW!-{qa-q2XHh?Es
zABT8SB41$zsxyOKVlB|q(1~_03FFD!mxuE$I^sY9Ax;HA)|1+38=!9ehnollbgtac
zAW0I2er2VZoQCZK=LK@SB#vwwU3St5+Nf!jun^+t;B(O@$;)ar(7mc9QF%N(sBso$
zq3*qQ65RqFn+ec)Rrbw!d6)3nl+YM%E)HY^BDfX(pGnb_nVGrNw0ZRlD{w>xKu1E-
zU82Va6*ukL{&gzvR25p#|9<=+%jQ_i>Bu%x|6wuU1?^A2jUf+vnArp6<Dey<eeLzf
z38##R?(%$(yQrJp5y>(p!0%LmJEajSRIDg-+_hqq-!Cfk7M%Vp--BbjMy$WbR0+kv
zD>pSYRq=nYWKzk9Gzjggq)2nvHiD@MlZx;)uKtyT%;4L0Tl*vCrA+T0uRqmqWV<OM
zEKENupQC+wwy1g(OwRYY>52Yf{*GUk)ILCPduM(IDh?>+0zi4YMjGUT&%cxv%5#=~
z|80ewyY*+s2dG^M0IS>HCxDR(oS$TVy9sW1)-Q$Fl7FY1^+pPTA0hfd1?pzGAZNi0
zfXEPalaQ=zv27)<+GaRRjAL$}2V)eitp~Z|4Q9~c@A&WDtL{K7e}nPA28u7J0_;S0
zj<IXEGUSjr;F`(vZl{)+9!=k9epU)PF+TkHUJ56&O*Ey{^}&$WV*gFA-!v_jNx3pZ
zf1uogxfK4&p2(U7{^jD3y7o-Pkj2t7f4f8|9zu;U5w}pcPz*CN{qU<{fwnCeBC7D-
za%f!tuCe*_G18*V8|(V#w;DXmY7G=g5IC{MZ^9f0v92g%7j8l|V&UNU0qggmP(Smp
zvP(h&1J3oV-~W!SQd3dY7&8bJK~x$R{kZQ^K4by$0gQW`s3&e|WyPbG3n_8}m4i6q
z^>@H<=t3g$4SL)0ESTxMlHS!Js%7lE0lp1VT7z2W9O9ZnBLvNMbw>*`Fp_Q{r>cFQ
zoxKb;3Uk09;HK`jUav^}jnoCYY;HoK2gkE@4zU%=1*z*Q(SQn94ddd`JW^tc=W}z%
ztEpNuy!*3LQkA47Pga;z8WNqTP{*M?gQ@D-50I6>{31xl27(J59(?hb{yZSYOnf))
zgEDIkJRfX_XZ76s{%_SVVJQ>r?CKbQAj<2@y?lI9m@|!G?xhi^rQMAZ{~~~yLrcyi
zpUe2+vvkWO#*oQCx>t(^9TaF}nESORYweJkAN)UF0HH;JQQJzd!MY2C5NMx)Wg*(+
zKZTx#H6bV7PO1Jkzs`1Q0&S}h2uhc1%J%#~UYvS09t>$mA(Pnt6Wa;+b0i>|r~|vM
zx>rxOMK0(y@5Vwzw>HrsI2zwN^g-K><Me*@i#^yhj1p34*x&xoB5b}|mV<0uE#Btx
z0ii=G7#_1|#{;Pbv^5Y1gOf|+rr#zuszLRFRVt|l_D=xh<o$xt7QSmC)ej{CeE~9B
zUj;+od2J7u!#$}j%Iz4DGp@lw1oD~HS*nEzS>jzlRrLp593)z`n49Sq1_F97ZWoa6
z1wTru`Xz;!Z8CLZ>UNv`FQ0)CcW~2uV_<JM1R`)mnKQC&>AKoDH5?#}SdyHBy$X*F
z`@Z^|b>J<PdYg|vw8rQZatwU)cY)p%l~`*4tCrEJVQpy1lsEN#M3|r;9Y~0$ca`gx
zM*d#&tj`ZxHeP`MHPUUij)g<YnbSCKnI8&Zj0CL6?^x4*5d~(YMVGV3%z(iq*7wCu
z{isU)iSaCJQC@hjfB^&SUPYh<j2qCUVSO1fDFIC`atL0=?_j|c=JE@wtM$MMmR<gV
z8l*39WP@(gn!RKEuS8m){1tHk)<hf%c5dfdfM3@CtOZlHWp5Fe6}Sp0GJd^)g=E<f
z+3!O`1)giu`7jsxSK<z&F%P(&Q4hI)Z@R%I$*8%q9mxDjm}IH$fTel!+@?MHB2j6E
z?rtb2;{^YQX?s|-j_Iy+!5&d?!Jj~%S0fG{Lfz8Ie*@&_w7W!t0x)hcCOq%XJo`aC
z$6PKePLpTH>HGP=68!hi<aeARY$lF#;$=7Fp<<JaA`^q{>GS=~F!BtB6eDUJv7A2z
z1_}HWwE2Cu#2)c&dwmw1c}plQvmD-Munx{Ha6{U`wArw3z@^u(U;ivQ0q!Ztqkl1!
z<GWx5{`1FbAZD%_lqnIbVxKKrBnKQB;rNyEdC}+vEye@hmN5H@uNY6v1&-t}!~!y=
zT3(1P<{_9Sx0G9jwq=lt6O+uy3KU<BIR0n#RGI;pYEt00H)gXV=ZWL3@FE3)27<x}
zaatNip3l;RST%GKCvKs<f*27G=@CLKk2ow7_t2sHBIw$<l@UOj96p4WE7kr-2U`oj
z!UWUq+9Q_5qdDi){(=G|MFSRbd^`gNMMea2_<%v#*xckz?i`%FmaupH2JlAjbn<T{
zrF@p{S@1;e!B#aaH4lQ7N${XWdd|UcXuqkyar)ly4AFu=Moc`7FcC-`T3Ziy^vRy1
zvjQFH|8cl|y4=LcFfIp(v#In)1`qCCM_E(<=daZS6W9mU$FSmbgpc<@wUC7Q#M!@Q
zP%4B#26HkjoSgl@M?s{?Zk9&QMMy1cOLO!S2p_<oE!zl{pEx8}3EMEY%I^PziFxT6
zcc2$JM!Xh^$`9;Ftk0{6B3LO2scK>Y0#L+QaI^Yxz;y%}zpkn5kPY)RMv&Z-#q7Tm
zC!ygj2(3aDW-TNmu;Snh9jvlX1|IN7ue(U%Ygrh@vTY5fCOU$}*}&%k>J8EA(X{<p
zP(7N%^LNx0DL3X<)I>aw5N;D&CbtgQG<R59t^=?m`ZRRF)dz_fLBM|>{(onMmj|G2
ztXS_df$44I*6nSyMm2RtBhkIQ0*1}Spu+b@k&X4V(OC^^$l+4joaacY!DhIzvB?`{
z&xfeMjFuY$*8kQk-VAL|l@N^oWMuBlQ;Ji174}6v%j9IJm&v@q51VCk1+EjS;-ay2
zr<FU;Q!t&5I!kv(|B30(EzgC-<tDk!+@&Dju{6&*@1}I#0u{^LnQEya$1%U%t>4ap
zJfr7%bAc8H=@|(Wl%_j%>Qv|_25PeJBO~?^pIDd0J_2fd1f1B<M(-QMdGjIJp~01S
z)TyX?r(=Sry79T1iRPQKOY`#zo}RViuil-3%)EWz2gj?j`e5zo*hU=ds-wy}p_ZeQ
zr9t=0N^Xz&aK`^o)s<Okauc@3D}k0MCIZ7(c##T%^!4?%;tC&{n$jZTK)@HJj|>k}
znVr+832AL@TCQ$BxatSvapS++W8S@chlM7()OkuBKsyr69lk38Y%-hsF)e8h&eigf
z;B1k1`hQR*V=IEQg~Mh1tt&GEn6Fy<e$9DXs@e8yxaHQ4P>)+Ll9E_-_4Klvn>hpI
z%|7*e`I_3vn9eRNv=ZmQ&n}jqQEmak?yjk6B(3Z=0;k~rX?53T?T;iSm2AmKqjxn{
z9d}WU{-IEJji;}v6vzJ6Gq)Ta1u?XW!<oNHweR20v`M=9DB}J5pC2BPBICvN%Pgyv
zAnRP?tgfugEGnV~7KJM*Dai>kLZlad7bdExhD^-CfodseCiqf`D@!ATKEo!-bI2<5
zl0N0}z`&36f-D_Dyi#Bt0fXOhpc=pz^sb<QZ+<3bYwFVdjrZME-Y^s_n@*dQo0*qK
z9{lKxugoQq<C}tynQwB?gvd0Ou-g>V))U`+vE7>|ONOA~oeazbotw9!f6uuwIE}JD
z2eGXc*tL+|U9Sd^aKS3Xs1&ax0L;B~^Jv6b(oqs%x)GI8j3KG9asMTD!hBr>Ij0Yu
zv9U2o#b-f75c$0TWJlAV*b+0SCh)v4+-fk%hX!(W_0#D)<1oN_0b=Y_A$0pX8~C5h
z!KnN9VaAh=>NtD#WX><V8#^^MH4yJ)wCHxf_9yh<3_K<zM4yZjjB4aoQ;_CRwX(Xj
z?<WU!FmIWGD{CTE)_2fH8eSd{5K9-8>LbfJjZ398``z+PNmTN916edDm%~z?r`W>%
zmG+3Q*QiKAbHyGW?3McLIQtx~rE_jpLb8{b83UcWxN*}E>;=DILjM!I$-rrMeeM+L
zD|oF1tyGyC__Qt@E;__;-Z!{>_3Bk|pKT$49RQCu?RgyUogpi?>Js?Z$|h+#->vt=
zza1T&4?DoLq7<*l0m_ZE%ySR^y+MO=%ZaQo0rL+U#s73nNjH{Uou(MAsuFYQl_fRd
z<rXX{mw(njRBCN+f1T6C>4nJ7fgit74IQTs0W~-{crq>gb$tJ3WeoFT8>U+RkBH2}
z60!8FW8Zws8pWtvu>ea~=L<^A!S+bh;&^;I%oP<$;r#>8!3mI6TkH8Ld=W@{*<WRn
zl9D{Sfv15Cel9RpN6)xh{t^@5E&#bMtK;%J^DRQO1@qL@6tMNmV#&(9vE~$0=LeoL
zfc7c|tr9OEU+00(HLxgpN{+L4w0IgA7@XogU#7|vv%20?m2mSed`M4^5&rJ3fx!vo
zyWgaELKlH#$m=&S8)w6+fe>6*Uw`-R-GISCLl=4t=Lq;-u8bRUY!1&$_cL@-4$SGX
zxl>e4g=T;E{{2ZT_H|k7Mt>)~jtXG%rG1GVCl7eXg?`zS6Cm{7n2TwKg(a%Lvx|!<
zz0TH)g?y#)#j>YjlhD!*n`SFZOBUFL^AIxddnMRDuy_8!#J~KSZ;49@g*o~68R^e{
ze?7InzAhil{dA-*c2A!c-dq9u7T#V$aXgT$xu(`@!**@E=#BI`UQr(m53|C@i-wvH
zH%e+9q%`4>P+pEB`laS%hw~+e>Q2g7LBCbX{e|Ykf>~j<5R_wpuCDGmItuyw_fsU@
zPC}eX&0GijKMFDO+nGAOOJ!G64`r>;w0Sfxbnp+KAnSd`$jUKZQWLH%u(?l9LX$D7
zGJpODjJ0oRDk+hAl%DY#Vv20*WVjX0{p{i5Bgj#o!26Cuz<8sWrZJSSduj2t5?6$N
zWZ{Asc9K8O#?kQxr{4Yh62JB)Qj`_fmVN!6t3|&6(inw+#Os4w1EUxCe%X+5z*}rv
ze_%&lh?IT~@WvconNhRSwD<Ic$DMRp`{P}$THjSR@16PFAl%77t+Z^x-q$YMFlv4v
zE8!LY^4)f!i$CD2fLQHJgXXb`X`k~eFIBRqCqa-8h{bCcE@W@ibIm-$K9q-7z65it
zq33eqKiKr}l#gkts*?XGF^x152mu;_YGB44fR%o^CG~Ri(V<IPM@T)`#)Khn5|$kO
zy}wO3+_^)W<~4t%0VtI~l~+J~0WnHxdN8ktLO`y(0)gBRy8lxY70rE~jsjtt4bxp5
z%~X04X(nWCa`!ejeMIut=R98qW4dP6R>$LB-a_o~@tQnj4-|qd5B!)3Ug?$1vVx=H
zcMheZxI-7RTT7J{QD9N<f$8ASpfr@iRX+tG8owrD`3-xcX*K-(Siyp%s1|Ek3tB^i
zk|^0B*oTgdL13SvdDuD`DJMG{$IvSD0bU3I8_n(K7NksafcvbgRu}ey6wbrcG!ro;
zdJdtYG=Sf#sIE@VJU28_V=9~|y440jdC{Bi(y6FhRA>#=!@XDwPi;P_Ke@NKNz|gv
zn2wX0gtSjNPqTjkn^sTEffou=ww<l*M;O}+2T)=LW(#5rJbrwkxcSh_XYny-vg<>#
zRv{I(^X{9s@ULE-+@DpI`v}SJ1JDcL%%+-iWpQ(J)Bj+Gf)+jifZXcu`-oje^6flk
z0)4Hn>^Rx+-74X6H@VHRi@f$b^@Kkj52Kh~Wy#+?1GFD^j2csLiwgrSFdx&Bz`^aX
z$0G)Dr)WZ=G<l+K=Y^!cO?%blbuHZEg9In(C>KQgM|O|l0IB7g$KO_<qrdo|*hu2H
z3)F<4piaD4$S){p1ts@`WoOa}!9aMQimOC7OO&?(DOM*XY9Op|?84sv#n_vGQ@OTb
z!^@DNkg+5g7GcX+QY4WjsgPYt88Re<%$c>MSR%qMGDf>eA{3c2PnnV;nKPEk6f)&M
zpZ511|Mz|WaeT-B9{cTh-;TYW^{nT)@9Vy<^E%J-n&iHna_7z_VohM#y&6vs_a4;0
zcUb%?_}d^zAxV+CK`s){-O|15pNDoz7Xuf<C|5VX-*$D?tM@vN87*Vy16Jinht2V@
zaRDiLc^w;@2jgz_OoA0hPkut=qGL-MUs_cQ;C}Mt32R_k>V4JyGPjT1joLDGhlpg9
z`-cDAzAY-q5iS?{J9YQK%lRV`wx7AyM{Bt-I@>p1$WvfrxRE~fZ_Nl!D?-v)#FiWO
z(0;VSNr8V@_7ew$k2nsyHb<^M*y%Y-j!KsjtPtnd`R1GI$tWNE`kkSP$&MmMfrDto
zEygb&7_-LBa!z{-f@^w6ir0#I{3=KsK@0s?&^xn4ul8|P)+RG<E;{5lyOz7$e`e(+
zsrhi%MF}HdSHM@or(2DWf}P}H#;l+aw@$v2n{She5~qqz;9A^d?6H<wYFxc2ul8Vr
zJ!@msCuO$af-BrAhZXp%v3A=!J463E@sK7L+7i=$`Ld^{tMHL^(ax<T>J@8>b~&O$
ziz~J_eh5~!`G5!ALlz2!!@{Z2(y3FEVuf<v6oj?(#(Sd1Rmu)5I_D*sN+>nAg%!sa
zmb+qt>*>y)8LinAXHkBXGE8Fg4x`md3f!!mTo8+EOe-kZT$Gi!QNHWOs#n+zXB>NP
zmDn>fZpdAy)f;*99>XUVse3qC6z++T1sHT=nz(Lvh(y-Qb^I=#_D(8z#LXAhI%ZvF
zKu&kHx}iZ4H<pVFerS&zG>&Gc_VHoR`i7-FH9|cU3T5~_@zsnl4;Zz^{ZXvzsB!DV
z@U1jjy)6F}Hy3N|?_vst=ftT~Ol4Orj<8OA|4E4jK;%Q$F}aYpCWj1LML_b57@0eQ
z4heMi;h1aQYLRde)adm5IX5RN6csKPu7=ac)U}2W=g#3r?cd6d;+rG!@M(t?m`eI=
z_?coF#L?+iuzAB9SVmcQ9OoylK;nP(Acao8>o4vw5$-UWT(FFXSsi2_ki-uxEayPE
z-nN}9S`U3zeD>Bc3Bwa7sv72LwL!R%Xp0Ui%ve}6uYJ4xbKFqyK%gK=0c@ya$*M*!
z+LS))?>7_vghNV=NaIe`efy#tN18UtQij=qHdgvgmNJuA80qnBYJ>bJ<WPHidrj5k
zkgSRL{eJo5ZjV`r7XJGWA4Fptp4X+;VN=|VT3%j8;Na8n@liqD7>bA!86zppm3Ev3
zXn*~P8=G&KTD5N-g&fUk%o@GyVsLm^0ZVm{U+HdT<=in7R-DId+tuC(OwG^D;l^v9
zmX>bImAPM1A{eBh;i000K<GVv?Q!)mgE<M0F?ha!GHl*aAy=|PCkc5Fq<7e#rvB44
z>5XZM#6DN3u;4_L3$YY|Jlt5sEeFB$%$>A}A|LXRA5$n+$hVKskqFHCm%n-i9~YNl
z%3Qt^mO-DTS%J#c(`<U}O+kMlP00$W)!Jl6;$pwR?(wYjOjvz@xb98^RrFnua17kA
zlTp&!Rn_aVx5)*$hh?Vwrs0Cl;Y;{&b!dGC0W913afhXPYO874tkZ@kuQIz&=5O@a
zj4$G_E;UL;WLi${4UJ1`UE1$@lAkMj)|kQETx~7{g`9_%*U--|?QfOzJXFMwzuuT_
z$zMVhqrTi7@=|iAim-{&@e1ROB|Z9v|D*?OLZ-=7)<20fwV{Q>`{~oC>rQD|Te5pM
zHa4z1wbNLN_o|HF!Y6;lg_o3?dr~(ZExaUa?ZpDLUKP^L+GB2ROZUp^V5ObXjg}UD
z5`(3fPV}ZYPsY-%^yb-<4s}(F4tomk#2wnt=T0v(VVg6?&UYhSD&?$B=;wb}*Q;FZ
z3`kPFT9BF9)J`Yw*Am;wNKdEZr1cHU{M$ckLN;eMU9M>TbhhznkBj050cp4Nm!I9c
zS^6v##uw)<XYx<|U4iQ7kOyo2Ul*jRgi220eEM`^V*EH;6{i_JiK*n$@Ovt8rlxRs
z`*7>%Bg=N&ItWM#)eqm6sioCGyhFQ$+98{bZ#{bjKXS7%T{ru84_j6Hgv>zX26D&o
z=+*e}Rwp`V1lLC$-5D7ftYvX@@_Z2|D)|a5!<nR+suW&DZbIMywE%{i2c>UceQ2Nb
zqOpRlO2I&4CGd*j<N*K8&Q25gHOFD0Vy{W4IOBKPtS4%tCpNChvYE%Fu2INAj>?Sz
zPaDs)`cV2W$}UY7z3rKrJ)@;PKOietrO-q@ZOD5%;ullTw3ith_LpFYtN^X#QD!C^
zlC(h={)`MValC|u;0$dxwAU9XJxaNb%%#~{K2pH_H9UUEiOuUbGTP51@lM5f2WgN3
z%g)&qNB#}s<-G}2l2=6Eg9-#Lm5A1{Az53Bkb9f<W=q%v2$~Jk>pQh-4ApT3D4F;6
z_Kpr=iz00^Zi-jb@?bsB<Iqi#>{ZeF?Emw=@QAu8so<sUy+4;ae&);;B|R+#{%e-}
zk#}WoeDeZuU!NOcp+Nik^<vw|@gSEOsfFi}pN2A<f5hrPB2sQF-22!$?qe-{X558`
zLA_dxr;|6F2F#Frb1;ja&Aay|7CwWS@0SNCiODLLzwzTcL^yFE43k1E|I0F%O5`Bd
zd)wE?Vc}w|qqM1ms|0Ooi|v#L*TuZUk}Ti+^Ov2YBSrf5=fJnu|Agxp#4+<!!9m?&
zw#}w4bVQtboUy(vHB|S~eW~>*1-Zi*QKcYgR?!Q@9+R*RIc8&HW4ac0gGO9YRN$r?
zH8{K>eS-*D+NeYKC%*=#%25$cs?4lH!_CZs4tHMpn|wNx++ioo%x<FLxY5YSNKHb5
zoFHPJAVKD?7MOAakP^KgqM@^A&yv`#U4HrU<<%YJJr-+bWLKoEYHiy&xeA3sA98ZB
zRuZ*JV24YKs&uTKIFb;Y3zz<!3!DyRio|&|kDz#ABZ4MD=YB<YD{@qZqJ+-k-kPK{
zmn4}?F0jM_D$b&Vqlz!YlTl)M>YW{qgQhyiRt@56)k3(+RfU1=Wa$XJ?-%*`^!zN|
zPct$YPgC?yojDT*Um|)?GDpXqaW(Ru|1O8>B)#%^<baK^xj;B1iY}ek*{U=oVuLa=
za4f8FgsGtug!-h2jjV$t1vkLZ_pbfjyQ>`?r%)isoPEK994q}OdFEd`rS;l|U^^h6
zU^eR%ZTxoZ(C}`e?(Op#+eRwSJd>vou;hX08*PS{0;4+G`JG`iGj2+|cZbP^)wlh5
ztLakfd9~pqHnm)+BDW5L-EYhZ2aLC;SH|yl*u<-?sU~Fm>gcdx2kUfZY$td0GRkWY
zPtUtiVRq+DkJ1Fc)<=ttkB?I`GG2YlZRF$#)sCYmo4l~&q6V2nK2|dB96G=Aujkz4
z$*0-bo6YFC_~Pd{q=<9i^Xwug3*)^EX)Kv;CEnxzmA~+?0TjaE8TYy1(C~4T&)B-<
z<<C~wyi3EGr5}iKhs7gfRBeRBxvi}&2nlL!?ID~^IIg+cq%y_4pFPW_)o-mFR5)pL
zJ!|5)k>^`ggL#OGYhWxxB8??`3CX?wWmUn~MwtrSO%DB=x5huI`2va;%H<sM#|`Jo
zMQ#5}C6MwDYi}l#<Bs#z^jz3A;Q0s-4=bgJ%}r{eI!x`dToM{!>Z9d57e2F?lS6a+
z+-ez4k)81pr<RFBS?#cA{0i-&Zp_{d`?#Wy9*$wuHpI_0QF;Zxrz-ez(2)Q=5$=n`
zda-8T`b<>3tI^lIlG<6yKNV@{<+TSw;j5kIe(ZpXuotTdlnI(}CfM&CY*b=nK%^lM
z74s{ro-42vlSirpZeGOlk7DQdzJ!<x#IlXRf+)ZhuxW@Aa60OWAOgp>`O@r?ne`yh
z?EO3=9QezUqj7TtHfhOUf*kj=vo{u+a%5z16vb;<jTc>FVyLjV#pA}sRe1Qd&ZDde
z{I+k~zK}aKWmfv^q)L{b?d6~HWhh9ue%sR%f&2yuX5!lAfpR**y)LBg0_Qrdjr*0-
z%oX6uR^#=0Hrf4m?~;lZ@lF7&1h$kF6cm6eunmq-%t-8ScusV%Z&BYGLZ{b$XoXFG
zNn8j!(B}kEG;uuSpSlr;=)Icq(qJdj!*dG!lq**bP=l;5FkSy)xxa33nv?xHowKX4
zj?S#U^1|mnXI=m_@b21%1P|Q?S8%P#{A}qjU-AK@6oFyI)tg)QSfQ#MGS6=vtC|^U
z%z;@?Q}Vvepa9hncm8!*T(^?=d#jrDO$oa<@bU58ve}Z6VFm1jKq>)e1rsJuufFGZ
zhnTexW&}MhRqR7qi8q3YNBZTPNi*8~<ri579<M_f*ynegQT}~l=<b1$;WJ9S4j=0`
zv0slpu;~=<_00{Y5&b<<DB76x+FhVdXBXIQa{6>A2yOhJ+--&J&~+eR?>jmo9lD5G
zo$%C{T?4+b|KY;zxVWLj{*_UcH6b=80;xmYrZ!d1;0-3x^b>=Ciyt>}m#q8Hg<bE%
zqeI>sHRp#LN`H5&;rw&we)Z}h*Z1$=M=bA6NecEVE!?bwXh$DJI)4VKsoEvETaL;O
zeS^(I7GBnsd+7fw#!2O9_xrZYMjx&gq+136!uv8mng~$k8mo)l)P{gRUj-AN!KMnz
zgGazu?;c}va&j8~@ne%2haNqH`2upu`#Ji7iIXq&=B@mgCI?vhi1RHmHVaY|w#9`1
z9aFz!b!U&v#G$ByqK}jK>dGs?$2X1SNlQzU*pjaCPV8zZ2ES6Kh_SM~rXnPS2~Pm-
z8*L+;AJc;#^PbT%%YWlq02mv8i4-d2np_n%uJ$upGoe`%Y{DC^F>EUdrkiAGNm5}T
zp#T)hZ`KS5*iRY&Yf}5EgHm&IGl}Wom*`V(2`1yOX6L|Q^nugU)9K-USqFw@4%#P&
zV@_tenModqHdE|DCq-(<4lU)x8p6dd-Cu%>@018<`M0h-?KsU%f`tTAY5;BI@h@M3
zvL;58^-q2R{Ml5uxZ&%tJ~mzOS<@SI@;7T-g1^@bIU0>_WAoO558Q@<Do!t{MH0?O
zy98xNPE#4IMs#OL)Z!mS4%MAd(^dQTbaaGZt(1UEW(P21b)o;jSN(ayd2qV8fAv!R
z7-;FWta@r=Cviw0R9_g0JBO?bP6fBb8<IwndnVmemj!_tgHt`-p8EU6c&ac*s39WJ
z&sJFp^T+8qZ<tF0%lg}Rj-n~xS9B#%NkF#VZM*J4KsY#U#C6@Orv<`b%XRCpIEhzT
zFyhetl)lbUKV@#cf7<EIbz{^@Tj&MTbjJvk!+H->32r0hoO9iqLR&?STa$jO?=(~A
z7&SXR(<Ir7vW6;{g^4f*)yj1cMc5XL5c;$IM|Kc+Cf>h@E;phd5ed}*ng0UigQg?o
zbN)T`RQn4{zp5E(s}RA#9Or(lLF6q9oX;WY`Og=CF<291y>PkfkN@2g64aGvE0(N3
zOHK#{WcA|Z%L)aypY!up$cAv_27wmVJeT}^3$40=xg=L+-xB?f=r}?h9IxUovGZ<(
znB0rE)vh~Kude7z^oM#$b%>l8ehizO=ayAW-}=jh;cZz5v)V<=bH(QyV%&er3f%k*
zog&y1Vn0e|^$8&EkwNMqGqc|3jZZQ&tHF`pGwhCr=Lu2y#KPF*^JkRh1<yFM%S%)#
zsJ)EDqZHQ_`!SJ3X;(4`eS5v9bCki{VxYsHspLWPr<Al?blF)Gc<R23w)a2RZ`_ot
zFDIMOxt3MxFQ~~Q*_i8~<K^dcZr@$a`&d}F4h-xISo--qyJ2C+^zv%uArSI|5F9^4
zHB$qB9#`O;$9{DTV_(Uf$z?<|*LHWGrAe0dFLe|gGCifEQ+Ms6OFEAGBE}iKbSw3i
ze)3sIREbzL2i1#>y-7#S+dq65`Lc<<L&)WFUWj*xOVp7QIZKld|NhHh(xKGjpq>?c
zw}p`kMXFqY=RXPzkxMSDhG~GrfNaOw@2Q}AF=E-1sPpslmdh_+5;+%thulH+p58)t
zNdmUV@u9ag><o%7OdI4;emiBnGa)1xD-EeDVdFdNYfyCfZ6hRq-Y}aEpD6^9s~o;U
zXZ~p=ligx+b9eqJcS3+31BO%Y-7ek#kb$3n`-#VWt!R%J&L(kk`EW7xasA)W=^8ob
z<Nq%@ZCH)-jiG$Y_0V1JqsIt6XL_T^<_VmDuLG7zq}{uB$M11AH&pNZ>%JI`c0(^l
z91#jxmS(#WQa+{)+ohuq$BP+Q=kNyyB9ok9oa`I31vv8@5_wWNAc0mN_1i}V3&&Cz
zsOijs2rAJ!LKWcM=YXit<JBLKf2|nnN?B`SP{Lj;y&*?mfBfGRYYN4@R+*iVH~SCP
zVyLJ%u&o9hRZ(4TLO%hFwjt?D-u?RwAav^>Wr1O2$Ty>62%(W#G65<pd$xO|su!g3
zEqZ!-zvc_d_K&(0Aj-o*mlw9>rJAvIXi5BU=<=86Jde5BhKIX%kp?0vC=_kv>C~Wo
zw%|)dLAG}H`$ro7(-YZHDTR!M@PfBmU7Gv&!k$Oj>66m<;+9<(2DU=Fb!)P@b<~#K
zJ@efcrQe_Be_OqsptD%}y$am>=HeuN@g9PBIJ;(Y_UsgNZw#b9-wD}F(4Rgo313;6
z)_k^|t*RV-kQ^)*%YGn2f<$xe<5-xT8<7?Bos-$LOL*6StQo36R2h$-J>xT54?rr~
zDPAT*^TY|Bl)xo>_`0quzWn_lS{V=+xuRV+WG%e>@!zvzY<re7=lXTNCff7Y(sq4J
zS?&A(gA(TmZ!x4(9N11eZp4RegToR_JvK!omUd`#VVUdZq3FZnxAY&SWk}NpQO9#>
z^-Rc)zkmM@G7+tAC^q4G5-mvwOD0wqT5p84f8yd$?$*2FmJN-J_<p<^;(!Wkn{{9-
zaaOsFys=E(c>N%O5x;V<x3i0XyFByt>sJCM*1Uf$?6EhV1W9Sno*`7>Bq}WJ+$6lx
z$&a&%#h@j#X=pyKUJF6DFJQrp#4vUQ6?e&l;^M8az85C3We(y0x73K4kePR}F`%qq
zV^kR`*n@QqvWHf?5{~A!ky|1a7H|DCGtjb0CqK`BMkY@_*FWo!+;oRlQsLX<n0%B*
zQ?CY+T7m2h_&RF7;yAO*cL#3Nb70dve)5FNjES|D!Dt#M4ysS&D<~jzJh-_E|9RPa
zB=B!X@(hR2p3SwehWEpXk3!*3MXsWxs}?Kjrr==E674I?6TLij{$vS|8T?oV5<ylG
z`iM*L5_xB0x)id#qAVN=;c|66xm_|vMMWeMN-#B!c~Z;x!C{LS@w;Mgh*T#WDcxa4
z#mq9ViF!`NlFg~Dwj2=_<r5r}Wozo{zXlj~S&#+vejCdsHb?H|+JBQhD!4QhrFzw(
zzd9*ca+FM7*jLOf&zCy4aEZMqj8-2obKe`~&N=Kp^r_l<dK~p5b(9iTa`G(s)h_1;
z&3V;RJGbJyRZP_fK6oadmJU<9CM!@)6pq8&7OAH@Jvky=_D=ln{A3<%CytLGuk+oA
z8Z<Odf2m5D9eCXSv_Z`bfCgc-lB%L@V^Zd}d4UId*4`PrrtkJUb0yStV<}+I=&BK&
zDFU@Dq1>^VrROl9)I9&)r*f56UyK5d*j)q=3OEu7A^jl5;YY=)ydL9N_VdjqO5T6I
zu+e~@4#o0J!b*^|WH_hT6^o}*>&11sP1(8ljsQvj)ZIM|;P5jrIpP|Y<r6||-uC3Z
zSFV@tz1r!wa=+{vjY5=Kld8+-|6x8|WBt9YKK3Zt8J4(;!gWKg=$(nK3)aNKH-F(y
z$5f)k<yd7UA2~25M8JVSpX-Z>HaB2OsW<o9mQH$qThmESPDb*ji3|<qJmN)y(jpFo
zyoeFaEG5*=$5PeJJhmt?5jTP+ZWgTkrEcsT7rZMWAv|>?M#-*E*iAk#oWfz->+_{Z
z?m2gEuqD3&Td<`N9SLfHaqc6V4CI~^7H+{_MqdGklG>S{Fis2w&_F_;RPX`t_8#VT
z3XYr~Po4hR=*<vBn)IDwA?_GrT`EAS%CT8d-t}Q}p%Q=HK4J-9R1y^_EWy79+Wa6)
zGLAQ@uC1d37#mW<@IQTpu=V!}3OfCp+KMSH<pB!O#zk58v$9aqx&6!vok|pUdj&w%
zFi^<KDX(J31IlNdWJLdxqMSHFTjGXz_ARc5&cXgBfvMazx#(SZZ`?=k*!+h8izFgO
zk)_EFiR^t=6!A!L-Q&F1CGTu+5|}^Z_adeMI8^j(eJ)#SE3;nvD`5$8E-L|gkaL+z
z1a}{%i0cFM`cqZ}AJmxP0B|>De#Zu}YC{%{MzUy2*CsJVhWqGj1=JHS$H~RVg-?@h
zpECsl4uJKHi}v|~lhW5s<(NrmGuhsv3@fN@!k28+0nq9FW!_!L|CU)2H(t;16VHDh
z^WxDOOh!>w9Kq=n9PGj~IxPP8Pl3lSna}o6e)c!_hv-v&49<Dp*<4ib@G5(i6Id_5
zyxyrw9`)St6R=8>!gV~P)TrCD@9wm@@rc2E;#xY?L;Mg{HC(+iifn$G5FnDMm`brp
z9?_%XK<tf2`T3hbDOV%{s#CWztPWocq~1L&-c(0)8JEvaROe94MRa5SYwuI*kAfgL
z1{wnjqIE+L0xv^(>cKk9)j>Udp7bOm<55Rj8h?mr5hK|IG&f(>f#&l&M~AYq<d7xs
zt?YslzlmA^P!J@Gb!*3@|Mc=eB&$?E5ZmDGd#;fhS5m9MW{{m%UamrljsuaF$fxdd
z-XUPxxXVKLb;Wya7_%SR19c8Nf(#5`@NT^wORb&#{ZT)DT*4B$_xiQapo{nMx8HV$
zSjKNI6W9Hm%&~5CE<iRasv*3jq=nwpxxugGb@sN61iq$yew)!79tV0VWPu{F4dn`K
zO)2|LJX9!^{xh+u)tC;^kbdP7s)dtAP>|LT2KxpJuFFJK1$`D9$pvQVgCSclahh^P
zmjdtuv&0_VQDP02IK|alpERn03OTy@0uDw+>{UQOD(L0@x6sUm!gQ=VftsC^UZ)#_
z0Lkj|dvPpWE{Nu6<)m4|Wk%_{Z!dtBA06p&X6E_wqvtCS9|mzNet!Er%;49pX=HSJ
zJjRY2Xf}aUO%&bQQrh4%bAKIs6%|iOX#Zgf<>cWT>G4r;Ln}W*d(4?Ru~+`{F~L@q
zk}x7(g5D*l>cK)%0L=zq>*g>DL_d@`O^(kCS6=NHtvKb%9})!DpmOYQpf0apn|tU|
zDCZdna16&ivM}I+a&bInkvl*YFM)eE4kCaL#_$IikExeHEEa6}^yq`H-vKV4;N5RW
zsyEl5)#L(dIf-G{hn?7?GfWfOCqfQT`u>$#f6{0~JBjs;^HxtY+t&k%#HS;1n+0bW
z%u^yZh3-B)MzhEHJHTQ-w$Edn6G0t(E(Ap-sedE|Hk9=PjZL!`@Gu)Ikr+DPt9-WL
z?r@+f56$!%F0z0g?KBY-W+a$OS~}a?gK?Mw59JW3Kd5x7V=gIJthRM&pLO|!?4R#t
zR;TpUQVV|Jy>R*P03c|*BA+YcE_{rqlIrD_*?lPT-omCoK^wi$z6<*c+dR!Iwm%5Q
zS9yIYibPq9wD{7Aqqkbec10y`S=oEyx1Z&63Xjcd_u%!U=^+;L$<L>5(`bi99QqX^
z2V^-M#!>0ln;=;{Yhpr2Lf^Ojk-=tDF1}cCT>`DUkjEA%co>ai<IoCjJzmKZNrE^x
z<RS;QyY0mvGs^jPQtA_vcX`}7eXvs%ghQ(=gKpoFS{zV_Yn=YrWx2PIrz=QSK%(xy
zU%8@;E+^x9MtZQ)ZrR+G+^1d1Ric!Hz>oE4)(I1O?KmwSG!~odq7PFnwu`?3a}F;L
z%|}aqrm;L1PNt*hL(%jwmO<O8)q0F|z<1p2zpEywdP-x=nsiY_j1{}L%GJ0TtXFq;
zluH@}#1Ahj0y5opT(H6<0bCtaV**S%Q!(2RiH@H>4g9jxs(gaDmZ`{r$_UMC%TTo-
zUnFTL35-<8MkDOoOXbRlak603f_mf5C{h^ua`GbD>`QvX3V46WF8+L8P>y%~Ah=VT
z%$T}{K{VX!Gb2v20PO`kir19O@XPd~KYssP&t%DXrNc@ByX7lkyn+l6=&Seh{S480
zXM(r?xDr+!uV@|8^+tk|1s!omq<gf(ocb%8;)j5}K;s0cgo8JX^0ups)dKA%rjj2q
zXqDqcawFr_aA&8dAM{{7DS{iL-JtOVaB``{?dE+E$+X?;d7v#C!ozB9cD$ft;3b>t
zgwY&psa_G_BjngLZX)=4;~-=b)Q;jGs;2JEyJJ&={Bj#FIXG0!wjPKAMR*$kD<ql~
zKmc7a=_)Nh|MIcjh`NIWK7wbC!wf@8QjdMLC8xK!5)FzxWSThs%sN5Wz}u8>DnxT<
zl*Jqr4mPHBrzsR$^au_T;bkX{rWQsw$xkHND44b>{nZGV+6B<5WB==Im__yW+D5Wu
zo@155)YEgT>({1**A&P54N}y?&aHgj4DM}U5C|#}pZ0buu=!&~#2!DgeS&gY>UG+o
zx_eYweIdwy=(v%isX%Sd??l3p-k7d<+~|W^3al=B`nl1U)wL|&cLVKE)IS_WS=JC0
zzJf*<ciQ&p(+z^B2nJ>aSaLa+dW+KK{h*cg9=&Ueg*yE*JAfV8_!gwxZiRwSFI|Ug
zR@9S7%!*h?)YK>?iSo*IO^~9bNOWL^)uInGhEeFtH4IMUHQuuqK+_0;2J92uNJF&t
z<^d0w27*uU>TAYR7r2s$@vhg55l1(Ca4=Lj1o$k90OTg8-rIP2cno0X79DA2D)c}#
zAc~tVd)ahj?n$k`&=enp!=&+kWFOQpb+972fzcVCp~>00fX2;G6_+-J(sJSZ2GFm5
zaD`D^K-jbbP5w_c!GRjj&=7@i5MCJ#TIJHYq%pB*7FbL%kr?h<gcp8;aR&qG5{6IV
zwr)=3JrfG}X()B$s=WW;2DTL1jq;+DKdWHYC4IM7tT=0K(+NS;?9#GoOItL7>7!C1
z!6MMIzEj2|pN;e!I7+%K-d_n=dL)G>h6Qj$u1neP=K{E8YF;ayOtg)L9siJnKo`c^
z0-o8%IP7OPOJ?Wdy1jmWU|Yriqy$eb`fI%gOIe8I@^ic`UA7q2ct!?2347lKwkeK%
zC0j5yO9GF5lbk6HG&Uk7tqKfOO|Cugd+2}_Tw4LCHN3hgnI483!WO(6Z@8Lc6@#2f
zc#+|dy<Gw&rC~QtWp^x=4Im$l&Hfh#IJmmL0hT=oQZMQ?zSE|Pjt>u>JmNj9O3;W%
z!c5P!8W!wN81W@Myy;W5^RA7Tmlp#+OXxZ8!F2>|H|mD~jnQ*K7%eaW?Q5;A`s`iN
zxK-%4<-tmLg~mqR$;sn^C@?=06V`6E!pmHc(ELf1I6lgm9}-Job0|+t{cds%{Znr>
zxo@-K6|zpY*rn&r36m6#r*i?+>+{u*wd$>vy775&mBe@Fp?&&M{yY7Xds@?QxaG&z
zQewngoi_(p+#Bb1Ov}jd(fR4^#y^wdviBR&P4|{ix$3A4uI$Zwo?7V>OYwVTV$WO-
z)avN!3WKc-J=l4*lIqi}1eqze!2*oSU*K@t5>Nr0psK+fVwNpHK>)9wa@cc@=sshC
zbgLHL?xFr!X=!vf(vNv}>5I9Awju9ATSW7Y&B;@zBE1c%N!fU>sCq}Q(bDU4zk!O-
zyf-7U_J|>G2(Qv#9c98lnC)_tO7`dec7k?4-y}_OcIk3k!^C&*q~zOgKO)5A&_rt)
zKP0TYe{9w#uyorF=geNPeMfF>JQ-?L>b5`S=E(Uw$^G#-WP!X7=B2L6;Kd^q!5&xt
zJsp)O=9TLE^@(O<(}lcWeG9Kc2|NbIKEm#$;vjCL5vXBevK5dqj<;AO9q7a1li5l<
zU>T4FRoobf#SJ2{PR~XjuIPPd4CZGU9BB3XGRF+{UniE^BQ8(l#eas@t{kNgh``E2
z=m-q!6r`vJ_<rm}-b!K!`tDts&>*vKQ~2E|$lq&4`-{<<S@~6Wpca-wQD{^M4X<EU
zpcabC*K4bj221lJ&R3SZFUSw%fi389`Lh1SMg5q=;@1N5=WhIKa!EyU3@oKTbc@>O
z?{DrdhNwvcA0~O-TtXZ0XH)%F?keA&hZz}T3k)ZX^*M|(zaQqGQVq_UP++TSW56*m
zwQ^aJKnN-}bqtA0I)^9K`u0o=d%CK@5fGG!H*ek$l!!?dWRoNkAe*XrPkSe)btF?@
z$J!?OFP~MqoCAWsTli_lMWzz`DkjnZ(a|2;zj(V8jC+_8BGe5qA1DVH6S@-7_GzDo
zheyP(l0+JtS!QZ8k6&)MC4biX{E~>xm#W>`CsQY6y26&}5@aOTR+D<ugQ|h;DiM8Q
ztN^r3lQ5Dk{ir}p>{{0Mf!h^~9kW*1S8OjQjBgv)uN}I((PwC@>@8qcdFR?)syd+&
zWgxwRoG?898(5$}IA?jXlA(p?@Llav0xS2xM6LRovv`W9^~mj;3S8{XJD7X*>-u+b
zMHl2~2hR?dG3)#4T(;QQ7iO71;IwW_MWVATiLJ@bw$mXTU}xspm+Q-xK4d?~%*Y;i
z8w95qJowuiS{4!sXNiqrTmWLwQ0nh5sn6j#)RcN)EotnW&j=q$!`gc2p2<Imz-^We
zgkYWUnh+DXX$7wE(2~z%7xaB-k6m7wZFeg;CQ-%8@<k9fE_F-$T@H#T$V3h#^ULSk
z-P_m|DE=O$0y+vnt@VHIjsXL<-G3E}$=ORKHXkfzP|whAV}({`r(c|>hld3~Nwh(U
z<^w`&i_ave%)reAh>lOb{BnXrmL`!GfblwrLrO7t2HXNvgIosci{hjKY^!haS2|X7
z<qR$-Uq9VNK%OCl!s9c%4bahjKK^&T=Ew93pUnPXPZD<<MFDrpAvCMpU64u8r<?1p
zEEina_jTd`e60pC6HAd~(RfGp7Y03E^3sjvj;fpPoebOz*tQNf87*77RIO{-;2+-4
zKT}$irG<{jMrnz>ZyF+0Ow$~|_ts*HQaGyD`J_(&rS{ZX7?IY_SjPk-rJqq2GUejZ
zJEokt+vLVHe^}<;hJ;yE_+7|bsufJ8l)D-F{E<={zp!h2oLx2?bF`_h^}t%m5Ft?f
zzmsU+znyQ8SuUAvsUn!RAm4NB^ZEe|rX)euBl-XpdV6}W5oTtu<lH#=4rx*-;-><$
z;1=q>X(EY7-(l{M6|uaK3hq<H69V1p!m{C4@WL$BBkP7xQI(YR_^&MYul4<Kb39>W
zbR7&?2EvFczCRCYm`u~cT#?2W*mfHiUzWva(0xA^Ql23OBL_+R=~i#v!@}cp4r&0l
zDt@=WkCC|_VIzVv4wyvS$w~H44lbViyk!JQ1OOv_9*)qq_wUi=f934S;!xfgZliPt
z9e#kz*JK)lcWxKhlDEz|FQD<$hYvRDLXE-^a~&rq{KD)ses^e5w_1|rAs~Pt0_ijt
zyQU%Uu793KZrTpqqXcGz7TbrFpi+VA2c3g(|L%jSt#$4K2`{|W)F0AaSAu=#sC8Co
zpS54PiHS)#gs2Ss@gFX~2$r3=U4m^~+$ic+3icT|MRo}x8X_V^f(hb!?Qh?`vkFXw
zlm6@R7}}A9IgPk+M*!ZrGFqK;nwI1+TmF#6<oetu!vqRP+<{43KPIv(u()o=jo#pt
zyX_j-4>PuVLhxgNt$1@CF_H4STxk6qvUbRT<Ozx{ZtSxBrf{qz?t2*-Z%RnbeWOa;
zx-p#VucQ1yMTxq!=2R$h(vn7y58Nw{keM6N5nN}4`tLJQ?1TAmEo-7g;j3o8&itN!
z|GdRt`}!AK9%o;HO0KzXPp?O8?+CK~8*@jVyj*4C;%j|j->!gMg@}}r2};rv!)A`{
z1A-k~ACfF?)_2W(--6SKLI#lqP$6`KsPsg-%i53)-Mw!WxY&3O62KJx`T07@RDW)u
z6b2+V$8lnG=0`IQuIT&qxsH)LKSyLEiL^~F33n1mBI;5<q<{|}N)W^W@UGJvMaeVb
z*cyl_#01OHwYA><)old5fpB5}Sl}#-5du*V-wmQWQMdW^)Vf(ug9EEdkdglRG5Wvn
zxS+ZryQw>P1L9RdrCg{J_^!C9NcE(V5x_uNeXQdo)Nwc%Du@&}JK(zmi3+afXEbZk
z#KTmPvZ(G$#m_h&;MqiidM6k62NkBbHF$A>pJ};N6{I6Nh35t~=gt%PspmR}qG`;(
z<3l<k^gedM5s!KoabUnjcmH+iU|PK^=ry4ZmRk7X0IpOHp6v8|oFV;+Oix8emQeM?
zs|Tnuco7%#A!2;z$B#5ZR)Eya&fXpkTz^uP0!q+OqR~_88%8^#OOQG$=_l5Z49wz9
zAVDE9Jfqt?`@%a$eA@IX;^O!kpVC0{hU*6jIube)MiEhBWH%*;dr$2J2wX5UZ{8c0
zTl70b@x5f#(}wi;%cWbp-!>C0730rr1X^<oX}HYnDQhw_<TX_Qn8BqlY?g3iH8N`Z
z@ZmawGv3piH%E}&M(fSU#<Epe$X`xh17u8lT;C#8ar6Pk`}_yJVh4df#R{mhh3Lr;
zg}f$Z=+==3Q}f-<3`C-hSLJgUq!d^VFq7g@tfM~%ei?aMLU8K!>U{U;um<P?!ZZDn
zl|?dSvOG`%Fwq>FHAqZ{M21~|g7+M~pI2K^7`FR_Jge8c>6k|rD{~1Pl6HFi)1lG1
zG|5x3Wh6`;<36U5d{vFjrWF+lKrhc_#@+SjMG#cto%N!b%lpsVS$FDhUk|fTaa}cC
z5%P>k*;>_!UhtT}$;wTf`*?H9yN{Rx{H|#NSV4oeMHKj&Waf4yc7nvi6}@1*@rchY
zBSE4FL}VtBpw4AQ5(CN-g>o4{3pu?4*c`Sp9UEiGuq+RUcR!leqS^njElg6B0z2B6
zN|ICS6F;V>{mLw}HCGb@Mx+~q@)2kU|J3^P{7gV((xld3tq?(y1mPTISQi(Oe>HH`
zb#))tO+pILci|AFuVNE>jqU3G*>okYXs^F+Tm_Z_2)IhYThxe(7YHUWJqayCf)PXT
zjeU4p#lr+a9FJab1vo{NWQO+j1@-{v1H9MeM5g5DH;a(Teu|0#S0*0*Pc8u24XklW
zUmaIT4gO2{0h_pBpiNks$L{W22P<w5-C-AOpjv9quS`1mR$E$9FvHQ3P!nSUGCnKz
zmGNM^(yZb}TD?ACUV>xW<M!5Sn<o6t_V$3n3!?oMF(m6QJ2+WhT{=Jj_8zJ?@eh*S
zqO8E|9HB?5AG&y+f#M(Z;U9?!D&HgKZ!zLNfyI@c)R~fZ`4R#*eTlqX;?PNjKM-*~
z*KS1yAm3?_Xc#=ZujSfUe~bXD91eedNQH~A;=lvloZqDrm{hi^;pJ7ZXl%&a`C)SR
zB{uITzk1%MgWu5VeEMU?wGxTMX5WjYHz{PEKxa*!`!GbXMD^m0DjZ0F>ciOYh*o^L
z;?n7VI{e+(`or!%M|t>;C@0RnSN(bcc$5DncJJ5*@wG#O_s}+m5D|!d-i2mvvzAxJ
z444r$hI=4KlN6qFYz-MKS}LuTrOPI|2Q|Al2bR|ALF50-{bQQ<$U)}Lv9RgS7TiWg
z`(O5{$_)U6sUea9%vOzBE(O%1UX9n&cN!#1_-R28@Tb-JNc=goyxt+uL{LXkgI1Lm
zS_wq7X(%03=!V#nx&0Aab}66(-)8fVQo*{CVGiIC5P*bO@`JUP-srkL`^BGhkn#z|
z7cPunQMFQla}(w`V2yoRb)}>b!j5GWHSl{H-+APQ381xIXlCvjKokPmhsV^={&`<u
zgOkKfzN!t=AYY530Ri-!CRgS8O7Ev~H_v@xB=Q$tb)&sQ@QC8~dC9HaqDfdlISCA6
zFtbePJ&1JYcd~*5h5>8{_K*Kn7<kYCEbW3k>bM4K!Qq)S)pFd39)gXEjtJ*~nS#&-
zzw%|o{`dYX6X(_h{&<>+(T*_L2LH?gyaS301OK??*z6*<?nzbbHR6b@&sBS(e)?oW
z=3x46kAEw3BVkZ{bXcfkA2;*{VT8>=+xFvhT|!eSdtEF2{TsH&eJu4RlMSBd_aObn
zd$=J5G93Y3ZyElZ&AY||!NcqGzikvsjpgztSm_iJM;)n!NuQ!5IKdJE0=Lizg(5fj
zKgpj!Lr`@qItP6z;xICzW9u=0)4wO$PsXf>H{qdZVzXN>I!z+o1<$W6#$B;j|MB%}
zuHQ6;lHxLt&_i~^&|gROQK`O4Yfyl_W*7ZUAun0qG*4Y1N>{XlNd!og^S^#%W4^sq
z7E)i(SU_A7+Z0Y5rjVq#BB|LSEe$;85?-hOcZYLT7&CN!K?juL6f|~;@AHEG&NQ%0
z_{76w1ThSXZgLR;?vU&LIvh8PsaAtv!@m0RF!4DGQd5}D6N%-;ix)+7-_a&lt0mP9
zX)!dKV9lrWATR@IJ@_9Ls*9U}IdMg^tYh`|G^@j@;R>V%e*W^MN~6&wP$Z{|lgT<n
zSSBz0&ASd6Wrc;y@s9+dnIeAmJ3;Xwbi^rFe)|+})}st>Ve|eg`{~mS$n=37oIDxf
z`)C7~N)?tqB~F84{T8s^{QPC2lY-h4yaDi<(skkcMv%~C8VSCdk^qS1(JuMSC5d~S
zg?N9c&wsQg1ciwE1G9_Ye(xodPg`v?f3pwdQ*_-)1E4VRnXutU)PZ{d^&M8)-DWJ$
zRj}DW3S}b^hc`}ZUXg*-!Hrk!)PFCwUWp`zJ72!SXsA0rISEQZaJm#G3EZ;j9;1i{
z;niJqaKO0$_yJe~y?p``>6UodeL_9tSX=y%R-yy6UNI^)jOnA$6o$ID`uh*v-1Di_
zJ3VF^Q0m?L>~h85b^?XLQT+`_1YyVl>MnJKPI}VquPn$a@dU<;09r@`PvbHSB3jcz
z6Fqu@EW0nGiv~e{s3kvg&tVO*sGR)+&z?Q2!FY5MDeZXuj~braY!h^k=gR+~m25w7
zHIDgF%^Z&Z#Ak}c9Q5D6!)R*Y?cwm$nVR9h^bI75Kcs;_a+hxxmJ^;6!dO>Wn_*#Y
zNA_rFkG2bV1I`n=^f<aUFqpfdpbQxuy-)(oYjm9xdEayGs%qmZmkBAm87AbFP|Ajn
z4nb!6=NCsGAk8iMmq_al8{*hzKF#JG-dJQdq6^KVm#dYDE74j<{zXs?*#_%Rb-<7H
zt+nm`-y<aE$(&N{Dztg75i^<V&aKUz8*=_V>P%31(MtZfs;J%B+e^qG2VQoXYaz4}
zOc&tbgQ#)u{$#QtH+CqogJWC`$P!5G3LQjFSNkXHKoN3*f%p(n3X5>^&(Ho}`N+xd
zTearv2}0rbr+;y&T-$9td_C!N?vHZ_Ln<eL#;7JA0+0<;%*!zn5CY8H8UFkxwH2kw
zT%2PKR|!uQ6;+8rB*C#jS;q}>(g7jw0lH&}$|!Li%W?1F#xlxZ-<&>o&KV_g$A3&^
zuea+7u>F7GDFZk(es6OCZpnnuzjbi|SuhU)nG%XwHS7a0fwukh2f_)z#d-I>&b~gp
zsmm7NcNrIX`~^l)-4087^y}PD3U}|~^oheSQ7(7|9HM>^<Uw%O=}1JH3_>tSEH|NL
z6xY3K$?w>9{QM$UyIQJfaSroj^%RG)bKd@eWJT-Rd)sWu(d8f&KTs^|w-3;O0O2`&
zFR^l=j{OXPS{oE-2ZtR({!^L+z_xYIrRpNji5o@VKoanu4nJ27zZ%Hb7(hNp!xOCo
z0;LS)*j8D(&=*eN2wQnGj>cSrZIv~K7!Vtkpi?&PM0*vWWDY@G?8;*XXR+0_+o?eB
ztCtRstyyTu)-XNFkGk&0>OnfW=vmObW)&r!A2uB4U6+xOtF6UlpKw8F4121WsLWAe
zDQB_X2h(V)l3tb3X|wB4t#5;TEk-;Iyaaf4D&R@{$8OF~mm`ibrQbgd)_75xsv{KO
z8DUAGo|}fb=0*tDkX>Mp2KSvdH;;6N(FO7B5uKWLe1a~GAXy$*sX9O+r3B2MNts$-
z&Dudb!eF9AtJ=hVezRUX$cas7o;0X~A9KgXeCnZYgRp^#mX5ScO6ze^*IU)q3b@~H
zovGhHyvp+YqASs-#dHvQ(xuY54)>HO3K{bzwh~Yj(Ug5Pa>qn42e23oCknxV38Ppw
zO)^k%8r+52e=aRO-O6-Q(r=1Nd#v<aJ9safcY_?b<-3FHjb&;*yCh(#C+KunxJg6F
z@b>N75YHkXa`lSv<!iNld@FN8ky$rlo5?>uiQnlaZQ{%>ERe5%`SIfrYB$pVNs(T7
zb@1asxqXSv31VK4kgbg_&;_5K37hA8*f2zg^cTVY6)bL0bXSFwQ4C&Yg#xBL5k?XK
zFisD%FAklje&1p@gDtKdcihceUF*6M#vlT46E4b{<@Xi*nBGLx{-wis6(lNZ9ZMbS
zVae<JU_dfuvo$0Lof$$34RIIoRB8iaz*vGQi3dzchJ59_LYXNKs#FJ=oe#+Fj}fmJ
zn{};DMZ2*<4l4mV84~rRbPcsrg6P>@QcF#<8W``L#<}NgUB2)CZ6JgKMmX61LNWE<
z?w8*nSKfGTZK>0eKcw&~W<d`GF$*bl%SPP(Urj{j<^V{k;ZP_QO-)3y1rsLatX8jF
zD`sb6ec_w|iD1i;k1;kHq8N$I^qc*~i_);Cyiv({^+>{2XF~{s)S_MQ8Owy=Cz2Sh
z9M$j3KfCs0HL&cy!^;*;uJ9T`GlwUiYZL{-Oc>0wogI39n%!DOO@oVzHts}KKknwp
z#lrYM96`8-y^e^zl-|3{=$c8Fo!qFtJdgo@6Y$I94_x^jU!`f*ISr5ebe!T(Pd%$7
z)O!J~M1UT;ecF7>Od)(@iOp)g<hV!ba=mopPJC1MdGDk-Qll@YDVzR*oB{^t&h%&q
z1qkLcx<XUSy{>z!shx{Sed!gxGGAsF1<jtm)Yh`G**+=bE&{R?Hz?RK;wrwycFYBW
zv&St7t{i5Xdu++KDqj}O>J_VZUdd#Rw@n;p8^0{;2WcL*bK>F>yl)3#dR_~paSF}g
z&(n3n!n-|`|G%Cgv#pvG9!by=q&*@3mQ*<(j*cCn8C`lMS@2+R<VswwOrADFG-`cO
zv^>xt<v1%!mZneXiVzWsVj$^5<?zh`?Lul0PtEkt<1*E`0qVl|+7Rb}2@}G+gJ^M;
zV1d6HO>hVah_dAW_{vp`Fm@u+NjTNF#_N;@U=UhGhSAdye`Li%(1uu_jofOyT+Y>{
z9V7n7j;Ao4xg;aQ_u{r~>unn(vcJG(lf^s~Cb-ZAftU$0b7ZMEG?GnqU5*_&%XAZg
zioEO7;PvjJt;8ES@E{rl$Ec5yZ!r$i<=f>pIYL|=RDbG^=F-W>XYpG9MALnPP^k9l
zNywEpb~EhhcmEFiV2`HU?YfnzYm2GYL$x=-UY>kT=ls7Bn{Q53yQ04?rrzJyrkGWf
z+~j+Za$XW%Av|ZQ6mE2!(rACdH7O!O9&0$i6E+}ke02_;JC|zh%=cnGvBR@BXF2GY
zL@-zd_)|k<;G^o%Z4Bl!<ZNf>SA8*^e+tNgSXOVJI|^3`=zrlWs?rdDy}+ygVw^{h
zVk6U`RE1iBA9Cg;{J@sSWPVWSjkV~^4~vl9Bpo);_08?_LHGrPSqr}*S71l~Beg~q
z6bYGj-nxqajjBu{h|rN|@-6u<oa;li#kRQee|Tr`|KMh=R0o&nhb_BzsnQLK5_HND
zN09gulsuiB{{xxtKucsr*w0JI`m|5~l@<4v-pL5t>4-XO>s_^5ITeC5sJtvk6#f-Y
zUw1@S`^CQ(hc6l4-}HoWJ$v+<jfZ)oAM245lsR82$-OmZ)J&JUQT5K^j>dI?V9{C`
z<=SrV`N9d6v4y%}^AyEhLw5@s&-_^4u3cp5-|9b^oEO-#pcaMm;S`J%lfC+UF!p(^
z+|TRV$24MyCuB~3W^wy5T!mT4TLT_ZiOM1R7<P{K@&lq({R!g*A$KM^YzBVL+j|1O
zK6wNXQzBqw$GO4kEmMXMV$WF}lOeG@ee=2}->G&D;suR%MW#tsG_$_KKkO~qt{E_z
zwg+_G9;lAdLy4D)>}wZ_vUb$DUI_T+o6~q%PzQv9TSC6$X=f(DZ)op)J-^##=Uw}?
z)n2cDVoc_C-9KiEFiE}CVYAYHU^Vtsw)y^vR=W3*m}kOL2G8Vn{e4cK)VFC&d8a&b
zD{RVNPofFZ_xXUI9SgUSKw@4hFnkOhsCyB(>T3pFei2lG7r#=?nsWX8{5&z){cD$P
z>IazP-$Q1FA%6O3T(pO<Z-pu6X2L^TZIG=cWvp4HUUi$4#t_$3em)+a35@Y_jE}*Q
z@7TR{*Z#3N!+1!eoN!6T5X;-*gw9paJ?IvMB*cS6utm0g!9PL34hD#`gwsDx?~CvY
znGI~k0wj#pLS+14pj!m`hbKJgTbB37ynrYA*Iy;170fF7M7Y<&dr|j<ffIa1YoE(J
zT`Dq_GuSGzLBktnhTGKrjo-b$3}fkIa5SX9U}txj@QWr#ABV>oa_v<)R`&P3z0Y>F
zs+|a<d}h$f5LuEoFf_iSEy+b9v3soOdD5GUBsD#{b=SadTCgtKsk{ABpeWptgQ#0D
zJp~QY?Rq!A=#+1D`r_gH>Dz>z>}{U+JnT<`YxlYgP-hlP1nb)1q@q%>j><Ch<S!9{
zCTg7f_;o)d+@6SmJ}Hl)0)O5gX6LHxATdVxR5o}AYP>MgKmMg}uC93U^;1kf(0vCr
z+!uN}Dhu4D(<_pyQ``%#=gB-(?SE(LzEjknxEK7mv%zCNZf2*sB_9PUmU-2={@O><
zxW+Z5+)6B6!q+?cupN{~up6E(7N=k^(@Vv&UAts-e&y^Y(P_bXn@L}P)yp5VS+Asw
zDVnYqxJoVXfzRigk)4BF0#`ZuKE?NM>9%{}qRsl}Mu#_QcqnRKTf^5|V9+ZuU*PtU
z2k+b3(vYAon{kt`;q!LJnqqg;#Pzs0z+BTqaa@>F0cPp%-hbSCO<lu-YicQq$5g}^
z-k)l%uwj*?YW3odH`2)qgi#BW+1fbGFkGhmq&^L`Ym^i{*9E2<$8twE&N9ZqJ^I{;
zL;*3)O?ZbUFV{t-jdX$nVU{fyt6nP<eH5UQT+Ir6%RgJJqils$Cg%Ljo%(P3BzzG)
zFJInST~2?OPQL17S|4Wlro35jIJ2o!>G8hWfoJS=&K@}h>*%X{DrWZba608XJ-vTl
z3&T2aE=;<eh4J;*=2<nkG9QV-`)&Pj%^>ag<JYiPox5{b{Dte#A&hK^EM`B&>OC~Q
z*FZe|2edBa##{_~ExcgPNZ&UOe^B~s(;3tb)EH@i>^moSo=s3uV#0+hzNzl6I?}X}
z&e<D|=;sK}!xcFOI*aang4Uw@ZeYMdsv{i6<7G$7=?OcfN$!I;se6iuHSlG8{P>fT
z>3R3AW8wcdcI63f9QI#$;rF&HFyX%Qop5s)ScXv=J}Ej2pVWL(>eLuTPswaTjUrdp
zs6m`_r_XI<XU{PwR>fB01yCWbYHRDf3cpUm9rw?Agh8c5Sl!T(NA2)$ir}IZVTc(_
zzh~5<>U1{Wn@QG%6}IJ_<_k}=B_$;MGa1W|p^j2^{CJN-u9#M~dX+#&N>%*AJP-x)
zf_+!6O<-IQ4VU<ETr9PfXR0MuMU8<c^LnsT_Y!n)&Sj0XlGea^IHey!!>S$9)NN&r
zDuk~z(n^CE4-iHtcBL*xwDL6z8*tp?`<VX>J)%W3CIQ%tll+N!P;HhWu{Bn0bw?^|
zR<c;Ut15Cc?W2b7D@v!`cj`|2_4>pU%x;L8D8gX5w8vXW>Ho_IBhy;d1Ab%U>KqE1
z!raq8TZoNhr$TsN?_w%qrj9EV8#G%oD-=L_A)Ty`K|s>64ZkqQn56L-)}_=|%!kCk
zwD5(kP2mds#aqk}th94iG3oJavdtMh!WiP;cphG3^+6I-$KbvAmJLHhE#YLgBA0fl
zY5FJi5Cng2f{*OCl^|vKyL<dtJnI(2_O>?B?$JJfet?EY6I8FD&8=%OuQPf}1H-Up
z>gYr+e!O6~Ln4Rr9{x1*H_I^BIE3|0&9VKIFL@&I(wloz#u{9cF6ubnH@;Kj%YKcm
z+sPYvK<y;===C|)XSJ?^@Y^zKeTgu=B4@8jT&zQs!ai0vDK?;81>?+LW(Rwbj+7nM
zsJ?5he5Y|Fgt-J+A#s0dt+U0~R0wNh*qGucs*WYO?^Uc|y%`blwjw+szKVo+Q7_C)
z9l~n-GeM8{@sFiNt3>T*mFkOQkX?JiLILYF*=O|bMQ7)QV)0)xagEv+`#*Mf7mqJs
zjI9Rgq$+mVE5W+yLII_!;o*O=BOSXrde=#iNS5dO#E1Mhu%u$}4hHAo5M12lB8YG<
z0&`q*|7mS7j6gQDM^FrmyC$*pHWbR*?XnNuEOmA8@n%bP(VFUfjbHJzx8n^otSk}9
z;irs<2%-qYYB)MPH86mig6WCssc}RkN`0!C$2IdZ<w$w=_~2fJ0tmv1LGf;aI)s|N
zn@pm_|9RQ?i>!$yGY;yyoz6}1JfT*W5|Wa0n3bU$1JHoQTkNVVJqruRDUFFKwQOHY
z-SP=yMkx_-64v*^mRRJIG+p-4UTn5UkG|H=s=+~3N+{~|8uR-T4}$k>3U_Te#gMn$
z_3g%?S2=a-=quU0eR>#a%6SenYUjDpg~`1f%jZb6S8oE#-12S@xvXMfQ|aPpV$4H{
z922Or%GQaxNn47KIlL4dxG6zAFsywZJ&{jMVP@m%Q=55ykbc7|AGPNR%(s%h4NKym
zI8u_jjK1#W80+;sFQoIOPrDTPl~0N!0{N-i9Cn?atlnbRO652?+1f}iIA-Z-8;)Mt
z)?cyVULoEFI;`@#Uw{1QK)Uar#bQ`@v$S?_h3RHBqr)7Wd5vT?-F4Y{&~7U|{Z&x`
zYDWbpdxfR>_^v^YxBfL&IZFw%!b-tihc%xYo>Pma9U>1r%YNfk|8Hy?W(qJeN>Vf?
zENCGE_kuUy{9KfUuY++mMn<flv(~iXa)_<QE*I-;*H=L|iBHPU#y2Xkx0SmOXlN)<
z7~ZOSUsK%MG?}^~;C=h>@OjVNG^h=qnHIhqyK%JvQvz3*n_X7;OCAdgoO*khv1HD3
zh2rq!<eR#S$2TTv)6r=TFucGdqGyC^F>4|tFRvX(h=8?Xx?vM%1Dm4awrwSL`5^6k
zdU-YGJDs^R;-wN@ImaeIL~Z~?m`?gPT0#9I*HX4jxutN!=2=7IZN-Kck;f$NgRzMa
zSIx<8Cnyw<B&?eI){!ylrdE0QA<TuqKY?*vTk5=~$Sa4VsZ_XQols7^J+#;3`HRPd
zpDq%$J2qdmgqLphkC0@8a`@YzX2+C)23JA3q|xJxm(ZfKj$shhDc^97na=qq;usyF
zRR@u;XL9GjeemHQ8_T#IWK)7+S0Z76j`3%JtIHZStI-`+i9C^<mOyeZdgCQF{ga!M
z7~trQDCK})j~JSiN@2A9o5aa^o5!-wb=XH}sFF_a?S~H!Ae6u{M8UR)le6wIi=8|n
zwxsmy)XgkW7&6oRfi{@UN}{t&VJXgP3{xas>zU?>=bX~VT~Zo>rs0$w*~<Zw`M2j0
zojEy2x1FqWML-o9ZhHGJ=@{P$)6UyAXRvdqn`u;_I*U6<zpmkops!RDRTg$-AYJ33
zo9w~%{?OI+#I5jHfiPk?J8=OK;*YaUlDp#5d@o1Pv5-i|x3)YSwZ{NhI45+1m-7@e
z#-E)lGl$Hw5Y5eZ3Bp=unsWEIw!<%&uy`FOJeHAZ-fw8oO?GclS(=F}Q=9A$_%&s@
z3QAlDk!lm8R`7(1VC&|wI1&ddj_il6>ntprx{eYXf2D~7<JvJIMrl3Wq}sQ>QWJt^
zVcPlZcL5ZW_VB+WX1IAkk0I;(E}}+mnJ_v34KngH!&<Lti>X>`JLs5NtIHUv_G0bz
zm9@bAzXpQ;m>2^jQ<bJl5Th2;x8FWOghf~fM;v9vQ57Oux-SS>N3UC>EOJa-J{g7=
z=_oS@Gj_1w-UE&C01cM#g|`2iYn>BMSk)8CUDef}?JvD1%>hK4&HnBMfcTt%dl7qA
zoBLO8sxP%g5htdp+yod^#LaJ7@gcnrp%!?aR9$rJmiS3WZpxJbYFXAua!HDg2EplA
zy{jNrA%<JOn~WhfiKe#hh->`o?lEKsZ8WAP>a)*`1k+a~^8Qh|Ldu>V7p=Lng@uI)
z(5E`)I?rVTi~ovoO&f&X-MeFR1?;6Z9rc;$uWJXqrv^%{OhUdlSI)5Ua10DECNX8B
zY$*=|+=7DW0tPpnGk$AJcYNQj@kHl6>Xl!B@wB0W64!;6w2p;JBH5)+Pfve!eN|kX
zh!y(^d&CX{Sz`VLv~Bkx!$|!!BV$y+=PU1u@FGw_lXWqxrI9%>rTK#A$nDcdVxU$1
zQZBPh5rWZ&?mJ-N`(%5rUoq`8N;{6+j{MQV5)*G=Ci3<B_mP8QpPRn-2>NZoY7pA<
zA;?9sE@w=YHGa$K+wS~ALZ{M>-}?8=Bq#9_V*Ul@BB#)p5<cd$gocisZV`^<iG0)d
z#Qfq`wb=;5zW9-^E%!59c$KRzf6IG`q834E*^EPgM%`nq)*3regrFZ7M^Cb&m&jjN
z`M_A_6lNE8v|^4W7$x_?|I-HdV7gfFJxx?#RAKL4DP7!VvS&m!hE^OVrtWQxw`sLG
zfBxSh_ivMvofFw*S;YJ&@RM?3|N0Z{Cgam)p0|O_dsW(iu>5q%u6_EB{5hc~KL!p(
zMeBYHzJLTNNpO+oq;9-kWHaVk|K|%pApV6}7}xuNm=ru+oOA~2g^tbdt8+8II&|l&
zaNM)mb|B2tIRE<d_hAF;Cwew|Oer6di))9bTU-}IO;Op-7N%bxDw+>qEOF;YsrobU
zttyO>#O%LD#tiM9?Xy&N>vG>BBy>lKPX-Pm2RaL0&prFvigW`k=rr^d#?e{(kP78`
zQ(-27)hBxXqG1^1S(t&CG|w?<k2tqUUE6P$j{%vHksM`>c2c|QSp?h@H@YQ1tIT&Q
z%-Z^^&SRgd>fS^5B!EkQns|)NsB74*@aqJof+skM)_V$}n1%a0cI7eC%@&rc#qrxp
zMswI+Dt0o@?&HAS@{9*q@8X(jNpp)E|Fy5=;ZU7(f-_5Zy-n+boif!}s6U@M2OK44
zq5q!RxON=mR?L9(ooSGqgWsqL5yj9JvLr&&?lt`+jL&&{R=T?cHGB!(@vcdx{TSNu
zUPtZsZySZG+4psr<$?i;mb!nRK7HCQn8J8syZ;Lrhfnk2pW-bGySCqSv1*LhvZ*s<
z)wZ>Y?0Rv_A}hKv<>+?t%VwuxN)<$O4PGqe;cPSr_<ABCKK|`O*;)Vqj({bs#j!JS
zb72Pl2LUAP`u(kGS8LK`TeWWjrFqCZ^I%nR*y;YRX|>i#?h8g;qg{Q$qB3;OpiL!f
z>NU9^t#kF?&w*AENT=2~YtCYF1~41vm-pu4yAJ}vZdNh8HKCO4U^1w-E3?v9>pu6C
z1UGAf*jK6b+e+X{`uB3;*%Tqkfn+UP+ehMG1}cK{ekZ=U0rlqQ_m;%ttZ_73EA3w{
z-Llq9FvtTS=;%bB<k{a(lqSSD^GeG;!-phXUddu~{pX1w>gHw)gHyt&X0I)>;n+;l
zSD0I8hkeW~<OSh!Z)|&<XV-h|t=B@#7(Mz_>kd($w(%vrVcQ8T)2=1#s_COH&J$a#
zI4gK+<@D-YGsapxLOM^Y31N!ojGFv=I_FypH#Zm^EpIfc6htFeTyOuSo!xX_F%>-K
zUlevs7rNY^FgC`#@c8m>%bFf|hjX6BduE<1?qe_XW=_w}ZO$uY4&E}(R5-hp#+?@)
zG0ofNcwv@J^L&8P=PxHT5>j+Ir#kRuAFLI+41VQ|B?tdM?7ew3mi_xSdff<RE@jA+
zD3uH$L#8yUXp$&nD3T%b5N<;xMI=+EREEl!c~(!U5F$dNWF|_;*gh`L@4J3$zwdtE
zwfEY8?X|XLJwv+P*L_`|&-ppe<2=seP;_a+{L95fd6GdYLc&4J^?8F4M;<$U(A=u&
z;q#)PPfJ@{dtug6S;@_}?Lue`(_enI0hJSWh!M@?OH~G~58RX3RhOqQySF8`CtGc5
zSmeFkiQ{zTx6jOJ8@Bfco_%w5w;V%Vo)hGCG$Rlazp^bV!)p+@b+m{f_=djvGr?r<
zKH?F^5U;#WQo~sj@d$>8GF11wF<YbJ*(uGsJoM_{0<}-XEEr1sZe|4f^Rc%#8Kvx^
zgs9YZuJByFTaJRV2{v6QWsx)8`fITDo=u$N#@72hlfV8sv~<9f&*+%xgeMKTZQ$;=
z?zr*;z;|cP{()8G2)cS_N(|aagT&?KWwvrxbQep3X*+?7)9%$#Pz)!Tk6*rI8882e
z_CU+yiP=)r5z!U@)2ijssv{rrSAS~KBU9q2=kJw{k_!Jo*cK&L#Z~rL+@P2BQ+&Hq
zo+H(VO?F+iB-*1*NO6FFZLd1ITBUL7NtaW_KqzH9+Lg}efsR5R=aQ!l0f?dZ)(9Pv
z80T7k*CAZvM+gVxlc1xTAUAgt=sB`|o;!B_c;29+UzayES5BG+XfGH>;A$m@(Fe|0
zo&`7eBp76G^Jll*)%1dQM>}a^M_TET!%hrQe-HWmKBP2NFLP<JJ6%oYL~_Wg8@ucB
zp7iBx=H^DG4J_ozx<aMUmS6IYJU(1EHcaK((ipX#w}{sZ<+{DMdLx{prgCt|kR}G5
zm0o3}n-iuetiabU{dRMQ6ZkYAo!FmRbUeQuc42w8Lqg|lmO%!X-o{~S%f+NsB4(J5
z<1AD{zhIh7Xke~5ouamGgDn~^=y*j8x{nMEO_uTf5lLzseZ^%Qi9Z+_`L>0rN#x?E
zgRY{yHDjm17ToC*4kCZRUiCjeWrf*|;19&l&mS<vp<cb`IC@>3rOr<a(Z`*b#Y~Pq
zXco3_@>+)TNhZ7d--TEVYrvDi8KkYz%jZ}x!x{z-noGcJPLx<HogY-A&Yy?pBg#7S
zuHAQClQ-3~-(7tyRj`jcg~F(YtuaY@T_Af#Y)GelpH2t~xgF!1JaqPKbdRXexB5_S
z+g=WO%1MakelL}fw;`&xj(Po8G8%{9MRmO#b90RV!>SjO3EI|ZMtDb4h@H2|jMi`U
z3?UT9onl&w7>l$T^;qBmAPecV$RHVol#os;g(jJ2egsQiDnlH3F2D3EQG$*R#t`gU
z7pvj@ghX?Mbjbd%zYV+=&M5dNqA7|79KRP<GGh|Euq2Nqb-I;g5$8m5FzPJ0S(xib
zL2L)d`=Z2Zf*K_Q8z@=^);G?EI}J;9(9`jbrt4L}rfEUebr0|v6uk<HiU1$7>mh$9
zWXY17bdEV1p{+D5QaVo%lbT3uMDSsfc2K15HaDNF0P!N?(XlPSFzXR(8L9X)vUVT~
zCzR@qFwu^(utGi&@yLAFu%nzMr{gt?KwN&cVBym8wZ3ECjC7}S^`2D@%%>|VH}>(0
zJ_4LZxPd7Bue}n)T8>tCrCjN1gE4WBMRE=N?6$(N#lhiJ;$TKag(2!aB%%b%Mf!5o
z^?IwR9qg04YY}Ehl_P)~@PV>yh?4`#Z~(rXG_RR`i2WNWc2aye&@n&~C^DSiqTO<H
z{SMT=M3W`11tqPWPz=iME$Odwfobp0olDqkKwg&6cgi^iSn-)H^B2Eipte=BvrJE6
zi=%8qPMvsgpCLTw`378%R&O*y$yLoRK{Ufl_{x~S?`x#)Q5Ly>{U;dRMla#&L;H@6
z-lQLSuy!|&ztD{=`>QvS1t!dT=!s=hL1x)T*jVBqMK?7SIb3>KS=U5FQ_6OEUgbtJ
z<}bWwD%E6U87}vi+D;EJb(h_?VD-VqHX8HAWf~AGm-(`XVwvbrB5w7>EW(D(`zc)n
zT{)6aqv_9qRIcfNCYc+|^mV8Az@AWYrdpk__odBewf#o7(MK>jxBk#q_xO$8n|m`8
z=70;QBr>|)fBNCor~>QDwgAq3vIIq)L9!{XxMUskB7YO}+%+TgYtX3B#i<Ql?O|8V
zN+D*AJ34JON4U1v`d>G@fAAq53>>*zVeTf9p|kz!y4?XVI+TLw$;8@`*L-(VIqL6f
zDls-O|8!G6)g^X9a&P6^sJG_DZmcEOR3)uO^d@RmKb7k-y||i;=MF%)owunIDXK*6
zRr9XGU9rb{X^pCiI$~pb^dXLDe%sAhSs;;7Y(ofyy`gs=ZgyZDpIY-dy-LTKJrhig
zZMU1?IN=?ALjHE8nWa$<gQ9`DlgQd&GP_wla^M_Vf{5+??gFNU>U7SAiblKdTO_QO
zmNA%CZjAQIDOvthIDV96z+D@o@P_)6AJHgSk}%!t=l0{`?lch$1-IRpl`FlPk%(!n
z*c;D-L)qM3`YpYO!&S4CzrR2E#`g}b=e^Eqbog{1KQ>KAVWyk%m!@?&1`=EuHcngi
zR_I^o+P#T?gN42Q9p|n8)c1L+Ce2z18f?G!Bs=6)ilp31h6*6N>9YyNPw2*=?SL~m
z){KvyRKiAo48pqrfQ1WsAR=h^8=i5#gdXA{#KA<n3QL--Oo5M~DHJWDxdXPFR%r<X
zjtA67SC8DL1Q_>oPq-T$J#KWW<F&#dCB;|T&DRn$z#ni1f=`VCeb`k|j}G`0!ClR_
zcIz9eF~TfzoKu=#dGE7+T^I5SXu$b-tAr!PTE$HJ{oTzXt$BsYfvKWPLB?CkS!og?
zSf$2QsXm2iXNe=tocHd#X7WzPb=O^SR3{~!U{m&Iw>ze#*T8x8-`Hq#&#R2=wG{Qw
zOb=WQ1(Fbm=##^Q^$4JuQgdCB*nhvo&Xq5%pT?-G&>0DK-BrF&M+8h~(zJ=P7m;;;
z*H4O_aQI|y+`x13kl^$9AupAV0w5y(Oaj7pN-eY7d1pSd)^8Q$FR!LqA}cg<e)*3f
zz3|Z3^TxKh!6Zq|dCc~ufo$9EJcruA+0aN2sO%6pmCt%Ch>$!~_`qiQ5#q(d4`Uq`
z7|a56CU}FWTzbhhHG5&CI*;FHiSYQb9owe`9G#7VPlH+U`y?oI8ix)Atzgi!)pqDr
zVI#LApuOS@b984W7`KL+;QqZe&zmpR6Gz<RwSvP7Sv*?kN5MOB{dmXoS~4(NYLZVG
zdX6Hb?>(80SFNJl5vhQ3tOt-$Jj?<du;iUa+;^iZuT~~?!pVN*qQqucP!8~r=t0CI
za`3ecoRDi->2uIXP5=58MKVS+Eui}(H|ARKg1buXPYv6mgz3S}n>Tx)g(;%xGs)nS
z3|M~Np!u5@3{l#lM|$trmg$#*%S{sw!i{<2RYwWxVd=&oZmx}tp>VqeTzo1+q;YCd
zf`M@rjgg7Kl`dOLp#k?q^cr=%lu0u$mIcGT7OHF8pO~$inVo%_mIgc*i2t$sw-}hX
zTAkl4mYs`Q|4C`^;aY{jB|P_YcisDLv9D%iow7POwe65>RG&gze7sJso{mz2t>sIO
z5x`N5H-Hu#j#Hu*T&NDoP>;Ctte`2JhxzW^yJYLsLZ7rQ%j<a7BT6CSsH{dpG3ZoG
z%e5bBkU_KhFY=$evU1Npqu1z_#mGbgpHvI|Je%`bqYe(M^Vv(U`)!J@J7brJ@&?Dh
zB?n!3fg^bwTy5cXKERNzRxg#HX!-)ng7Jo)I2tDTgap;^b`=}ITAAMd)X{I?mK}m=
zJut&tGVgm8ZkDD8gcRDyTrpf1Yg*cI;I=(^w52%s4VPx&5dF?*NijG^V=0R3$FT!s
z@$OsuEZQDvdb{!Z0nS3JR<tWW+K2z>9@tf7r{wmMd}qd~_vRLPazyjXSH~yqPqN9@
z!k1B2RaHV$X*|D$`z%MLOxRoH14Rn=CNjg<oqAXhS=T>O!b7oh1PVOT6z((`$mVG(
zB!n<W*}slrm-BO$!dt6cwTW4S{ZpY~_k`neg~C0dP#fOq7j)LkPp%FA=N0vLQ{Xpe
zCQ8Izf8T5(^_P{d+%g{v0sev4R3}`wX$>l9Ps^FI`=Kei`<|It?eW$S9`Uu4yNw@P
zcTOog1m7{e5$MhfGf1$4r?ELNqtguLMLV<_v%n9|A2>8@tv_?2XVg8>2s6|eG*W9J
z0TAH&V4e+w@$xZEZOJNAFE=H%G2pt)U!#Fl>{LtB@+y-)n)SEQ{?2|or3_dj0Ykms
z(JeX`+5agFOiq6m!grVJ{S=C<ldcyTB=^b6Ch<IpJ|g<6@1H=^8sp0XE)!oe@7q<n
zcN%v84+tSNisZaHLcHuiZy=>mv0E>7`t(%!Q~{39CMU$^L#qBTd11`ZsP)o|ZvcQZ
zYYJ`9=YXs7>eK4>C&Mmcfra+vf6uLSc>hkBmN5b-nI>IiNDZB}WdN{Xn@>4=fipw8
z376~;nq;{XZ81HT4SqZ7)^Ts#*s`<evv9k!Gj_KXAQ&uYC8wb~q`V@o<q*>PFt~F?
zrAm}V&Dz468-_Bl-uc+`cd2Lecgu0qzwNNz6!09ACPB<nv;wEY!>;XzvQ^0ZmmK>6
zP0_F$l;)3ZZK=>kM3iG@Zl+#*LRX&Y_AEP~lg^4U0s|veM;Uic{dwzh+XHR2a|M_0
z^;qOT>)YeTAux9RcyIF|K|toEn3Z&T$s#Y7pj`*!%739m^h{fz+voge^6SbKE(5}i
z66tp9yR9_(?0Mv25v0Ck35LG-D0U3O9b^NTmXGO`0|G)uj`vnNf9}MC6HNe8zz=MV
z=-Wl(E!=xR=8OjWRP%tYiMLUqj1Vu9zM)NIdi26^Qu%n8@~;~r<joTN<Bod)0CJnA
z$NX^G{lc+xJ$n=tw`ra$vhBWA@Gt5IFw)&epfwB02U$EKJ@`nTe2SKNfS~!6-A-W0
zo4ki)2S5}C!@y8uk2(0>43ufl)m`J(5NbhQ0ZNq{i4bNOL6zT%8P=-<7uWE-&l-?Y
zR(_`OKuGhss_==``;t6=l%(#&qk`8S0{iiZ-xv>n|K|&r@wJl?P7tC<oOC48*&M|W
zDi!+<z%e&gTV+~@T+S=KJ-+=0Pg&pMc-+r+yqajfdsqa%QpH<XHofjmc!^Y6B%`Rv
zwe4P+!yz-M`D&ea$#h-`=OL1n>2{cyoe%Q)!$7V#uul+k=HDLoP5}_Fk4>1(`cdjW
zHX2|f9v@wv8A^<#qnFjIj)Ddz<3hmzj!|(bHS|g53&#JrKP#}g??s=(xVOs|F3sAi
zbLQxf&siUyl-J<QH9Ectm{7m;a7pKf9WrY+gd9)|m6<5xD-bVhD<$ib6(9&>(P<>7
zr*i>iPd=P*$Q`6=>iWv2sC)Nl6DJ8Iz!<quv4;(Xl+&~`nrKjl5Bn%<-2eKvteinD
zBs{&I(d!lKc3t#+MIK{t?b@P=13>e1+3SYb+`DhhPZync9LmO|v?{G$fRL^HN-0HM
zC6!%56$2rj?R#Z28=%Z|X0e1Lk1sCCU$h@7{ZUT)<y@3uSSj|mnwjeal3)_J1;qq_
z)BC==M0(JLTs<!vBPuPC@=AHDOweF(DV8Z|!|T8ow22pq!^kbFE9=ict?j3ICS0}*
z1kY(`>sssMVJ!VlQ~L-rI~y{bSsO3J0me;pQ_jLdQ-NT_e6-IT*ZAj>(>+T+TrnFf
zH#(uvA9-4K*REZCjmQRH-3rlTcf3K<ICai;Nt)KEF7Morsm+Uemwe}>1viUt2+`_x
zsW;0UE!FC=-C6e}u{N)f*bq8SUvT+xZcj5WUu|$}rF8gV@%ak!Q#<7dD7)FW`P{AR
z7ty+&#l#a(8fDS`>k9}2q`W}Hqip!2x18(f0_+^yc<mrKDWpgy8U(8327bOAO0xE!
zFCU8t6;M$cY^2<Ef0<$+(daQH^rXH~O;J(5JJ2`BvcP78p7@lo`IRfrGWuS}I&O_T
zB3gJ}$Q*6*rzLU!P^G|g3h$JM0II}%g|eVr(2+mx-`|Am3qF8u(;X4ACdESDIpGDg
z2sEOXd%6@vv0ZXShBML<<95qzr@qcGn5?}ZO&`oaWrTQMGj?zR=vv5k=16t{Axy>N
z0MGGbt`r-ef>d9XJE#z%np{S1rY%v1b)GvmeK4S$>c6C1Ns$f0fqP$$$sP8|?5a>_
z!=x7M&&Q^Gi8q5&u#5ZvfH;*Uc2)A+zoomZqN3t>#yVo_!RkH#3l#3Rce}L=`92E^
z;6Z8S?N(M+HnV|NhNP$_kXxs<G+HJu-`aN#u)e67QYQT}JD*<)BS%%M>L^6=V^9Gs
zAqDDzQkzGZX|e<5CaGttg3wP|NFyvg(5U{Optp$;0TmA&=vhB~*Pn;xQ~724%bmGq
zLF7b7xA(%hvoQ|ZuCA{0{DohS8qY_3!6Us~9N9QHjkNe!&%zg^h3DbeU4NF|#23Al
zU9piiT_4|GVz;?>H4120&tuRpKGX+pNHn2av<N+YypJhIN@Vp`=o)2UE`1;Q&H&U>
z0jOy$%mEXyEgeqf=@~7};;W6FZuY|-tMjaaN8nKQ-}&rm(3ywr{?6E4T9k(}!LVOh
z#K|&16#Pm~he0YrD1T~2wFx+csuWLsyXGy=ZW=IAR+^Mm@&utOLu(XTN38<rV2Wl9
z$~N@5NYLnC4#E^*DmD@~glu#+5HV{^0xG$T#lUAQ8?FW#kp0ou>P!2WH}<Ns2<HH9
zbrPwFx`IKXYlo+tfuS>-&V0?Pg|e!rI4ULPWlVvsKXv)tu-jK4<+ka`b4*+jHKg=7
zzzmd#IORKbvnf#2pGwWn@kLX7VjHf~l`n%3m&q#H-i)^U#<l-zK3`eeEZ_z&R5Gok
zvfQe_+OnZ@ZmM^ciDrwgLcZFsC!44kSwm|gqWCQhk%@?7iaraTwRqhrnDwPgx7LsK
zT6p~m<H@@3vyN`#d1iK@fxWl%_xTPz3AkZ!n0}wAWj1pXg|OsDN+fT@H2}eSUOv7^
zkxc#P2T49uiEse_K3O{2lKl$smZw}0_~%nNP4j{3jpFyJXunD{DeT*~5AP4s%DGY}
zE3JL;l<9xC%cpCkIj4cQ)TlGq{rkGE!Yf@-qoreqeHouRF`e0Rt}o?{F%XnKa$&=h
z_Cu?A4P80on%9QL%Fp{^Cpe}}jA$L>C&BMh$mOD^N&Tfj*XM|LE;^B;GgDv?PjlNW
zvBh3y)!tVnvoGvf{XL#KO}_5C8oFNf!i_+C^kx}j_6wg(Fv&EK*b#ed-hhMZvB$ey
z&F_Fub4|!aB-*HZ@K2n92nycOt(-^`RoV8wm@jq%ohT6xH87aDu~!lacQ6v}m_FB3
zT<-4}uDwoGe?pp8H(b<@S|MWq4d1zLQYrMbS3_Z-P{!xaC{oQFSc)}oo<nE9o4$ed
zlx&islL&wL*BV_~I_7nml;$Gyw`-B1b`&<O3U6dgXu6WAv_E7{zhh$4VH7U!m8Xui
z)g3SyTHksasZ*`$jC7VSt*}5C*iye{897kLo4E1O+^B>A5!h{Z%{5GFI!dd~Ta`Dy
zD}3z|n7k?G8C&I{R=A0;bc3ty&d_{0a!IbsZIXP3cN;Di06YY5;NRfLdj*>Xy#@%x
z{5RLra+}lo7i_pY{SGCMmE3li?(X7Y=XaW<X(mB>pj>XsMhWsn4eB4dV_WW?gyUlw
zvn&9|6RJsP;pGr}mD^V+SNtfdSO>%CEDl*MQAX(?#^BN;F%7GP*|l&l^W(3+=JAfE
zt^&)C5Jas!Vtp}WRh%T)m*BC2=C<MVYK?P`{oDCM{UD>pO2V)TKvAm?4J1zk@@(R*
zziMgNetWSvpZPk?{+4ZN@xm+3qVKb21iKSSIopiaBv0pGsD?{jzd|VT6a>ojyKbaJ
zqMx72F5*wR&UVc>rM;<7R}<&Z*C~7<#exrw6mYVP83^s0n=WMpK^LRwuf(rKG%3|@
z^cdid!f{f40o)Qo)zpoiu+%EFL9*LX*u55bFlu;ll#P>`7H;&x0E8_?p`LV6PXIq?
zWdShqCxG=!+W=s_<}p3-0Y^C^F4&%nF$&GRvtXDN+K_~q2?uIaKJW<i<#DI)YAXes
ziRP%I(Bh`qAs(Rk^_ET7&fH}MEV+cY=I6f2n7u=SV&n;T^AAXf3ZU8I!}Nr<%~<G9
zeMrBqVjInrw6Zer2*Ad&Qc}IHiqA>vSiN;tatCS{hmv@a;YWA1f0ghC91!1N*7N5)
z-YN!ahsgK~kO;s`9t9pkUsat|vCj50Ui=}i&Rtl#%6AAI0I;2rQDGC1%&_3v9<{~J
zLjK@sl{HC}8%k3&z*2I$X{s!`T&6SPgzUKXX0d!VF>+j`^>_|xbaO?BJNz}rHu{86
zxmZx1e|KH|U<RtK;LBk3zUuq=M({g2d(+M(2DZV|wRz%bv_CB?J6~du<x$Jqe2KO3
zq!_1gZe`H4jE=;R|L10!){v43kZd1x+xi~7YAD~-b>NQmJ_YbRtB%Ip<C%Qn6WTWf
zoAM{SMH8Ic^u42H!7+Jzczk17SsB+77$dT`=Kd&dkcd6@1eY+S+&?w<=J<;<tMwmc
z<Z9~suv;V!o-y7Nx(=W4S<+W9)IXr9k7gId<Th7juiCEAMl<nL_7t608KZ*@Bf}J*
z?BRNw#%mu~ma!Rt?L_!>Gkz_VSa{uYp^ifH?%J)q6JqR21jhxJEiv-)P1o*CRP8EZ
zC6b+2$t|uSD?SkWG*92&R^XICxu|}K!A=V0Ie%rUfdu#l10;wqCewu0jH)h8@|Tsv
z(079K@~VN^<ABdHLA!djSe*3~T$!U0$gvp*+M_T({4W>4?c_R|FD8iGPHzGkDU$3b
z28cfcG44bPN>|0ueGJ)++k2|WmbSV-lwNfZFHyr{4(%*v<I6-$cpd)&bo@S8nGP}R
z!3B-XdLn!iuOEhw0T**j2GX6~rLG=}zB`>ck@CPn8r(cdnch`R+(;ows76q5P?*E_
z5SL`rHK$4X)BSc|-38jkH<3iaNS?_LH7IYk9;>Oo)u7&Y^=oMIo=ShNtrru^?C)|G
zTzK=$WLH?n=VTUDj~-ttJ_1@QXba;~>G;)}dAB}qAc1p@#0kL$o$dFa=duY~>B{b>
zpwP}4p4u1d%hScn&T^$aeQTg4V`?9*k)yC19ou<Ub`xZKh<Hklq6~Y4l<Gq^)n5Hn
z>7{;l(<1nAE3MFnR`uFyenNDJc(J=F+nd&qLy9VM-|u=yZ_-55jHICBa3-NRbpQ6D
zYonQ#;BwBH<`rhO&f<fso|Wwfw`R61K>ge0wTwCDyY#z*Q6SL>326Yr{@C5U6L{~&
zAM2)s6VVSh4~);0+Rqf`uz6?&sbBrzr3=3|Nkl={s~6^-^Ovg^zTZ85xu?=g^=v@N
zoedO<E>BQ<p$*8(pjlQ#)>&Y569)mB9@;n0>v(#W^@!#xKlhIwXKygivdd(j{3RyO
zZo{|F*K$|T%}1L=|B&4iGKWSm>4-g8PtsOaXonyaLopJlnWYvQu~qq<k3{kB{bBBX
zfmC^AQk>okMxs+?w+=6WI+I1J!bCi3^!nkb)OSnfT9)O{?JwK(q8?W*)>ScoXWxx6
zuwP{X)g&5rYZFh(+wyhihCk3ZT?5SoDm8fq?H1?vpN=*2qW^gz<IDC<F8z+QjB0pb
zX{fMwiD>$-5MMk@Ow=zWI7fwjOulzVfTDhP{hI3|qdMFB%|z9$L4p~9)>#>Iema<@
zYxc?NsBZ2qQ+L81s!)7)T8ijZBO$(BGQE$j*?~RUbOEat#bFt}3mbw|$yAzdJ&hhE
zF^QN+H0{Uk53<C_A@fW04xg-bDb%R$!UH2Yo9Gwv2ABhKgcSB!$${Z7tEg=jfe9zs
z)OaQ^A(5P$;<*KUynp`SKpDn-G!BFuJsha&>kwl`tjX&zL_Edm1FVo%1wX#5fjz8+
z1^gr%zdqR+kk^X|)FcEuFv%%dU<$}$N6FC>R0Cudi-{%!B+eP=JKEJ{oi#*mHV+1l
zWaCsT4(-2x!v3ZY%TQu?_{*ih4a0RFjX-Mbl|@fb+hi3K($S|!$ioSA3VmIy-26zA
zG>&c&x`M!~<t~Y<?-1@y2m**c3dkaHw`wP}EAzpJSHTo9#3;))@VB<+jd^_RcTcRG
zDabyvkG}TmhG~5wUJvJZp7<7h|1aJPh%PwSNUAyYIA86_T)oJDVm?s?!R(b&R2+8Y
zQBRkC+qwGI?c2lP$c2xVk{x4gCAQ&1YV+Zhuur&*79m+4K}{_&RkAG6Sp{?k?7L`q
zGmaeaVwL>@9wGTg70dIZs4Tss{duo^E<OoM=~75y)y0IQyKGem-P|~J4WAt+w(8XD
zHSLFMzPcwOcRuoX?QJUJkS9%;>U=u5%%af-DI0Bz_hMZ}S!7@+4u^FS?lIYiM`V?E
z?wEdJeeIe7e2U~Th;z{pPR>lr%$A`zJ5j!1L0X0V*&GjAKZ#0P2tm?iR`Ai+M;RsH
zw&6-4J^3k{Mm<D^RuuV)562|^Az)pgL_HyIG<ZJKWqk1jWrFYjnhF;;;%C#AH>%$*
zUp1$NtZu0zX^Cgg2LkecAKUNM))E0pHdQ{{1AciW2JIg~<ob;!=+zf%h@J111*Z11
z;Lw>esmpsKN<Dyw`w|y~DH9PZknm)d?=wjht4I6>=qhJ}PdSQ{^8++wv&jPJ{stjo
zxBA@3VOiRgmt(z|xBNo~r1qr7zM&;f?5*CThWd*fT_R`XUu&-tzIidy9-z>FBUB+(
z|6#jz<y*8r%);Fw!3L82?u5d3XS$25Aa<eF#p*iE(~iho8_0uGo=$+joawCi203Zz
z2LNOOV>KE;^`v9)39}DQ`jeEOg<GyU5%gFFC}l({tNp#$ri~S29<Ah@)^^YvUT?IE
zKU*PF0_1HP0<@Q3C)t3lK)Nnf4q5d1qhXH{3fy(<jI@Ls2YovZBpw#Q<kPsBoVw$!
z(9jVOmLgJZzX!NiusM!*T-n_^O*@6m>Pa>UHnlCNc+L!gpqVF5a#Tph*Gr$RIVT3#
zEr*?ya%uyFK01$m${iVntu@_~-2ul%59izL`8^up{*2|<vbpN^<Do}Svv8O$i)WI6
z1t8%Mzkj^#i>=8REuCU@TLF;9L1u;2W)rEfMdDFUzr<one*Xki=WeOt2i*K8o!fxr
zq7U0Rr6b0P&Huwy)7g2kdvBXyz8?-8uZlbO1NL$jb~e~P1TCX<-uIPMSV5G9oJB>p
z6S(Mrquv(=sw;U{FpqffJ}Rx(9@G=xklx9zdT%;Izc%&a>V-k+q}{*4s7iXge*e=|
zIq$dcc<$c*&23=SHw~rjxXhgs$V%pY{!aY0LWDkxHdXm(f~E!WyJI<M9uve*D-&a4
zCUe&k(=WKom!+%hrY4Bw%6i9Bg>rQ3_sNRK&#4?3f2#qIeo|QBk7oeWC4Hb(MgNUf
zS=I2b2Ami*y9@=%rZ;bHkOAjodF~g?2{|s8|N107@{+IW2gFOl%RF`UFFnrK$ni|g
za@Bx`9cw%LLDQ#R;8)HoCBY}lJHQyoK9m7aBVda{KDK4D$4WKONQ^Ts@5}v#lJM|w
ztG|MbH*<6G<bs2Ps|D8i>9qgsa_Qf8+C;iOO~o%iQ8KV(CtAEGd<3Sp>}(%uX#TI!
zi}w%E^!lI^v%Blzb4V0Kh-4a;6<XD0^;eLss2!T~PpjsivX!Uho#q!5Y?o^{*sMAK
zTE4P5l8H4wM)BadL3bDNqVLaFrzT~m$rpA}7H4SN#g3ma|IwA=;>g<!cVEorZne5>
z;1bWxU!I-@{z^$!ouYnqjFzv#Q#!%(jeM9%LFW*&0nQMH$TX^$r~NyCmMExUP)h->
zrZS?`e0_hUr0k1#e2$7qjSC=Vcxa}zkc?!DuH9URw4i&yn9$4MSP}Bez5POv3~d~G
zB39_{fjPgO*}~Ujr5_+_LXSx_3E+x3;zvn>DyeEa0+r)7FrpO32T?1Kj3klaZy}Pl
zoq5*o69`!i`qm_ZRXKTG5;_!G81_8hG`P19;0ZY*1gL#wK4`-be#IS|F5qw&L!B+-
zqob?I@)_#A-m~3tgIglka8}HbYHM8UP4xh>GOKx!_=k?P>eY{W3~>kz{Mx6gM_l1;
z+>Q1saz!O<my66-NKWd1rdhkAfks_AtvEdJeNQn7Qlo!yw$}8Z0y1cTv$w;-j&Gb@
zNA+>L!ngN?{%p$Un)4lPpWYJ$W727D#RE{SYla8SlEkX5>UHn_$0=5?<mb-V2vGKG
zxi#Tzcs8`JUP;DdC%F!YxH`U5{y(gRxPb(3#cV5ooq%N^D!-K`azB!6+r0bTZ=L>7
znlF42yWLsIRLxnJx3a5wVz&O@7*#EmdS=ma`x%q^Qq0|Nv1Z$7ClNpU^i~@qS-#NP
zcIb5}eTwzl2vG>;fbYLifnX>8H=vZjgvu98pZp$r7}i$KN3<Yc9<j#3MXW!`=GMTf
zhD7!pTF(Qn;KS1*&ymHlB&e}5VRCyyAMO|aY9=?$6;?}DsfwX5%>!}%viuXffsp+-
zMF1?4^prx-UGK=<_$s&!R3cb3VX_ZZ%P%+GH3aWUdewP4gY)hQf_Rc5FiC^tN|7X1
z(O+0JB&x4qi11XxSeO|YR>TI>l?3k_J|Fo6ZLDfS$P%NhKrnu|@({Aqt!FGQs0X&v
z5CLRmE(_=)yK+nKt7C}%Fi{k)y&BnY@ncI%pQ|Q8<j12Mapr?gu_xx3=M7kzlyDj&
z`*|{T@5-l~nu!SN3E&tSxH?emGfs%5r=<m5HKcupVih$-i9u+so%H6~t4OoNtVYPG
zOo9qJ!_mY+45l<ifF-m*dl@~QfF_`{4_D@N`)0{OsU0t_pm2le!GJ3dV~*S6(!Q%1
zI=>J%iF46NSNYPEj?q#FHh*W?&^U!20mdgeLtb&~Ztq1B&yKSEBUPT96QtR+A>`GQ
z?7IHEuzUHT5SDv_3k^RE<D;-NA%ZwVcTC+ecF^>SPt9|HmzqEi$&wbSp<^y)@2X$K
zhRs8OS#SsC#12pw4a-CO{upRkbMx>tBlpD3`sTWgpi*<jGme@mxeW$Qte*a-Va8+^
zt?Tl^nnMhd>>2~+h65r-#EWf!u?!@@@<6v}wT}N8rN?jp>yabWKP-G*#UHR$S0ei9
z-49^4jYgi<2K-3eey!*@35YA;gi=D*4-@cWCS`)GO!&Tlq*c`EY#!>voV*M+u$+?8
z6YSEeqZIA*4>dcQ6Fh_idM!EIGEE-%%NrEuQ8eBF4?DFD+#OU#V8!{Z5&EZoJld3(
zw{Fq9!;a>50U#HgeP}vjaQq`c7)5E;p+%1QmBM<^{fehL3znJ2O;cV>DEc4Va?V=(
z`Mx+Rmu39BJ2D@Br*Tq~h128JWCCRNB>BF&a3j*>+ePi-9C><;n_#}M(tH_^t}>=E
zqT9PN2iN}RkJ}>wCi$}d)ovV!n^Halv-^hpJxo;AY@zvL|MeX`W5%&9r@b(d9dKSa
zJRK|3MqJ5_Ll!dh<Z&lL@;_TbCrn>Aa0amKldVUxG?JMoaMe*%?-N%b)1~w3hhw>@
zP-3Y)LbRe=x&3~_c7ZSw3^mk##G$mwGDV4Ap1+*&KBkaKVn(5j<240}On})9`c~32
zw7*nh^mMfp-_c_G?=%`@Bo03eIwCSmJUag01P>pd4j@CNge^HQNhr|@hNIrm&1+-K
zS@xO*CY>DV9lW^a#g^^eQw`N-UFk)b-2z(4Jk-m=^}eu4if?SA9axy*!#Abdx$5#X
zDwYvu!FW6zH&U+*5(5~1mgPBDo)56IAd{1I-dq|w(QDJD)G=f>{!{++r#5QmDvLcz
zl}{zboGE+l7iZY-jr-fEpm=ST0_q50kzIQ0?U35O2Sz^I&9)smb?MTjbTuX^V^jUc
z+Sji^QO9H(j!t=-MYC!)5d6xveR^Z__FBrL=Bhg74KB{k*DnfE==4t%Zl1bmg+YIl
z5gNbh@~e=R(Wvr@2YdUYseZ}_gAHd$v<#Uj*h4pBw%!cw3-DM{^!RN#?Y#Zon(0fF
zfgR$b?iiv+aWAR~UdVNgn!%e*BlQoML22hL-5Fbj4I%&A6)f=G<F1rUVX?0860hhZ
zkE>f((fQ8NF?BI)cv4Z})jT1JNzxT03va}gkTn|UvBa}+<I_rQk+lz&#jCI7CentO
zY3u5UZ`erw+PbUpfy)w81HpstW1Fc;jPG0`wbKf~iN+##QsR#(r3xYI>vxD9r)45b
zO$yAgpJX>YfFh=Xk<M0q8A*HhN9%4xL7h+J1;H+b3i!k!RKR6X7<iKjq9q16KT>D0
zO;Mw*pf6-7zqo*j-Tj&frSHY#-<p#J7Q0iS%~{$Lj-M;Va%>O{BlL^NV)+(g;K&=k
zM1WX?l;AyT5$a~UwTR}#)T`$&@m2#u^LR_@95*-j9ZZAP_BoZ|rzRkkQj$YsOjrtj
zDIX^>Q};LYmh1T>iNB~T@f<MUf-Ss4SfDv~5PK~{=ih1NdNUevn)58|n?m~#3j8(+
z#;$uRT_&{w{mM~f7Z@8l1)>1zNDQokfeU2uH+p7=BR~lxhq%dJ5r|D9nqW_pW4>6R
zjDo=8X@Con;~nVuB)~ms1#oy2B_|~%w9R8(fAUXMif9v(B+5Ah7J|rp#V7E}*$|=h
zTv$i)KQj#Ctf+E0?wN>1*P--Nw|K#4Qn|g)E~?G~L8_`Q7z8Ws9!^4PGtHK2H3Vp<
zu}{)KbdVrnzB2IyivoEK^>54TrA;USM@4&Ak_-EX$)W!qP+(jBo70lV7?qHw-GZ|B
z8J(pHU3m?<<*8)3bb4RA&?oNnbvK|^bY}br)DvFdaL_T9IkwMBk*t%yE3k605>z%J
zlej9}VxR~1-M*!!3rt(4>}r(NYHE!cZxD5PI4?!Du5;h}%1Z<=QSAC+dbTDs6VyHq
zsRqDP3;CFCBgbdGqy27};I#;IrZ?z?gu#2pRuIv0{<%w)m0cADpF;Vx1%3ILpzmw4
z;|VkY#WTpG??1IOuByueUqrcavx$=LJM$NGs_9&gh8OZidW8^JmNnG(KWZ5su@;D?
zdGVH6V3-_JUAR?SQpF<s0rU;>QQ`KzFbp^%3g9Ur0i9myo2###kR&lk(JMvbA>%4^
zf0!s7HT$jA|HRg8rr3Gm@L6dIZ)Goz^~>4Rt0FyBX8L9`kY<VInvHz-#uM!2`4mQJ
zgoW=N0fJwWbrd0+(`<5gsN(nQb1JlZrExU<3!-E6p(*g;KxK#xi>VoXdE#KRuPUuT
z$HC1_H5-rJv-Pqa2%z;e7ZMh)d&1gMjZ097HD&Cw_d>WH!Wjn&jVY9vuX`-W+ocA`
zJvuo0B!D~0f@4>MAC>ENH8;r4p@-fV3b4jIQ#c~tnxkd8EjYL>4wLI6=DXB)Ug^^H
z@u`ADM3-FxKu|+Q#+3A9d2ig>8J{DAeXP{Md+}nRT>x%rT^=cRf(DiQIrW^(Aht_}
z{~N%`s$oy^xu}YmSync-QSi9hNU8~$E&+hTc4&p3(QlYF9|8X2p_Rn246QHS-S^Mz
zzW;b%X7Is7hsc8$#|k;Ah$#g$64=pnj_ncMe@fH|+dlr0@ST)h^ztts2-Pd7l2s0f
z>DoMh8kyZzcfa(dy8btHtL6#@J4qyn<T{a5NHC?e(B%Gtd9f$JaK6&kDLH(<@t;XF
zjDfU+S@q%|pGNGlK-!0F+sy<(3Db7+C7w8S96gp3*n9H=gCzE+%CS89eEOBF8w}<R
z39b)H7~PJWpmSOD*&zhD3f?OXih^SIjtpa%s1)cz!KFrlZWN=XkPV9~MN=677)cPB
zhi2vjOeB*;XH+G6;v?kfX(cr9M^Tiu`z*El3_}6g`1dS*e#E;R?g;EEFH#o*xp17E
zb@6I{^^kWWAtds;G0cWGb5rcG5i(aQV-_bozpo<=U{S|Plzd;?4a;t%fPf;v^eP4U
zA5FG=$3bmqy%cBE8I5^Hi|eOIdJP{{`nx!VL>H*s7HndSCXvE5g<ZCHMMw;r*_O2)
z97M9soGM{HtV`pEM$AtAy!}U`CC-rJ7yX3H&3Ei919O{e_X~NDskjM{B)r7qd-$Oh
z2Ekl)BOJxD%6}cv`~YG6?tBaRyetd?m(A}1`nBP=QSy2c1r3&XX+|dPAkYNV0DX1_
zUyB^o_$<4MN|4~u<fMIpjNShXP2GN0SH8JUVMldv>`NKY?}!a`@rugz6g!~Zs0UK6
z>_*xOnyPvN>OpXAbhaM=vzU&N)~vzQ-u^^HYggBv(Hd)v!$drS-uP}CluOpIC5%Y*
zGkdYU_n+<So@w5yVj>`?6n>{q0jC|N$s#FM{d}A_uNwAM?+GH~sAE#Pu13>^T(R1#
z172Luqb%f~WPvA}?Xxr*@rWeKY+alsdAGki|JqHwfB0zo#A29_?f99B7Qj8?G!i@Q
zPw4DQ55JKtU3_bcZSdV<3+4CmVves@g&uw1X!(99eUr@V;P+R*iXHs?%`ZGz=pe^_
zX^9(tJ7|PuS-PFty(cEZ5=(Vk&aXG@F0yT3a1O2QGCXtKt1Lb3M<#IiscVZ@(~9_=
zTTjNRc)Eny5Ih|_x)!dfp3%`!(9`;{>*27d)tft;3_@>4^hc>OaXvpG_6SQN!={RR
z&GoS^h*}q*gA(7xD|7K!(1G`PJ~InTE>$*`mQ5&x94@y{%v!^tN3orfZ2zg9=9Q$U
z@kU0vp`syqy7KNf^_$wdbyjuMWLU*1GL%<Xl}V25(mG8tL?DkK2i;zNs5q%Cv%sB5
zH`;M=e#tulBfcty6m#>cLo!T%*Oxgd*@X47(ED$flyof?u2!@idaj<kLsAlB3OZ8U
zkv$W3?b*=n&c|#Nu2IeVT9CK?`SVX-vZ)!I>bhqV+hvxk_ueK*V-vr_n@#_Hl?x*q
zv)hkW9jMyWJAQDW{^$&2g_lJ(=h)Bhs*kVQbxJrHg2_wjQIVC|a7)@KM!&`@-H4Ty
zf-a@zDGQ6oFFz8*XWPX;KfOAaAqyf4FhJFd6>H<S8qF75$lGg1RN~~5n7GSR0lng1
zapslsHe7iUnArobd&JIX;eM#%05>^q?)iSI!qEBS8Int>RFRpQdJw8<?=$$-FZ;Z}
z2~+ZI={feWXBSYAl#9jtkb;YEfSl+Fm~ZV!UG80C`@OR+PuBEBK>S3{hYufY6Wuaw
zej8up&9;=iJmCpBU<cc-zMoA!(k5sM*>snl0j^zRk~1_AnL1%AboB8lGJaxu$*(3E
zI9-B-Rh7A9ByU*laM7Ei&d$@`YEjgpjj?Qz80B?R3>ycHS5y<Xe7vZT>CD+JpS>X1
zBnNvyL!YH-_uCo*a+h_AiY^5i4;ceIfAC;0L}&BFpVs<3y<ifTS^*)49lGq<Sy{UJ
z`mtDG>sf;gPs)5Y1;KviQo1SkOmM}8*9(F}o^u8Rkl!rxdiuy^YsjvU63w~exA*(h
zs^8O28(*?h;A^|HX&+x!y2`u&ecpay(?&UCBcmTT`+DA+XH7<S&b<14{&)1}mh!ba
zhK4qhiy#}eX-`1Ln^*G#M9Jne$2D5-@&C|{-E{yu_KntbyKde$<?bwTOe1-J0|Nst
zdw4jbdM3`oytwSZH|m#n-f9*Dh71JZwZk*#=uD(5F>f^U%{(AQeVNg%4APCz65OnL
zuH5ZvP}jrebdt%e`2jw~?RGY^g?Q?}Nl82p@>aS3GAEgI0cEr&2G$;{%TqE)S)E*L
z5tesUWG$=iH*aWyc2;;6VJN&V+L=%<Vt$!Z#g*Gg%tzPu<)(!^`m4ADXATy5YptNq
zv+Szn&S8f;n`yoX#6t{b7xO-HS|sH_$e3g&sbU5ZJ7~$s;`xAD=<GJOBfC6>oTmgX
zb-JeOOHb{^7u{&DUH<k@a4FX7YFCEzW7Rt}@z7XxX@rb#_3n&+{rVt@m?!xxD{e(E
zOXc0IZ{4mp*AZTlGB=|+%HQL}!Nv~W@8`qBWy<<R0E?e&<TT^03oM(3cb>~n?;d}6
z@b(8_oh`#=S}j-PtMoUK?`_+WcVNfatVj_rxL}Zd0Hwb#MR&p!7JUD)UoM<Qu(GjR
zcbWL{g7M3Xx2$hskE|k^)@8RN-`|F;Fz%v}btJPBB;)~t->L08slca;t=S@KN`-5m
z%iqz<6`OunjB&JfoW7J?{8RFtKh*z6psv}3djBzGkVsVsUU}PiSTZ~Y>muW<pc3L#
zk+X%WY*`uqk5`OF)fMGN+4zPQ`(`)KZR_rA3~ZKAU?7rZ7T(jods{PsIT%llQWz@G
z7;zf71n#XKjUSs3q$~eXQ>7D%?c|JM&EdQ{iMlyyc&M~Ob9CX+_>$My4UWyfL53e#
zQnnYw2QraLcT=`CK35YY(<|UzWsLc$<uKaMi~3O(g30sOR6UKrcPouhxe#r|+gazW
zq9C!?_>C%0<LV8A=Em2xja}KY<SFzyV^+MVKK$<yZ@8u=XN<Pr%ga9am5GKB;#vsp
z?alLT`41Q1l`ErHQ+wWmU~ScV)0_0JfU@XwPKcS?$!O8`N_6;|D+{>htet(ASQUe6
z?M}#8Zv;pAxJ8gEUHK4G$HWfp9EGh51($Bh%%Yo%_vI+3E2ox8`4!FP(HgO8Q-~5F
zlM!Bu<UH@@p^$wu){Trtv%SV^xT|3(%nW^O{hJ3LYDd!wvBj!?7@W!->i2%f`Wkj+
zxOC#PFgv5z+F(#GuR7QZ2B;2M<AxYzgJl#p8{@BM-jMJonucELr4+XIPTV_t1@XE(
zJ@fSFe`y)(nd#3M8D({Lo=ZFWf;e;X<OA%Lc#d5a)r>}jdL<Ngu|WHscn$<}ln+_M
z16yfv@@0a=!|xW_1Kfdoja|XFGNEpiz7WtN$}ASVU*o5n0lrhM)=s80+N*!`M2$*t
zvNB0h9&cJV=}}=sf1`vk=^b$=XsqPc^Famr6W6e1*^ZDM;-Q2wYI~olqM(Tok}~Kf
zCPVtO<)0E}z>nWI?dfF+dWmh8yw&XWt#Au{Y@}PN05_v|O*uH+<B+oF348zn<g~hK
zZ`@_7vXmc}J=aUA;#-^ql?FbDGj9yF<7w2de%;7Me19XWg@(VTO|<7aUR2x3I=2K=
zI)?!p8k&6LBi@{p<t-T8bisojM-=o~n91qXL&q9d{*Smx!lB=9;7OBToEy=bCSP!`
zwJ@FTwCu_sQZJAbT#Aqn1!x@c2nJf&LL_0zcgOQ(6zQ*SiT3)gHPw$wX-+4jIwPHq
zK2+f}v%XDxI5+KO*v5PN-C*CtbxUbD5wx_lkmVz(xP>F91yegx$>dZ6Ap6shOfob~
zpl1^pbJA3QE2VYX*n93VXk6HWtg3%Z(vH#?_(sx*alpkQ$$LISBqJ{`f+~MpAnXtV
zH6E=)DQ6^jfDe$<9f<&HfzL%O2|4!bDkZoSuT6&4kidb5Cvgtaw-N5Wyk^P!VT-{N
zxG7`lf+=zHybya%(YoLOauuH7&D;0*SQAt7UkIv__-KSx9P8QBN;oLv;^Kh0DwOnV
z#IsV4z!wvYC9G$Z9*|TR_uEXo3Z}~Y^ms31l7|Vl_NIM-Iby<ukYoqNZu>oloVCfa
zFz&&;k*3X7{bIjNpZxs%A@qMEL|Axs?T7Lf>)C*7>{xCP_upekGjr0=Fw;_Yd&bVm
zovA(##}Wz~2X{Q_B(}mR+zZp~5uF1fJK8Ew9&;wOD?up0H(L7_kMQ>XHJ4=P4c^bW
z<~rXqdNHGp2U!;S2cC!Rh*F|f>gwvILY-aH#-IQ5?=Ow2dpViu4bONUJsMJBh==#2
zb8mHs>OUPnzMa<9y=XUqCj{d@R^1k(nKiGXEPgIimc!ad3yMgurb0W;(}o=_O@Ee_
zQ=YKcN8r!*eQ{E1=#N&HR~866)&k&AEN<z>UFT<pWxCkTFpP!2lj}3q$iT53pRj14
z*Uk5Ai<<4S(S07~Ed{GS+CK409%?9WLBNrTJ>YvnInO@@tya#DRVQa24-Pw(reCIw
z-@r(RaT0vD;>(uPC|4|1^}eFQw3{*(7`|E0(s2gCUZK&NJFIrR&Y0r#eQJ(;1%@Qy
zL!WK&RU~bu00#&Ja}vXfV`k>KWAX;OZES6IMb=);iCZ*JQf@F3?_kMe(2PmNEn&F%
z#lY~)wH|iL?^tyqlOzs2_wn}<G-5`vCz&xp{OKqg7KoQggK{51UAm`uu~7W!cP{sL
z0<YTdH(RJ@<%NA!m>E~=@Xy+{HQ=iN2u#6m<2mxgL&En^kD520Y|yaouJiWNyF*Ll
zROR{CpJIyvzS*N^{)s2HQ7Npp<;RF|F`6tVF60e<ZK$N9RsZ=-NZz$0n{6|~&B6}P
z-{&$I$~%125PJO^2cM@M`1+~HCnnV@ss1Z>X99WzYwxP~oSYFEA#R?X*}Gxtt&8Ns
z^l$^8E<ybe#cv)pudxrWj<$KI#vN!{)sAp=nrVhM^^(1sb`?c*3+I}*Qp_-RkuZK=
z1I>7mf}O{(I&?Xit|b4)PWsr|3NMY)%i*iHB3Kb87!h7w-sp@h7#*B^u^P)KqVvVU
zy1YLEZMAs|ii*B_ehxE7awj#W&Ch$d+)*U+FWf%nmnXcuDq8xJ_hXmb=9Ydjyvu+D
z4NowJAAa=Rznhu-Wsp~LE)9r?@LJEe^f@6Wg(9{$psWN!i8B-%1-=1iS_2_R`;lo6
z9t%1`onjs~w)qc*J{2W}T}xwku$?N=bbai=5RN1Aw7hPiuhdJW1x6hElm)@{Egd{7
z-i%Hw8L&_*rLKIEBKIXS9BTVx5{zU*jyAJ9dWiC>QHt5KayOy{cWk&94yYZSR7!o{
z&z!@#As{fQb75sy0MJLEJY}XI1-jUn+%`l@FpLQ$L`xjjj|xka^`D`2Mm&bOIYzRS
z4C#qv2Jt+40&Ikb$gI{rbVWdi2*FRB(o}6J74*jNWlxF~4AC(?Tlpa&PaYAZUv*~}
zTzhJcto$9K^tG_Hda#VvX#`9NR$u>TIm|-;39@fQckt*vDzu~N4qt|d-hK^y`>LzM
z8DcnsAh>g|XEA+)0RZGx?(QWB)1t5P#1}rbOEJ$2HL=hbB?rju-hCM)eVl9vS-iW<
zZ?uT6Uxj1&?Abvl9(4u`r8PrNZKo!$l9L!2h`lf+W(_C2h|bh8VFQ}^v%=v`AbtsJ
zK1okcW~1pgxk;ZKfS9HqC7Q7mNk`l~EybN*1QA0r9!o(0TosX;bopRCOB<Pp?KitO
z)XVpWv>%t4=~sI_Jh41bajRD$A#ZDK6$z_F>=mo(nQ&4hs1#VY;}@XYxH6?@5u}9#
z-bD~pX53)`wmwd#Z5u+Ypsx!jt0&K%>HM2Vs&PyXSOONnC{`LdwDlx6_axwnmbY7z
zo=y1tU8qP*dM<A>(P*}_?gLaA3m|a{dHwSqaPesE?0f?Ai6PKW#y<AYxJoncg9RQQ
z10)d=+t;aI#m)aUcMXRI;?%X~dZ=A(z@@n>3-CkErCfrb`oYf!9~QS45W1N~a?*8;
z2QFSxp4GG-l4z<uetZDS5Wj`4T;gE$eSZjP4S|w%*D7viLUnO+pVyzSK^~ZSafai;
z_-xQIyG~@>sM~cx(O9Ztt!lJoc)WRSuo&;{BN6b@Wyn+yXQKC?5rcNwE6Xzl4dEE0
z6)C8SWFx3wa8oW@S%p7}V1hN$Bdl64XXb^o3Y*Fr0prnpp+V^~<Fd`Gl4bubn~A-S
zj?ckU^oUK(fL_uek^!&`R@!Ko&*H;_VFH<2_ji-RXk)X5o>O|r-%w&(!(^v)qH@6r
zBY_iVm$<ow!*5hH$g-&aG>Z!i%`;|TJw_p}M|uetv<`gj1{19E&o?Ho#h<lg=}HEQ
zt7BaxL#sl|?T0O0xr*CN$0Zbs&iEUnDIpT4#5f~Nxt~0J+Jd_KynBUF=DC6+IG=5*
z3cK&u3b=m{Ur|wpW!~jaIi}+@s*wU%M}~Q9<#}vxj_A+Ss4}DyRE>A4SM@Ga>}KAm
zW^wzS&RY<o*nPE_bqsyj-Efa$2Z{>_o1_d&)$t?%W-3sCiFlN>_<bwCE$kLZVxg#J
zcE!xOs)Ful=Lf%Gg6vs2fMdNbk1vZ+-Ie=Q69@S=mEWek<e!?(ET?jsGjxs_e~lcQ
zHOiQ-S-vx;8m`~O`lCONDh~*7yr`$GLB8i1IZ+5d$RJf<da7RUK_Tn1o8j1e(8(jo
zEKEf=BHv?r#M`<|a#X_#l^cY562rLEuOelc*-f$Aco5jx{E->&$E^n^FHRV*_U?@w
zn@duh+4&5Em`+MNGsd5Xx)*k5rmH>HYH7u8O(CZiQDf#D+5jUCeHgYi!!~`7*Jr?Y
z(-6tpG%<Kz)+&NjCh1a>mz6ynfo$$dnEmH=3Im^z<~)j?UmjUs@^KIPu}!`-VO~h7
z!lph(I!?+|BS}Sm4UEN)z(pXUlhxok?ISNaS=mo}d=ue<lkx1Cc-jJhBCH^-2wBS}
zA)F}FdX8CpwK4FDi5-Vh00X)Q4jj;)T@}yHOyB?0%{(p@5K-yaj`i!;uL*joSP*;U
z@vpe^c1_89`74K<r%O0)9~^ItPDEnJBA~1=sDzZQ<$(`qTy=d{8|JK&jV)%+OBB&N
zJfEOB-p#-D(s#$PZJPT6ncr^HTYX*QvaRiVclAI31xRY&JCAi1)EVXL%u5`b@~BW0
zY@zkq#LY~0Qb8|A=s*#J5Q{=wu_(N8_S3Ilzv9#Kj-|D@9NeslhU(00r-{_3uX+{m
zU>x+aSFLRhG_?dir`UZq0y$^rNB-db4)fRc{oGzk1-48AFCI`rEWmJyC*yMQ)jo2T
z^zuk+hKh@9U7o{_;*LDmixFd?F{_X6xjXkO9(j*;S{bv!?8p0s3wc@B#Er^}>$VAz
z<0t&s1FL+rf8Vb0XBfrwbyRI%c>CxXcX#*b*`>*sY>tUfXOsO|KjfIcbSTYk9Bcs#
z)#mRauzwfNznXWqIG?{%Xerxi=6ULTu@eH%4$d{wM<@q&t_`-tOlm~wj~z(&8s~)U
z@PCTa;kPio?lR<KI9-e7I{3M;ee@$(c8*znXe2ZtuNX^0EeM-KM{ng5@BC_nHh=Nu
z!@+<8hbJ(+<>w`9Yipy0j}XQgbaKcmFE=<}Wcv{m6o<=2DJdEN59&)op>>52`S|6;
z=J3PsYrEBr#Uv$*+7tf(X%{V6m0@Vu{5cl9PWi!;e_-mE6UfRR*WxuhKEc3euU^!H
zH?Vg6aU2g-(I<*K@BvJ0zB?Ifc9y$82Mg5duYtb4rkZO>il!9Y5YLkgT9Tp7ZxYHa
zt#K*QJvCHiMh_K9|K~P{1IXbdG-{6=UhL*yyvUH2YBj+3XeM>M@)^@()GwzF33}zj
z%8ai}yf0v#b`z*JeC|J(TV$u<cJ!w2r)6ip6(o6^mA>8B3{SOZ(1~nnRw7+t4kh^*
z`_r>Hl+qE^gcG3m?d*&S4%ZJ$Kg)<{6w^$WJQgFh7k8r-vLNvi7S-=A<*VI1eUh#m
zhjt6>aYYJli<AR&l=YhbT`tiw#uUqOt-E0=Lgkb~62G@y=a0J5Ul`fBGz8|)kKcv<
zY^udA&#T0Y0!_cm?kdX4%6e%zn}vPZzo;t3ybtH%-Bxdj<R9NV$>+J3fj_dv>v3T-
zD;lALRqO2i8V5&by!e^v9e-@bA7!gT5)F>UDZqG=I7dcG{Ax_z+iVOBZ?Alqn)<r(
z!mGh4E9}xSPpT~{umy)HjCVOq1=Cc7H!7{46Ywl7EOb$PBP=85XQ7Y7=_I!Odz*!&
z<?@*j2y~9xFW!$G@Lk#wf06rO*{ru}ZE1uPWX1-{liln>A66J;t5i#%eJ9)CGTV6U
zc~pqH(4F>6nyT3-B=310JE5x^H9NR!{-9mz%WXHQ@(V1+sFrTtyqUtS7|kQMh`{UQ
zo>U+DWpJur?nt4$&8&stoC2nK%PLeni`D2uuc_HPsBp95KovCyn5GigFekwn-FSvL
z<<0PudfLOoc=n9o+jwGi(92!S54)BvxeprI44$5<jE;FeN#FE*j4JO~vV;1f?ov;?
zp>ciuv4%10C2yak@v~)0?=Stmc_DYgQ!Q!MwX(%+eBvsUadOkvqip%n<b%QKB>snk
zEw)1Jmh!a<Jw74+qqJY{LS^?5-_-y#RkNUYsZCOvMTSAL{s?0hr`D5)`ju-6bczqO
z0kaa-1fDkJAkYQ6`O%2iq5SvT9j8~;>%U1Q)syo;wKaDjtMa5e$tW!C^<(;tfM7NJ
z;W4HHGvF%aqWJ-*0*Uf##mQ#W-dkEz@aCMea{-Mn6~a>x4eK1Hn+9ecF5-FbAA|qt
zm)~!YV&=a#S+gQ{h7e~B>I<pzL8?YAElv1Di+D|V_OxIs^a(15xus=R8_vt&axbN1
z@s8wh27uVyf%gerKV~dtpJ3*9Z-f&plc&#jdFlTPauCGq&|eGnsorT1xvSsy8$v5@
z<uwiW$y$WW59GhNJ+uXeubTZ8B8!ox+~j2GhMngg?HAsZVdLE~H#G<#)(YVkiN)yT
zCsO5;984fbUj2%YSCGsGd~tI#?`XH63ptFm2woJK$cOascP*0CdXOt*<JY))bfuqs
zNDAY*D%t#lL@?i^hDIx1;UvhCc56Ca>;yK3ydss{yVUNdT4isERY{uoT;fP(H?E!D
z5_;w=b1+{;5`8DKOW2x-m&eJ7JQAHRxH+ZOQJVZw0!x0$MK^wfKCMA=VWBQ^WmzrE
zQ`8MR5Wkbt4zy8!FoSAaC3xQHwCwj3MA84CmHl&DwByrB%a7wK(~+1*e@u#Hd@jC<
z)?^e&laI8y6i365%6-Zj5PQE|!Svkc-_M==sMvu{^0O~T`5Oilr?aMXlD!hQdRKi8
z;#h$Sxhk6PbCe%YS<zA>Uy^K1B5NyOy{G=a_MfV0Ox4xE9e+5jw&%!hx3*)Tp|~r@
z`_SA>KvU!ITs^~0<C)6c+{}Q%o)Ra<&@cNDQDMh&UE|ln605HNsTU{t=a~h~-y9m~
zJ#%@1CT>ZF`Yp2iIQhb!%7t^<sWd-K;dp$mdR5+R=9qPJ5^#Y_=%gh`P$bp@@;rHR
zlP183%QiO7#lnONkF4b#ATQ!|-(&>seeD&V<>l?tJcS$;S@O&Elp#$Un@j+)R&it2
z6fT%@-rO%7@hB>41CH3B9e(%!T?};@B&U-6vfqYc^WU%k|3Ck8ACg*Dne-y5)Q%>&
ze$G`cDBw|+`w(1G%rBSKyf%n=dsp>*VWp#S5u`s+PCoFwiORM<Qv=cYAk_Z-P8{4Z
zZq}QOJB(%B{}QjlRGEs8r}lifL~%7SW-4a#+JFsmcj8v>fg~$SOZ!r~;=@whItLD1
z7s)6nI72pQ&eA1s#rYqBT+KrN+5oKtJ8RGxUEM93<Z&pz!I4}C8JU^)Ou13xmbr{O
zl=2ZO|9RFxNWl)=W5Ds*e6qJ@VM*`x&3|zPf_fCt#lN5N|MZ{I43hKAJ7hKsqJM}w
zHUc0Tu3LgV<W*<p50>O&RIi@K7`z=WqB5BeyekwB!6#DR6>DzHbclOH-8@B%1_XkS
z<?d4=nwY(hC0)vuFsGdu@6u^S>RbX1CIkFR#8O)tK3)o=X;IB~1IOE>xR(O?SK)}a
z=#LWwNXl=JIrhLQK;Sb5?MnYHFNkVFQTRS7OcJz1BLiT<|Jw!o5@Yql!bCD4Ym9aP
zoE20x*nIGoJ$YaCi)<5c9n3GAq-**5fG~JWA}1KaL{_B8Z2^^YNI_Qazdl?8Fq|*M
z-W0j*bUX+NY`g8gJcF0dkqGc@)6k_LFBaI%e_d?!h5skljTJ~xMaW7My>z%^lc)N@
z+r0hu2Z%hfkA~^ZiAm3@o@~^ELosHd&6t@^L9>eC32oC|5YEXS7&%*&W9lDw|Gu2t
zq=^=4x*qthJAX0_0@RmQM|VW`(t<u9y1oANR_OLCeZRAhY$x94Q;euw!x@_udUsaN
zd(myF)@SjMnohtekeaKnqJs{PMsP)=L`dQY#zC&XJVSXCh32hl>BNM*P3I#L27rjr
zM^aV|hu!~C+?6mzb!O2AsACi{ZfGnKV;Vuh76nEHv3o=UEvv=^H3BLEBC>CZghdlK
zVno^?ia>W1S%hgYve~Q&xCCSoWKp&tMi6OQ1sbJom~+AT0aH^oLs=zNcD>iW_kG{J
z_uO;Od3bDwQlKj)3`X0f;o1#`t%`7<F(ffch7>l)WPv3WBRkQ!F;~YcQQ{7z(GOA+
zG`d7W?MD93EETP#iYrK2UfN%0WQ~n}Z%SnCvsF$^d?`Vg>@w;ow92P=VLEJ0@erS!
zwgEDxTB&45XNbhxh^)~sK&1=)qG1|AgV1;bRAEU**ZLT@?6*})2RD%}JOCjU4=!O%
zu#FSJ&>q+l@D9Y=3H%2Y(6GF7+=(RUZ%tLvVux}(cV7gg_Qe+$$ljA!HdIT{2qlqr
zj7NK~X^7Ohd128Pj*V3;DPoq4KvKw(&o1~=nMj7%z<`@NHAFMG?y_RNM<@$=>wbwh
z2@zq_05T;!#63!-t_CV+65kPOzd`^h9r{U?(vd2CRbU>opwWiO)Oe@`@5i&R?ut@6
zD0Meb<bZI1PI3{cG#y6u%TI<a(w8W}ZDT(Y&a)OMYvVZZ1%O|N#jS@F3+a+v81(w-
z9W{q(%QR8bQum65w#umtrG2|El1LS=9?YgJstZHM(f=T>8E2bHwg3@HRPb~Lpekz|
zB6b`_4z+)xrT`tZD8_CF)}@^o$bTPRN}B^w3ln*kU_%6pH!&O6*)%M@4ZB`n_aBvn
zGzke^=~xz#Mt}zj?Qw1R;0SnpG7-k{iajvVv?|+vIYxYWyBTWfnuLw~WLS?>-xM>?
z?*6I=0JD$8hSAO0>Fiv)+=e>`ZhO~Q9u=qOQ1w}`m^p!eSi{4_@kC%<%y8H4i!ox(
z;Y6+j%ME=ML+`c?@(}~K2X)=i91)G4a3I`&scO2;a){r?sq5`p_M-5PGi&7%wU>M=
z_Cp9celD0ARR+4877wJAqViZq2ld~!&`@-v6DpLVw#xkc-I2Uz;DRNZx?5Hvb7Jmg
z?z*{!KLl+)`=zxZnikYc1hA5c=DfP6aujwNT3mVMj_F<<?M9wWhj1;mucj^g%wS<K
zLI&N#P#`Vz+k-|2u;G5)fP}Wk5!UI)ODfP6J%{lfyUp*CCu9K*hkyw|LS2EKVIv-x
z5)s27G*M>`SDRL=0>f9?5qWx5Y9UhbOE!%@lDrsbVLkg|8Nlf*50v<74$8thEX!ia
zh+HFc%D>k&-HL41z-`DtV-wK+)p7}hm=jo3I-q%`a9kS6?l)1uO%@4>G3u$|D(A5s
zDIR@LKrc3HRJy;RSNJ_EZ2iJDk;ks3r~lEh6nPFI_;(Fmedi+0-0&1u66aa;W?}81
zY0{Q-`#a#)q=aky{F2LP?q=ql$(#G1Y4N0`<0TmWLuV2B&~;A?>G6}RudL5NaE-C7
z=!#<IS7I}1LE_?c$t>nPJts9x0w0ANLBGJ{ia;mSyB4qe_cQC1hLi)8m1`;=xUb<s
zL&cAl&4Kt7b@pBzp5EuGOg0c|Zt2&B!T)4=Kod=j`b)e?AH&9tq>hcvkOpiBP?lo%
zipWiqn!Vg*LfF0g8gV@gRL!tda$)_PMV0{2kTUMiE*BR?U_*9}$<{wtn-T@@914KL
zh(G`gdNXg;_={N#H)h|IHId=@-@HD+tH<PQt0e7vOV(AutycU`>v~@?%*<%s)_?4T
zxnyH_`Yi4kQbUTr?TDxRWX<un_s;4lN0n8anMab4_WVG&%su&y&h$s>>hyp?bIS;a
zaHJ>XJ8Un`6wXWWI9!`#dUUXQebAi+4!x5jXP^b4H1RDkX1F)?Gaqb?W9nwpBrTwx
zUHqZ9-O<2`a7++<4SntH0N(N7EDufD@Jv)VP>;C;t@QRb@nxZo%bRC^5J%;5#6q39
z2rx98syX)RZ$Tj~oZ$Xlcz1Sa3Q6wuF$V+l1V%Qu%u0#)NIU6u_q~zOpBUrjMr<t-
z$o_WI+1#yNYt2<6!*gMCK1tPaDbVGeTH=2iU<}%c$+es1)_{h5H65l4wdE(p2<cy-
zFXdK~kd6}3wt_LQ&{s|LuUu)|^~1a&;QWU$GmV2hnot_Fd=J$p^WxXKoIJPT>GV;y
zJ!_qOH_&cGe98|X7oL`4n|NdUwa*_q99y0z^;=ixya<DAlRxY`1-}1n`ttwH*o9@F
zV?ek$5JTICkzt}aUr(`;5<3tb_{q>z)ctsRFGh_eIX{7(wGyn4kEb63HU$j<&iG#N
zvK;WZaipl3^k=I@%aQK|L29tK#U=VnkjG$2Zx|{&LX2C?hTG)_aj>mr0d9T$l|wHY
zz(5E@YrJRgI8@^eLk9NJLw9v`jsLq>JkDg{#fulc1@yEX2RCAX0sW=(72SS`VY0_Z
zyG|^8q8P)5X~3mZsfR<u0^l6t&#KtI91T!rtygIAa4ONLBS+D|3Hn9^z>d@!U&va0
zd_n|p*d-Zn%6H)ueH{?rPSBfdQ)@lqQxplNXhMD47=iQk#ha=%_&B_9J+XMiuw<r%
zlm_jU)E?xoWg(MGV*Y6~D50rf{6pbmQjgmh@EnS@f3*$3|GM?lw%8}f8_vSfCE2W?
zFblH~w0DV$R#WRZ9fDzYr+n9cjm#s(`tnsMa}B$EiT|$(jq0Isxs3Q?(BWk9P`!*o
zLDHGVjZ-{U#aQ-2II9`YlGKBMh^*=fCa$LW@;R&b7TCl63`q&LA@(s89I-Dc1(V1-
zVT3b^hk*7AiOHf<)(eftY@hKLb`du~tULa``u1CxYUpM)w+;;BK;46L5fFn;`s(k-
z?yL!a=wpP2rt16&0DyZ8_Pw1QPWTAB<m~|<s>n%^uW9&I%ptK{+YMbm6t#}QW_Sp8
z+eC2#nAYw!_`Ld7KC->t)iD1BRv&{3379pU<A<S+Nqo*&7V*#b7Wi_;D;2lA+r`0I
z7Rb@JTWDM-tIi86Z1^nr6j>oj3n*<ua{g!V<QGx`RiovaKNjeBCOK`O<UNF_VVpy2
zKu=BMrL`{LZz6sm?T%w<M7)0;Z{Tj5I)Wpd67x+-d0EW(AZPr7!>nSMn!o>jX?kJS
z1IwPX7<`pE?53BZJo!=>0NT0v5lCl(LOXGI>CQd#tPMRaL1M$;7_JbIF9GiQSm_mx
z>Sk}X&JrSU2ZhRGES7rsl%gXPMjCi!ku>E4LQfg8S&0NkO;D(Fxqe{991C0xK0k#;
zUJ#g<$?#x8k9fjlSr%-#xcc_V4vMnk;&x=20DAbzl&EpGp(b3UkM=j3*~-~^x><6m
zNwjxJk&{o!Ux6zP%*}lMG8kPSS860^rb4pw(}|j9#y;$QnG7MUv0K4G=6swZbr*y+
zbt|y?9Yal)sO1-~HaRF+S=RomtCdJg_V(zdlypd~p*{le7y!Z{Gg%$Ycna&GoP15V
z3wTTAI7)5+-d|RG#WbuO>$Fo625TkiGg|Snnp*Z7`q<d~R9x)YEMm7(wuV~4U-TLe
z=!Hy%gr&BlCTR;*xuzI?!Cy0kXd}P|I+jI?8`7B4q3{$BS;3+`o6(EJI-A0)CuHux
zohY^xU|}<?nXLHi()VxF@}oa7xZ8m=G}XilC8MIM^u3jnQ3qZ*>MFC;eALC{?kcMV
z{j@TJ%vg88lMDBpe;;~6hkfioLM7!1(;dx*Hery&CS(oWj3)P56PLTHRL3C&M$o+<
zj*gEQTB*6f2YLKXX|wr4$49>;el%}I%b)_X(y`<cg=ChMJM|{1ni4*W!bMNISlccs
z)BH#E$5|G}iA@AMN&%oAC2>U_67C(*Gauo#3QZ?MhLB4k?h^wyOY-w$7rojN8y7cO
zLlBXgT24W~f22+4aO}=Y=Cvty0ueMKI>+A8Tmz2fgNsfg8Al!3=>6n0A&0Pp*LVU-
zCRy{YPG5bVT9qLFx-L42OdqGaK5aGPe;Rt!Cnvlt9PP`1z(U|}Qx~>mox;XV=3U4M
zmh9nq&HTgkRhf!hPNFD3lk670G#52EMux?SZo-lhCm0bsV_b+8v^ZxV`ozqIea_><
z&V|KbDmB7Y>MD%Hva95K1FDT=KhI9Qj0a7KuD#CB0bZ144V1_S*gjclX?5gTgX-yd
z+X~#{2~8L<%v7ZNLlD(A2fg1JVp8tZy+Ko-%0O@oGx5e9z@Kp1bp*Wdsoy#w+QSk0
z5E>(-XkzGEIUi*hj?d~#Yt2d_YXaiB8mZB?=XH<Bb73H7iXJh&d?X<A&uk5()(h4-
zP1ooMYU3QdPVVoeM@$ib1PsxtibVQ)&bUvf9aLi&X8pLrE~)Jb2hME5gP(lm){8IA
zM26qoUp`GExMe@Qt^^TIR0EbTSkOW2_G{;ixAAizVL~h$q`V9>Z3T6NL7Jn%mLAUd
zZN(t?(h0f*rieZ)+d6E7K(GpEAQfd1h$N`nLPrJ~5$>rvgyO&+K7ISNyu2oqnK(V&
z-q|G4HWPkLp92JB=+jZ;E$taTI!vO)ik%DS-m5;``v_{=32|2xd#P4*;8Y=n#wT=y
z7$$K<-{2vc#bJ*1<+bR|I}qB?zPr(V&cqhn`XsmpYpz_RYV5d*uE~5raqv6+ymR?a
g@Rk2*3Y9)SyQ@gO@J7RxH}p+;=C)f!TlO9Q7l}2LC;$Ke

diff --git a/reference/plot-14.png b/reference/plot-14.png
deleted file mode 100644
index 43ac4cc7cb82fcd42f1e088b097349204c5e263b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 26842
zcmeIb2UL{lvM!7|;uw%UIwnw<QGy8>K}51+!Uh2)N@!3?N)wcvK^=7rfHne>F_4ph
z4U&x@IcI24a%^%FyKhy4bLQOtp7oz|?mp*V_g~kwW;27`UwGeotDbu5soM9?E6a1N
z<66hY#>PQ7drFm!ZM7pC+o~%+vcqp${GX4)5B4jH@~7Y@{FyzO?Eyd5UO%gC!^XDj
z9{#h!r+dtpjqMjU!l~cY9YcoNoSqUjGQ|fBH25|8xz|>2e)Q*`?5{R=Rc`<3A5W(%
z|Ga+Od3C1zKmPIgu<hO-U;O#Yk1J2G=l-xqp--AO!r;-RQ_T<8{Nuu@YbVai&neX9
zd2~gcJ$I7a92}}W@3futNW-NpvtE~Ap*f(HMy*qiRo}k?A1vZn+0-HY7dFqjJsqp@
zU-mxr+=f3R&ac$RpPnbyI<h{yj{l$b0rh0(ZSTgF!lW~){G%JWk6b;kqpR!7rxvZO
zBrh)?df=9|licr9j_}V09_eGbE_7w`xj+&A(9lr3q#ujB7i$}JbaX^)`zoCSR*-%;
z@K1lf|Ar;_FT6!vJaskOr0pKM#B(bDaiX*xy>99dY+*!U+O7Zct^F^2uYaEX{h#@e
zQ*jO}>O`(pziM@(x6O9)GM^EA1&oI&4dosSg&syeGSgqO+ZRW=%Y%<v=oEPG3cknd
zehki3SyZrPkj!-Bp6uatT{t*#vYe~~t`oH1Pa%?e+EzYNwWz{_F`?(NI90Ms)PAUO
za)4s%zgQ|-(KVD-p=@f}t6F%p&9)}QgE4!JM2c_D(GYVQ@3IWPow|YR$lk=nL;^Wh
zEjsGu%a`R_E^z#7%e%S%hR3ae9ny~1o#`W?e1Rgy#UlC9Y?|J5HAB+1YekjMPYx2<
z6MlMN@LtNK^jWK8r&oVnyjI?1q_FY)aE3~_tXt}(_qTEu#!GgHTegby=Fav6&kkti
z2bGtX?{pn~FELqpvZlV?zd_$csC~BFVSHhtqIZql%<p+~b;+8>JovK5DO=iZJB)V~
zWsVha`|`;##(dHSld_|%+VifNnJLhyxuUK!qaoJC`%3h3=i?0)_wU-ZtI0TkXkB|j
zJ`$b*K2@AGm=wJJ(A!^xgoK>>p5Mc*h$~%%mybz*p*8BgLsWoZUHSVO<2JhnV!xB0
z0>4A+4S8}`<w;KBfMdS<Vk({LV%qxl#tu1mSH^@KGfj7bjqQ@s%RB$@Qds{hn{&_=
ztx?VWeytwNMFJUi%+Rk#Irym1L{FI_+5GL-yCpl*K5mraBQQqq$=SLry_nF?*|!|#
zCi?^BJlx4faOX0c3Ymbzh6P>s1;frV8H;Ikn!3jAZuByj_IYY#PaZ4)fqePn-N?vD
z<G@}6tK2CvpK#9TiHu&Bo3A~QvV!*JS`GQ&23=Y{lc8tdWK1mVDGv=^!{hi1r}XIY
zjgoy4t|K<#x>rUX?GPpBY$-UT*q-O+%Qw+mp#uIJOia^-I<Dt3Oqp$IXQY=J@-Ftt
zr_GOK3p*T=srk0}VxCS5SBO+CG-Jy$?Ra-{>JZw<O6N`V=rF@ETE?0U+dCMwmloVU
z>OSCb+(uD&CSt5`D0?V%=k<3hB)T4-6uI=yvou&-IDc{2%4W7Zz;viFrO7NlF?Fh5
z*P+b9%1VX&hC&gdFFb7DD=8&aQ(c|fJmKH_^&W5eOCQ;+Qp1z?HkQfG52dwcPP&5i
zf_se6)wIojr8lB-yuUTvS%4m0P}9_;P1|(I>w-Ba_4NB|rlzL$jhFB3bQwz7aqRk~
zprD}YXqE6GN`pS}fJ1WbbTfa?A)Py^d-?SJy}iAY7JGVnEX$UR;|^@=My?|<B#->X
zY2BV@7haw*p6sh~8vDGhhDr^i=DG+H*g1JZ_?)Kdv;tfh5;i@2+S>;Xt@G7#ppJ*w
zluHDPk?KW!atDdD0kf08clNdCdo(ok2V3Q3e&XD~rqC%{Z*$PyBj~W<-7TWWtKbE^
zSFB>U75Wq;>y~BEWc6EFS(&2I$;vOdoOcq7KKy2@Jx8atQ{+4t3rSYz>UBtod_{xF
z`FX9fneZ+;WTQ`i{zObp9Lee4A?rG`&POFode)-i_lndRxwBgjUi_tF^@DUcl+FRi
z{ouv}eSOzGRzB!q)Y>e}FBWnB!d=oL6nFcKsyXM)^z?M|*YDX0{>;sdO=+gWhuxq2
zdh)LARId|kT!4hbw@>#8<XsM}DJ^go;E&n!Mo)19+8AHa+Z&z!POb|xV+692c$@A0
zXgP5)F|=hRGFYBTtdmVoX%W$(&5y6isW&uRX34K@eX!Bdp~WI&!}i~P*(`AGi<daH
zv5uLZn=8X9J98;1Z*G9ED5a1myS8=J#$&%IDJjjCS-Jb_T4wU20jIte`XJF8;=6kN
zRwHnjC2;6<{hU~jR#&gSFK7L1iEw^Mjf$Mxeftd$rN+PPu>oIjXrfU~MaB=d$~3U>
zE+=mo&5h>LUlXlT*ORzk#*w=s;`iHG&7S8%B{h6L@?7bX6|BwvA?9K9NM06||2+SQ
zij434X0PJK*xv2nC>zr3qs|4sSYvhj+TR_d+n46uuu1Mo*&|u4;9_QntvrGQ^qu#i
zIj6T+1~1av<+v^0R9W|W@01W}AUY6PVSvYZuUgc(B32lnw~A)d^5(6%?UP5O@vS_~
z_PejPQZZR=)tPc4&NTJJgEPi=)GNL+YDF{Us$#&FHXd-~4&IV8>i@!JbR~Y}h{y_I
z{CWKO4>|ag?arp9zxY|<-Fp1*^LzjA2mbFmu(AFB-5v%sXe%$~froPrcqf&RIhz=s
z`6fL*DlIL|yN=D)XZ|7_qaP;u;B*(}w}%7T+7<sbw7#}h9plQc9mLxlJCD7F^Dn<R
zPfvDO%hvWywjNI1ibxxefk5^R+t1TBIo_MSDVlToT{gQ*r<eEuaV72b%~flnC=@Y*
zh(+@oTU*=O%-x=Ky$}HqiY?_zp^|mt>Z~}h5dv9#bu}4WZf*94i+g@_Y`El}`zl3a
z{}zK)$Ez0P5_)Io`HCJH$BW?KO~)72I>+c7R%4IF(0>0ba=-sh?43J6WFtjAnGb=5
zj5rjQH9z1SfbaFXI^my|%l^~fr|&s{Q;%1$!+=bj#)||8i5tZ#F{i^nlnb%y)l{lJ
zO(f>m?}@MRYL+{mJpb_?r%I%1tiz7l$6Nl_o|*QzkjLd9T=<%WMOi*$Dw(fnqCzfQ
z*kiV2$Kfldz~&x9JQd5JI+Fl61w4~-o14^g9sAtmK0jQO@4+<Bo9&sc&&cx=Z+ErH
zY?v>UV^rDY3WmE{-=UNRiV7MQJSqa16{d8wiO<5qLK`xQGyuK5sU?R7;{wDU@~|Fj
zbR!TD*Q~5)&H4RN!PZ?*<PRM-fiRD__>2!vD8#n|2(sbj5%3y~v}O`J!`!BuTsyhj
zzeEP;2StZD?;lLepsq{K*qskB-+JcjJ=5u-CdC{^@qy$nKP~;DTsk!vVzQX1Y56&T
z!CJj)(Px&n-Ba@IKwP~4I)`lzfF-&g97y_dAUTKZ<39OP@sI3>c+gTIMkd+s3NUbY
znr`=CMt&d`1LlGC_mrHO&d3i1{A>(BK2XBO1lB3LneT|W@T1|pald3Ez<4D?O=&h&
zr?+)aH5ecwZf<FrXf*UOh6^i`7y4B^ibX4=(~89wAb^)_%b)q2<O9I^o%GnJ&4Hq(
z%5cRr%y<B)J9XS#Gagvs8YLNiOe69%3s2Vm4ynXABxgp6l==bZM95AgW;$+9yyRWn
z8y#Y;U{VnlD$nD@M{oD=kOf<E@FH1Smfw?G__{FHkYC;tY!$|b*Ug&Ns&6t4Z=*18
z^Dg`n9ctgasSxtR@g#@b124~b6~|n7so7ies^G6h>$tVv|3`;y8BE5!!!YeljWVuQ
zi+Ng^$kj?fYk-Mz6&k|Dt@JGcIPn$X9bNks4Rj)B6Lu^ih4X7z;KN6VXMqRMbjhxU
z+S;d_Zha>SWXNR`tEUXyXS<UYYm}vg0GGud2^YJ)#6}HVf;^!_COu}26?p^qQS&>!
zkcTRgZSC!o;7gEcBrR%K48IFHxdcG$UZ)-~AyHD(Rby)xxN5_eF2YqQIhasGVM0z@
zpRy@y`Li5XZ*v^&YtHW?EBaroQj}PF(n`<f=4R_Jzuw~`@HllnS?Cg!i?unOuyG04
z9EpKblTTu<H<&r$eeBzx3p6If3U=A|;82YeoPl%Ce*3j>mxG>3S&&Aus`g*sqUX_b
zGy7gV%#ww36rA>Xt1g~%c-3pLLmlsLr%jYg^){1|6AdE|`Yb<mV}Epb7@||~?MV*%
z64;j?;cbO$kEj!vlr>NQCTMl9FMq5CUk|p~y*Yg5)S;}t7ieLMVEu!{{mdw>ToyX>
zJd;v(wB?m*Y|il8`uci5t^hu?K5HKiURl@33eoEw?{?ah9+v2NW~gZ<O=AD<X8ynA
zUy%Di_gCqw6sTLo%&XF-iD))O)JHwiYniJQ3td_(v%WjmLNg~-CI*QG`&-Pf+MG#h
z)1-+waxYPtF%*T|^%2oaVD&hM23dXGIY-5W%%7GR&Itegg#D}UM)KFUdM=x*nxDwW
z>XnL?r<N~}vZ-a)(~CBHmM@X$_sq}#qRp!1{}j^LPl!yTY9!o^0)A8BA1qLO+{^Fn
z_ViCd+)tLDvFv1h%uh>uSgZ47h1E?l@d%!6qspO@VrVje63`2CTfV(~r!!h9*gwUv
zum>~O+E}$T@RlaX^S}Y~hd*%@=nx`O)EP`>vD%fg_$bxbF33nrKHc9eYT5es(JrZ@
zq9&zZ!dylIZ*&$UPC79d^!#3Qc#v9+)AN}!+*Ua^Hp<R)?DSX|gH+FvL+&=uRwN<x
z9KYVtetqe8@~f9G4exY@JM{#HfCug$a*Z#Vf}<W99qumKS>RPN+Z!GZ5Gan7>_)86
zQ0*~m+(gnp?a*eYHQnaW9;t8eR5&kGA(D*TiNSnxgTLcqplMi9W^Y(9a24W6Z;(hN
zjRB&6Pf%#)#sVMSaJ={$NcUU3-6qP!C+d>j#en@qH)jKxU3WnKf^kyTK%9AMk$@9p
zws!(NScDlXfxLRjPa#m3vmx7owiC{8=GVL2;~&-w8v#WSe}pO3KC7HJO-YV~V}~IO
z2&X4NBbA0qiL1!Y8Yv(9`UjT=!=GpKXnBZ)C^#_AiTf(9yCLZ>Z1HBoWB5UGKLpWC
zVj$)v5&%aaAw|+<^)|i7up`pIMFk$YrmE3{=2{uE{|=*J0?wy?=5Tro2D%;@nv)9q
z#+`4gZ48|IBV88<4OGYrvlSlY7&Z~w#m4S~$hkwp+6dn;D2b}%eE!}1M_$Hn-~h^j
z)Oi*a6}81XQGI`Ar?(WD?;ekHpK9U!)L;NaJgK4&8t>ae*TBt>-;V$Jt;YLjhYiOU
zX&!RIxs$4P%5T{vm<toWe5Rpxn$&qJy*|;K=ciHGxxSZy$8KCHm!2q{?62vyC{+1>
zp`E;vsMjpqRH?xF^*|&lE#S5a;dJd-l{wu^nt=bXozI!7Z)yqzj_nY^y7fk39mW6K
zDr%-pZx3*kmg!XILGz3@`vYR8RokkT{8nn|VTkDZ<gPp0^!uh+xhdC0<F8hC1K!#-
z+$h`?057^1(2F@^JRra8ZKpT-yHvPA4@A(@;Atrez1clMr0VA8tl}1Zm*KkM3ga8e
z`3z?yulNw&2q=d<!<>H=`-U%fxh<ao86@c=J9kwhIa4~9`Xs)#5`aS>KICMLib1~w
zoSn1dn%lDvy3J23kxRqn<bV}~lw5n*sRU_K<RnX^(}7@1w7tZkuEKJ(D<`*80q}og
zhV{2(SZOexbDkH*4B#6*+I?{@Q!eEU$>LhEGB+jh`Z*VQevgHQ9EX?}MvFIdQ;RCM
z$lpFWXX{~cRZ&sV9+qtg-ZS&tg99R$Km4J0<MS3^J<e%8thvo8dkjIBJ=Ljqa-_}W
zPVeB*&^Z^E`TY$kk~W&?0~Aw7GZy51cyd2|tcAt#1L8oo!N9|<9Acm=J-a<|fWjnJ
zt)KxC@kM42kpqxWIn@(e7w@D~gqU8G8Lt;<X`%qc;$^IyM!>qd1!lt+hfLE-4rL#7
zICP%-q5IiYw&8TOiZ>#|ui5I2STu%OXaCSpg2USQR}PA)6Qf%^Ld(OL<?K1!8_b+q
zf_e_cjBqMLxkuEjHs%qR*nXsNfVT)ojxcj^y2S=K-)ILKRbu3Ar}e^Ep(z-Jvdu*G
ziU~CR%_`C<adwH`z$6J0^qso>Lau*UE3r6P1I$aWQD5crKp7W%%dGyW5Zh`(=IDp@
zNIRNV$j}qCg+8<A3=a=W&~he3fa}#@tQx2zeT~kI2_H#?G|F~Zc<zc$8;j{Vk`y>0
zHPObDiliDJy!=hB*9iVexL!Jck782mVm5I*CC_cXrzV_XE4$dQVp{NMM<(3H2Kc;Q
zK%b>(L_i200N`t-ltkossLADEYiq<@q*t8YMy$0<@L)0|0PBeYxDK(a*OusfBs$*Z
zt5Tofe`OUFSUm|Kjw;|7;7QjA%DzgW#V6-wwH~(VIk>Qr2IP3d{?@wC4C!#@La!)f
zfyO*7S6_N%>6~Dr?W9S?@X(^eUK-@P6Xp8C0HEGDV8XVw7r41eT6_NB{0R6xwTE}v
zYg$`FfF-QWe&nYW+UqxU5we0EB1UKH;pikaY&AHQDZq8q{9*$m0H}ClZwX|&>a|U~
zW23{0bDSnL8@EJA0iVbVzLBq}NH9F7Hsy4?Z7H}-FsDo8u5S;8%LRhX$$Z+RL<;$!
z`Q4lx+02J?S08OZ{9=%TkB`vtJGHn%y<uS%;`aB`lRw=|rREhd4Y@C!Mm`f3+}_QA
zN~MmhdCbu>ytOVdP~n^-Nz2B?#YN!Uv&Z3p&@zGk^*y*@#7A-<2Ke>x3VGS_C&uOI
z6#N6%<b?w7+MPE)GN@VhP*?hqKIA3=5+-0X#Cb5npGFT=w7A_)N$5OzcLM?9QBOs9
zIN*N?B-B(5(RJ`Ofg4(489mQv?z8c8@*|D(^?l)G&J-?8M$44}C)eQD;x_IZUh2W5
zdzjY7T(IK4m9&pjd$1wdfXZahdjW8l!Da>l0yB?24Et<w#WnF~?op)l^+r;V&BK!j
z>})t-KI7O<0EbPGrc6wPGo>-R0yLQSX$&ahE}#Jw$U`|3<!xb0{+?9nnE?k@MSAQM
zW>N7;S^I5z%azFH0LN#+X@;n{4XEuna<x(++)Bymiw-#ig1Q7_rqc#09mYr(4$Qpp
zD{8UKwtxdl(v<gcUDx;`Ea$*LnajHsYhFQQ$_K90x<n|$b}^{DtH_si>fYa4-2{Ln
zbtG$Wx!U5DpxROI)}Uvjl-^_#tOs#U8_EpUK=}67X&IF5@sTEf<dk#u`_Lv<_LKb1
zC(lpbgFt+E7SfY2uuf5eh!r)b_4-E+Vy@FR4>nz_ab#p8MUHr(H^fHSGPB!1Qgd!2
zudFnNvbv>k`)-}yJ6*4!H5+$6vvM2MmH77O9?gM34cZ&+G!vn!0}dBj{)BUCbzdVx
zy!HCYjZ#Ce`Fp1OYXm)JKJuov*|&7VVwM1toKZUqE~9k_I90JY_vd#x9DR+tzC6i<
zHxvQ4&kFKTlE9MK5TKYQsK*7gymql;RZuED_xq{u@0R)THJ2E#`^;C3m{;lk4dD)B
z4FU9D3hB8tx7l7fW+3{|j8<z!#zL>0Y4X>Ei{G^Xh$g5mv`|cfbj?#vE1Bu*sYQO-
zSmAXU>Q{Ez(_h?_V_GQB!;sI23$K~Ke@5dWee={|NWzlp)<eP1t>s>;Dmz_}%Wqz-
z;;`dr2ZRVI0@U4}cHb4AJO>V4F(FH@X7MX;nhkZ8?*M{j*Wq-%nb@fTBD>Yg=UOb}
z`cCS_7cqZ8!;XKU;iof-ldJt(A20vHhCd3?^7p@2^n04Q6QS<oWvF}ga-OdKMV%{U
zUNM=={KaeM<#aMA?p3LK`rp1yYTk8z&oa)AR0v!&{~I!odUG|xe)*;(ev`jLbVEM{
z(_F=ib~;CX{+*Tlwf$xtz8Q$J+RthZ21ME>&mUR(fQa<{ej-{Q{|47t3BU%$13{x(
zm%hm+rgeq&wDL<?q?m@;!@Gdwi(?eu&4WAZDVL&^;^BR=M;;R{ta=SzKoJOAt=<zM
zu62+~XvlPgLI}dRAV`p%e9qZjttd?VHUmY7XUw@|rp`CY@3+~z#|n8XT&NRaW&x~}
zzS0oB=Z3j{$8Amlf;Z$|{N74IN3O+Q7bhS_h2(r6A~@%%YBA*CJ$a12s?`q=KgS(N
z>0z(c8{P~^Ae1-{%*o)U!lcZeAig35=2*pm;vk++GGB7uoY@H2>+(%Wo(j)(>2~(^
ztoK^6X6Iw@pTYoj%GSt^-VAY{>qlrEZ|!wPbIKFzG{pJAgtYQag@^KHzCkRJ!;^<X
z3^cq-aRO%j%a^v(Bt)uqXx{O3pyoI=UB0&~wcUN5h1&Jqrs@!FS4dBs!|UWgp`^se
zW1(~Z!VdAIxHUOgifnM#)zxiaKyLD+$XBrUnV~nqW<oc>jXB@mPJ#bn<~4WYsvlp}
z*uUDcjrj)MQqL24o+2o-SVNpO1;z+aNa-tr@Ay=*hYSG<yU_=q>`OI^;~6avm2wD7
zX2S24$gbe4^}bEXy3N@RGXW+i+e@%69C}bE)a4_Gco&q}9f1j`2A%>yl~g-pIwM;r
z$D3O`Da?3l;=l?TJGXfyMUGUTkhF;Pc0wv7^#>2<rCv<yfsNM4-ig<XiN7PnDfzHP
zwSQ>3R|ecTUlI70?9#&kb_~cy#ZULOWohn-cYc=*beHY09KF?sI1d#sHL}r5ucJCw
z7QT5n!RqQ#N?2L&Km@qr@VVM(G9NhmPa=h_;1>yGAGaYxn!2`(a@{H_AjG#MCZgnr
z!g-Jb6d1t<vXYi~tz`ZJVV$O6zI(guFD_ltb1vRRuC1;$tN+ErLkc!FHr^0I(aix#
zr$Rm#E-M4V0tXvhS$GU^T|?EwB(XSsD><XOA)s`e08oYkgTu0G2>R7_LkA3Ii&Z;u
zV+6DtAJX5g;!Idt+w)pVN`(CSEy-yK>g1*gDeChFWU5fvwc)xq9}Xu9wHa{FL6Ry5
z&_>RtV27y@5K}te8Xy09(z=L0pbHrnu(>R^#4{RGo*9tFA%H*Q#pRd3&@D_LL)rBa
zkBoCHBfGQErvf95(dD50nN<zA4H1Idt24ALWiJPB@eVwiEhBF*-zcSiA+_QQ#b&6S
zJNbI4qFRr7w@!uZyg3gf#F6^pAf-V5=N#liq~u(CjikCi!Gr|X^S+q|8mY&CIj4nu
z8ys)Pz^@jdRT!6JPRJ3+fHmq~{JgaHv1!c{%vR|X0Fz$>d~$;fWTyw#6`H3o-sxG-
z+{?0KMQXuNTU9I^wgXClplJbVOb`?hdm))B1;`bIG(fV)A_*W^s3kG}LLFa+($=FF
zs1PN!j+uZRCD(PMaX`VEK%Z~QtI5)xNYGBRt|kcb6A&SB9Bqsk#C071EbM+5fI%4~
z;en#%$M#?_%t&g8iDN9${vGYL^OvvLIeuxznx@fZUh3zMZJ&2J_dO?&k%;o!FI}7X
zlZM-BnmFvQt?fbcAE6NMk!=FvHdy2hZJdgOFM`D?ph92tp?%UI>~!H)<J$>YaJo^S
z0rYj4QTNB@x<Hoi>)!iY$8gk7d(Q9?!T^utKn7Vdl;Q$!BuF^py)*N}FVDJbUa;)m
z?CRx=+X_!~9n9ZUooR9&O!VcGqUG4|xJ_0ay!iSTf(YO^77#<qj1Q`~Hd*w0t?$gH
zA5%0^u1;Bk%wte*Ek0!J@lK%*B{H6yCcrZjAQ>?QVu9H6tipXdjhO4GZD$916uW`i
zq(uzbb_qY=*UX-)j@HY77gR=$-Uy2Lv!4WIv7r4bGgE@*RN-O`OaY6ofr9lbnUmtS
zq}4R!3Tst|KXdPv1I@uTOMUyht&kf-HYT%8-|=z-$z>$d)S(z((Nj#-p=DR?Ns5^z
ziBwW0&l`$@h5fyTgtm2#8I4ySCQYv^Nd@jOxq4PvqWkH{Q`Q=iqE@b=p0PM^At)>3
z>e;RA@AK!|XsN&!)n_l+;7#WYofwse>}JWj2ON3PwEfMj07HS3JV!>}8ME4+R>(Id
z2OER|uxkQG(HJU%FMY{NV1J}$R@;lO?UXlfY**(XFm+I772CVx)yWWl<1gAFYnHWT
z^bq=+i^FOdq>eW!=zY+=QpOPjeiC*|Yx&n-70xL_a352bVNvRATeY+YBsBT=CVwkX
zPxH)p`En1s4G|!)q=A@#Bu`6oVEruAC=?SFSi#RhpMU-QsDBhdjq^$?sl8z?hj`~3
z3&$b(s{J;J&&>9UeFY7uj5NU6M)e8W#AFL_Tut^Z7N8W;9A>%EcU*~Nh5dfz&H*XK
z3QM6o7ZUpH4{czKMj!rIs|^%ZyxR`E++O+nZ5!c3D<DoUHdVOSQ#`;x*Rx2;<2njn
z@G5Q)TC=9*y?1I4*Wjic*WPa|o1E~D%&lreQ33T1xBYO&n>Xt7tOq>*Oo3c}LcHEq
zgVPaY2^~p`0DRM$qY=q_j;pV0yVwc0l9}!D&7P)#hcydl$$cKE?HiR_YqbM>kR_af
zB?d?^%<&3lGiC;?@_eys?9<GYg7~?U^_U^McJ1LU0&oxjU|kUDKH#8f#`HFa3Lu<o
zYHQzom_^=&qR3j>x!6~wuiwB!Uw|O;_2a$h+}zy3PNYO?)l`W6vwoZM8<%6yz!Hf$
z>aVarR$US)*Z5aCL98#X-7KI!{}dPQeEKGjYq@8{@?mqDjC_?;W7Sc(iKm>BdH!|`
zggi+4wU$48Y^_!5Yned+4F0ndy@ExX67{uZzDrb{-zf0=?amlrUT6hl!dg)txO^>U
z7ss;fKul|&ZG0F~Qa0i9dfp(Q+6e+I4H<_$o0q8A#5$?yM31~+<b8l+%bcuMgauG<
z^A&#`M_;l<fAxK@*I0>>@*sJ)=?!%-fhtcw$ZhJ%6**Q?fqpm)6i~bjGg@x(+Ln0M
zk#&xLCBR`pN>mW52ogI0EY;O%@GGtZYT<Fk5>AS+j#V$^9R~XymcXrZ$fi}){fVr-
z6$RXYf;SBbUd1?A3Y2O=I#nV=1^<A_r>~&Oh_l5*q4cwmkx~>{_simLO%`f!z2O5v
z?68Zza{e?k@8lMK7k5X;jcrb8@LLfZ%Z%g+z)38|#~m<^5pa+r)3X<K;Av3&c0kOs
z`V|2jh?+Inu_<J9r5xg|kM(ok{D2bX2-83b2lu2!IVNpgp{=c^YOF2Z!eeeQezuoe
zRP&h0Z3??dFK0r2bHg(z_-usw!6T5I6oP9p3%|D|UiZb^h(o><P<*<$7`S19hYimZ
zcyafxbw<0G+`C><a$@qK74Q%I=fY$KeWNS1zdnny6^@=S5YKyp(ndIfP!}1g&%BtI
z#V9%|Ce}?>JVYR){4o;>C4<@<@k+I4BP%VYfM|9jP|Yf`W~SeBy8lKHLD$_7<~m+v
zNld;_CnMQzo|fV42G}6W{SJQZxR?5Vzi;VgGTW1KbKPPS)6#5WI}^gP#B+%`v=NB3
zgOZx8<3y=ktb5Jx41V3;We8-FHnKDCIZeJi{xwk2Zb0gyMui&P(_y4FBdkCxwpxW%
zrM=`;^=x)zozOthpp#yk7X1BKUbfWm8?{{rSUae*@9<Zll%4`Lz|~`n9S68X;b|8j
zLoX!=hSpBaeVfN8tW<kC6Fdlh6%hxdsiu^m-ONId1$h)lEd%W-%ve4KA(g$t+V4n1
zu4hjVt4Ip?ud&XB6&yS>w5GP?Xc@TKNlHp;9*c^M<VV2}ax+NT_P1p_P)hKIo?<5=
zRSA*f(-uk%M>HG#pM%?m5JddTo0lqp+eR12dP@fJ{<UlO#l(5g4;+c0WXxR6Rb=yI
zSKXJ|_WS8~x_>wv>;RTJP|CrQ{Ax);=^TBC1Mg&~xs+%{XZj0<14YH+@*$%_;pZGM
z*+7&=LO{icoRl+OoC5)ma&XD*#H)@La3!0ppn+!W2|{uNP6*L#?GMeqdM5{o*X)OF
zd5^IQW-2$Yomjl|YR7YZb^0w;G<>&mpehqBmS4ZcG~6u}LIhG9K5`64$exm*rf9-g
zn4K*(mG$M57<&ED?81_rT?mwMgdz(ont}Yx8f0JgZVQ<eZinRNs(etB$X`$vCL?7U
za^p<W+iwr}ijdntA`lp*SbG9K$9V((r2<nN^~G1H#3}=QMK*#ebQ!!#w))#(zQwsI
zRz8$emkpi<<*M;S*8PwB0N|w~*WCd4n7Ig*hs5g(P<Mdz14xPQnU@BEA2MPmUCDNa
z@5>Re9OaN>DI^cj^I1u`V=~dBhY!N0nJ^&Z*jo|aPHC^M9#8KK4sI`g)>|H`J8X@A
z9g(+xUs|}uO@8qgamgKRR~`ywO$(}oEu>>js-7Ze-Sy@B9kpWBAGBVa0JCTVh@3T@
z4B2PKuKbzXbz8$(1xFeA$Vx9UBDfwyADMI#lVmWjgL!DT_bzba$}&^6#^q>Ho;)w9
z5{a%Eh2RRMu6+74#J>3&k(ZWMAv-%$C>I6^lMsS^9$*|>L66YnLd!)Fuf3NPgvm4j
z&EK=FH0S6yo2X$41Dx85WK`lgElxFLHw#Nc{m5zb{h9zN#^aM4A)U}ueF(i9OK+RX
zdfQh)KjZ*Gf@LFw&fNg=%l}xr*<WG=%IjClAuT~xh_d&kBdZ*{w(@!oEy6hm7|NcW
z4EKIWuJjdMH=O693&w;w=Y<z9(9bTNk}$aNLuBh_+i30qQ0DLH4lpQx#q`OQ;!I0B
zSh{ARluQs=`^V#dsQ#$C^^f%*1y(;L{%}NW_tsCTb*Z0pbQND}Tr=~i*O`oapdfVU
z>IGG8{<D1NV@~{7thw^Rt$inc-aEhPY-yN8FI_(9@UOl__R8fs67({=q`{Kebkl;I
z@*-bx8vVoiSj~8>g<$Cbx+~s>We8ItDd5_*BFOv38j7K`MJSSXXXM!$54y|#K77Pl
zT4TT8-8L6rv1?j$D6Iy~#|=xUU$c-lY)&x1rLMl0KIM>YTZ2SaO!CUufX--=2ry4b
zkJUG}orM}7$@Yh<k_EeJy~o~6p~#WGArutbiKTMqS1j<#g-6(PNrVCIxPE?pu-S9P
zBj)-&NQUl__tqO+gj$CPYB3JwFN~iHId-@vN=eaWerxQ$423)!w~Ja@S}*q;uQ@yt
zVgD#Pv2cKQ=z!k?rp3({v&`ROc6DAe^wS!(wCsWUTnGdNiHTCti74^heS6M4It%A-
ztFV3u)K$8m=2O85RG{lzO?WxP_kJ#8OIf>pj@yLwTXSWgDju|pWzL^tqe=z+ElhgQ
z#Xzy1IJgKUMAVL>$Q@HH4@R#?x!l4>xo4lV@0;cAX4OmKGvHc!fMj4g`!FBo&paII
z40rr?U#E~26Yj<_+Bb1T<_^g8&sbQw=#u@I3PWHbNwl)`D`LWqR)8Q!hS6aU$}m6^
z+)oXI+%&Nh&6mxsa-hP#Z)nK;5;?Ulry~?XEq$6XPs^uv0?=z884kA0imktaAC2IQ
z**X^&juIO}Dr|sputj?8U~amJm_g&W_iK2EsMN>NJCYc3RLq2yAFklQ4Gya0)|`ji
zgv(jE)hrzd?yR=x9kp;b004A*(%_unkAQC{9a#-U<{Xmq^pbnZk~s~ATZM^XRuyke
z*YhiE+@4{xky|dS!LYf7#PDJI@)3zdiT+o<<CslDk@Grrp+~%-Tej>)&uZGAKkdP~
zhycO=5HSd96_d^j<J~3M85!UTzAu|@-ivtpA~BbiOi5C(%MH{F4$u|_={?d0$P(j@
z&sGQIsm})=2@KkngZRwh{XE!fQc8iP3a0rZ!x=f<LKJUSwB(>%ElPFmyc>R0vVe=n
zd8l7{vau$YC#}6P)Q0-0xhx4n8d{J|nOH*?k>;$0v5P}zg(tOcL24$9o)hgdAI|6x
zuXkC13Ql9L+qe6I^^z<PengL=BQ&mYAilWmK+@aYs7;27K>!rbgNXToxNlH{rKfH{
z3II#qcA@z@4&_c(gKAieP@Qa$k2bdam~O`@l?^?*xvetD-vd+iDmS-h{rV$`@QP3=
z$C_c5$D)?Ulom@H1Sza=nrTJ(Do*Lgpmh?#RS-}Ls&#9H9tVGmjJEV8>+oa;Wfs<>
ztUqtj>H^JD8890$t%)gFuHq_^{jXeSOAN}9SmetD6%g_>2@^9OvjHAhm(cgwixyUS
zsqWR?l3nJ9g9Dv3GF>2j>p|~OCf*(balLNu&iGo3EuHy{!TfTBSz`5FZn^ZFUYXf&
z=4|-Fa67{y_X=L;_(_fUJ4x@e0~L>89+(LkgfCy3Nm((FeAQJ|Rh@&t_{DzC$l6zD
zCp+1gVrT=@xp5Jh-)XKtdIIbzN+;0-XnSnq$Z8m1MVLh0o1%`S7}Z#iza-dr2=VsS
zo2P%dwGAu6z><jo*D7>F3h&-`GRWrsU5MU_H?mh>tlEyh8%X#PmiDPs>Fj%!Be2uD
zqN{VIr{^UL4|!%CNI58U*k<CwQnDeVp-8-)QMID+AM03+bTOifho)Zq_0dCr`)KYK
zi8~qZ{^h#ri}_3iwJSX)ua-XU92Zquw^5<A^^c!f`nmH~{D&OQU+Vh)_i$&Szg>=~
zp(b>l-Rgc|@CDLKmQl04zE=7#mlAUnsyhF}l4!7&c;EOMe9~vBH4NTx%6$E!)<?@X
z(y9{)xVrbS&F1m%iqOF2cUMpJ&pdS!zWHSNn@OKi-+U@9lYhVdylLo1c`->u&m0<k
zTsc3#=VCqfaiV&BoiuQpfwFGSV0}+9zl3a`=GrU8y~`s0GCNDhSgU=1u&I9c1Yj(Y
z8{a-rs0JxlKARW>`+a|fc+C=5izu)qNMDqTKo=tl_!}6yME3S`nx;z^IloOX<oXHJ
zbBq|s!4cs&mu9EBv@t{tCQTz~lc+fF@)D>~oEkbJ8Go&<f8^>jmS>iC?WOzh@a#u%
z@BXQ0TMq1Zx?XOTe*T3~*XVF+$(PHyJfdcWRC<m0-lZCYH#^7BmoCxvroqF>i<TE;
zNBgZTpPwqa=|X!V+U{hOTv!^Me|>%AWt;Dx*C)Qd{?I>z`$|DVQ{Z8eeRAQ>*K{5U
zhR{5G&wT&)2NZ*IpnIuDptiXW?U0SloJ(~1j+)9HLf4O)-+c1dM#Ke)v{e7?^>ptO
zafcr-?^A4$dc5D6=*Fvc=I%Xr)^EcKXj}fsz<>g|6{xWOzP>)_=%|gAn5tD1pLZ`?
zHul;waJ>!l?vbkcE2vt_&z}@Wc*3dOvQT=Pon!yR3Y*o-wjZjoUavAe=TjMB$y%l~
zStmsR?b%b{$kk6tUfyhMN>QKEmrVxWX-Zuw@LxS?OpsYz>=DU=GY1aah%RqluHP~4
z^uHa!XrV~izijoD&XxDN|8gKDYiO#%%iEZ}>0hmhnwYWjS2!}8moMAqchTSfS?28W
zdzCAMY=8^>b-v;X%r6d{{{AjqwvDj7;q_;btO0(k=Gk9oLZEJ?=_p)i$3Djuo>v})
zU)Kj<a&qBYlp^=86@aKBVITkH0i6%r_*XlZ@F@w`<IBl<ylw<l^Ux<<YCJ(2Bs;Tx
z@^aAMQ(#mX@jI&&k8Mw%A8ZM>&4r4TB%}(td5EcljKpuVnmh-JxpLf^Aoz#w@jkX2
zqHG(j4Ok#doKdJ?3Zxj2f2}bj(0Uy-$b&F#HuVpwWvdE{Tb-Y7ot!hSfQ0~CJ3E%3
z5E_3TgNzH3d}%DBBJU0A5gXrRCMDTTsLqU1*bv;l&=cg!VX}16LhH=Zd;HZcZr!Dq
z7%;5;g$=@3_0J(wxt2vnU)4+GKOn=}7zj`%46qPlYi_s2`ue$wWU=M<NUBo1QS$7A
z6tpVo0i87)oIe|!)gG^t*z?d};R>sGLK@};;R0i!f+<a~uG7$}lc|XMe%LL(u6bEE
zL@~~jQVcj$hC#?7r*jty9h4y>Uw|q@^1NF(Ft7)8-t1)|qAABxl$!9Nmuj2%eh6Qm
zwu5DWntL%2@O}X1P~!;!iy4@aJ8Z$q7eI?S?MThYn@R*ZiTUK<ADfjg4(ec~i);i^
zfNNGXC;k@LLy(j@cHC5&lVp{xpfXbW?&c~Ks`Ns6t++H$l$2$3PwcPr{io1%nHDb7
zB+YmyP&s3*9(fRmbBbiXHM?xj3#f#J0O4i?ltixnOVwDT+3{{}H(FRZHkkSH85VjI
zd!0aQp+q)<FKL4_m3-J?he8?pG!R-(fG{(`^A0iA2E=xz?>tv<+vSFRcTs0Hv>Z}d
z=o--Xak@q&In{ic00-EWL(diUU<Q~@IhVf7cZw6BI|I&Pfd;4`6)OtoJ~iql|HwQ0
zMcx|z!URf9jU6PYXzueBP>LfYMJfge5J2bL;B*ee%vl-PSy@NlcI<*(VF4akTB(t@
zhL8?S0Ko3X8VE9&m^FdyxCGi0kJ29gx$MveK5OZhgcXM5F;-=C@U@pDWl97XI4FWE
zeOnP0I9*m@Uhl?OY-h@#aDeQm_^Y06)oG-@L_mU6mSrgfyTwPKOXkSH+MK4_>*}_!
z8@K@ZM9KyJs-^ppX!)1l*X(+V@;X#dp_~jDJ*!g#6gcI0P`%^o6#NhB*7t2EU8GWW
zmt;H3nKV7{GFbO@3gn#hX=rgx1C|qe9fRH4bhbnKMMhEkptK3FeFRPa_uSdiAG5HT
zFz`2$e<;^?Ex_dysAa-luVCV+^B4F%@Les%cN|Nc>)v6Pnmn)D5tP16eA=wi1*wGx
zmXHw5X3h=7i^7rd%F-9&O-%<e0Y3rRFfo&XBukB1)N`13X;HHQ5dKC$@NGFwS@NUD
ziErM#F#_6i79=g9P7^)G(?d<O9c$#CK|UczgDz4LFg0XW@qlbSr1zFOnCSeaIVwlr
z3eZ*?r-aL7mAp$%JLXb+MR1hf?r6V5S%j0OLPkAXu$$0dam_2-tL2?`PCT^D^x#2&
zl$;DPsb{*ysse;CWK@N*q*9Pe0z{-0c`P<~*vYXCPa%FVqhwVjqg*AC1eJ-dWGJxv
z@}WYlp*bG}^b0Ndi@M}4Ag&Xe=$B!=+dhw>gDT4l`;K;W`aGr(S?`1r63`Z;)azeA
zU>h__N34M~=$+~GsgR<Md1tk`*r7FJS&lj5!~7!0=#(=V`}UbY#$E)Jo1~dHrGis<
zUT9~X59OEXw+kP!_KIx<*e+b}=-CpNj5Smhat_qXDnNpJbe?p68I(-j41DW$iJg{(
zNg_}L4k+3~3K&byBvn_-Kj8@dirg){B9wlsfgy&3b_f_NgyN#^3p099TbBuIsjI8&
z9Or}-$Y9~ywlpZb8$-;1`VIj!h4c;GAPMM#hDRIhA_P~x6dxm08i7%<nQk)e<!+xT
z*qYsowI?Q=nlhB}qfieBbL_pf?vD&G0gHfKck~R=3M9*=!0Py8|2QsMD(GCJ5)*|`
z84Tel7>@BC<m9lM)^}l8ER!Xdcx|bB^{Ba$-!rGp!$^q~YvqxXUAqr(yOx7MDilbC
z>_RtGJIMpJ<~sS3H&Nr$clZ`sXCu?%yifRs&0mn95o7n-<-?d1w=a!_hWBN4z=CM8
zlv9}dB)d<=W7jaSaYZv@oi-oV3L4SiP)Crkx(l#WrKc=70PCz+HoqXh%MUZNaF({p
zH#Vwr%3QIu)%Af8vS#6F<&vaCx&gGtv_esdrS1h{srT0ULl;fIN}863xPIFlel}9I
z-<^iyZIC;KK<(u)8YlMA1T9H)C-&Jf4jJ!F@34SAPpo<vdP@l4d&4yYK%0M1H~k>C
zapaBO5$*!%T8V}h(GPo16DPJoFh)T&>@zAsQOq0V!DOIh!{+OqZ9UN8S^{drr|_8w
zfhA$mCxsA1A1o9ZYzBFM5l}f}*h0Spgnp=UfbIdt;(RYe<dSySd_e*#Ng(_u+A*M*
zR#R8!hhPGAAE;3W^Fz`<e7?1}8v5^{N+pHr4W)9-IU=+**1KJ}b!9zustNkpefjFl
z#MT3XFf^d(lySO`-sUU9hIC5#Fz7V!gz*)HI#4s4mL9c>tF3OpF+)e!;(hkq+}mUW
zPC6>bsv{MB`JgvtU~J*!MhQNGHSC||^pa*X!|XY;ioPYAoi+y{RIsRcw>|h<(vS-v
zC0{;hjWXz6)yU}JWI&x5@`W;4D<Vity>n*1J(L;Y7Pr&`8D|=m#6}SVCijfIXX4$l
zAAk<~eR#Xdc`2`*D!>YIWYYzJ{9zfwAL+%n9GX&%S!6TcEXenA>KRzHnswXJ;;Ruk
znrd894GRV~qy)FZqFeXPH9UPc+OI{2`a?9Z3hiAvh3c&T#l7O0CBvI`I@JC;JDgj?
zz02_BR6VK6!Q5`O6^v>?VJNy~5Og#F;aV?g0s)^;ub*|<E}qGr1f5kRQcKs87+!xE
zq~y{uX;p_C6)ZGXTfy3I17ia6dlgdK)&Lo&9rax1i-le4oU>&52L=YHmKGLm2^00*
z;5W+-7w)X-%oES~%Ax)C`sL)_RTR_Kr4=dI3{2h+l$yRV5EMq-rSi@_NQ~-8)Qrde
z7N|Ha@U7u>z0R1nV%moerfk?D!bcc~N>Jj2GmuEisj0O@vGopW6}kb)2{$aFxG8!5
zX0xu?-J4~m%c~CeP~KChrmGu#Ya2s_rgm<9N?dC$sNQkPLN}bXKD+55P=Uh$1CSDN
ziCSvWfuIE00LY$~42iAw<k)K3K=%mBoA#+iB{=j9;n@NeKVK1m*P715K#&=*W;C=7
zEp?y|Jny@eu?#QKLWVNh-DmH4%-rK}j?o4m%94Vj&>f<T<JdP4F76><|C!a2A+k-K
zH-DcU*BUBD(!^E<6kT$wk0gy_K@5xtd&aP+foHp@*v=~YTp)SN0nM(6+Z&6&JlR9C
zUYhHWxMlt^1=}Ei@lYVMibE)}EjPqh#SR2gjyS|&=m82Zdb&@|8fg}&f9n}Cu{i{#
zKqMkbjnyE+7DIt!Sf>$>Jc)T7BkU2coNNHmUONifOYqbp(8o^kP9Hg0P}mPin*Lxl
zz&cfI6m|3WQh<{{1LT&LENb5PI>dqMgd*znFUy}$9a>87*~*lc@}Tp3m*V^J=Szr!
z?aq>?hJN6G?+0S0S;<v;&ig$MKpcbe;@t)oi{g1=$<4E^i-cSf{_K6a6w4#dFGaHd
zzz6E6kWK4m|Gvlm8s38K(gEWo^5B2t!T<MP{oj9uX6{|^)Q#ipKCLgSegc)*ily}J
zpRqRIS6AJI#ULFovX`5yg`z;j{bd5^pR~ylR+5to$xCDr+iRYs#PL6BD2>e=Tq&fN
zd(a+q_x(Q>fgf@6CD9E$nY!?q%1=u?(!axUSTAorEUkeWkeF@%s|$NxvE9g9*Y_iQ
z!0o{f7R~jaF!z6R&*O5dR$R=07{X3<JH3j=;Z@UkV~y^%&b4OIvYkb}GC{uKOVKGY
zvFG52{p+Z*=0N9g<jmV}<cJ?pHZ0s_PJAR76FT=uFr0o>`R@fC?m4d2R=h_cM}Pnn
zW%xh2^)F+!00u15o1!k$@8z9(-*Qh#H$9>1r-=UT4(Fwooj*oLs1Ipdu6D1u#ScpW
zqagm>M;%{{jqPY{#825rkSjbMe&s)X`dHUWx4$hvQpMVecbdije0TBvNc`s=&VS<-
z{yMkI&f)5nfAeT1fK&e2u<}nLQdDdaDu<Vdmw$&t9rl8{#_Q*2Mm%g;hi`J<zj>4Y
zbR_sUkH)_{mi!-a1q^t5>rh;NttP4)8a9id#|E1XFwth|b|D?;gCT=(P&S_*wY-HL
zAlT)L1cM8w+`@Gp=ruqs(JZiQ!GPk$fBO|bUAf_4)$1on-Tf_FKv<9c6|+6!`GLR@
z%0N@;MeG+oQ^?Xe26^6c*rNopU*HW{sW7Ncje!?}EV2kN$^?oYP(THo)dVzG9+BLS
zw|ecTHfs5Y86}8!p<@k!jLpuZVcNff_ovg4Fkyo#LH1851wTm4Izm7mhQ|k!OEMqG
zu4r3JSdsmRAW+Lc3g<6alZ|FZzXGZq9D61=b6f5#85HCWxeYFmRf~ZC;-L=%D};lJ
z1OyWAY+-pm=(^vPWF}n|0^V;IxXVJvTg|Z*j?KTZV7&OG*gC4$>$pcmjc?ONkR}kw
z#wGrd@$sb4)7QzaQ}G66K#Wv&=3x^CXt1b{-UO2_p#M#S{I>pUT*2nDBDmt=FsL}Q
z+-o$4TC6&ky;UFS=N#_K2*@#$N}hODx4umwvYPb1C4lJB{N;bf`URWWyk2#*{Wv^)
zAfFgt|8<xEwQMM+l4y!Md12>K<KWWj0G$EIH3UB<lkI@5Vn)7$OlZ-I&-`v(VGR(0
zs)?H1AwL!ygFnV5ulb6#1)I%zI`3;RA8-YQ!e0Ije?QZRG>1+>XH!}jloNADJRqSz
zD)MV22+Rf<ZfRGmUWhS1ur(fw_lf@ea2vPE>mB%p)6>%<C(No|oMxQhJ$ierwiVLF
zpL0e%oWrjabKT|O_UA*kv0-*iaBsMWEJ4J1vQNmLRVBNsj7tqfnw{LQ;~+WiAA1I7
zBTPX3jAaT-jB*@OdCNZ_oy6Kc0sGmLC?4>>C>bMnjeg{8f6;_y;|Zm<wzD*MdNz#o
zA|o4!?RJo8krVWU9Y95o1RkoR9_{Sk7l+?C?nY6UF|mHR_-$MM7l6smtynwmhGr^P
z$Y5VL(Z?QBdT>weg;zn-ku+z@&9Y0dikS0Z9~h96&^M7Y{@~t;_a?uE1wmhe?%j>M
zn)|j_3ceASJboV$KDP3JBB2Dyb}fA;eqdb7pur;?Gn@H+HTWECQ?SdYKHlZP_M!mb
z;FGH^;IBOej5+@4xc0v>%={G#Ar!pJ0ul~Fxo#sECvpXRAT@F4d}neo-8^vY2eH&S
z$2j%<V3>D-{WXEw1!2MURv86Yvyri<GonW-b9CaMh*0&br=RR}icdGx0u;%5P-ovE
z<7|6-^Y58RbVDZ>KLPU=7*j%de?LNx7dq!`pmz;^s}ddteKN85TaU4A9#59ctVT)P
z7}QT3%0F~NLyam@?|eltIwWW~qdo8aeKxL-n^>B({|I!6)PV!HYvsj$V9CnBQIq+g
z+du_I#NlIV@~Xg{tx!PSaq4^$I%*(x70E;SOa@9x@iRb085o1gj6lYIWT@WY<2gfr
zIF35xE7)Aw`aNK#o?=yS|Lv57l`6T@SOX7%b`&6>D%KR3Ytr@Idka5rGYDz1%6Cho
zV~03)KthL6IgI6-#{O%LvunC_QjG{Bz%8|cf)b~XV5oq?tSL^EKoSgTZ*ZDk-fP`d
zRpf$U$_0zdMgcK5b@c8|+sfau!v1JaBirCy4&51<zx;)geDj;9*xYX4+BWwmL<A7M
z%mcXqB}UsCB_xv@JJip6CnY6~j7`kZv4>O|#vP(+VZ4I5P=Vs!Rhv#YS;9jAEocDU
zMB1H;reRJ3SWQ9k5ekbkVIYYJWTfnof8@nShbuv_asmbuOaDD1*|c(pyopfEl!3OD
zB?DtGzIWf#%fxWHK5a@agDu^DoqpdQB=do_!4~0CR3l&sR06ob8X$$w$lrc}BQKo>
zb=#X42ch(ZiX>Ji3@C$pm{6q;z$Q!FzH@8c+VOr!B@(QGT4ZTMpno9{JSQkku~mE)
zri9S+V8Y341JKwg5H>Tv)b~-7*`%JR3E=}{87t9u0RC7S%CFd~KT)LOE&}9(M`OpE
zN_sA1g6DWz8oTi1PuQGJMu{ADN`tI=9GqIt``8HSWTs#e_VQxGJDeB@Ls%%pVR0?b
zuAl!CM#H?yZe2uqBTinyy+Z~ZNdH&B+kz(Q3-4kTZ})hKaK-a`Jk1@SDl-xpY1itu
zb4l7>#bGC|6T$7JYXw6d^RXJN!C=pP{^QEU>K`sC!HTZ(7t-~|j#$JBK(l7cb>Vy#
z4rO*-AIM`sM$1DWLtqg<#xY3*PH-F1_5!&}h0&#gGk2z)U7?A70!Hk(DW1BWs0u-e
zCAS5d`5Z*Xq<qHQ8TplL>t1%;k-Y>gG<X;ZFdW=19Cop4&Bjusd2^up;pM$X_YQ<!
z85;-KElcz5eILVMGjpIRoIrMfS&AIWtID230}+^&3AqbuOy?*%7T0&PMeyNkXqEv(
zwJHw8vYQ2!&yicfZ*!Ea*1ZkJ4CGrkj0iRJKd4QHkr;&^BO<opEF~5-iRu)t5cZ%#
zm@nzerwxZ|tB_7@#}T&U1Zg7<E|heYBVi75CY(vpM&pH9D_DA9O=&<WM55?w?mk6>
zIaTbB<IDGWtVG_h^Vn}#y<+tFFupk6(w7FoBY_MfJ;aW!JraQO9hZ6NIv?A1u<~J@
z9Bs-7Iz<t;rC-|ZSs-$I8<%-cX&`|NRl--4Df&rk5Jwe{e*Gwl5$}&%UbWz?@hn=4
zYiEv(U^gp-vB;Z65j0)OnJMU0uIb>b)N5|>kfS5l$Z>9^@x>{p<mdonWls}BcHtxE
zfmcK+6CZ9yLt_I(XJ2K>eQ;c^VFAmv?<B9;sZ)Dd$|0<ew&g=Lv#omsTg%TuQ4rdw
z8bm=-${zJc9*G6V@RV^VuPieL(`{=3Q7q}wroMT7rW0pG6EF}Qlb}uNhI2xMXdDD?
zEs??moS;HRrMY&xd6U2Qn$ymU<Lkk5VIv?`Gf+DR>L%9VV~@O@X9ZD1X?&jIcvHv_
zTya|q`pH$k$J<Rag*?UZCP!GEdf?y3LDyxPxERm4b^+cLg>lH5>mK1Xw?H8=Uy-o;
z*hApSBjMaZ_@QCMB6ne@{EO2Mp;FV-ajJ4Bij?_^JXfrGB*k{=C1?gXQf>RIv3vmX
zi)+I%bq$=0pzLk`^V#dV)VyDTspA?YSVQ;{JH~P39(W%p)C%0cH@K3GyKIC0)yG(l
zBcuYB=!J5ycJG?m(+{_HgV@yo_SzQ-VqgMvMFh+>;BmwG2_XKihROqb^v%3juxgJ9
zaSvZ$M7!{yqpgCUcL^;N23AsTwn!!6&X2({K!WPF&uKz7f^+S9Uim7U+48YP{AMzG
z;x>+VC6J*(jMAXZg|k_JPA2mKWG&oU#I766ujt=}Jra3vE(8hCs`PW(%ZP2*lLu;A
zG9NmK?2>3OI=%zzWAQ+t9&8*N|A6l_Gt%3u3`{q$+x~EGLChy?pt>$VIP|vT7EBE|
z%Avf5?P*kGme~`POAnm@!CHgkUn*C?*w2WD&R$ppBtfy|2h{cIpMRe0z%zvrg-_ei
zqD|vY?S5!|Fp!;!Rnhj?R{s)S)?e4Mr~$^qOitgpfxTVX`SaN@uHef0sM{iwlWDP_
z854^_;q19EX$gpPE#G!L=Yycv;EQukfP#6mqU`-9j!HJ1IY!<I3E7~KgKrmymE%cv
z4z8ydXP^p+Gdw~!pJ>uI7(jX%3SMsG=^42~&>QBH&v%of5>8JBjJ9RTJitb~Atfq9
zuZBHt(6fFLr|RH@CMTGNcJ<tQ_KQ<#vEaH9T|(FkL&Z7j0^rTXJAXV{Jt#+8aYXbx
zdDY#10D<R!fMWYE&Yvsqfhr8gMnE0a49nNhO~(qJIH?uR6h8q~T#)r(A$NF;3x~L(
zEmu`NL3*4p4kcXyoNtOo3<GpbpeCe0m;-fQ@mZMK6^Ml*kS7&G5gto^Fy`wJ8dg1N
z1eM`!bUi;CB8~#UI{<Q^b}ZmX5*Tj48Xbai8|Z^&NoKx&xRXnToF?6#MSQ$dKt%~J
zbBq@!fRJ*SIHLzg2H8My9j7xCgJLSpb#^@AV#<rcL3m5h*+>&0RI?@_s4yp08nPN7
zf9DUM`6M_FGl1N^3?W!N)wRlHP2ek+%RBfhFk?Sz;{Qhga{&jyNPy>e$uWh(VqWhY
zT=>e%ahxLrGl#0V3R@vt%!Fgi!VIkWOJH0aS>m7<vp~|bZxsM+WoY-K6t~sMfEYdy
z1g8+iQr|Z$!t8}S&CKA0gUVXiI9b2n!N3Z9J*!0s5;DutT+&zcGx4p&Hkj1GdPlf=
z+pSpOjA&TJo?HQDUn-Q*O6Kc_aTj?|<L-r`oe~^mNYTbYt+N|Dj-edR+YNR~?HJUD
z8B^$;QNjlk9{?u7A}v7hU+T<+UeUZ7w*^z#i9n3nfFE&1CqPWv19FEUst2|QgMdP8
zHc0$auPi+ihCU~bEndvBy*2_^8><{e#jvi7p^Eg20wC#BL{RPa4)VaP44iF^1FcXA
z1P#RqfpW&YtV*$G7I3)bBn4(78uXy}0}DowoQh2^wzV%3>C;dO6@_66C|rUuib^0M
zhrXr&9IykOh?Zm{s0w0lkbt{ia1SV|^*ZN4t@=bfw_gcGH+Xvaig20G;2-v^Wb@2B
z4isGFrS;nm4Tbs53wn$TwncUiQo!ZH@PM`$5=>u~0EjH_Gmqv0VG0`BKm8<+E!{Fa
z7;JHV<24K7Y}FW6@6Q-F?(~>8^{AIoqfcp;+3I)K>sqy`rocv3;6@#3#L8E$TTh_r
zwDS<i5FkQjzHb0Q1EZ58i-s~>x^pOWuiI^V$0;*e$wiu}uL0Fl-5>*4k=HKsP{)jM
z*a-`2gA2+r4NqxjMT2}Tm!}1uMm%rKi`#V=%DGOV^ZHOC9~s8!V{{#u=yjYv{q6HZ
z@nau*8UZl4w6%kt0!r=z1RfA#3;hJmxpO~5_2~F*s8)GNGNuOmra?|%Vc3~a6UWbW
zw4tDs(e^mm&M=P`_$CU<4UaSelF7~p2XV_34-<4|sXcU{hekkTkByov<-XW-qJC;p
z@;f(B28$p0=A6$9r*0Xt$@_pMu4H>xe}c6M*rr9F-lC;fHJE}eK_GCs=sTa3=lT(W
zi@pyL5S=Ax@%a$kHUyPPK{w;bFyJNQ&;}f7joF9g^#b%Qz^wqsk(HrkZ~{(o&>bjL
zMZCu;Cbu_A?qzkTNyAVb86Ie$sh|JtjaL;&I-p}+gfI@hCVSy9>Gh^ypk(FJc*1q&
z=>U61hH+ANAqme)w*jO3jByS$GQ!70c<6~2Q+A^f(v#uYXS}kG!qMkgvic=8I=Uv9
z|JD~4s|zUu9wk(`phhLDjShnriZEe>UNNzv;uF(Qa4bc+Kgd{)EeX(JS`mqe$<mNM
zjee4M?mVjB!GZ8$xgAEPG4pe!eyqB-zJraCKO#Q6A?j7s9|^ArCe;W35qtm4Zr{}3
zM7ed>E2TX#F*P>4nHsj+Ys=Ove-!v{cyZo8HMPd*aFyHo+}H*~o-leIZGhg=Hqzh9
zlqvCuwn+tXgI=!l)IHJZsVQSKGqcMwvL1{OXwEPy^xhR18mdHwshhpL!ZEAZJlj!}
z9KkWvmMy+&)hYscc6RoWw>NSQ`yT8z);0mQCewrIZcn3zTC}9|&x!?v4Yy{Be*E}x
zeTAW+p*F=#<H3tBK&+)gk4OGa?!yuifq+%=%WAG3v4(jXd<N_KzkdCSCEU{&U_jm;
z85tSqkyVDV+hi)0N-Gzal?}t2K6-R8LH=dcSt~0m9y#}ru9DN%zP`S4*}$@%Cqco|
z^XU1XPg}NSQTh(A3Kq92pPaOfi&G#cZ%vfz1Yais`sWvh#i^x!2ed>aCFjZxLl2hT
zvGK1Y_BCrn#l-wf(q`M53JMEt09RP$8Dv@;o0yuWfi#*HwwaC1U66H)D)XbFGBTks
z5w8c13Z~&{O{FkU$-2KfQrczuV3~={*DEPr+~N!}&EGf>+)!6pu($-ErgY1;pwxq)
z-3h#T^Ja*ghb)kb6<@x5c`*Ue(#XU_pdpD^($R6rAC#RSiQK@!(GBf;nJ}w~%d3FX
z_`rkRu$oSDlh?d?<-%Zw{i}l|lXQA7Y?Kpygg7@h$0(z%VmqF=r7m3eI2+ql7x2Q$
zDk>`EE|B6#T<<vkaPy)wct3)NhX=W9xFth7!;<)L^9r`(^Qlm!h!B9m9Z!L76OxAi
zBZpkO%Rs25M?j#vx|u%GRsw9sI4mLzkUy-U5g8X(T2|Km{`MMh&x$;<uFv>(aB&q6
z)Wz$DfVlrEg;(ka4mM9_X=|&VfPjF=;ln9?Z&MH201e69w*`3g@*%iz%#)Fk5or&G
zG^ruU!VY@tBRCq-Z3t}KCHWMlt#v~r2n-A?f<CZuAW>``986(}B%CL&Zsd`E2E11~
zQstEb58~qDH1Dy1SVy76(sYA}sAbMaY?;7|*|%IrS6Z%ln|-f&s4dsJgO;mpwcmNT
d|G*_Y*rwe@%~IZ}u?lyD)5@okPF%Y6KLBk2K=uFt

diff --git a/reference/plot-15.png b/reference/plot-15.png
deleted file mode 100644
index 8251241adcd1249d50abb1ce7f87d3da78eaac65..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 19275
zcmeHv2~?Bkx^5JeS_QY;wx~?CMzA8FA|NtEH&qNESP_s}%cP77G6o3hF0?AJ4FV#g
zQba^RnZpz;Eyx%UAxue7hA<^zOb7|&z8|)`XRUkBy64=r&b@b^Rn}_bO8)%$|L=Rh
z_kEt{eZK$4b7n?TpYHw?gTYAQ&wO(pgV`92!AN}f$tHMbAo1}Gylnc;#OND%g+GB|
z6%k%GUpr&tkHN@CqyN^$jn2AaFkfQu-<-S<oH|V>zUW>IZ=9vQcdX$1xn{nUN{#9v
zxu#ujfAB==$lK%Bu&%DZ+<#%$+p{umrTw<zMndO9Z`<qn{!zU?Hc4d@trxL1#b#BC
z_}nWp{|Hn&YuK{exiZl8mZ)NV^XSht?!>?Nyc)b4%q8=e#|#wk@$vKeT=fi(0#>Rr
zQz+1S)aItKcC8OIyo+Sk{#7^rN%h)4PHp}VzLuKv*4vKO-INe~kfA&*j9ARmxi)kn
zV!?=b)Fer*?at>abIv%P^WOGF9`9*Rj&fBXyZQcJqoOOFPhMPn^;Jufy1AX5T?=dN
zS8cO9y|hx|YkqijMJq)hV&O8V;T-?Hwdjj%HmiKPhQJ?d*X6bj4Go1fexYjnpe;o^
zCo3z<6~-9;@-G<77|gxd8?y91x{JX`-xe`*J486ZOmA;*Pc;ZXF66OWhU0MQe2RF5
zwUOc680)&Sytr%U&NtrW^u&Ovp6slwy+S^hTEl0gaKl)-A&t6>`!TZGWiHa_{F&Xx
z+}DJeq1Sq%nR-0QJYp_hK+X`7Zv+HncV2N>UYM|{xOUlxvckJu$OxZ))p+F4p?|ox
zB%)uLn4UJx>U<*-%=<IL`PupTs=7g()|(QNwnrq<xjQ_vCAekY10<{`t)LAKQx7fZ
z^BU_kQanp~VOP$n9@cPgZ+&i=N5Y!pRu&$99I*z&zPbS~t@&*A@UQVj-~D=rra3S$
zkP@$Ku7oeBoP2)S$$vOTTM3^YGM0Rpx(@x&z36$q+WrsONsh)uB9Y{>DiR2I<&-q<
zfwIvlCvj+HRh9qBBF%kqvOA>3j^W3cery;`!p@Jinb_G4ykt4OymXQ@B4A}G&dZ_m
z)dlaD4JHr>@u@%NJ1xIWn(Hv!CM7>LHPymn4-|(Bc`qEQv?#N(l-cKfSw?Y6f|K{;
z&8)4hNk362I0*@ftDc^+i-Tu4WLA3ElpS+7uUB34v-*g2#HjV@)QyaXyJH$Us4GK5
zmo`aCDu<lHTsZ9AUt*dPI3_b)9gUd@O%*DjVRkFlX;NA5Soof^C$O_v)xPoX$Ez%V
zeqSFw57*<`_DEAO>a;p)AK2%r6Ef!#INf{SJnX}ciHV7(MVeEFtWU+|E4+8P7I;!@
zZ0vR_(dVJOoSZ4#@H5uwIXX8&TfD*)>rcNbuVH&TFUxE(t@X#btgD-PufHL(v+`Ut
zM(wOCPBWeG{a}nX;~~F>8=kD`X|+4%^l@EX|Ec_PX$P`!!koBlS?pdv(RW5K_2$i+
zj~m@__Hlu0p>;{`LCrS{MiC=MQ3~P@>Z>1K7KX5NnaR`$Ua=rUEXYVb9&n&5$K0IM
zny#;lpZj@>`n}-Q9m-}{ogD$xhLO(pn32cP3cF{mx&8h9Q9RMc$%luXPrJ1}N*HM+
zZgUJ7O~5;D$5=wBx{kFb<4H+Q5jx@F;qld9VV;k3IA-qd?j((Y^XV&6E}5}pe^zX0
znb(lMtX=%b9hHKeJWV>CZWt#N3ZvI^`-qp+YnML!AROT^DLA}en<j0mzIH$Pe(S@-
z!?zmQ$r7cSDQ9tw82b``rpDSNiDn2<{J%W3ZRYIh=<K|ovGilcnYr7t1uHN83sZH2
zn1ee0V_SdihG>z}ts?L_*rv1E^rK@sD8M3T&uPslHT9Qz#$)4ft4l8%Ga{BJuuV@+
z-E|D8FFKudzdO%b84qhGkTLk<4&}LDc9>)FKDC0ee@6V<n?ZuPQSx+A%bcQkRuScg
zXdXpBVDdt(XjGkLk+FP%G!o7wFhhzxdtTsSoyMCRB<I_eKZO{@`@xF$>M!{^$2_C`
z!>@NWh^wN+*0r@%Vw+_X=9~zZJsPXa**SaqG==9$L77B~FfGemI<4j1e{}lQ_giGf
zh84w29ZIA54+%BW3VO%J{4qQ7va+1Vl{^cp=kBDuSa+?nva&LYv*)^K>AJR!#L$y~
zprD+*Je8*8lox3=JEfjEGpa+f(p43M;}yMPUPAH;g9~mMO>#;{K^a0NFyiwHYnN{o
zM@!2ao12>lat4#@qqn(tJ~?GW_H*?ik0xIBeWKMvrWp=>AG*O2bJ}ER<E4@fj?a9X
zB^f3Obz2r8``=5f+fsLBbM-mq#r3`u`oU}W`Si%mwE~g3|6k<w8L!C^lUjPYKB~!a
zJ*K|5$hE~q9d0FZd&RAPtvmTw8Sy`_^S?H()*^au3gk%?$kmFkHzXT`FLw;@(r|O>
zKT7)!bM1P2__m4h@iT;~K?#c=GnTYp0_;KIAB?-T`R$dXQ3~c62J|F<j9NdLyGO&V
z^-OVJ>Mz}x9c?)c+AkqfnDhRMsS|ED{YODT!Hu9GWxQjAV02+|@$B3k%(Ev|Rk{ui
z4tNqIo@@$*NV48g?FgVicJT#MxE|%{{l-HZ|9t&(9R7@h|4ZxOd3cdbhlf$DT$Arg
zk8d`xg`yU<7A7E7r<ui?ni>Rd3mmHaM&8^%Xj`&h#p=bIxzU!co}Qk$kHE@`{L^~h
z;=a~K1KumdkMIDbuf3*m=~dc6j1YGK+|lU+#qQ?3>Av_zMWSAcwy&9ykx|PSTILwr
z2~z))_Y*l8?dkf2B0!q7F!n3w#-i)P4O^w9$I1qLTK2~2k<`Q9ZxIYoRtBmk&iXd{
zQ}vT16@oK=XKaB@8yLflaje&pimg<!i0c09ozDve%Z&<zgTM`J%DilP3Y{__ksaL~
z)0%<9@##TM0C0=^S?zMiu4V4~jrgI{NGgUM0)$NEu5!3fJFnQ!lmfwxGg!Vta-mlS
zu;epz{N1ebZ2=EkzIINNK)0}SThcc};k-WgV!)hX<A#dH?(Qtl+S>ARZDsRxQ<8l^
zw`G0pSc9yuDM@d>Rl_;`x37&#_&PZ`tpZU0N_+5{;TP(TVc%*X+(l|l(N_2N_O7pW
zdwpvop2W%^Mznss$l=E@Ux%!S^V-Dt+S=N|>y7d<`u+>!omkutuj(fzCyhx9e7^go
zOJBnGv(jaoa&RjXKth?jc)hxe+1TnS2X5t%cpK*7C&e(@)BEZkT;s4BU~Eh?48s3v
zEpcKWn~i&evNV)|hbXh>-x#7-1^JfwGSdfqDf-3k)Y@^fMUL6Nrj6p;K^)2H*RNlf
zl$OSX`d|?;FQyvkPQOA>oXy&TiEonG@GzLoW*06ONn>PMoc{9jJBz@(QSklIEX9>8
zU#QtXOi;0|E$Ay&Or&yhbU$*MuBT^ZyMAYe1e-dXs2(>0Y_l0f8O$P0lfd&}$~XpJ
z{)Lu$?D|(t$<GM91A{<dT!tHBfTvhT=?DJ$x%E%w<@{Uom_ys4*8MU*KEAV`GnHWN
zh$j-o*uqaD546)0-vjtnSVn{euK6A!EGvcL<w9(eq`}P1`B*7wX(fEDA}-x&IorvY
zB<glrv1BYf)!v66YfnqThRpsdV>GB7Bwk&?lROK^ZsTOcqS|~T<~PPE>IKhe0%0Cz
zFHCgB*KBzB*Y=f!{@1Th;s`)%isi97^C`bF3@>?nO^L39zYwvfRtifVM@!Z5KjPeQ
z58G55E{xLWa_Cj@SZ(@^u0II7O0Z(s%<JuOh;Ov8%bDpM4#$gLxmZ0Mb9!%{bzwUT
zCl+`Ol&J<mh~&j)z!;h9&3;1~8B6h1_$TKkZscBT=QI%>*M!hs*u(=1QqBlFVpr}X
z-#<J(^!j#R$yE!p41=^^@5;CX#aB&Ui`*-vMb<O1-P3)=un_q!Fjsw*0aGE?%WC!&
zhlrv^;$YElsUfB4rZ&Eyic=XjrjddF_vu99Z^wHarPl1-<-1>Un?maJKGSDXvRbBq
zCw4X@s@f?LHXquDL=wOdct6q7?EeHRKp2qrR^I&XurH(fTdE_8FtY|JOgFAqN9**&
zw|~C=IS&6n#Q{6M22IGrt+HBP52yEG4&73<D}#V<zP9ndS;V3{J>ApUdAQF6gW(z-
zI@v!k;I!CFRlpod5FETQyMdwo(x%w$>~G}Z7Sy8~fq`bcBN&WQ40n%=`B1ooZrIYy
z@3aMk(AL&ujd-j&MdW9eu7_kh%tJj3HsQ_Z*7k=dZEo_VyslllR+NOnFyb!VI-1gE
zt<Ni##F)AoY;n*7X#@@Y{}yCEAH7*lp-{SWHlV{V?Nr@X4Q!?5>DiRI6)q9pN4h;`
zNDkfy8H~(l5d6C>tw$ZWq1Nf4levIk%9i8S#E*w_whF=G@wnCfh%=GG1T9YOfTQMz
zajV0EUz82*L-K-S*DmW}5ge=S%SZ*5G>T-qVwWZow(0e4$$3fQjAXJtAus~!zXnq@
zrfs*<_%X@pz%}4e(Wn;Ex$PQMy?AZ}ze@1N$bgir;jV-?5cW~W`aKKVN@<QM#N`~K
zpM@5>_{eW}C)E)OK=e5t#0(E|44udp0noC8jGKrRjA>JnOFe0B0N%(`1yyrIF`qbG
z^V!-LO`0nt8O1P>#eSpBdW-KY1gm^X27yCX24rCj$c9rZ<k&+kCNK8b%_kIw?E^5j
zqx9g~my47R&ZJK@s}#1P!DRaqb#W_QIQ<0?MJ%Ma_Z7L4MRP61A@gq9u4`jb5xIF=
zH9#>ZVLxq-!LQZc^2Nd6ZYdhN%&T2bM|)p63*vEtcza-*f+~uz>PN?~&<~y1yyFWb
zEYCNYM;x(s3~^<w^2*((dh+ok&?ntm5><8lM$VwglEX#l5qa*!MC>j(xmK>IojY>Z
zN%Y|x)g9<OnfI<sRo}jSyRX#Kx~IS)rA3{hSI&$eI0nD}9^0f^K1h<+4IDu&jC#N-
z5Sq&Px82<dQ2JDymVTCrYM*BjS6RbkQ*^ROQh(#{Pk{8ASq6M^*g}?K8_45P>0F5F
zXq@PMwRnX~y|r<RT5zmoo|TmT>ZCR0xI)WX5bn<(y~)?-hL|f5re9g>^EGhu8e}1-
zs@S9ZIY79TC++^6DFEGem8aK!%H(8=1eYRep={M(1@(C)Z05FNtNH4PdF-S;g-@W+
zp1ZCEsJ$Nq9Vma3g8?)LGP`WxEuF6a6hwR{all=xHBNqQ;%dlHIB4vk1K5V!qSIp{
z-jrBnoEu#G{2^K&YwwzF{TGq={uG@4YXzr&mGMQwSlz`19&ZeFuU6-AVh!eUPdQ?}
zs8vF92y*bLS?sA^%=fK%=D)D8;6f{KK$!#e4?kD2a)HhO^q@QiYn<go-UF#>qd0}O
z<;5u~Vf9xTr`F%vSO;y2u(H}da)&Db)BlI~ajw^p_tD)pl{c<31nog%s=gs6(0Yl7
zoCg9L4y8js@e1Fa&!Gz%0eg!k3s)yY<{M+>TN+|yHI5y7IN(#I3i-|iI5&^W%ph=U
zv;no-D+CYS^ct*q^*i~s3f#tk3Orh`xOy$|pbg57PocjEY(idM9{s$*-2FH#ULj=U
z{?@HqQ?<pt&T+0lGa-Yvz-@N*^(6y+5X`e!koBR-XNNim5|Yv-zjr;GH!)~)#W-Ox
zp5#`X45h7R@*>CvS#buBur}pB^pN8JWpA>6pKLIduE>w(*8_`eYl>Hn$67lCbibU}
zW&}HvMv~prcm2+1DqJMEj2dVnZhi&)weWc0c*kRGQ+3Gv>FRX@kWWciDo!l=gG8mg
za5w|)OAbZ6ifxLLl5N+G_?=1X$bJufDh=ofmX(!#ON|+E5-&Ox6UD2S<7yxXP!-kv
za1dkqKymf%`AcCa^IFEq#x!6HHeoLL2^9Qh)yQ?aF#Kq|k-9S3oi}fRTeM)+jG+!;
zJG1h=Ju9m35ALk5uO9*;LbN^)l{b^bgS)POfl)FJ3gQFkd726x#VrwI8!#2`)=r+Q
zx*J;W#tAk-zk31DUFZp?i_u&~2+N6tURr5+#zen6ze~Oy$YsYc#7xZ9-x`JXN33VG
zQrh%p>Ldis*0rR4CW#}4ipw28P<pfoylHIH7WL{w9|UHp$L^1qzSAds&bb0T%qhz1
zR0zL{2|Y~Ux=M|4ij*NejOhkIbcdgUFnpt%0}~V-LD-79bl{N((@?Eyx(KM)UvKig
z2I;SyjX>i_g;u!Z%w!(~m=fMGZ1&gf{3iV>ZH^|*USo`W|DZ$V!mY+_3VKVASkSfW
zq0zMaOd`L2+<9f|Mh2(`f<6h%(~?}C&6R^U%S8(7Sq*gPyY5NR@(yiNac9BO05Kbj
zhxMjC?HR~rq+!VuPg5H7*81az6pSSq5Phf~w#Lfq`azGH1^hI$ts*6reEd6TSK~?0
z!%hnl{qp0EkPkm9ld#&%u=zhirjDS+F#xPX=2TA2<p7S7`8<ef3&<6^wCQ1UO~Ao5
zS5{W=?Ou0Z!gwP=?5>P@JGA2Mes9soxPfFoed=C=c$F(2lTchbLDI19alAsM1NtAC
z%qukD3Vg$cWr2Kmz0}wuqt@DpTb%_sT;|1DeB?ox@#%ee8Gvowp)#+&BWr^L9X@ES
z;@bx=xIAa;O2)qq`b(sCK`u|<Wv$&;<Kk1Kg;crW`nBs$riLj&xWN>-ODh_rp1S*m
z>u7TVwD#;ySAP{;^1Ba$_bDT{?(=F)^>))zicWOXhff_J2G`!kR9`<^%H88kzEwW0
z@Uh-oq}6Rfh5y`xKVSbFj>C+YO%G4C3`molp}Jm+Jh|>y`VUWj`So2`k5yZV?QEZ>
zrb-a~tO+~8%`^M-F1!A&F%t`$k`0R1C7N$LS_jXY9_;#(ru^dKzTuOnPMoj9>~eDx
z4{7e*t8s|m{=IHv2u1uefuqAB#JNW>2{`^BbKU;^>pn8zlbb&_k3Tj3Ewaw7cSIu`
zdF!tq)t%2CM1F)YnHxUF;>LZyDd&Hl0`p%w7SDb2w`BSU1O$|-t-~zr?KK7KQWuB0
z%ELr5S_zK}cFgcD+4<H4<OAhj`NSjh@#8&GQc|UA5|~W4FzJco>wLGB-I#c53lx2M
zyO$E!n@m7OjmN537qmjP^t9iOIl*2iqwcvaxve1B0S@o!?Y%hlSIi0At+t<hnKb4j
zZB0$VfCMH~lly&T``I287??jJ1#{xhRfG8>gDmC;0K=sR0Jx%6ZonKr6d_tdLFXG<
z@Mos@f15QgFl}YdT>$Tr@l(@e5~Pmu5W#4YhR3^Sa-rjo1$B}J{{TOeZPFTeTUfuI
z@oDQLO%<orxumYa!9iXxmnasBKy{)z1kn?a*K&eZ=p?1GCX|FqSR8$D6?mDBSTqlZ
zr^bbi>=kszMLft5jVP0sC-aKwfdj5c*#b+Khl3JI9vxdg32wQ;N(*|LoBOq-{Ec&F
z27-|VE)$o1{_FZEsg_eR4iDgpkag3h<jS}CCTqe6V`M|M>y@ilo8*HA?_&kMjRs|m
zxu~cpIUq}Tl9PCumQuPVwq^3S&m_Kw22_)0VND_y*-9Dj{dv$<!!|+k)V6a#7jc3+
zqU7m%vB(8aavAV0KMPzYni>MO#z$Dh5VR|6P?rXX*VMJXnfLX@Al6u2{2Ft?95hHs
zwwaj=f59Bcj=qr=S-v&@vX|!%1NYZSY{KHL3LVp)K7DF-dk9OOYgU;b&x#AInEol=
ztXYSZ+7;umF7x;%sofVK!|vY?jzr^$Jr2r4s)o+a&ZM85C=qD}?(`GoWOB&ub?WRD
zK2aR>&mVhC#LT#T3CA*PJ-1m;=~c^L_3-cz24xt?rhixh?_?VoZU^5NfhATK())`{
z<IKd#$;tkR)2QWJrXGd4<h;ydr>K_oe$~{!WlPTs8;jvPnEiL)jVG$bS5{6PK6fy`
zc>RUH)!9XX4hpUkAP#03UAvnB=m%w4b0a`RqYup`Ifmx8etOe(3gW=7t}1kK5)8sm
z8a`il?lXx^Td!;hb|SD#U)YvD|M-@qwAYqkYiQ?!`HxodRi-_CbFg;Uk6*L5${#<9
zFAARhU`CO$13NCsMF>qcFO*0yn3{MPSi`8TSF4V2zn5_wpV3*<bf<xUrR1xjBQk^j
zI+OVL!hz9;&arsXVCYo-l&TZh=fH=Vu@5f|tgFpd79H3jMDYeKy6g)FT>DF|A`IR?
zRvI#qopc23r7vuXmAGKaJL1**a<?_Ijbp0&5^Ukm!{M9G)>Iu0;L+@BeY#9RyS-o2
zz6mCUSk50-@;s*p<+R1h&N~5X^Xj|r!5P@$3Ip&41j=|UsC>^qz61APb+QY5ZG$dS
zpI(7j%MYg4TE57l^@5k#uU<;qrTM|v$mlqqL=o-FIvvw|Ia)rkGs`Fk+&w+})|BU(
zNR?p1A?umH&4Mc~(wO|)|Lzp(TF*SX{oe$ep+t3pWt@E;LcaOkyLS(uXRU<aEi1bs
zFVW!&cF(I*d(NIc8~O>x0YFMGk2^NzroSOH`Me4I<=@2w7gwO>z^Z_yj5pI|m!yJ~
zTC?Dj$R`zou@nuT!3t~>DA@j@*+vLM%$*Na%oU%ArFumrd_$aKhJ^I7ObFPN_^)@P
z5ctnP97~U@SFh#<Y{&S10g^%!^lam?;L@W0uzjxw0IXB5F1@~m+y?j~pmmAZ%vJ9g
zmGznXk;DXsQzd*L#OBsrM}AuS`QMSYY3JHyocs5&_;ODrCet0IL>u?C1X!`)KoGQt
zwxm3IgnAUFDO#$~BxpvdaVr>EQDiX)+_V~Azh_NVczH<)(%!#=G!T!3_U!YyAC?dF
z;pAmC+)R-ReehbnR2sv9e`|}{n#p#mkYa$K=D=`<pw+zy?nyKxi92IxXlNs3d!@{{
zl|P;hfR-^?yegb0Q^VC1;q!rfXVG-Xy$t<gg-rg0v9E)5?uBNp0k8P9Acjt~oZp#1
z@YqWeJWH2-gFJdupKGQEW#f8`%?qTS<LkCy@vu$7X%r0SE%09z0~S%@=%7Ef?dR{e
z4CvQF>Nx0FeQX;}bd5A3WQrqD-=SPpJ;7LKWuXi5#Y2#c^o3js@l>g0CGmI<6e!5e
zQU>neGM6{V$<4i*HI-O(Hn@18W2t{xyZi0ix18(4?o=0%eO^wt#DamtM80nc?0`V7
zoNanDCt&Cb?XSJwIMJ1}689*tn^<L0KXxI1wpwN7l~t&p_qns{9lvaAYa_#f8qiAf
zf*l6z#L|)ym+9VDXv@KxzWAcJz4^`ky%tVR8DnE(qQydD8!{t+TV9%SlL2lMRBgr{
zc@{K92-()W@^^GP?u4rMl`#6wVP(vV!n~Tr`3!NGw87F4)|2o7wi?W(Fhxq>&Y?QT
z=_evJELt&QDRe5JTx^xKDbjCq@J&6~ZGlv)_G82M78!NbGe2)h)(!GO1;Ip(j9n4(
zeeJBQt&>jP-sB3_9V(}a%XnXAl6~G_bvBMQ*P>2EMr(auZ`O<c$|~Z4d1xn9_3W!l
z`8leA7IQe_3k%<!I6}SWhR}inUQa+$f^qiNork_eQ;h5<pjf#>nv;*pqq8a9&o28P
zPOndij+A)=?(qj$WQi_j6oJ)b4<4LRRV7;_n_n>_gr)2pQsfiT3W`;vnUY4JjB{sn
z#GoA`3-xh?QU*M{x?y5bqmIn270xYmndiKEbr>ND1L1T$BvSWZe%PMIA%ILH5KxKf
z$g+g=q8>ZnG_VvY8_8Y&20Tzzi|M!^&fs7;pB}Q%X*3V_><%JO2vf*NLWYZ0$n@Rs
z4SLdAuA|l&Owk-<5Vbt3{AQmjcH$0my~DcGFAb{#(BiqbOXDe9**IhA7v=t%2|3+B
zuK}Ho1VIHVscP(Uxdgc_<4syeYXiTZ?ON)yOz{^Dc6i$JmkMl450ujb?Lg>DN5PJ|
zoMy+~|9u*&y6j+tHT8A?e-Y%XF&GFcGHpRI{u)_y%Kw&ZH7ZAgW|~a$q04^@qzBMI
z-eO)<54Noh3o%&CHFpK!4SiHaE>zJ`e_a0jv~QYTh~F0nFa9GjF;Uv^180Hh(S43n
z;fNQrz4W+~Wz7?DP5rqlZtbbbXmpdomW5Pn6e4l@yf=p%<>Fz7P+)NNfeF>oo=vBe
zhrV~@fYYaaknCR!1DXwhW=hppY9nO2u8EYmn5$Q=ym|I*7Mdb=0FAEU9n<Zr+FG>p
zQ(>mQslp;MSqIs$R+k!aWrQ%!AQ^=yLaiv3Ox47~b=v#4X`(J^zkcnt_dlb_LXA0%
zt8afQYg)0hwM7MW1T6kn_@a>cF;|##)EV^$2a)@`XWz=#mO%m-B#Kv(k6r(+$`x|N
z8<;A;$=sZ=is9I)g@;GAwF|9@c`<E`mYfLfr@D>nL!ZaQ6gpfM9gXlYRc)Y3)im@B
z3juD8LSxk{=mi=bX#S3k#GAJpfqpOTWTL}5IaLcS*Tl^`luGPaG2*yp%ckr4(=Y4i
zhHh^g`%yXl4LyjN2C94|omVSp^K}C+onYD>SH*~tQqhOG$vy4Mz60~YdF~?0EfEJE
zO;4)IFW@y|k+j$&J1hedZ>~j-+vL04G+wkZ!9=I@WasAK$4LaZ6A~76keTo2dP4FO
z)f^tdRg7e;PGlUYte+%6G^1t<5$aw~Rlld<^HQ;&-s;k@0Jc3m+`jkpNQ={|U(`0m
zbnq;@K$=9}D}QbcKXB&*5ll-vA^~8S8e^`qnMJTfUF#y(UG?#4hm5QdF!?TU3Ifl8
zJRGZg<hx&fNQLFKOUKWptoOUSFYL)jD65Nw^RVMa9nG>&<-}97!rBy=2p$YaYCw<8
zt|KmIpx)2dNo+%#S>B)xKk!ivdne~*y)=2gUDv-|hvkCk#B77Cib3M&bh*m&>jyJ)
z`g)i8EgQ^-Sy2n`X8oZG=|H~oe~VXk?Z`ZN)yu2(_BMlu;CFO=`x730q;YMiE^z7r
zL80UkEQ!-`3TcqubO3MWHmAN#kfWW?eK1eO;Sh__xc=dnX(&G-0|Sgi7k2jyDypcE
z8gA1s=~}T8iXydBJQjMOCMe?r0pCf*$|UfZoSt({OBPqf>@zw82g-4^qpH45u+gUo
zV%IG%zswk%fJM;LNB9!f@KvR=j~?)A`8D3oEZKxEyXSpXk3aDUsydx%6udl<6OZKv
zH|id@H-O6yTkgs*E8;1Wb5i|`4FpZ*V=$?XsMQ2j>Y0gfZ8s8DfO`BzlX*R>Ewil}
zRIH8AA`_<fr-A}CpjmcauTGu5xAVS*xxS{I581UU!0N@t`}laJeR#E+g|}b?6dCT`
zz5D!ug!51nU~d)+g*TYh;;|^-A$vo9S^ojpniPV-%R3q^=fZ((5TBs#1h<0$vl<Fu
z$k0v6<N1s7++d@ISoxIva>jA^>BYrGorMm=d8Cgw_4_t4pDmiz&;=O#rU+e&eCKug
zZ4C{yf+G>Tak^dcja0M)PzMa18o@I|Wwel)*OII?HUCAB>qRo_btMm4WGKwM8%wGx
z{q1FU@1E-}jL^i-l@^Ap#>B*+46+I>6%~-hplQ*%$@9gSV4d%2`b2(V4bK*>Qm9SJ
z_-N}p1$l2*2hw&QkS89Pprqxx*g3G{pjLPxDuLjccNLtY&zG{|_j;oY)<&0p*1egn
zI{NNo;wvcKTsxa{YwGy&c$Pm<i?TLcDOgG%A5-3dsq#^mIZod_Ex>AZAqVGu4d)A6
z_DC0an{syx)s3dKtNh1WC+3dBuBl*BssGIYt)O?6sMS#dmGAF6%+3JojHc=gMM{vW
zSEYzXh%rkuU))pBKi<@zaWLPpCNyjDLEa-_e$T7O^rtGrp4K~Y0~!&v1<G+5iYwzr
zx?$Ce2P!sh`J#kSJve{KK{@Tlc}rYH5WfNvN`VfHVlkkpHDgCVtv1ocduX7e<K^pc
z+GO?LpX|>mM;i=uAWb0j%z`0rGr}!*qCI*O1LeN83)$+fFq-{C%k&9KLeII0B@3t_
zrkdrJu|YLcuob~bF}$RU^<cm*Q~U5Ki3Pv>7=rV>F&h@Ow|^sS4OOS=&6@&VyRV!c
z{M9^*N|mXd`92JodpnfYp7+bQH%S?jP~+|M?mqfT(lF!@U$kq|H1*hFl*18Y0$t#+
zVYD=RLBhrJ#M|X4PiI@oE(+v{0iVxiPv_77tt|sb>0;mDX|y2<3=7o(A#jLcGF!@{
z^nyR7`}h|fT{i*{%yK6*P-%-Jv;HaGWfL>AJ$bc`KxzZyk_ZYfJeOwPyA~Gj={{C~
z>N{kUxDjwgo08<f2Z}m(HJ!hWTX}ilQKu#NEOmv?BxX$o0v<bWm_2O1Z>3zLd(g^g
zjGD35EJpK=putfstulfGY8~fHC8E*VKJzzqTIlCzn6(bF(y$_ny3BN4bJ7BvQSlvZ
zNn(AttKJUK<NjT>%Kgwq^N)_91PFotPg~U_a42$i-ldiWFf*o)j(!c7KtzTXEFPlC
zVBzt$QJ6>vuJRL5IZQ}pvJ|gYmBMP03tKk1V6gR>25{N&8+{0@qrEKM&n+FE=gM^t
z`j*CSTsS%!TbOn@qRd?s=3uUXo0^<F9#&oK{{qDpeMOd#<snA{4QCpR${|3=;(ww`
z(X@WDV(t6vl+M3JuVO11XzqiZT->V9AB@j7Ng7je5;SiTI%+Qak~J?JU%l{_sB6*Y
zUDJvDKai`kTMDOQVBK}3d;;|@udt8`IySOd-|1m`<mLN8ff#vUVb+9<Ux-jcNH}mq
z@&}k5l1XRD;#jZow)&38hV$R6hwVY@F&)fQNbuqh`kO<rm6;|V*|$T<^v%o5_5I!V
zt5??rINK_V#cz{pAE3>m4z@XaVR-GiN}Bn$%rAZJ=x(+CJ#vD=yj?|J)FnZscGr>(
zC*);;2k~WL!}V~;;QpX@m96miCE1ezZ@Xlb_WZW83WWR~*sloo#Dj>b&F7kjqwS*z
z+!eh}DYeWJC)ybXHDve+xnuoSc~ChG$_M>SCa0$S86hlJIBk2Poe$b#^b@3DP}IQx
zA7(SK!O>mXaeNfvS}Yn16~^f)Y7(iP_(j+9O>c`qL5Q=EJ3YXK4zbUT0Z7Vyz%~5e
zeQV@8SaDqdIVufNlW`-Q2frGeUKEqJi#pOqF}oA60@!3r%Vq^VEhfJL^4P`mFFJd@
zyw`(MZVO=Pk;lElA7|%d?qw7bG-5JF(<M{R2gOYRNAY|AJ#bwxOe6tl(}u0lHzGvC
zioQ4KwrNS{KBKeq;Ud=drn|72tf&GK{lZF+q2>i=d#;C=J~#mGF+esD`nEhKQ4EB(
zf_>T(&u6Ni^hLPS2R<<-`OwdX(?fEPr^w_62=n7v))q!Wpo88)H1?H<YVeSZfN?^F
z;FRxwp=@=es2e6f8H==+s-x7)Q&OKtfMswDtRx4kVy)0>8VIwTT84U(fb9!mrO~RM
zL>r4i8t6jL1$djY*c{0b{ZdR?34HqdtxpyBitJndh=8vHSQTZE!_hJY!t%(%OlKb`
zk>McKReOODm3;K-`Bm6I?|W^<BVBRE?LrKr5(p4d!l*E71FL1ni@p~6x@&sfxn5Je
zYS-o+kc}%d3+U`y*bUjnsC;VbjAp3zXE!5dLdiYa2h$v4G(jn`526RG!F4r<eva%I
zN)+#-726ReP+pV9KyIbfvrBv^gW6~4ed0aQawII0XptO;^~43Yfw>Ms%Q1)9D~2Fw
zhBo=!TNCwC{haLVeTY1PzM4F;xa~>)CTY2cL_rb}PfCTA6f*uey1yXGGtJj?Zlnnj
z&VhMzozVx_rat$KG}!p@k}h&Y)_JtO9#XI?L{OSShI>Z17ITFYG*DIrD_I0qyylAQ
z!!X^jfjD%kgpnqKc^eoXbcFQd{ygB{06us3O8<kvVLAfc`RRu!A!OD`YtdX1rFFJZ
zF=LBbMdq~a0sR8B83nm&rc1ia1T)dNR4h{QY5|Ta<7;`Pg{5tK`tKXJ#~C{*fA+0r
zs3(XNyWCUnW08Wbz3iEAcKb}A)$fTr!7C5w{^BtP2qZ`=9;j_a9LuMAwgHkyw#i~b
z-tHRNJ9To^9W892v-ma9UA{f!U2pG0geU>CMHIbe-`lNuAVWCFI_|IvbYdSqiSwmH
z{)%G2yKl)Wya6!kq^1V;U!BqygJ<Mim0;FZA4GOCXj!^+njJW&q~p5LP7Nf1pyZ`i
zIsQbNrVpz5!0O8TZNy{fu?|QqzpGwb(X|pjpTeBJ(U+T({s<{K2#V9}0ue%vmJhth
zokctu7<w{LLM8L`K`m}Kh-LHX{OX*p@PHtS?D?2KuvvktX!>Gy;}aF_nZChh{cT!L
zfw_S+_j%iGho<WeAlf5EZFzY)4CEVcZT^_F?({zpL59a2w8O+6fl@DBZrvp4Orvc+
zWs%RQL}d;58M56Xk4>4n7R?f{vyW85Jsg1P*b7DZ1q+QxGy=1WNrYI)WPt=C3LVfN
zzNc~I^3Q))>%JK4vEZ|R{W|ppKgjuQ1JLOZ&BhVYlNfrc)(3&Ttj*HeApvo<SB(lS
z834qPQw$)fZ}^m5G2rsC%-hn<;HG&y5o3t(a9^=IB8{%h>!rHFLZ~$0S3!>ma03z9
z208~QxgrQ=Bp+}gVs$<vsDfJDeq05$G!&q<f}lNGDv|3Xj-YM}>?P3|&%AwHboE<^
zFL#`<Pwx6J+Yx=8UmA`pr<Ys-4As2dNUoE^8G_5oNe}giJ@(-T5WMsS;3{RJZB<Be
z9$@gw;F%Q)2oLs;eVAN>KS=g3XM}nxgfDa=u2sdYC79{W{xV%X-!5OE0c^}30LGeZ
z0ruBDpig^{ZY;%jcyNLY|4E@fM--~zLg+zTR-&Ay#|7v%@4`*0@b!A%l|Jyd2W9a$
zbPZ19Z1&UZQ_IM4i&VC421}Q~Q^gMmmdN(VfJ|l)!|0#9r_uff5@!-JGOfvhi#dqZ
z0yU~|MAT^Cz@#zpV1lurr@S@+X%}l`9=7d!6}M^i!F~G!yKn{hCIvf`f0`Th)&A0B
zd5>vA5T8=SOL5{iz^MFyW>|#_VTVkMYm!4E@>GEW7SvtNlnn=})DE7sIS`Kyt@Xtv
z$G-_`v>tX}+=Zi6B4MEGtoG=WyK4O|);XO4?Iz=uXIJ|6brNrWnAqh+lh7;$Mac~)
zB=R>sUp?yvN;BfEWqH1s(0I<h!Ul7#Y~Vb{K<*8)+<)#^%3QI@6Y=_Au|pf*LRC+N
z=O5U>V>qgGpIg@>zc_62VS)O=z457Jq;}rN6C^Y=G$2l%vHUJ4)$YW5uV$*+kE>fg
zi;t8!Lt3*b0~R>~EcCf|F=St+c$JT)Q|?;bI?DmI6ZbD1sojn1dw@vxAPhDn+Hent
zPfwiE`qe$@8}Y_FEoXZov7VI86i8-ym2E(6fe2d$q+m{2G%b0EHbl02-}(59Nzk!C
zs`F<L{_`UE|Dhv9D+}d*<VyfhisDtufDiF`8So9#M`2a^Xjp!NIdT8N$BqoN9T${n
zPQ-Ae@MD{P!pG0SA`h!!wWjH6qssayCB+})I~MH8yK0>Mc1-PaOr(MtGPb1H!0aJ0
zZ4jUizQ?g!7v%V|Q6Py3w5Orf4qisc4`yY9el4~MTQDcKx!r-?2feM(lt7QEKp+i&
zBy?f?g=0;nw*W0+G;#PN@O+B*(3DC;ewh<eM~@zL0c8eHg6&~cL|)^ua3s}lu^XO#
zfzmXR6EyR-V<OY3pGkqE%h1~X`l)tME3r*9NX+QzD#p+eWWY?bR)t(J{S`Dp!^<Dp
za-KZ~6>KeWu+j<r3bMzpT|y8R5!9%Jb(nJ-)((1wbV0O%v5Uuqriib!&d5J@TWx^-
zSxYt0E_&Assj!<v;X-oQZ3)-hx%RXLkMSUi&;hA2*e1ZpYg-};K5WEfdfLB+0n{w=
z>MN29?7fU9!Ey9tZC}TaZB-y6U0?9vH15bBbWTE5RaNKx-*-U&vk(am=mLPd%oS&>
z90Z}6iohTs&1^E6+>^Z?({UU5k1{tERD8H&ns!_h?}u_MkSu(US=UTg#_G|vJr_vM
zd}b2^^B*<_o`T4>X{cpu(koUlqKr$1suoeh3gh@^v@FdI$72CR`=;c;&Gw+sjBL9C
zA{!E;kee7}OE*9sxrK#J+z`M^O~8xp!=`|3NP$9o4ngDW&ioF82|Z$Q6rP;p2Au&C
zafL=~gJ+qLuwap79~wMMLP8h}J|3FK=(#Pa*M8o7-5UbNcjVLG3{-x<Z8G}*-7K<9
Z^!+t=CrKKsg4RF&l-V~0f4g++e*hdCLm2=7

diff --git a/reference/plot-16.png b/reference/plot-16.png
deleted file mode 100644
index c2d1913002842f44280160ef5665a7eef270fc66..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 73236
zcmc$`c{r4B{5CvNsU*HdL`Wf$bx4sV5@Q#VU14OMWEnfHvSb<SkR=9@rLr4q$~wr{
z!i=eGjb&sVjOD$izQ5)99q;oT?;r2+zV~s&G;`h8HrI81&d+(Cw+KUhO_pQa#~=_0
z3ry?gZ3yI$9|Xd9|HxtR&Uo^h-{9Bb`#PF8!EbPFsbfRIucHsO%)KBG-f;RqhS=$4
zM+oE`1a|X=iQnr5Vqnl~NA&*k(mHPHtD14P-T-^t=>XOEr|+YV?Y(EZd{{>OiN+02
z9}afwk@feK19Otw2dw|uEm@c=<ya1TFY)C$Tn(F@cwS`gq}7ds0S83g(_Wk&!(RzU
z)3$f}530iw(6Z1z-4*{P8~=5*e<`gKIi~CMsArcDd+<*u|G)gvb~;KC64rljvhCGm
zQBjdZydc!Itfa&}Lh+PgX_W_YTLKoYeatvtcPoV8ANFW{XSJ`7RR9-<nVXwabaip5
z`jT-}MANd$?V5CC&VZ0;WR9(V=Jl=48Ysi`+lGdQUL1EIq7*IT*v0ta@Bt|o!O0U4
zJMXu}#l_O)Jb?)1a}kFnrF7vNrm%Lcx#XKqIk|+9`|r`#g0PH?j1ts?-v?Ubv;-%F
z$@BQe*(5>k`sUxb1zJWY&I8{_N{4*U*6?KdkNG;z%3|~HYY}5vKZd`rvvvEwANu<;
zOy-NmFnAf3x;b;E{h3JZT}rv24>pH%Q30Fo-UgLy%U}T?Pd`2^X?VEa9qukgdq&`d
z`9?bg>h7tu>|i76e8k}^HP8c)Fv8Xt+VqDZ%_2_ul9`$e4AsfeX1dVfk-0nixk66I
zKISNBN2d8Ci)zIDsEkvHl;wqyAykG);jNWIi9B4%Xf@7aCcWTQtl~uoq~jHhfZ;Dv
z6bvR<N0@Ow$iX!w5U<bA{VZUFKvZkrv@j=1A+o$~PyLXnJ)vm#%KP5?W%*jF{w^4!
zka&gro@I)h7K-=R(D;;$^`v|G{>PX7U?7Gs!2Aha=J~{eDhe;HKst`W8%w@FhCrHc
zc$x{W_8qsbS$aBtj>Vc8^i}#57<Ep4@QkN$gWSk=$J)qq9BBPkb27Rt$t26TL<c1)
z;&2vJo2bHFuNu*HI3t;Kq`RCC0%5mo%PG9Zxn$XR4@I+*h&~RgW6S}aE*1&TJF_kQ
zO3aWA0+CWfogpR9#$~|YK@S}TRfieTCpzb3Da+O%);0LeNAT7^kEY^7<0^tlE6nOp
z^4^mf1_(P^tZ==kYa&FqZw5WPnLL^QIpc`xRxYT&Xxq6dc`(zpwpLuyjvF*_V56t4
zGL&o`axd;vu=w|4jeV<UcOhZt-yY4fzAEkY#ziK@C1`W|NykFdQ%y^?@#5*m^%}J2
zETdr~eX6?Ld~THr!&crnZMj9~%q_7ioPt2E`WZP<RchDL8BeELNhGXEau{b>nAST$
z!qgt7IF7?Z8g4gjtEpLTklZE2j|l{PtHuS`pWOn}B=L<cS0n25P%Z5Mj0M#@*eG@h
zAy+<xO1>g5zrUht&%?t*a)isXE{peS7YsD<6@LfYD=V|pp;W5;7HmLB$rWDhB!lKL
zE55s25X8rN%f_T>L-9@Uu&8kGX0z#@+LQumj-O}^`U0`5achHSH12-$sux>UN|6XM
zeJpliG$pstwUJ%@I;8nXcU4bN89KNtk1?zHM;ez^cQEaP*TPE{#YisJW|8LO6RDWn
zg`Jc}4ZJvNLd^LE#sya9Nu}1(o)(%E`BMmw%hKH7@tDY#MndrWgnhUE91Syg`Dj7U
zLOIIjqH{fFlDO*?aq}lyNGI7O=4<O0DRgIUU8PygR=A*2HOWJMfHz)9zubAAFX93}
z1j5Cqc)vB~ZOEeHg;5?MS`lnth<s6|N#^!_-<%M=1hg9A8Uuu>-8vp^eSrl}ShO!<
z&WLCd4Y~x$Wr6zHqjZIp(t9Q($TMte^A*kmRT9>9vI*LGZRk;dN=OuD@Y&H*jpxYB
zObRhr+kQNLWoav9-{g`{64v%Yp(UQyQ=C^@yOsCi#fvU*jZ*fO@2Omk7J=4QPy#92
zjbrdkRFWR@xe~1WPI_JG1xgT@;mzWZ9)!^D70Ovi*a2rn3X4pG@W>;16<d7eA>D&^
z>^Utw35t{&`*e&Djg+NdK-IEH*3Pp*RjA>Iyk>iM@pD7(pp3j1BXjDVd@tM7Z~kKY
zB!^1W60||iW*|$K#;YGqcS*ovVRhF@lR57<dVZQyUkbwK4}d+VdZr_dS8)nQ4AJ#<
zYxxA>s=S-?=H}4I$bLN%T2xqwmPNIn5U>)3Q9`WuXC%NKV2W?R16R-Xo5!7n+>+z2
z7jglI{O-Q{M)aU(dRp3f7=Ijw?GyFZTn#9jhjY`M8ix<pur?e#4}n||>bGy|E||D#
zf2vvP{Y~|VqPK6awi1QNoij5t)=v{{GPU!R&yy=rV?w*1hzJ*F=UQ9TlI>2%h1fwC
z(icL7Q(q>_t5>fI(nn2Ey~%4iXyQDDq+2jh{}$5v-essrQmgI~$rHX2<-a;3epyuz
z!qsZU@gu~PCS$s}6vE?h)E!-^=(e1NUNcRvql|c7yqr&?V3+o5X&kk~SQ_dO^Jxq-
z`ltw!N<@;<3Bl`ja+FM)_jWDsl4o1L*=d%IvX8-ui}-7`gG|-w$1+4c-rXtLj>kl;
z@H)XsS@H;*-ZJ~<yPuA~I9cJ)iSh)|W?pCrOFJVjF3!4W3s#auviJPCaPtP&k<u4K
z@64KhuEl`8oH0ddk}sbn>_(v)6~Edcn@aQYrukpcxT@U8H}cs&DTXWZIfBg~CRj+F
zoma8%Jr%{mbLBNQkV+YuQItUIHvS$PY4jazLaX;mB$!+f6(eWa_Z~<rEPX2+wEVR5
zecenm$E<RV;x+*hiN2&iKU{*fTbgKqJ}Sw&?!QvBy}exz&%8INpoh1DJ)&lzM{}3k
zM0qW%Ju-$SCK?~SPm7F<ENSNt-Rb$#^$c(PqeSy;+U{VWo)X8?AxBx)*4`=V64&BI
zOO^FOV@@OzY0!<da&(+7_nN_(-+8I4@v)$)YJ=G`y!qGX;}a*#!Q9+5GaKPW%?S~u
z486=ygy3+rH5@Jmz`|IHxaH|ymHnLvfoHdn`=yv#IXSs%?K-<A&&k03;*>RODD4)w
zxe`eVrZO{gN}RQ|w%+AK?#G~muOw*`)Ab9M&h$<m5qK>^9%wxiWd>dEs4y-)`pC46
z*GY>KlDJ9j(>e0v;WF}tna;~BX1lv5j*@Q66!T#|nQ+{x7qTy?&+F!n;&@kl(@549
zzf<FtU6DFEF4!PWIh%JrUhBB|B*pmi{8s9)eE~V|GmC;OEmmgnS9SX!@~3-j#+#Ir
zwhDJ5Z|(MJFO7D!E^{X>dvy}4VL2^PPqJGq)zZ?^a)yV8Erl}d;ni6IbA?&D1x(6f
z;0t&BvPUKQ&t^z!Z+l)!3b>0uMcIp4?Nwa*hNaZGs@v)vd2=!HytD0qoxhjYYWM!N
z39mjStdO|7o7?LL@DW&@4IXx3iz`0sSg6+ICvolVrAoy4EK&P1Gp98VpK&Ycqx_IC
zGo8%<)8|<yj4GE>x|e_04rSauH&XM7i4hWQuH^gCMtA4tP)PcvPm+m-=~kv)V`qId
zq6Xa&98Dw!E(x&PbE<0t%P~=ehT6KiZK9r@-o2~sAj%lz-!9}p_38<pbkYi_ZJDR^
zQOEgySCGId$WL8cv${zB3=YMjBr!AJ>2!g3EkZ^Q+_?26EA000iSCf^D;*<?@_8Le
zYwg|-Hlz<JZa4CF@&^qBTV2VFE^~l~*V1|##$&NGm9`Ag6Qgb$7f`)CZC}?>dmhgw
zvx9;S@a(WNu%(yRm<@eTbcSpNqh>26^0xF}vD)s>QGR)`o>nj;5O$%ci8R+>uyj;(
z>Z%_`7~bklS-Sz2I|4l#+Wxx0!5fKex`9tQ{LYiAqE+|tYJ_1ILAcSk%-WYY@k<6h
z+DF)ZQ3L(*23gx-yvoa|=BI^E(Ns3_tO+zAoRJFX|1Qq3n}$u%vRKCJNLqix;~yF7
zDMAB36Z=>M2B}Rd@mk(%?%`}#SY>@_8;g><sZuvkCZ?ux`nM-oCN9KW{b>&Cn^8PL
zbEwAcXx7SgO>uvBAxVg?yjnf6Wd1entu2izeKRO&V`HOkV)Vx&>SASOrMR?a<d-7T
zcQJ9Yzs4*WJ*k@QbEGpiCAqowh$Z;<*#jN657O7;gY@@`kBQZ_j|6)Ss*HdmuhBlS
zomR(b8{|3aT#&Kx?!YZI^sisPQi;{>@Omfm?5xWE)(ery37NZk!U_H_o1j{^0|RNy
z!A(%u*ws6s5_d{0d~z_e;g<;p!U<bTAGVQ)k3T<*z-_ovCd$F#_T`;fS?}2ePvY*>
zM9;<^wP1j(RA@3M3iCpBPMN@5H~1}$jiDb9dauFBqQsPbXlr@njE{ip{)pa3hBPd`
z`y!<VW3W9u*=TP(Tqbt^=JES`J6kHKwcHvnNHbRoDbR{4y`1ScQSVPet(AZ`b2~?!
zvUW9Lv!62<MJ}$+_U%Vv%D`(by<610GdflXFKYYP2%(|6v9ZxpFbxZW<HQHlcCsed
zuk-G<34+M|bjGtki6{qnNq4gNyz;l}9R?l3eJO<s;X^69x6eQOtZ|6-5D1ye49d-X
z6%&`s0w7<0@WgV@TE=Efr^g1r!oFxsI|G5(52mJ;7)EZ(AtjS5>OgDv$oRZ#CGGgH
zUk%fBh4))<AybrjJb8@|W>NF-hd@$N5_8W&cthSk5s&}qyT6R}U95Bb^z5i@LWN+G
zwzWOn*4@KnVhG%xq$?~rXPG`i?B1A5bnMw#_ZKha-&udo%m`84iXbE(T8*7GS(wvK
zC7FKLf~<UBEH~IFZsfx8B%0>w;*-t$oT|U|7zx)6#{@jGB%EP@hzQC0+2hB?6f9|J
zSd{l2OUrSJiOY9al!lRa*UEJF{x>8xC+GYlssUBbj#Y{x0OLGSVZS~%lmWspLbg0y
z$a;}t#`-Y2FK>WMTMaPe5-=J>Ixf9Vg`NVZO^DlA8WX_7@RZXm)(f^DwtaZjhZ#g7
zaWm+pqb*I6#?P&GXO(o5X&2ScTYXP=Y9i*B_{%gVf^6V4w|brczA;LMWs^hA?~8XY
zfCw8m&r2KQN$I1EC|~U^Z3{8-Eoh*tLWXiV0=}ebVPP?xHfjX-O~SMXv?Ea7_UsxL
zF#}{*&Aof)FV6Tkn!ob%Vumz_6qnDIeGKT?Xo>RHk@X?UX+$2hJ2IiX+5_KU^j|vU
z%5=T3C-&%!c;hriJ(BwbB<y4x9mfukFD8xnZ<cw}VO5HE@Y2Y8S$X*-eB<*Tb3*;I
zo1+h%Ugng!l2%v4R4z1};=X?S?wyK<U?@k^mvK>jh=%<}m!`!}MelXvZ8lf+M=a!n
zuNhTugu}O9Ut!FS``A(n6-BoqVC{!UG-=ofhcZXoNm3BO`9=gXZObjiacb{D;_s}x
z#JY7)C&sWAsi&MG(!NvMuTa72>Ol}YltW1FP+4HTNDBla0`-_I56q>`e{rzD?y4`n
z#F5VW!vjp3eDyzuEV4eJ9LQ^x?^VWQs;8u<h=-V;O1CS{=yp`~Y{%O5x615mrbxu?
zoJW@)ZOO!WWNtZ1tTAOLf`nJwP}XN`R&dwhJ$FJ`^agydN-G|;W*@JR!0*?VUrMXg
z3*0u4^?i&_C>(TMJ-cE-{gL(0(Y57vgSR;NCu`J1aF|*6GhFF$cj$gb)-_08B*@jC
z;O;o(6uL@!j?#n{Rn(CgQ~Z^M^S7@|qI<c1((~;L-K%@wpR>A1o&SU>H9)V03iAFs
ztL>HIu<>Rgd6Y#X`Y`00J&R5D&Bb)<ikWa!$gf4{9HvV_Bfush%f)`?A5h1w1c&p>
z?Qxw`(x3J^0`H%_jMTWNvlj72S@Qiv8)9k?1`|7$bsVyy4yKjy4%Hir*#L(MxSrcp
z8dN;58l~0-+2}J`RiD0kvh7Ka7hLq1v6nb<z?>3^V?TC=8M5-_Bt6ZCgndJgJ2}pE
zzFF%O!OoJ7@g2#vm4S-({gV_B(kpPN^8kbiB&hU~=X|d$7KSVRx(@@`w<PL3q9QkI
z+_Kn;tLHMQT#));#aNO<@EGLmLGJoL6{F%_S8Wu;kBdIq4aX2Am!#ZL<gvGSgYr6`
z`AMCZMvh%|EU3)RN&+U(oOVW4^!{JsLbda~vukFecLZPqGRLh79Ddy)*3CK>-@~9g
z%iTjV97udQgF@(9YqMx8NSx$<G-h9=aKYJ_a80UC!+0dPecVaHyLxqKydzknr7=)9
z-xZ`SPdyzDhP`=z6J*`s;!41=7km8FK$WH>2Kzkrg>A?i5yvFimCiz@UY_|g1c)6@
z#&t1%joL`A;ZaMxE4XzQ?MPUs-9$Q-^R6y1dpaEqYx(*wv0|UeTN{##L43(z{Lje+
z%^2o`ocd3L5LGmw1OC2*I7Z6-<!0=kU@s3SJK0RVxq&y5S$tR>elj95Yb>$dHc}B}
zh^;rCa?)>+n%e?wjl=k<>YIGMJF6L}PL;DVMB8pRw7>bIxw>iTTL&R_cj--qz1|!c
zU}YXk;+G}6rHr6m({<kE-kiDhF0f9Cp23}ciNv8#%lr`V8e+X`S0q=h2z7@_Y7duL
z-kAcd;aT=!`he8CDteUVO)ko7g*0@NB*-it3YPh<YFcg`ZkbyXbn@rnxy+;c!o1nF
z<FTC{D^=)9jh6=@%`X8>1IWgq0bj!P1bH3b&1Q}(6F84vH}r1lPhm!g-2*z;AR*rM
zVdtyew%rXz28h(BKiy_r0lO}mohT?OZBp|5H)3yXtu>O38H~qK{l16tx^TK#wg{kE
z4$_y&L)B53H6le;so;KjtGem5#vzu25Y<fGF4xXO)SWp7$V;BT#b)}?_@C$~;{(a6
z*YlFC!SqBkB?!b*r;XPxx*Pz~YLJa&N)+}JCF4vkkTbm%h*!Tg|1Wd}^I^CDq$?i1
zUj`Vmzm-)bC=(rQi02XF6(*m7iS;au_w@A4pV(+&fZSrtN;RlNyBte82nqY4b=TYW
zVTu)s#d--}^d$TM#O|Q<Q~6r}Th!k?-#<g?Chp3>5(G_CN3&m1O)kle605c9H1L5!
zJ#i-DWI4*7GBQ*uNn5TjLlN0LDJv^$Qdw2C*M>_I5>J+`MD8r3l*trV<HA8rO-lR$
zNLTWFZG9I}p5qcAgIn+UxL!$Jv#pgy3N1!|j9B8hdJ>#Yk%tk{YDbh%$>`BMRLo*V
ziz3EBh79+`qQ!lwxpVXLVgqDvK{6W@t93i}FNg3jY*`Prr-s*uG0UKJeH1ccWY9Qv
zC^o-(!SW9<VcKG2fH+E^%5Vh}9Mh((5LM~BiCb?muhbK3vAapU>r++t9l+o*AdQZu
z)XkS-g@{CC;PTHe?a$d>oJF?tBHa=y9QJ7gbez$ByKqnkbw{f%fI_IXwP*<Nfj8k}
zwNYs6F#=&{g{&!NR=N<kwgsq+WK4OEh)bfk`d;jDh?K>*3ZK|#Rpw$=3i9D|!=DG`
z_N_NfAdso{xl9jg%r(S{?gl)iC)DThmD+K~PG{H6?2`~f-v_^A)YjM6RS>SO37EI}
z`EV3wBLu5Q8{RTK^2R5789ux~aYB#&bBL80BI1AAac@k0#SC>=u{`%e<R>gncI{#k
z1oFArC2MS<wA_8(6LEr1uEQtMN>S;ftx&**?0A3Q;w!}qe|9sHeXPdQK2f)E#{dfe
zgalavHVacQ$G;oa_NE?&9TO;)u+j!}=li;X@do#q^d~u?Njo`aiz#KYto$G9So+%H
zSs`Io=+3<LCe+%ynsRV9+@=qJgM-7V1}msEN+8s`?(f1koRJ8ra(=lhSMu}q(l>tN
z^K=4}FPqDJC@){VfI6)e;76$OWMx-Wgk;N_@7A4wEv-~d?hCl0xANcW-xxn%O<U|J
z$h>sxMc`5c^)i$eaw3njCrc%?xg1Pr$CqHhIV569DtBLDRqE_oo*>!908PlG_7s_D
z{?Vbcrnk4(Qm~V_ACR2}piqHC&G#p{0RaJ}^5wQ54-ie#E-WuMM%g5;MSyHR)UwXo
z*|+9emD}i}`{(2DN)eVtbWUx%^N;rTFC#w)F?89!608l(8O+q_+xPb>JIO6s_$F7K
z<;&(ox;$d9+kL!>#h>Bp*w~nuf`Y=HVYIG@e>v?6Zb6Bcm)FZ4_eA=lt7(9piwrHD
zjHk@m8QUy;SRIDDwdr_X<atrP?JfzD2`+w1)cMg5_L{rpeaeczT=iwpukX5)4T!i?
zlygZvr#83FcqzELy6!7c*tjyll%Lm0z6|sB@gWKn+TdB}oC)vSDl9fI2Av_(T-qwQ
z`=#v_4{O4pG~qI!1i=O(J#5ci;1%23ef$&-Ke}ucV6LH|q3;6_HUH&@?1=?^pLrqn
znxC?UHy>tP+M$Lv4&-@m8~BVVR>&_K)fP=M`#qGH+vOg56x4$jv?+*Zf(H7J6G<ZE
zSIc=fKGf2#z-0Z5*o+7aQZn?so&FUNch5qxyA=caw5i7~oZ=SVwP@l+G8<CmLPDnr
z$Z|yDwa$9L`BS`|WzazZfX?*!5=a+dRJU}tPZf2B{7UK}rZ}1|R=cWgOv?ChDDi;E
znUUI}cuZDKPNSrxL|&(^gP8vA-8&L-a8UNUDUMUZ%I_sMR1-zm#U&>vr$~PE1xG}q
z2-()S(4-PmR!dt)N$~3Di$sQ;QLDLs)!rym?TDNadvGOnMge9U3uQH&DNXWzhKa_+
zw!hAEaF9U{2nE}{t;@{JMCb$*|Bf&`tSm^Dg~^n=5_fA;SM3sY#$qDV7TVJ>4VuR-
zW@Fz8(;|ixPwww!X+-)O5_a97SVI(U;WRi&DZ0;x2L_;e_}Org`M9;`=Rddn@_Zf8
zCn8;L3$k|USs-b(@j`MrStevg%4!26FdUK%Yxw8j%Vn}stS%~=`#jN8uq}_IdZlB^
zV0Cx#G(rEyT4BAkHBDG34tciAEHE%I1JigI!Q;xz&^iLvZ|vB}cb4lZdu~HTrm--E
z;J0;Pxe_<GN1FDK8ZV!n1ZAWR8dxUSJ}uHECQ|B!dx_m1nd|?c54%$~R-UPBCwGp@
zH+1Khk}Q7PF%=?vh|EHr#84wkORVgS)*&R=pM1`6&f|9ydf@xb^Rg@b6B8kDc8y@2
z!%CfftYPtLPphLBK`!YuQfeJ#R%VlN=KfvNr7R}5v0Fw{0taE*(Yio0JTbAewHhi)
zUiX{@%7yem?X})$DOhK5PPq`#qC=M}6VQ!rC(Et@MW=<k-m$2<mS(>Q@@Gq1TT-gJ
zmzfc}`!C}4Jyq9*7i<%w*satT(t&jB!WBjE_Euh>$)rc#X4MD3li!5&5Tk@jO3TU+
zcfrOtkg!jYv~4uPFM$JtUpMb|L2rhlyeJH8uWPbN+91x#2T*M-Q#I*8C|a2^hMtWJ
zng>YR=D85}{fdV_KB)sD&gK`ES4~R`h0OxIWgU%zr0Q5&SlBc|)LmWY-9SgR@#gOC
z?pP2j!$=n*bBvQEZTBK8ANCZt5c!j2jc?BbwmeZgZ*X)ga=j7Z4%pnqbNqk?hDgCv
z7X&){x+;~s#?=4lL2ZZOzuU@kNM^EMci~sQH5pfsq-~Zt)ZXthfgFo2SB#{N(uljW
z;DBQnR9d9SYZ$HNdn@eneYj3XpK;|P@eM||Z(UiVJxB>DxiGTVc;o|^9Fg|p6I?_(
zPgc2ffY~1gn5q*_R-yE8hgxf$_yfb4I`fNdQ;0IFtB+`rx)Lg11AR#e@294d)h}=X
zeqIFk2aksRb8@JJR)i}{6g!)Z5s&n)(hFYskaHNco_iYl2YT8)TS!mBQnG^PKd7*b
zeh#-%XvJ_RWO|Kubrdxrd}1jvbXAT^=`Vh5K2k2qGpxNGPERw!`eTKF()8#0U!+Bi
zzi1VF(8c~2&V!f#J00@>&%#1Lf~8*~hzV(@w1Azf=(@zWo4I=aJ9$90g-G22`6~T$
zsmn^1aLJB}o^8F3XL>C=oHzE;;}$29fiQWR6tv*_vzH0dd{^~v;kfFB8FW@OA=Cp;
zF8lVQD3Oe^i_i>Oeckj@6exuDhyPBAs!hO^J}Z%OM8{Own6o58?^sv<E}0z80R@Mv
zJ<ER-%IRpl1X~GNBs@;lV;Sl+V7~wYtUU*(InN<F-%Lj9&K!ffCdyjl%i+F(!U?sW
zUm`g|yQaA~tsdS8HZbUej#2$-%w+Gn(uWcByfje^krnKABt)BXVIc$6c=x?3*iN3f
z@8XE62mLu><)aG&@C|cVS;FJX+d#Pe>jjXt!7o$xYHyxj#-W6CE_q$k2Lze{ED3cf
zpou;~-%Vvk7Z$FE<1T3dc@rqv2y>mh+h(=26d@m-y3aTlJlQ+3)1fF=M4=Pcm_F$_
zu}In1#E&p@gVhScOcaXfAJb6{rf2H05O>$$&o5E2K*PBJ10mA~IpIIM-5C6L{ikBO
z!xee~<BY7q@Ym4vD0>Yek;>NZztX$WWpf*?a!^Bp<-?wbq<m;aGf4Eg22xur^VN=+
zX``^VBTcpGWgzB=6u!e^X^32rfOU?C^jXQ~Pp=Dk2NUxuW$lkWhGv3<eTX?z@A$T`
zP}IfUeK$<-Ia{MVVe+y?#jiz5rc8NTAf?s`c`1pzHjdu#Epe~lww_t-D<SU{QbN+-
zX8qnh|AiY5QUt3HK<~<8H5!%<-P=K)76Q|AA?z^XOu$sUri+lN4n9*U@QUn}E14c}
zGnj%UJ__zzh7E;g=&L-+w!dsu^P3IXq=#1mi3ReNcN?Hl?Ryp~;*}+|Q_y+~)k^+A
z718;LV@!l1`4Mmw`?A%67k$K7O0YyHkyPa~e@9~N;PI!}ECSUE<vliBVwvKt#XT!;
zqav}h3C+)BNUt~7vue7ymBfQp)K?CjAI|y0Yhq?*_5~ciecZ)uIN#;%r~~MM>K_#7
z<~1J69Fe(9Aik{Bc-eX9v{+`&33lQ#)1~fxq6w`xrH%=JR0X!=>gHDeM+`cwpz7nZ
zUQOJM(-mHwP&*i%^WZTi@A=g$+F979VVp;lB9qiHL4OH`3*ac8|Je(PKADm}I*l51
zPAOEZQyu!zUbIHz*g6e6>gbkSw{YEOG-nbMnU_4E-SEZ8$Vl4Z71rSUI;VzIA8zIg
z?}0C;)V>J=7satBa(>-$*x04ravMCdXMu-EBqo3rI5;%q^A;U^7r=n>LKBxCdy|}!
zHbF12O?|l1yS0xTHRC7g=Q5Yj6JckAwWAHGo09GZgtEm>S|!vop)QA6h^Gz>QmPYt
zP`_@&KgtG7Tgzesq$A|UU@F0@&jY3dl<bRXqdtI8@V-sjqYDUZpLkJxia=ySxOsR;
znBYk0QMu;|SG8lY{O8FDfV=I}{&}CfS*n)SE~J;3C|3>`U?vfJ&KlZ$Ky{NVeDOm*
zL)Qooao4?JTny*-J@0$*N56;{okTG~7QbYBe(1bRc^QZ3$+kGQPyO%>GZec<|9)Md
z4gqa=9bR>ja-6FHgj)e*Xd`Juh1JB^+<a>iz^Hk_jBkLRt^A|3Y6~aWCdz)n^}4yZ
z9DCk%*V2;q9uv;&6O@=qr|!WP1A6LS0l?OM?ZD<@1`u^hQFV)=|B6GVPP3Gcw>K_%
zO@XdUj0_L&A#m|w9|r;I5rqY_OiU+SG>{n2`{sdNtm^7axEKnUI3(%wyyUSr4Gp`v
z#=~V;l}~+V{ved&LMJ0qQcRBsxbeM({|~_-MG!rDjUs{Ew+Bko(3`;h>(`UCeW=6x
z8bDggy|4qe7TCzYB<g<Uc)-Hx3`|)3XMj;eGW3_?*AfMV5;_YRhXw}+br;4<-@fJg
zB^wsW(3r!pLI2XhC1M0lL0j7d?_BKuD~#>&_xt}ztW&V%KqF^iQuq!|CR?DQSmC3{
z-rK+9PXL1l{jzfaN{aFS0}@<C>GK<#(`vsg6GoWh0r3uL{`Et`m{vUSbiM>7+*zWe
z{jINrBAV>S2+4j4zP3UjxFWAyxzgB=<4g9g#M&w~r`2WY6bw+?*;whfi_E0Y;5t2y
zlH|!fk|gI+Hmf)8!^Gru)={JR%qVF9UF;9L{!ab?P<SQWHxX?uM3#n`!_ozbTpImH
z!63%`PSL3;F882r&oWk{%|)>^>@HPj_Sv)DVR>Gp_lI&22~SHut;Wd#<`}h>G*R{x
zrC*Sh=)lBFH=8xTP1IVuDR64{=ItJ~ymdaIkYA;T;FlSgAgTh=|AxebqRgcXLGm*z
z7i(pM#d3Yghk_7D+x35GyVoY07s|Q|lxt3(z65xqZ{hz^jq>+q@A=21-T|sT<h$d)
zHm2H$BIK;;(w^F56o4Q$|Gs;T{NdXCN+B|ePsHRj1oHl0|C6eI-Ki9%%VwG}(*{EM
z`ZRMo)dcysS%O2*qm*>zbDiNLi!nG7qDsLSkcp7!|7hu&&Tv^ncTZnRMR4<hf&7j|
zH@g1*Mei`A`CDSNA{aVYZuNo%rLG%`?R-3Ss3kcv_zto4#GcU1U*!6sm=F&HU9jL1
z`<MM&for;b6ChhQ*n%J_e7XNm72t;>V3hxTohbUZci_K&NkhYrPl8q}mMI@=68;p?
zKE-Jc*2ZVS-Q}*i-~8GyNG69B9jdEvbI08Gy1(}~3C@jNmQQHUpcBXtL+(^S#i<_c
ztm@hG*4tRw*0#;-@TC&1r$27NLo|WRW*@rqtm`VE^1!9dM28*PH{spII)Y7msx)hq
zZpyO@b&W>#Al(Pkeft*pLc>me2X3wE>`+RBoZYL2`gY}+MKAqJw`J)?RF48T68PD1
zYmy{vylChXY-Y(UP+vd~#o%%bqmNXm+3Mn~sighiQU7@>uLZR=6sU3U(Vt>nNB?vr
z3|rRu&c?#>?PdDAVJmbRH0)&Dk<O$?J~I~lI1fn&02x4|{@H&TeV?025MC{tG7%$Y
z6kC=TeMQqhh2%;0t?wCLl4#mliL9!sg5uobF<TU6o!iwIO~GAF5TfYn*|_4@Z~u++
zOEi&meupkeuBZSzKjn)*J0xr`r*OWMWw9LIy`E{d=nWHwZD%;1T9QuSy!NLX*?`rn
z)&;}{5O-aW`-#}B^JF$yxQ@j+avMm)duFpj_ttjX^Zt+KqFUK~hU^Yp;;Xq~n>xkq
zwW{J60-~U*+goo6OS-SAR4bnOj6SHa+v-oNgOtR`_M>jTy`|RJHlSv`t*fi+!s#ck
z@xa2Lv++G}aS8a+<SB#phGL1h(^#4jAh?pq-U=x1m%vz9sCnCkF4>9fsULu(c4Bjw
zZO7v4^%POn>o+}|G*0VB=kuATq}FY$UU6T|q|?3N(q<}`?KV-!*dy#w#VxCfE%CV^
zmNsHCC?O{F91CW!F+}ie=N%UmKl9x$P*5BIK}nc}0C}#~Yc|9g$Z+T<N=G2g=XA|^
zr?0s8Fw#8AV&wr27y7(g5xJ0T&8d0YKCVd^W%Gzix_Ol7;4|;ne-IKaiu=y{JRrR?
zSdZAcat5TE`>PW@beazTlAg~%!k$jKLk*5t`YOX$`0IiqU<hyL$GFmCbTtSaoD{g!
zwGQ?+1HBe*Ztfz%o@Bal^Ww@GtmkB#2s&u1{mDR>^h~iu<(w{zzZ>})y_hPGnBX|{
z@?dlTbs}(S9e&0AJvWrrQ}4gpXkM(VT4-76GQ=s=15(0gzn1}Bl)jEN+6d}?eWs;V
zQjpg(422<-5!a@eCK_CZ@<P8MY*ST2_X3&E{sRfi$!$@Lq?0H6t1A0YrN|r}K~t^m
zxsQ(BzEfV&KwgFttWlW&FHFc0E-x-F`k-QXD1M3j$fbQ@Y(5B^(ceCuedd(u4_~Gb
zv68*W*VrXD1fr>y4K1}%6PSdCp_1mkMEMP_FMk#ic(&c*0wnBc5X$#O?+8b8C(zup
z2on>^%2k3FimCok=*YHx%(a=OvGH+vPMK4;B#~JYv^qgp`<Jb;#G|i9Gy|4Z0R=R~
zztMe4-re@$PfDp-PzV<&-eg_l-5jUCF#eD#YX2jcZ|`!vmV>ZC=Fa%P5OL--;G0lb
zmCJM>t#{K%aM3dJ$7_%Umj}gE{_8iY1_}HAddpN~&VzVN9+TJc#L9$Rk!?qUz>D0v
zItwgviLdu-+M{5ch%<-%kJ%#X(=Tu8oiFcLXFtR(Fzs@iI7*V*Z=0G{7E^=IKYXvu
zqtWj)x!HEKQuLVQ%-zKzm4%QJ*U&Gs8~>CEi3q$hVwY?Cix%_ELN<>_b4^UYI>CWD
zx$9vdUk;t=&9F?GX8<fI>et*{o%-zt3=fcL`MAk(sva^ZHG%+GZHD(x{^+ltb8SP2
z7@h>1f*Bw}D>@9pH!>fw3fY?GNWbEqc1ELrow65tQ8)Gd)JnIciX#x!m7enE{>>^H
zO!3#?zO=%E%<WEZh2Z!7!>OaJCpONB@Kcoa@z?-FRr+X`Rr{@pJ4zC?#Syc&>s+gx
z!dmf6?*iXgI+c*pYZhRm0;8WWd|x|J6LCxDpxvpXvCMHy_Ho!IWBeNmUyjJO5yAgh
z4^G0DdP5}<!0&W4?^@KUw%4^3<pv~n6j#-L1`c4=Nm%%l#S*2vXW{vQTc1Q`C>OQ1
zbpW?!3$r+-8G{+-mn7?58_^83_2F;IJ*Q7m*1*>`$d9C*fp)dCAwxy!81q)2zl)za
zqIE+6R-hALppen%^KvX*$oeOx2e|F9cB3C1zqsY!OMHMplttw{&{^tXa&dL->vDXJ
zNb`cf{PYNxaKdoq0lQi*kkBSy9_0N5EFu!|-rZ6!s`>!sJf}6=@{JBEqcQe^J^*IU
z1YNmRYXSD)DO3zUNL7-1(&d+O&TD)U=?A5<P!cFXWN2%msPUfsjMjyjMjC~dgl)Pu
z5&%^U02Yis4zSS6Vq#(pVA}ypJ*=R{wvX(V@NS&OOx5*~ZB8l<zqG%G-a|Uh_GR@6
zUi?hocQ}zqzC=MN)_G;BlRsMVR9OFG&T1oio(Ph-iLZ+3zpuj>ulUjAkwPVY<h&FK
zH)rQJwNW|C8>8)qS&^M`o^c-&)YvP5JV$x}%_Lv;7KUO=uvo=e+q2ui%rjf@QY6f5
zydjY43>e|3A-}B5%{%8$eKjyO9eQGo^_{LKjH$CKt<4oitD<2hFKFs!Io)|hR`5%_
z7Q1f7AmLv>#CjN_iUZWNO+<??<-aq2qWZFSS!`<@&R=IdvJ%uxr!|FwbDQI35fk;%
z8{!9#XKMk?1swE{>6vI~Z}z<?>Gqg!?=OsBnYY#5#(f>yQ*(sRvdq5r>V03D38MV9
zALm_j4ULVB9pIrCVQ}AlRhp%xWoU754>53lexApflJhNS9LL>Bif?~O<&dCj&0RMJ
zM_2t*_!b)cv!QPT)Et&R+GjcvogvMsz<Dd<-ExR?5V)^`E=T5AiiwG3*#J4f2IlhM
z!CPRwOkMrm${I?<O`LPC#_a!6sj~H=-iN)TxAmg#2#+Bfv40t5Hf4oS=(j!!^-Q}0
z;3=0QnC7)tn&QVhJ#CY`pWEyzF6PFQS8fu=NS^_Q4!isHySO@VCF%(#biPwRVutkf
zy_BTgP~S&%grnB(vuij>*fs{4?SBA7cX?%5tr*M+#P~&uN@J5qq`rA$&=x=Qs)*D~
z?WNUvoIOS*GcJy=fG&Vay%+_WPtsa-qtA9q`DoX3VU=4HY|gX&JI2HNW4Z#AXrN@C
z0wi+<$iZg0d3c<36jUN}0B*QUHiYpA?cTvJ4h;<<B|w-Be5tO%?SPJ9Qwv#?KgZum
z@#&gTer$$^`hJ>J-_->X&@TGDXrEjBAF~mJO`cB90)Ji!rffZA`ZC}XM5nJIPKrR%
zS8&O?<HbyR0XiNRb_$;B0m+oST5|Dh5fPh^wud`W&RD7faDABcv#4T*mwKNTwP82Z
z_n*`$$uyO60YD`p2dvc*qgCd%T~$2+3JzH0^SMaQPs!$%AAvz)2zWB0=$bX;Ct&te
z-RNeT>Fu{cUk@+!cDR4gvu>(<&;ly?r_s6h_1C1$3{0O++5}>Ry;J%0#~+@g>+{8a
zc!1Hf*@2|a_U>~<B`Jb1Q@8d#1RJ1Rbvbg!B)tm09}GfyK%+Zc@8zdsFz5*@OzRWT
z#l`YiaT^)Xa^gv#Ub88tSIscM_S>uVvx;=5bu}h(kA;2`E+L*lf>rVJWL>TSEK_H)
zO88*Y9sHN)y+H2>!viCQYUi(plPk$|g42+DN6DYPskQpG_e;HCZ)D3)P<giChd<r{
z!}~##Mw5a)ID4(6VrW(X_$d11pKK)Twz22Uy`_P20i!UG=0HR||Gs_F-k6^#K9GO?
zg@Y;YFz01@zN&rt57Cz`IO=}QOdC1=505Xq41JqupZfd3L5;4x1dRK&#ErWSW*iFi
zl<z-38Xa7>%<V8nHA!iiS9g=_eOyCqE9^iC5z&8Jbo80$i@g5&{8p^sBDq@3{<aSL
zD7kanr~79Y6BzI2zXa&_X1B2yPC~m$DoVSpMf!XNBJqZpf|89wYv9@e@OtNO{^HmS
zzdj&JUjE}T$*%Yh4+WU;|CzLaK7iZ)7pKGjKEwGx7A|xkFhI^;CtY71WE;(!@qH9$
zmRd2KmnEEEYJE>Q_RmTK6&!nd7_s&7|5nORoxkyaU`j9PdPlR<Qy{5A+Kcbo@2wuy
zx{*q&Q`a6ZB%4zM181y(uO_e_29_ka43AvyHw3oPLH~%GIRFr0X^tpiSdmn@Rr<#N
zpiln?Wy+ky&(vpq`oz-SV=I2;uk6+d6WnSmHcP3{H*bIZsT*R*F6-BAa^#IBUIvOa
z!GCCg`gsA@eCZ>qq|RKjcLaI}*ttZ2Vy9{3jY0=kv(-vJ`siNH+3o!S(`b%mdZgCz
ztdEn<57V#6C_0g9jU@V1W3V(2Nno<CVN=t$r&C_uM}Ba_v;|L=VS!b5gt}-6$gdPk
zyP&o5#|UZs`K|Ux7FHudXbGgsqD%R3=7IJ<$RDEmkMvXI@niqxt?vrQ;`E$mOzIH<
zOE=xe{M7^iYfitiH-R|V_OCej*IfUz_vyXu$)g+!Ie}Kn>M}=FiVZ!rI~xG9#-M|N
z0Ek?gAOhiKBidWo?lL{9^9WgAHL<>u-mu}02po?&f0N#cFr$dWOI)jTbG>#^GDazI
z!|j2a+jSzY6y;ro;p|zkLGHs9GD4>88v6j-$qJ01n$fBb$G;zf$z>oyTG&tUyD&^B
zqo)*uXd7a-#&2~&_VkEUg*J63H8$9hT%`<DcGo%;eHS)qKcX{0{o>=*h|qf&D8p8l
zgA5F6jNvSCn%5*{_J1Z?CxxzC9Wnms>C`=R3Lx!`($Z3atSC)hpFv)57s(`Amn-n&
z=)z{>A}|5Um5vB!6C7pGtvwqJWZCiyei96=r_|nhk*dFDV)?tgoZURBQqFkH?9blE
z{a~2ij6qD@AOJyXLb7oG%isy{3<@m3*4}>72Pl#IQUt_(IHyzjXdV+M$5S$_UN7kf
z<qfltqnw^3dC5%7^7TX<HY*dTfhLKFs;r1p7hT$8e&`*34cOX!c+?v5;PL$1O~W+{
zdfMBQ9H={Yw6!S>X4d`peQu?6C5;^1yBp0TuWxN_tv<~As6(LXr+@hUBLc<Iij0P(
z+rW4^MX~LKm)k~2pOCV!O6-A4Mk*eRYzMCKwCfH6QpbzI0;r*FPN>Wa1&`nw`-wy)
z(F}b}P+xf3Xt+=(PB~pW?{pjY>8Eqi5=n_2C0CY1L`Mo6A@}7Zf;3Nk%({c;5B;(m
zZ}YY0H{+0lq%p=_I`0otm&gtO9V0F9EK?*B6P(d94Y=_v$~dx5iMjx?^$#CBbI{fo
z$pNs7EVHy4nlMa-=W)l?FutCze1ZG@le>ZKtarbobmyv|V+uYL{=vXYhHp~M^DxyL
z9!+d&el^*o-w;B>9lU7kSy~tt8S_x^oYT;civw+2tOnnxk8#a?qt6gupZ)IS+*b)i
z^6{HlFNN;xUW1y~^~=GGEkr7uA78rk`;?K9nz;kasB}8=N%!>66URp0jT|bW1~gpy
zSZl*(RH|4LZ}RSlJA=l}$Rj2X(mdGHo;knC3zYFvZc>qXUeP0yFoo-Mu<2{J|AoGm
z<<*Ooz1SHv6WHiv`MV*U@brPf!i|U2T*01>H<Ug~V=g2fI1(4`I+lT~nMu$+R{kHu
z*SLj;Ft3jWSr3sC(uWVjpP+YD0Hk;NB(<YScx-s}23tKF@^G}3te;La7D1^3mUOx4
z>1maPg@v6+OiGr|usH`1U&;I2<BUKO*uR5EFlDmWY9opiNe6=UO4X^B_ZoFh2fg_?
zGb|bV(X#ceS4DO8$iW+5g#lIU=i}pZ`9kwj?QEtQjQi_ZJn&a5ElnjDz)#qG{$$oG
z#(L4`dQ$MPllyr$w3Kwv?yJtR<Sfo}V_}!A>(&}j9t&kaLU8;6)M)fm&Umf7+bRha
zODiL+pergTo%piv0vbh3$T}f7AlS!8&iDfZ<JXCaJq0MueuKOOoCuN;a%^G$fUpn&
z6FgZM+<lMBW3F%pbSE$bSYA*VkYgL6Ry=;6M@}eUFHN@0>lLk>b2IS1LvVuLQBIlP
z=yZM;N`B|GFRauy&sc8OQG#Q%%=kv!3RJwmJ1kPps(Ft~`JWd*2#?*y$HuxY7QNI9
zcr@Gj82L&Tr;+{r0IapD1w6C{_WShikY1|PkI@64>4#(3KT)7h&Mq!vz#RQv8&7bM
zxn_EM^|4&|*+(6l!N`rvVq)W_-(Ly=vuirLuwv=aJY<^xIoM3T)M$Kf#SrmiqLZ>m
z!tjiDwbVP;Q86-dR1@~Dtc;Y)*I4@ahr~nP41)%-1^<BLdpV7bA@49KS{%^vK!|*m
zUYwNw4wwd12;JM&Fci9^W_M#;@mO-vGe6ota`$p38DL8rjJ@heq3Hlbjl5@*O7P%R
z>iGjAHBt#?wdNs!K%w^^sC!4p8iQ%Ef=ZIk_te!}IZt1{E2WR>RhTc11cFBt6}aFi
z`@<t4Qc@I|KEcp)#u<Y?VtJH>JkFTqb2}aa<qR(kzPH$Kx!;S`OaQ<O1XkVDoHtq-
zo^5CBFI~E1Q-8*U>%~zZCqQ|)xoyt+%u_(vs^4@XR$BwOoeAPR^(Hh>3tvFo9bjQ$
zVdDD<1(rPO%}p~A>-3q%;itdXth9+c*GaOptLeis+37DBM9x<Z@EMLn=LLD=b^N*D
zA?%*#=`E}NV|NhYDK@`xBp`r@3)qpjsrOrUfu?)FvrucPz<=T-NIZM6r9MADe?R6*
z0QnfB%G>k_rv@_N?MDcgGA_A*xRZNrNINQ_(lvzNN0?q`^Y8W}6X$(^NJ$xNF{94r
zfkO$4+3!>~lnU_+Sx<IS3cH{5dJLcHoUtn)x|u&Uw|!-aHk4XHxFo}g=6(Ms+!~^|
zTA10>XfZo;Nw!L1ImvI`w}zP`(k&F98c!AZIX|CYdelJ5=L@LQv%rd1h0z-hyZ8K`
zdez6ZPnh)f%4$bwwz)V)yG_`5eazy9hNu=&7nXJ1nyjSg);55}4wuHr17;m$4tMVz
zlr*%9=v7il5gqAA7T)~$b0i)-qZ}>dSSenZl7^+1f-J-W-?N?Xrg0y3d>VD)wZ*DX
z;}YwBoN1XuB>fq4aIL*OEXnrN!F_cfws!-q5GYE1NVY{)idS>W(uH1dMIK|ZIVT)3
zmW*I}%qz52iP{r(&UFwIwB{8tPARtDZAiWF7i`$2P71rKMi=M%mtaDHQ26idr!M^G
z)681`^)&PO|8fyT{x{3P?SIt)07(AdC~P7_`P++|otCj_X98yQQ2hDlzVQXjJf0eq
zqo#?TqNf&zqy9o&yPGTG<*Y{)7hV{mk(~}G808qdh}%0_cy)M+MNb-SviB&Y*-lF1
zPfWA>yzDglibc__7K6k(C?%df^`d2^ccX5A`78@0OyJ+)N&S3=j(I24pgmWMO*cLL
zY`d^5{n{JV4?w+n^zTqFL{|3{C3_bJ>Q5(p>iS5%m3{;g_Um7Ja<s@9p*bO;;@CG>
zBHH^F<}n2Fl1<)R%qZ%#_%XhKUvIo{3nyV0wbRZyH(-R|8zpURZOL~T4?r>;;PC(O
z0)XWWG3Oe-G(ssAOv%^a0$RD2^pom?8T5c5cBgnV3N5FtV1PFP_JA0a_j7ES$Abr(
zcEtkp)6Xu9;O1(x?$M{2bO;jm7KmmyIvpOxV@?S+oO2FQ$S9@jk3g!eL<b94ZuEHd
zUYqTxk)f}MW>uk+KloP*<yY&0vzJRBXk+=TbT=qraInI=6D1TVv<18gx4|<WX0SRl
z)RS{p<$M2#^HOYg8uGt5G|iVu1hQ83o_o)e3$i^`6Ho(Rg7-_o=l5!I+Mkpb@;JZ~
zj-~$bX1|E08+(&&nCs_z;t4AE<a%70lYk5PO`G1ZcC60!D~y+5gBl*n#8Hza1iQy2
zA^AWC()_;KZS*5+VHSxW7-L%tGJMB^4@Dk?aQ#RYxAcw2bOsMCeETq|2o5Z4T@&2c
ztYlTs=VEw%mX1Pxc+W=H+EUA4GUE6>E4$CY2}zECZ=?lPGq*k*05K2|d;XN+9O&^m
zD4>>9x^a6;KyG$Vz2ZKmE>H9}p8fQkxqWH0e0CQ#qZ9C8@U0PJh!=(O_#^+zTv@k?
z`fL4*g}^#Favu^r2Kei!?$$*s3&5-lp>D04Cg=*6`A})h>_>kMfv(~j<`1|*nO}@=
z9b}IJoEoxerTqIW9&pop>iGIb7h(O^epu|%;yqKZ>@xlo=ceT#<83#up12^a<xk1S
z274+)1Zp(q9IfQQOOs1Ub0HEOi-Gx$-N_k?l&l`4A{0v^hMv*Lzkd&ou7TP)<S>}!
z&@;3#pplem2hbSaRcUWqxw|h`BkIWcy^kp>;DMq`<op(ggOKAYu}T|vo8)peV~2~)
z%a6q!h8P~>3*GVbSK<9uSEpcl=gw)7f22?~sJ-`Ht$|5jg=;S*R@KhEhjLi>#99_x
zR_g<7-$B#5!S-OHURmYcnSMgB4p4v+0bT8I{~#h_4m?oi3NKr@H*^qo(G5U_Cm55*
z_J7ivumeXbH0+^KiEF!vK4XfGoVL6x@?qh|!6i0(#>YQmL0q%|&)o(RcW=RX2)mWQ
zoAU!a0`6VE)q~*UfAK?uvbXLa<G=Uh5HlP&a;sRUxG0T7xJ<5}zqr~6O3rYf(E_ZX
z`%7ScoPz6IG9IZX8??QiR@Iy}Y^6eubB13E^fE`<I>7_)_cF4-gP~kLc9+c6&6U~V
z+&p}b)!5Y1xM93t$sK=*jg=`!18adLu|cl%^33?jNtu*h-M1z6?dx-czkXXDTicnj
zZgHfs>^-8MLA9Ou{{9TChw+;SM&uV&Kk>Bk3sGsO*Pd;u044DPq(4B<8o%ld1LZ_n
znCt>m0o$~$=Ia@w*FXYreRA$Ww{k-n@I|tKzibhC_fnFDB>5qX*VNRML%Mym515XT
zg<049dKTUR{b}DOTT8VJ<j{SB*5?3HECMo&=KYw1?0LY+oriJ#Xh1RpcYZDWmMmet
zor+0WHys!qHTRl9J1JXcYrTA2-KQ*id@;!DU4nlBmhb43@9kH7hU#j{r;7kT>vMx`
z#|Rial0}X4txlRH>$(!F1%tD*WKtX+)0VH4O#cX$1P{osm7|nZ!dmypgT@BIyO1iz
zudKIzC)0<U;>2mTK1k@P)3auoU>0gHwk8cSK;qCDyf5x$<-Wx&6sk=#KoXyY`cS2V
zIo!spUd4M}NCf9_i&$3^|BH}J!CFUBX4(XIK#j=29co@Se0>(j-i9tu^>d-rNb%bI
zlrSf??&+e`%DV*{oJdi?;=}}G_}DGJ5nR<YmL-e;_(&z%ttaba6)aDhTcr{)IrX_0
zA?Fmp<aJN>0(wWhATwaj_GfRH-)0**SEh4{UpGWeQs`){@~Ntx%yX4icfLN9UK#AK
zM0?M8L-kTzE!kdBT`jsVz-w80wZlX1pG%?+5Edz~XiEN*JR`76Y4HQ9qQATjp175H
z?#M8lGZw7l8+rzE_qZD6u$!q|4S!C>mp4++uEP@6hY3l5DixwVbQ*uX1^l3=LVYK`
z=~@zB-BW1jpQ`TKZ4%nQ#r3TfT~$YGaKDpfj9#$6cU81ceb$mSyBQo}sr0`Mkc(-8
z0omunn>RmdjP<VxiNfg5MUQ{|`Y;0HT0Xz8*odJ=+pycyAwg)}ly$|a4s8BES~Z~<
z$cOureZ4$9aD|labsOr_vMM>U+#4oVKTBf7?xNU_;bGezhLsW~Lp$AmdRlZD8ERsU
zbhefDAY7G`m3`2cx$ALxck)XVC+W0-8brj#x+!RBH;(;sVo<5BGg1lep>o`%J-E+J
z!-+26KcD28)w^vTP_KZj8jUVMwgYz#cY0NW?qj`<_2J(VuYRRp55L!LCd4bbehooM
z{py27yjHI6Taph{<Tr5~^{s82-q;`6k8`qu9BMl<adPn6w{MyH1+36=Bo;{QRocgd
zO1!9+PtN)285<8MeqK8ciQ`0MC`sbHuU{HX5md3k->nr+_5O%M93QXRA2s$W_>yr7
z!X{3g?3Qu&NpZ56&pe*MYpKRM@}kpdymhz^t6=Fg)SkgAa+vHOb*)dY4Vxu%p6P@C
z+%JpTy&<o_aODfS(E(<*hkQS8E6EVQBD|{S{W5N3&wDAzgnf(dK!p4vfnd6!T_;C_
zK#rrT`HfD!*kFQH2AlBm!*}ExJ!W9^TN_3gAdyKb$f-@;apqUPzaM;hrnAa;Xcbyg
zu95=qYTw-h2Q<!;(N$6-(#{f{6{Bwc3V1{5P7JglJ+Wgw|ABB|0%m{lo$xn`ZAWya
zv2H-{$bKUqQc#|jJydLyfPh1ue>nm?2`^NCb@sjMoAF4VPrB#06`o^o5F#RrL@(*>
z*s}Hzce?~>L3KX8ip-~vi`<fP-4iI3D;O)KOj-YvACZL4pEX{(HfU+xJ@IVmy73I@
zL#bE*Yw-iG5z7SQbVTp$H^|RF)`A-=L*-&q0h)Z4^%v`8Az?L`@!8jN_$MJ<x+vW4
zyG9SVBg1_LjZWP+gfdE#p+qMmYZG^HQEE-qw6VuAucB(_bYG$Ts(Se&#rpZN9X?M9
zjZ-VnA1|dQy{_8kWMn>`{WUd*0TR~2{caT1HuK@jbz3Fur~iwvw~mVH3)_Z=5~M*&
zS~?VvZcwSAk(87U=^7eD1teu?5Ri}#r8`6fq@=q`x<MG?+oQkdecrXc^{wyx;aUze
z=bRm9pZmJ+>$>)q{+S%<U5L8ZYs8Ejm!P&KNAEWCqxW1y>a)Sqn<nQq)~ars_UFcq
zT<g3&w;g$<FEkE<Fo@qSU7LPbxx$6d?hIb+EZC(^C|>Hf)I)ITA{Q|rFTczl)4dje
zk`ghYK)!c>&H<6rMH7H5jIXDTsw8x8$WqFLG}k|&PUr}=GHJ{ltx+tK4pb$|8~S`&
zXeb-%qj}Amtb@Y#_}|ICRh%kX_%us(1~e4a*gWq&>vltdU_Uy4;0bXxJJ#`3X!w?y
z;^9ESMAqG`nujkW0cSh!6s%u+SH~yUue|55^cg&#3X~&mvA{QGhSgy@ZX1V^P<jWM
zSmsd>yzY6lIo{x_;$wc>37-wLG-3pd55_R4nu3o?*(n!7{;f+8Q767prN*$($^a30
zTEn`FQ@WcS+iiXBeGr6+uATLKs<~+RmXAg#lxG9tOf06^vgZ>Iov-;F{dEbMTJ%1s
zLxO^_(XN?dj<gt>U%jv}g+vPL%v`<vFfXB7cAhpv!xh-Qs$_Y&Fa5s$y>0&lZy1D4
z<!|e;CacyFyrphjNxmjR1;3p~X6(0$JTP$XK-iKkPPZDHO811g(nsV|lz5XQP5CnB
zjSB1`av?kCQB;)#kPqMgPGIHinhPq^Pkfd>t8E@Ej|gVf6jdE;!x^a{^gdRuGb7%y
zjdKgSDY>_b<7GF-S*~E3#6}S{fu}F4B6TeA!wE;=S$Bgn%i}JIKBi~Bvw5}RGyrZ$
zpX`L48u+g@T{@d>gj(J_fXZ9xzT>(cLi2xss04xYO9f=`Pmik8>{*>V23Y+Nl8k<X
zsRx_-Fr)ml7uwi<O(iA$MLG^KVt{x|;r%bwFOA;d=vlQ6#Qi_gfe;XxCHfn_Qyf@Z
z_Yb`FtYTt*sAmhp;uU1!Bm~$XA=hd`#s#hOkjM)G(;uo|c3x#Q4wrfmuzXuo6muWT
zG{eJsseo?!$~(nD>zfdp`d0G5_*6bqe)MhJRQNz$B?_D4zl+ebfVpt~cDSDf2j9?o
zxVVG9iM84GtTApxKY&7B67pt)tG8;f{=56q*w5Va<}1pc#CHzd;tBKo8;m`+{jJe_
zj6utuzqtPrpn`s1@Ku^v4yznV&L`CqTX{y*(wrWl%AxVR{Z`Y!oq~X^Zr+tBk?iXW
zOt39(ELm)aL{j{BA9rXcH<+0%{YgAZIo$^R%ILEoxP|UBJ9nUTNj<)gc<T)(5P2`U
z$hiP_@$Em*N0JPd{-2-u5H=S(Cfw;OoEr)9Y1nNa-24R**86m8?xi@h!D^&zq(MX>
z|NTN=wlfR<6G7eF*s7>9D<RMz4GBrD9Vp|Sx+^lF<aE@MXO9wOd=T{)Ct^D|nZf2~
z&}&$OnIPH3c0L&fkviKSp|KVyJlX&7aM07V`B{!aiI_(jI{m3yS}qWoPkpr;&HeMy
zG+;5{`S#z}Q28|3YWLT^c0L-PP)P-mwf3uVsE=UGqZ=#v66;*apr-%6NP(WP$qltb
zii7EGO`=2zg0d|0B6F$SN@2B5@)VeDSV&eFILuzgjXUC?ifIo2oG-rm+BV`;QKWrg
zvV}+Q`8w*dB3B+!%(uBGvs)!DcZ$}g1b4w~J8-_a*qHG2^ejuVfM+<_*%iGP0elJw
z&+-eEV|^0|#2;sS=R$)&g2#&#J#(J}ldo9Y-AILIni6PSHqd!^1W$+3=L5dqPvP-S
zgF5~G`s0LC34Nc}AcM<Y62gdvXRH7Rw5Uud$Dky41RpLqv$|qQnPqvdlLZ|_j-vPl
zK(Ut+yZeZ@@qf%d;bB*~xh)Pp#P1JCF0nQKNj(F(^X3ls=&xS~f-tEU!vgS|M|<MP
zra~f_tgvTh00)U4t75R!<!3y5(syc{HmoOaYQLug*m{rIe__ta<$V9n6QnLcj?F1e
zJ3Uq)c}4?1Ysq<Uvw|h<>rp+J^~jUm2>~fVoZv+rsIQo)C|5ZsbQ8g7A#K0hbO~$5
z{W)+?@AGNz_|C0j$I*!KrCFJfj@B3setwPot-2owsJ?(Gtf^;?s&8jjY`@9P`q{H*
zvw(2-tj~z7X3SqcFVMwf4(fJ`8WlTL&0QkZ9I!d8Ha|U4rYdndY9g0J!e$gBagsow
zOnn?P^`2g{##hY&@uM`qY{23tpTt<|)I5;qff^~6%u-q7o~Q^`@K%dm-_9QRwDjFi
z^!Q)^?o?ps_m1ZV=ET+qGZ#o?Ss-r!KTYOuMsN075Q^lm-mrkq3$n48-Y|T56zGz)
z^J%ZFg2rTI^5-euy{-~_uYipvV;FOQ6C@<QcF<dv<)l2V#qB79o0efrRln%85W{Zn
zQk3S;*T}|NrBwaNTxovd_usYm0(?w>MphB%-d;Z-30VR}1cR2DlqoW(CPNtd%MV^T
zI@uhYoULnn=D@&0R%0jdc)g#z9x<k|L)D@4#i_kd708WBT|!X84Ykh(E7~)ZOuY_Q
z-<EII*CVpaQtsoYUHP64Evh&h=U}P)uJO|apf>VR56aIIY47BVnD=?nZw3aARt5VD
zuV4SraU9*Msoe}>al+eRFyqOaIT`~x-{WUrM_4ocDjF&)_eWfRMBeWO3Z#rIVOk#C
z_<aNCa~$yBDVMjCz`v^wu<CFvMtZJrDdRVQq{-*nl8OQs*x3ATIv((Kpa|LLuRMSB
zqJoLE^YN^IJcZHoirHqvD+JoJcK$H89=|l=rPPnxT`|U`ZK9+;FPxY^PO#})35x9D
zBG7$|&G|ADnsNpkZ0{v)znkM5YuU|3({<ruPz|-9d$T@&JGyTrn6<$0NjU1KB=*4N
z&1$O@xDopE`M{n#AhYH4USzwZk^|HbfMG>62;<WL-<?q;x)GK-g%i6&-H`yLOfni!
zn7(gL_GvOIfuyeS&T%upY1*5e7FRQ}qK98KeMg#GlGYKT;o&J!)8|${=U}NXVveWD
z^exxBNY}+UF3aPsbOg4h$c-7#ujeD(xDr3KW#a$58hUFRYN<t<w_6uE^XKUTNu}ZQ
z&rJ0QTE1<>$ottSO@fgQn6*_S)929NUedF~KDIJ``MGnC*9y8N?*wN)K_fwLKLhE8
zRUlh117x@jX6>}8Z9q;Vk@NTnAn3l*Wq=l<>tJMl-uNE+6N9tOYk)$X!UPD_zXP!9
z0}V}kArDA6e_ro(Y06K84CndPKtpqU4HE7vlhNaYF}8+tbxZpDgYPi9tWU9y=%3f+
ze&iY|dFZQZ&XF+`EirQ5JZh}1-TAwac+tTv#(B_h#L}APOVKl{1zb^>|Fo63(T8<)
zj)aEiV5PCJqAM&aYOrNHYXn&(G3HYd%>58NDr4WR!u%V5wrI@nrT-V)#RQ_kdZuV<
zq`%a+*M04rwn_key8+KgLcDomb1UnH#@TB+WU9PO)#COeApZY~EtSgp`t@)qI0U3s
zqS)1{3QNmmhj$EriVd4jeCD&voDpJmf;}N5@u{NVf%o_2*I9kbEMJ(v91pEjONppL
z_nIk5dPh6YIF<WGC-vuQfGo@7OPK}X>#C;$-i$}2u-hjl7{E-73WxL#+XI3IgxK+1
zci)*D(VezMkoPmX?!W#v%L7qg?ORRJX?${2zVX~7aq?#B;&|hhXUh;$rBwFV-z514
zZX36Mxx3Wmhaeb|`{uj^Ua6ri=j3pF$2~<j%zrKU;#J4$?{dXUb(9snLa-z?0wGC!
zHaLybhtZ8tPDnTo=z@{S$v;K&<G?@1yX&oK-SgpbkkEgR5s3i_Q{eHKzV+GW!<qGI
zK_4FzxX6OG;@TMm7*4uN+wyKc5lW9U_fFnX7M~5gFBjgIJ5ZZquNvsF2Ty4o)RYfw
zFUnXfI|C$$FZ6q$An=W6R*025X$iFr%d|$^Vfw<V=eE4-)M(+j)nPEf#3rwM-3s#n
z{%d#L@RiEtXO%!#rFd_irexPwHwF_}2Qn<_l6=F`Aa0v9M+a1C#*Zp9RZ7E+ej__n
zPjR?-CN!|0!lr&oFR3|frFN|hV8z2_9FA$`B@q*QXX^oqnS5%2CG=T5_H|cZsIB)u
zFr{35>HY>_bT;xe6hv$~qEDOn{qwwZU*GDy;3b3Y`MHtLD{f0QjI7ove52Z|PfLW2
zZPgven+FXw@?;>dPl7?&<fQ?CY25R&T7U*YCU$C69@elOth`;qcQ5e|c@iC~=I6%o
z)S2=(pN{0jPKMH5He{B(KVhzlKAO0sUkZHB3wSFOH~&J*$2@ULuL4*%VJ_C~2SmyK
z-0rsoDa1zsm?q*yTLJ_mOrTOgQB&9RS6YaBPiLe_XI;0)vO>}+V3sI(z88-dyfgoj
zV;~fA!KsPdy&efo6L<FAOkZwHj*cl0f?GtQEO3Oxm2HaY1hxzHUwBjeLCl_tx~cGt
z@NwN5ePQ)CUj~6QG6@d}!TcA#^59NYt+O&45y1c*nWyRTztlW956S<<Z^`}F3h=3P
z?Ei7{pvMk{ZgopWM~U=?b*hXbuMG~FcBX_*8nfKL1Nrw85_UNNcz^{!vQvt`Lfr&c
zT$OrWJ$#r;0+5!`2eths)=PK7kR}1r+=2zKNVpmPf(OXU<fAi@axAeOmy{Oz5l$v;
zMaoyhP`E$m$I(OU1>ucfb#CCa^pbxA*C5kw<N}qoz-}u?XW8}6vSm_8l}@|B5L}qT
z_}}e!s86O`a6`wg$r_CBv&aYXT*`nV$=d{|Peg8}f<C8hd9DK}&1|v&x6U6YU*=5V
zq7^abzIwMsj@#!(NMZJEV%76uN2nU^n2A}g89u=1;4Tk6_&A>(;~Y-{PpY!_9AMPh
zY^&k4iTtNQ6TQsMyofrj;Y_CHn;=R}UAH=+xd}}8Jp1N2njObcjG;cLcY(9u3gvLF
zD^R9)k#Qd<9RJJ-2Z66*BD6sXQwWxG5I{3KoABG@EaLVF_->o?J)av!?IPfo3w~H#
z2#@aJl499Jhpvo{Y2q?P!TB|ZJK?l6t6wFVZFN0O?o}dccW#C-kq{A#JJp2f3u6QE
zS7V$onRqbO<IDDrw_08p9q)VA$`6O^tb{&uZc2jc3!x*}*^y8X{N{ezc+=EpU$+J6
z)^Hv$PuPs8m@O-M%cKA3O@y!S1^p|vgk-&)$!C7Y>n-(JESTrG^C|GXfG}H^H{>aQ
z03Cc;oExJcsxiF!KysWr#4C^)M48~^e<z@nV!#Wj-8l)a#<sjt{~}~{M|$((eZ%_F
zbxZCFlM(kO3N)=UMop>ce(;JoA3A~!1U_K82bppJ7+?A;c!bHoWl_}a`T<t(y}0Ur
zrR8qWG{v9y;PZ@yhQmbgf!GrdP@8F?FH{+ByyG3Ie9Qd%t`HEFfYyDjC@fA?6m6eb
zr;WwubBo8~g>HMz7ERQ4BK?5lKSt)BPIWmk`)ds+z{_+duoiqf#Tzi>-_)W%();r1
z$J~aJsM7E@)0m>qcweGYLm_y<sJX0zQ@6DvG4nDWU?))+KlPZS-rd?c2DSlQEyABc
zLHWI{jZO0Mo<!7{aNpbl`(`?LRjB|Dja1;p(c0$I4YVL(fa9zzh~H1>|Gp>wi|;*e
z>jX<JU@&lj+lGYoxWMsa-D5_#(^2GL1{svz4AgH$yp)uY(g#F+O8Um$2rzBv1!kY9
zQ1iYWRA@S!68JKlPbKG|O1J;XV<f-<i-ZRSx>^cC&GOR3P2e5Pc!q5$BD_59#eJCE
z-;1<AfG0x1(77?I#9B;wkiw%A+;wX&Tgmxpm_P+=^=eDEb(+C;>2<k~fWR)?-51ox
z@jd#pjy6Glb4}`gC(g^<W9#(+yhlxHM;!IWm)f#$r3991-@J+)Ch+3KO*=f<n+!%S
z5Ym3pNu#4`{hrWPE@?9AAfl#8HBb?0Q8NmDhA(2_T4D;JH`~i+>Y%@Rm&*Y~2*4Vp
z1U64iD~{}tE1$G&z1lD^rrOHi15)0fS@@vH&3#r0ln(SMeuurmal~z5iMD}OS<vIl
zVz;Eeo$HKLglw^Xbs^!@o$ahpT8<e|;6D_J2kQLteOpIQh_{`c9i@cl&ZPPAq{S<L
z|LeF|3{*2$;Quj*9sH>%vw{_naBIn!2q7<dc;Sxm?!~#sOxKmUnoLniRmQ_;@BkLu
z_YHa1GPpJeW>S`xmgOEWs80nbjQh@dL+BOBwr0`fchB@DwBx#SJNTIWxu><BnFFhW
zd$ZSr*T6T}65gT_q|g=y>wXa02dG%r8L+u4onaA;HI!vDD<Cy)yex7#uN)f#Ade(e
zV#{vz=HcM3;O|ebn9q-$rM6?yZZMZ+ktybC;K(Wfx0}1Jjv(hA*ueW4Am}uIMFzV)
zA^g$7v$1DsRpT@|j1Mgay4uy{4}FarEY05bWv9NuwbgsrT-@KW=982-U@)P|P|S)q
zfT^tm<o>NCtqq4OT~3laUpfSFS1F{4gr5(8-#NM${%0!}SUp4qg1F6UL*GP4XWV5}
zLtcW+k|om6^5%Rf=PG~f$)8E#k?7o4e%8~`b1kec8ov5&=eUPUc1$WsOnftodua#`
zW59`f40M+{mqmqTWdRK!e(}fRL+yN7l3PEu*1u)vm*s9>TvU`w2g8a~j$C}lWV`h3
z-Ft9<H;T^z4Hz#EFqZ*y-{_umj!g3Ix%ZIX;eHImBKMiU>peTg(uNi_vbP@TdT63a
z_`<Bd3D*G<mh@ea@V+4pI(n?5#^!g8NfDVz-D$71ydvb@EzIp}YZhhC?w7Htac5R=
zx}=;<VhmBwT{}ZV3EIo@tm&i+mtkLsSN04*p&~XSt>g)^%6zsnb!-lk8tDrNxH{}I
zA%d|$*xpvr#OeC84EjJm_w*%Ti7rRKCMx2+ntB2fyO#W-Zp}ZRSxJG{lNL{K?e|Ce
zRcv}+%{mE~j28Ez07u@*pvp!=c=B2`g?DWnlq+xDu^6%y-kR#spg)NK+J@`Bn>?WA
zzh*yf?T^25;>B6LRxuzE-?Z$Bi`QY<GTdt!?z2+Z`UR2qtg5NeQL@`CcLd7QTcFcJ
z6S>1Pn=*AickKto?l?GoLMTo-_BHrrF-^7_u#M<|T}}Macx9QWAxfRIQA3v#HJJXH
z#us10CXEMoAMPHYmBMorc|5562q#IBqp+fN`s)&T#=^Z4|0ynin_uT+Je35E2y!GN
z$IQxqKz?Cr>Vn2&?rpyAASj1o01MexDtp0C$!|^nq|#&U0oeiKbz>mgo-!L-j2J`?
zX>D3!;C8+X4`?*+iVoBV#FY>HTzw_ajH2$whXJwA5F&9-LK7ekl}mM8+TN=EdPO5A
zeWT_<m4*9tCZaB8rh8}Uz?b>zlf$Qt=YYp=)tCG@^JmtH<A83YR-qP0v(r?cr`&1v
z-nNyFkL6U1-{?($uC=qq|8N1Ob9c2)|FjlueF$yx*zSoco|!*P+S~5nG3>Yq^|KJ{
zZ;`RO7KAna`5rz%H$cbdFhp?2cUFr7_~2Hg>~8LFdjeI6R6OujvjdhrwCme+pMl7t
z8DV&k{t*olf(-(94{5jaf!&$f%v!&TBlo}=ls1KXqx^yF_Gw03QyV_)-tCs;(^czI
zWc-u#ptu7pGNeF~Q}K!3i!+B)e&UI<>m@;l<>rks)w=!h)v9$9Qb8Iy5}fE>Ye+JN
z=f~V9LSDz(CMK<V!^=<UK!p6fle&qSZD#g5t0A=!21FqlN^a!Rg}iaBb_SAkdR`@~
z3-R0ga0lmF1oN%oq9Ess8Kl+;l_CSAfn}ikdzCgOITKfROhn4XkJKr?Dob*d<2BtD
zQiE0b&>R$)H^*n2)rXIOl)BOpS>LX6Ly?}f2f;kEzcBFA4p;ge;Id!YNNpkKqu2#R
z01v(?P^|lD&QX1n-cLZ>#@Sb;Iq}r~_%wTqd;$L2s2QFXCrEMvwkm8-0hg~YoRo9D
zpcp*sPBsn>b5j86nL7}omp83Mi`{5^aRzS(*&mkW-Sy3!>{z>c0Et0cT#dbj3s(Z4
zZAutbK6D3=YQ`N}j&nYpFfWS8T$|>?FkekM_57V0HFr*gezVO}6?5)eQr7HCcqfYJ
zN*lb~c{)h&K0o@PH9o_L+u=7~4lWX(eRW4!PmfFkAJ=UuHyW7s#GDJhD=TDSG@wl_
z#p>V3d$U*wiiQir4wWX<QKkZq0&A$(flQS+u$ZR7ZcN<*P<}e77{N;%i9LR<h&}~p
zJsDW-hXb#1AA(YmpW>j;J<h0}ZJNw7Z?6a)zR{^SwypR9O1^>2cDjK&@P|e>Txy;V
z<J?Q=>iS!<8XeDBgB9~=P$8dWc!H5BOA30SYlhmV(BU(g7i}lF%xqPQCZAi<8vUK2
zx}Rs(n^fTTKYnfm%Uo<*c&0-e&x`pTk~VSPr!N&E6OV5qK};0>*NP4hTZjc7sc_9r
z=ti{)_YF%rX&2X!Gf{+)6PQSZl{U8c%(m{7y=8iF`jh5;VZZL37B|I3pPa@UcC0rO
zm|{&Fk3CMcbtWoc3a&u5p4~=*)Bv<WQhKhZ25`P9PZ`12H$S~vugnpgbHcaNHKBSY
zgNtcpJ~}hoSTCl$>%po}e0#L)nC~#nygC159|+4y1pYydH!sJwXM!!9;ko5(zsAvY
z%3aSR-b^Q+r{X<{Adu}@+YU98O8K=NVg^>&mll8s1}@B_FjYt{xfF0^J=${2_HM}w
z<1Ernr!@a7j++3~XYLb>1U>zY8i?xTDE}g-Hp0gSL+qPh0f)lt?$6U*=kl@-g;qQW
z=-W$#x+REb7%?^fse|9zyY}Kd`L@7d+a-OzbQG#J(`1JQ*cCxU!T-|XGJSUXH`H~;
z4#RV3lTCa?zg_c?m{Y>^AH@vlS86WrLi4?6W*K=iE&}0E8Z2?x!QWdOH}RV7Yr2fc
zU#RqE;-rv*4j}l4>eQtHJwOz?eI@~uT7#J)M)u=pKC`6Lv$MZQ>!U)$FsYc<-u77l
z#Sb=*{}TUlR=p!|$5{7>M%3y)uF8fn-<1In{D*^lGPoBA1xp`^;v;?-3PgI~Kb#6a
z{r?4t5ZUmw5@d8HaxgyC^|qpyDV@Q#BG)rz4_n;j&6s4p<T&!_WP9-sX<ROXGRMG;
z44bbq2<2#)3l)cZS1)Y}`iejaLTOnAxvR$Wf5fog&n5*op27X0EZbk(v>hFV@(rw8
ztJ{Xa+5U;$rJb6t!zg=N8Io3;`MoO3Q)h&lcPAgeQ$o_}{vj40QY&I|ManUySFpgm
zt;n;FRDZdQGF<GNkKh69wsE<0a6yI3cezT;G_)pFZ(QO8z;8bZy-ueeY@VG}`*2d(
zeFQ$Cx6@>jVA~8sGnRUY5SBt**+A9Ecy!DE;m&5Bi{yDLtKXIpL$U8bU$B|w|Bkmv
zVM%A3BI4RrEwb<2KN$huQ=&mo{i3q<y%wq+$?P(srn}I4Of;OEH1Kc9ytMynEx&Md
zzo=f|E+v#lJOnYHv`fXvsoK6<r7YLjj_>2rs`c$srDx~8Nt@)HE4Y*W)KKoE(FdDM
zP_f2RC8g<-vO5RE-{iMx@aBgttTX*RRe<$bb*>xoRUfL*P+3^`-g^Z!r{XCA71|Uj
zh(H@ce9;{va@4>?Q0L#NS}YZ6*dC_URm^8mCdgRY$n?k$@OH}TIQ_(h*zlijtrshC
ziK`|CwF75WhFBYqbFta$=g>Ijgp~Vufnewy;hF|}k2m*jwo108q$+ItJIrVSdoXBe
zk^g$6wJ2$QPstt_i}&(XJ3lvlerhX!Hkec#c9|!8kK`FEZigzgHf^dC9^Ds>gr#jL
zeXEJTn-t2%;H#K@<BQxkn*<o|7tt`H!2t8vKN|h^wvTL0$03gIsFwjRK25$NP)F%9
zBzKykay61Djrq4w>qY-Q^K}3)0v-@^(xJdl^Ueap6GB>Sys%D%FZQ^E?Y5$4KF6{3
zSz49+krT7ugtkt#t#0IF<hK$w;zb5XlzY9=DluIBCBG@uPjYPXLQRr@ie}|Lv^Am1
z5Ws$iFiu0KY|3`Aq8W3xSKxO+!Id4EK9Y~C*P7ci1~`%cKWra&ZV;DvO&btI11oKj
zj~JwZZ;I%`VTq6NF1*!A!Kd0dHE&xTnxqJ}FgS(lD2j3=S~Q25q3>tGGX}U8ST-{e
zg|~^hFzY01I||QsKo#UdT21x<$dQ96nrehOA+?Q@S_R5&3MYt#+W=+Y`Tyj&Z4MF7
zw0Y>rt<tnXtFrU^x7X6ih}wcqfghLz9qGNEFKAC$@bU@zZaIMKT&O}S>*g)guDiuE
zGM5ytno#wyEND)4Xi=hk<|s;JG*FjYz-utceNG({>34}0y!Z-`!FK(}0B@!mZbiyQ
z%)^k7l*HX<!Mo5~Eg7qDm7hB(O939~PX!hh1k7vK<W}#DRK}LpF`$71k}WJ?oydfy
za&LNi^p8;?n({UHy#_aHt`y3(xxK4qj;6BP#TY=EcrB6#SBe6LPZHgzyJH|&X$~^i
z<tu$?B=to*nLTKOpMeXb5D<*l4o1D97mjoz{Lr_9J~&U>-&Y1?`XVu3{4Y22W4u94
z&P{Us_fQVE%ID|h-BQdvS1D5^{u}T{qz!30jT_S^y=E`Vu*S?E`>&X-eL>2*(<PKG
zcdem$m%rKLYp$DHA;GLuwD7R{B*Uf|G;MTjr&N1_y7_B(b~~cTvA6z;xVd@eh%;0e
z$BZ~+X=yr;Ur0{4GKs6n`MhU&fsO+EHNJ4tS#$edL?bJ~EIyko=11=Q)k+7#Rah}L
zUoq(hja<Rr20W$tDKVWD<PA*k>^1Qc5@`^IzMUbxe;mggwO{{@ImM!3m*6IX`CXqN
zR-t3BziOr~o5rk#gIVpGVF(-Sw%^yR9K<hwkMJM+9=;L2UJa|I61wgJFj&Z$BYVsR
z_41g%^DFMa@^9f_l2!U_Y8GHaeJ=~Sw`-aNg2lt0@TN3L#P!UQgI-r9M33W3>7%Pu
z(@ma|{mX@;2!W+d@k7(CJnhdt0L^QAD|NC}*J|9(Dqk*zi0#f^Kxu9zFj1Ok*z7T^
zodam?9b7W)mA|}v*LmL#r*|lNxHKsF2v(|DWCjcX%Op^b$OUCA7eE`62StL`4<{+f
zj9Edm68B&sb+e@=H>VygFC9Pslp;(XaNER!Hd^Xu{c811x?NVq%Us`s!pRiv;z8=}
z7(P7Z5y#Z5F<d&eQ45{c_#Gti9*j<qvE9MI(2d*9SuR55<z5Uuff<j>^7qngQ<LXN
zvR{MIK*K9g?i6W#f9Ff97ROkM(FvGXpaOWUFkIA9Gibq^CW46qygi$RN%bf~wt*JI
zO_Ce%N^4|{j*HfED=XdZg5fLD;yAUh&3EuCMMDEMm-*|BpMnF-@o=d^K$S1=J{`tf
z?)dz<u@xJtYeDD#;=!iGP*;x%3EsSkVgr4nUqVB}y#Iw(?V1=vHyUsp)&PbO{?G-w
zodCZM(|a99;Y!Ma3u^-F)62t0^jNLtniJwPx3odAr7bNtC?0soT&VPBojlk&<P<*e
z!M!QhcFRs4xbEr>t1-{z&*tSl3S*;eQrSvP%z8d-!0fL#EHvh3a`W6{&RZ2O$~;H=
z1XzCjQfVmNn^nJjZFo-{u`<=T7w4t_yPb8!!DC-V)#!TVpRI~CuHkdHGZ*u|7p`O=
zIqV7b@vYqeYvZR}SV8f&|HoXMF#(3<JVy`UQuGwv8{C+VHR9t)pGT<SJs<Y_=4EuB
zk+1Vl?SH2!8ZjV5K`#?>V-y=C5m)F?shymg6?g!N1{w7tSlK8vH8c*dS~g56Sb&k@
zpuLUF@5%FqMiT%ImoqzKi6A^Dey2O(JSY1cp#)z2yK;eh-BOde)ic<H$e0ttBdrm+
zIJa>J>amp?S1Z)2HPJ+-E1!%e@jnuY?5j%<?C4o3JJ%RD5ZNs#$MMqi+^i2FLkY<D
zhebq9h$@9kR*(KJ--29yC=19k%7IlY0u?DXgcWTspg`1L0`;#()<r#>jtfI<?z&yZ
z>YNyv{3?iXSTFA{p1-b!J4UmC>?(*!v1H~J=d5FL1*X6in7cB?T>a`;4@-ZEi>Wn3
zT~nZ@uA_M%A|hbQui=-Y{N{Jn(pb2r#YIDG`nQ76Mz}}})^LV(hpy<m=PZdmC!XiD
zN_i<WG@B}w3q%ft66w0$8K=lg1%3PLUGx2$^Eo*n*~3d>P^zpvE1f>N*=UKNs$B_Y
zUaGe+6#;4z_-@wYWOmBvA*JKGvYoZl*LqWdVD;{7|9gA7`$?-?w;Se`$)lDUmS1g)
zX|WA9t_Ieew#vspW5!iKg2@5;J2-gw?~E+^%RGPKIpyQ+PhWRI4te>#+nq^Ouuwck
zMK$7KB$JYRQx~6QU?P8IOvRbZOY+phjl0aq-$FX>AcJZ`_Ow)!<_Vd&oCxi5C!E#t
z#GM-238mJ%yaB;u{N+)+f5+ve4XzdXzTv6(bWJVb?6FB?32t0y9*RI$a_)|F+8I2D
z-%J_AT*~|y&QA@dB&Zf4x^q5x!kU&N4q_582sU_m?TuW6$e)Vb#knQ<`7|k*8qsvF
zpigL9@UYn5wK>-S+vRW1Nb~>YxM6ujC05g&&dlzn?hD?tAYNdT{`+SnMFPO=FpM(J
zu?bQxt{zLce=s|e^8xlGr07EeA9_r|Qg39pyRvLbJ>Fb>gX5Q!8-s*&zUr8+wH@WH
zm)d0L+yqaRl6i}*>}O7s)*{G0V^*7hP^6I2>Cdp~^B3vXyfi64`&QKa58{xPWnhc4
zzk`c18M$sfd-~%9Z1aP8--%~L8Mgc3s{9GV@jXBWkxag(^PU^a-wb)>=!_E`tK`U~
z`w8#f566$*9b<GRx3Ph@o9fp_AmKa{J7MIXns(DQr~1JV8<kl!xG#3XWW*-R->*MG
zx8^&rUkZBqzM)-A9_bIxSSa^%Q*xPHE)}RTl9PKruQxKy3B)7Pz2^k>J&z4Wj|Zpj
zub;)GOayuiT*d_KNQEBNfv3rI-9tBUw7+cIRQubSlM9ZT*ilG%M)MgX9m2f50|i3(
zrhu|Z0KWIZeHl{o8R>x(ZDr!S+Z6dusNTZgKiqD3!+PF%_szDH>)Jjx8Q2!$f#)jT
zL?7&Y+hW~y|3W@<N|?Z-Yp2|Qw;-{XTa8qh0c;ydF8c;nk_*h}HqWM20S9lIs{Ig&
zOE&4Rs-Twjh79-Krlc{^kRbs=_A+e%l=pOccCL$`z0ciGd6*SD4i`wvWrh};pK=Sp
zG85hNKqiH@{VOECXvR^AdaB`&<lRz_W4jEWQ~%qgbgHsHe>z;`(XfBa?S#A<=&zr(
z{}DT3?E7t+v)qsEryi)InT`ug^kJ>B0*{isV<hBJM-9e0yNGt|-GE+(f}J5Ko>n^k
z**cWB_t_$Xwf6Q)vR<t1U1qiglgb;y`uDB<u6sw82e3%KnJ(N6gX+-H)Q^XLCNeHy
zAhkpbr@S3qa~&HW5kW!eF(ekO`+N<Lq$t~OP~yLIls8`H@{EF_RY@1>F(2{f_c#3B
z8rVNa%stl{RoJ9N!@;bUd34`O^k|l;^c_xz!Z+|(M}pS-XV36TDWl+~q%0U{-Pb$N
zOyRgJ;v30&uY8dHbPaWR*!5z~m$`ba{soB?0(jLM%2{$Ue|DR1I|=KnOC<pmDA_}n
z{jRx0R=F>M5M~Z9)4q4`^T0Q3Ny?sQpo0)csNYe|Nb7Dss5PfJgdv^NH1q>INewJ7
zJ<3kk8w7regPIsQh!mY-tM-ReA{Zv6f{00dob^@TA@&OG1AN?VgO=Lmll<i_`aJga
zsg1yTnVoR=Ugifx4YlH}sNHsS8Y2O)^XDN16QVfx9`kOjdvmTPgYcao_^Re{QO(uX
zYqP2|e>!i)3Io^TGgg*pT9`rA`<uVISXuuYl&j-DdoG@(4>D*Rp|EECt1%r7TqfxU
zDfAym9XiKrGD<PCYo>KHi^tKnhsO1b&ZJ8{8Rb1uItiDxt4WJbA@%4__%W@C6)GB!
zuWq_So_4b|Lal_RK!$0eg;(s@ul4w?MbALx!bQOQu&|HEQ#(^scpb^8W2v6sZc3q^
zOY)taLjzMvGBTr__0(JE5|AzwJ=~YfLD&#xNYT(C8)M8f5h4w4)?gZqNX@rTYsHfA
zGzN!~b$pFnpI!E3`>57YymQp`=-2aDOSsWnmjfr76Eu|C$m|G!4ML&q@gmuxH2o79
z0g(N|#qw*|xJ2hN7ZK4Zvnr_<9s~oTiX}1lWR=nT^9muuR*o@-I*IKU4r31<@@!i3
znS@SX`Kx+i^2W<sY=#Nc(1$X=?<HM%jl@{H-54=2C}L&a{;Ao8IHEaIg-77cx`etu
z5ny20D*niJKnZDr<O%<RWoqi_dA%)~$9+i|#3*Z#;4q^GOfDr1G&G1Vuq!ax*w~WU
z8IHxUsJ2gi7PgdcWmGk0K5}Et-)(Qhe}i#5?RTx>WO~;BxI2>VM$kxS7P7dm0TlU3
z=~n9$HIGR`nhjLES4QX-zJCL)VlG=Yqw4HD90(2zBkt0)G1lGTg$#ZdNeUCzSoBy~
zT;4l{kfBcE)BkR}Al7;Eq<yOBk`hvBnOT&<%CKQ+VPR<fVqC9$qETo%K=wQ2!>Dm&
zkt>tHsc@Rr;Jm<9EM|*SK~vC~JYoU^BgusX^Hi29@N#6%CW8SCF7n9Uc<MIS_rAK9
zdP(N>c`8~BH|jwX+J=Hx9E%F&Lx%EC^>x00Lbx?Y8<Q_7B^8)!N`a$&sth?s>HEDS
zqIGJQIe6{`TT>>%Nh@#lUJ46cS0!JBV%?!5rq?U3kvOj0cDxYA-z55XXc%Z&t}>oy
ze&UgD<Dn#sKNsrHXy%-^WkzM?7_z4QVzecF>;Kf%^#~6F*E$s`pYSd2S}8>P1T@oA
z$*HNLWc?9X%9&uh*MotK4j$ab#v|tN&nlHp`WC4@jM~lu=!0zT?DT3+-PadKPc()4
zTh^MrHWXz39sP0=SBsIcG2(TUJb)FFJ?Z&Q;fvL3hg&AqMPq$>nS%x&tK5MO3dQ|t
zN9RwsEz*!a6o__n<LQ%_uF9E;k}ed(SE=i=ki7_(02!Csn}+Q(h?n@u<o?K?+-EW}
z{*xL2@V`t#kA7_Lw*R=gnu;?#LabU96*OIgf%4*Dwvp7wfGSiO!nf4*b}2?Ox%Q$#
z<o3OX<NI&#iJBttC-*j_Z|^!?o@~l9oH*%%{CQp}g|-p(r9MT+Xc%>V(Qtx%(=jW^
zBL&1<h^dY`G+giZd?k?J2>@*{QR3a9BwCTO+hR?_q}35QTB>)G7e7(&!U-lmRrQmS
zBe8C>%AU9Q&PZU0z9@+Gtvr^I;^C9i46g<M&9A)=zY0GNK6imsKAYJ<)7xqjA=U?;
z2<ja?^b7`^930csD9f((g6MCagmwSs`nZMNTTzDU%p0T#DKFa6l+4^G{gh5l8~X<6
zjGt?Hm{?Cw*ZAWH2{no^GjwTdX?-j#x^-Mp=MB0Z*?cQGdi>+z<81V=4l{!G3oX}k
zY^ln!n3_>hUVmO^PJZw;*w=me1>z<XtO@S_*ISiJYo~BtqiovZiV9-D9t)j)GWL_;
zSJS01k;v+pu8D*I1@28PpMw#XE?Ll)qw4zthtAHBf>|`kFHdsCP^=JvJc@}RSRpL@
z9>#n%qy=ZO9CCxZ$OI9i5mx)gtHP(LomI5m(HW-TV8bWC%GyO*MmU@C>#4(|5VT}w
zhRTP%MFpj$>4q8_L!0b^fb-ZD)?F($li~9jO@4<CsH#p81MX%sp2J;qFh~C?p!635
z1VcghqQ@H4yOUqdzk{tCMI(Fr`kLy2n9Ja?!h1H2$U8wm1zJFfM=Snk9ef+Q6fM=b
zo}&@d)VO63{7i~SYVqe<9V3B~|HdhI%K=jw%Y?+I`8{0viizPM6tn=aS)7Me=xZ7O
z4@FjV2PZoN#(cZe)!PN72DP3Xp|Onel6s)0W?Yy&gop|U!KIn!y}u71v~OCsjmOf+
z3*PJI|I#+Yu1exni4u?UhdD?LVvpi{4=mxnzYe>9aZ`5}>S~O||5~4Sehf$h0ZGLd
zT}g?FR;XchZN){6va7B{FH+&~=bMLKJ2+f*FM}LrFp4f&AtLwh-ycLxo}b!p=ZP$b
z)a6y~7Ebh%#j0Lk$%+PceXvc9YZ%U8GgcmnWK)$HxEFg&ts2U<gpNAokMEeeUeDm_
zXwYS^9ba;dF5Ts1$GN>t5a=?Jud5=l_orY$ZB$eCNS_zi+?{-%mq}3%)3~@`tB1kW
z)%w!55W_vhl%mqBpM<Wcfkrx$<A72yI-&wvbx5)-^BTq%l#7;ej(HHuHhmmO`HB5R
zV2Aogjl=%!6sUr6Q>bi!Dg42f4=@+2aN0%^Et}C^Ru`%E4i3^G?ewCCPE^1W65lO3
z^%X61Voa9W<ZO!NOB;9x%J^%!*J{YpsXYLlSr}nIQ^t#n29bLg@mQtOav4;jlm`i;
zC$^2uUyMJ8Xey70V@l^^H!)O@NE^e}WF;~kX5PqrM}sgNq1*_O!?dmf0y!Wi{0^Rf
z(!ONpQQlszt_*IYzFxH$nRJtBz1jEQ$c-a2=t`eF^xv0p8K0c=5)E2#C)qeh1U_Qm
zH2XlT@l<`x1R^XYAV7+T7dt4vc8=I5jjgDOzFu1)IEGyD54YTCYQK2IKrZj+noX;q
z9RV>+NKFj~6hBXMz~rDMkA+WK$c=7)%q{}^y_k^1)~9b;EVTHuqWy8Op1y<nI)$B`
zeMW2~u|ZDz`f%XEcqILexx8nk8DD!KwGvCah{ht#igc#GqN3tY!PW%=NY~pmlZiFg
zsOQ>{x!YcDhW(-~l9`%Hv1rGc4-cM6Vjg%4<;hAQHwh(rRC`&lRo2(nr%IZ2a~KoI
zL~x(i!X@`mshA_?g<kzGuE7|f^FUIAyGvf#iJlGA&OdBA?yt296!lfjU$0>ZPXPN$
zI?5e?%Y_7;oSfWc7$yj62E>Hxh$7^ZxCD}KGy0KHVbkhHM*bL(ll=uc;HW|RTKXl7
zY1b3%h~s?nd+63~d$J&O-AdN2^@S#no|Hv5VF1swW__S%y$|`2m>9CUx_S<JO7LC8
z957MDxk?8`OppkDX2U}+V_{%z!U^=+dfW`@l<L)<yVFT5%E^R73(vC4rg^Lf5{U=~
zACbTGlfkpU>YM`3-1jhyjlDdFfO@s1>R~a8^cHx$RUyvZ;hIHuC_!N|hFB5tMHxiX
zB|Cd9??qjg-jNHU)X~EyV+r~n1-Ffe<CvxC&OiyT=L+O$CmRnh({5+ey@@^Y0V}1n
z3#IiOSe>Q)YWibI&&S7?J%L8_PRQ%vB`zhujVItt_!{jh;a*D%gkzU4zA4LyE6Atd
zw;{ZPM|q-WWwqYqwefT`X>EPoQSR-8vdp2?<PSWKvl+r%55mjg{Kjpv?@Wd07T?8C
z?BY<p&>{9SE!C&(-@biwhe)9Yz2y1h2g?u!0Rsty^W>y|ZzrulEd2<vKzC=TD>btk
zmQ|`w6@!bIlL+l%dJF<H8>+I#O-9IV6u+@{h6@i30|P@gt;0%`Pyl_1o;seCtD!mz
z-s==Lt{U@X{imj{YeTWoQ3+7Lpcp<^S9cDb*;PdgD%8xUEiGDuz)>DLydtUs_p+~_
zIA-x;T&XlhB|}_fq@tw$-Me@B28q>yM76-Y=ZqT!b5;~`z1@lA{*nu#@+~6Sj$gRH
z)PfvtG$RHfR1KS@F?}!`#AX>`MO&wko%O-hnbWO2qv{L_3t8~^kD5C)1q@)TjK@L5
zA?^AQ6i78{2}vh&=3Yr!)nUHW+UNtLHra8X0+7qoCMO|T4ewSYVh7@Vwwi}HHO;pj
zkpzGvU>#!Xh^l}sfe(+5w|Fee6#YM3fPyV*ZQ!|`XjzSqk}AJ9@cR<>WR+f^g|_|O
zDaxTRPTNZK8Zz|wkvAoz8|i1UcmC;!Y1*rm^r*&`iQI(=8yR+e*_5yKySqpRsg>5I
zC+@_M@OaDU<;#;fsGC(JkK&~zQ-g|y>P6MIHZ?V6?IWJiV1s!0=us)Z%CRs+D5Con
z<m~OG5$57uB9Vpd)96E=X5Yp`g9lrl96oydho~~7YM>i3)XH0JTWqOJG#B?Fb_Qph
z2eH_WJW;F16DBYt4{A*&FrHf}db7Kp)a^4J7k({+D28-CGPzY%C+rZlcVaDIp9DO}
zdI+sdbKIje!nuXFD)^R9LCZ!jshqB`+G;fn?QyB#;tHwH&mEf&fA}8`M1+O;)D$(J
zdR<hifHxlRIri66w+>}*hii|C-S;qKnW}})S@PxWsN6)Ssw{qATy(!5*6e!lP=Pz?
z4UgCze7$?9`UZKDBq4`xj)hAR9mgqL$O<G<lL&n`FC>;=@VXSgpwWNCvJ=lmH7?BO
zf$EU7O!IL5M8CJmYYZX`oS%~E{>R(+`CuX-=R%$afuz^Bbf?w}i-I~|9$Ai{G)tJz
z6bK7d^!wr%vyEdgj2X77cPwrRQ2ekyHH-L-7x!j#iUj<P#Mx+F!y}BghKXBC-%R6v
zNW`OwvzYv-7#o)T1K9-qOe44Lfz=<2A{>{P1$PW+YDoW7E(o>|>6ga+`~A8g!Il(F
z_=sq<7NqW|VByLtc?2LavnA>#nbBhZqqxjb)VlTLu7GR4%61PfnP~M-KQc#8rZ=Qo
zqK+RBBd3anAA3nRDh8y_VhCzuz2wf2J+8ONoc;D-N}zTRz3tuT-zybZ6PXM1AKhOe
zIm6n3KggE(&QAIF58$7V*`EGdOCh5o3D`u;EM#8)!Hh6!hVa@PyO-Oo)Qa!Ih)Hmo
z{`HxR4F27Q16{}sR>y=rrQnqS>+yr^?;W7Q|Bl#af+oJl`Pn|0g!*!injk@(E=%v)
zDRO0H?!E<``QL%*(6p528$&1jyXOl^Q8jAH*(RP)Gi3+x{d`iloDTW#WExNJ4P7K8
z1m3t0?$I@UaeOrY$8S5D8*I!5xi*wFcD>&dQ`)-Mu;rBxm1(~Bzqk0<d8h43-P-m-
z(8tq$*=HfR7f+gyx7qf;%?8tm{eK(#|Nh@VJ2F^69{&1VT#~2<gFLt&9ntb8E5<{<
zexe?93WGVy$k3R+PWrDSFF8_zvJJGZIw8-sQBWWn6j<|VoU(xmHQO5xJ1Tzs$%D#}
z(B|9F|LcKlN%lcDA9aj#jt|p6OrMcR9Vy+C&pM|)R*Mc}d;0L;@8E*r&E!Lotyk~U
zDNsaYU5LFdHjg7E(l`M_{6D9W#}GmZk`^u4M<r~&n>9rRT0;4M+Et*&F83AXr6Gw*
z$cE_;h5Vxj3|rX${nh6^pNs5s+F7(V4GS3=qa#sCMa&PxMx6bwO!pbGxm-KNQ2t&6
zn;TPvi=4NWV2Cq_CI>}<Hr(^sBF^;Xc7oSZ4ic-4&U}o}v@_`nRe``69n8O#xbQGD
zWUHFL7P^CtgA=1ZmWHB?=A0W=QsIDj53ISZ0X#u9`prk)W&tsWSfCXU$6`Z{6_u10
zuqJ9geR`P}*0;Y%V%}`LZ;;*-h3wr1d3-L4-^3G7rQP&tC?TCN<yain0z`Py(uETB
zFe7+E!!_*Hq=z`MmBUKjIeOn^rux_)9#-HcEc>Pd8ykB-XKWCq4}}L?T1s0&P39&w
zHuf;l^2K%bE_lH_D7s`0vW6I<zGMp0f^-#ZeJ;9euzc}ZL`1}a?BFj%pMNH1C=-vo
z6KWRdvJaWGs@P*F$E5|W)oLdjD?|3%tRjLMdk^Z!P>4IVybKBE!vj&9;aun&*;P^q
zOOP+%l#MhSmz&O<eTCc<7g?kJwqe?RV91peuJArgS$iy}qOvj$d;#TCYnUNc8$p|*
znp$gfa{vJA<rAC^5`K~kY3Y9wkxynp+G;kNN!tGSRv1cqP8J4O4%wbsdMz<y(W2$f
zp{SxZQMa8f{NUhb$hL6_zk|ul%DND=KJ;Y}tl$O)XWU>zx}2F^T3|4sj`#8s*aC-e
z8#TDQ<z90c);WtoVB6Qm<uGnYjO7atNFwT=7pkfQ=}UR9d2l<T&wP>dA$yV#t&)L`
zlZCbQHGv$OS66#H+mp4Iml|Bji4<op*!+bKP2}i_T5X!-`?FymJ<I&Js>@e8IItW$
zvW^dn!X&k=sU}+952%zY$dKS%#V?2iuiM`LR0OM){C()<iRMKOQ66b35L5e@o^C`o
z&Ckori@!h6f^0jRoR(Sy2Q{&_cr4Z4s(8-t2T2r}Pfs3e*AsC_kq&JbnmMHg-B0>1
zCeM&hnls7utcM%Tjt*yxO8bPH!;3Q9M7m$oxW?qqQrf{z1S~^_L{kmL#_rSx{mb3{
zk<aVd^?MWM&gkXO^3*Z|au@dZF(JTf;?;Dx##kqVYiSV~YF&BbK(l2pS;yM+ZO+_`
z(KRuV#^6TC%q#}NjP!yws2oK>qub}%*EWLN98$e1&H;8}jqglF=Uc<#k0>-`z~e23
zfs>USEVx0M(4oyErDkpMLtG?VxS%b#6he!-@5Y?H?qmC$3K1GgyKv_VbsGt!onoDo
z45ta-*501lj^d7t#j84YVneQpke57IMEB>?H&hiBMdi~r!b~)QJ@<_2Z_-3dBaB50
zNCL!Idn_N!9L~kn!%+;dLk_~u+xw!%W~OlK<LvBgt61zk4e@wTe!p{YdS<3|8U;Rl
zmN&N<33VozXK#RfI3}3o{>7)|9D-B!>PK<o7wU{akFJZbWmB-Xk};78AAB-K^3vwl
zd)ACFG|4}E4w5L&e!1<azNK&K)q|%nno;`6^Q3)qKHy%cOl2pWnn;}U<KE|GVyP!$
z7S-<Tq6_rX@b>m*IxeA;k;#jp654yjdc+WpG3PV9#B;?+UqK-De7u*PPM}jP-_wtv
z&E(k(>LL%nTFNC;(0fP=!J?W=Pf3C?!m&1=@BhjPC>;v?mlL4ItDQ+72dY(ENWs?m
zOa$Jeqj59gpcf3hy%i<lfO*&h_KCzx@6Q1!|5cd12B9K#Xtn*H>zf<MaL(?_wjdO#
zlA<EYsnQojokLq=#a-xc+CB4sLRD314FH@sD4s)mYSJ$lUCsWBglGMu?*>7U&_tvS
z>rPm_R^{?3yfz^4+lw4exnXV%wMw7UEw|wI#YF`(hZ#8@3VZg0n@`}nIM$Wh#)Bl>
zHhM>)713xYxwMJT5r5>*qf}J=uM9&+ve5}`uqkvux{s()ZZX#7=`1=7`d+KuT;6@;
z%K}mvUx$@O?-q%QpOCh{{6ZHLFmx81lHzRys0e=?)uO3c84$dDe01S3+Qa4B<vQxu
zS3l(rX0Q2v;5$y=yIQWh2%+rcb$3wl<Fg-mZGK5A{U|lxi2`Mj9O8(A)AU|NTk~m}
zX}*Gujm<%ABpXyoi4VoSux;D3u^(hZ-a#r*#J7$w=5ga@9Es`;WfJppbG3ZLX%WFV
zTof;7JYN^x+?7_9llxH#$%6@!6J1C_Li;3?{ogpuq^B;-@b@1wd=uh~=Xna#tm>`V
zeMofo{Py$Iz}GiT_0FV4La?ih?ga`313{-w71n^2NreXHEf)rFBbx4-w1uRI?wIuM
zE-3l!GOFLnCF|r#d*Z<pRO#1rvoIGQMg5=UW{K3Nb|xj|GBTZWA5@BKvcG=)y0(eI
z;QE^JJrT2TY?<2BrfWUdUUUH4nU;rGkmQR+JV&u5IcYlk74<vxQWdg~z*IyRu~TuX
zr?P#Viqq7URj*HT-eaV<=%LyxmiHyIdkO%8g+KtdjZ!Of>gp!-06V=B!t!3B2Ow8#
z-{yHibAVEY#}WTuc?zSR2sRzm8+yqA^GAZ$=X`&t(1gvt6Ml^K<rv01m*THs4~p7%
z_z}(RnPnnJ8BHUBdW|V_y(FR$Q*ksU#1MUbK76jx`ilFKV0B-DIpby9k5KNwV1aUn
z?*oId2c>~hk>b}BAEl1dggthkhpOBhdA;91wi>@{jG%2#qAJ$6tH;!QIUN<}Qih;%
z>HgqTY^3d*9bUdr&%H&U3zl36{Xys9MUW)c>!%&6cikf$?ld*56_o!RGN>;wW&P5&
zO*m6ijN1VmX_w5DN9^n0H5YP@@VyOhJVr4rDEeGjSV&GvP7d1|K?51@9~C8^A$iGp
z-uOWSWL_q*>lgtA5rTnV1g`g6_4xyiy9`ukbQgGdM40Pu!eXC|tJL$BU+v%wL;`(3
zre(oa4&)^Tcy@5*ntv(L6S5HznN;m&G+G~Sa^Kjlm4q~eydgs&dWOmxBVszGmeFFR
zq1YP#a%Eobc&In?avVWQw~m<_jNmk2Frzz8FuiSB&QiK@*_n`&fptC`#V4xAZ2N_8
zRuk6EjX!HWbM}V8RX8Xp?1~5OR_bfLLWh#;PJz{iKV9Ip9%Vv)<+aQj`~BW26T{If
z{)JNau=LD*xt`r#^=8{H6kpiNv06*pc1G>vowqxI*L`l+ac@HL?d^;G)#ANRmg+7x
zixK^ehbiBpiUp%i5on0p0KD$LKl~fZZe;yU9$SdRzF42kqJ9BJlvyI%rWE^cU(yz)
z-*=lpw1nXl8p~qF+JWN<_*Vh9ISfaRFlL#@W3T1|-O07Xu9<0V0)_|t_5s1NT~)D>
zx-DnGdfZa;nF;4FG`W4$U6njFBVPZ@6Z_VMRy+vUdmwI_W08!EELwBS72*V@tdr-u
zI3@|K74M}j(Atca4XhdH>HXo<zOFBtUp~H|{jF#+Dp1eH5%jwf)%KO70ToAWQo9?0
zPg}MZ&>_fEgd8$Mkd9&U{{H^flGK03{v56-<=FC|@NsFxwr2I;4pv#hY^4d(HsP1#
zkaSm@*?`RYQIDGlFGo#&5yE$x#>O}3L44B@M6oa+PJ*4jMidnn3uSpbp5%c{lR|h6
zoK;pOZ266WhO*Cn=3d4q)-Cp*y}g;J&idQe)dHtw`DK?~x?8Y4CCA3Ko7TG#gJYWc
z<T-A1AzdRWbquzIo^<&KbYySqKci-FucY5QRduY3gN5sGejwV!Fuh#5e!Ri0yYk(M
z+Sd{!RmA1*F)6BpmYL&rco}Jb1sS4LAUaVQI?F231w^unHl^m^0Pmf`8cI4DYJ#S{
zpT@9^I1$<3P7o`UI+C_4w6pVP$L@KL5G*Lai6*qD+qXXep8`MI7dmJ`qbO*1K0+2(
z#vlJp9zERa6VWGyCBfufgyMCT9%ok+`)h2d1R?xLjj(`FvQYY@Ox!~#p?xF=`!Gi?
z>XMC*&#h!+qo*TMgX1U{)zm6$y0!LK{N?AM=<QFv=BUJe?WKE_;s@@w(<BW44`F{D
zR#o?{0mGYCy1PR}Qo2Fi2pEKvGzik&or*MWK#&$ty1Utk2uPQ7m(-@a-no5#&pF>Y
z-*w)1{qbDaBWuq!*PLUGImSKiF$RQE#?J?uZQmG(+CcP?L>%ANOjZ<w;P}~cXN<iw
z(WGRy1wE-D-n`T8{JVVng?1{4<ZOCsyQXnE1ANnC{ZVrUR(g&k$3^gcKVr+2sh#7q
z<M%O31@dOq-saDm>KBFG+isGD{Z(`*WDg3kubSC;W}f(d4qf;fr}Ti1-?WQhDGdh}
z7gXc;Mo{f3jzgC7q|2{)h$PjNaJj#$s36K2zL$a{#hKxo(Q8sT3&d#pq_~&>NM;V&
z84(1Wz3bUKlIk<Q)O{VZWB~ZUm{pzW19&e699zoP87AyxfbXRD;6$?Omg&Wzx=)sS
zROiv+!)L|vs!#0DKh6xzN31>6O(roe&31>3G$c@|ZijVc-OrNoxto66|Mm%~F3H0}
z8PZDT=1e&ZE%8)h)fn>$%U0Aq13_ErQ=!F;C5(Bu``=1p^+@&W)m$RNuV|Rr2t9|`
zZhqZ;wbC7Y`PqXV_(G=&@o2eOQ88g(2bba0oAVH;<MpZ}xd$ET9dvr=p&o4_DJkj9
zqFdt##AtIgU;aamnF3~6Rq^{?;48ISYiq{6)#k01c9V5HeFH(hX54yT%6bfIE@&BN
zu9F~CTUNH`(o=CVuqX`pXg3%b7$dmt7M^jZSp)036vrYjXPNIl>5p+>O@}CuQznY1
zsVTFDD-w0k9kHQ?DD8-b0w_S@j4>dM?i~k&(rSLq=P9(D1lqm<d6j)un$8)kno&!8
zxTO*789-YBarUmUpZ2_h_WrTtB!8B!di3=UeP^b#3k^2uqiuKRhs#GEtn$!Nt)v{f
zeNeB9u|Tz>6Mi~)g{d|u3E35el%k$71rQ&7$+s7i3l0YCH%)yNm5-dvoN{05GBYzL
zD!3+2ZaC(p6<)qfqspldyY+qM@KK{sGc90ki3bvSy%fQ_$=zc`hY3*R<wB6CHYXzm
zef26fv%p#16(91>^7LK+6_HQG3L%XsXYwr@bT8^`;5!XPd+MLj!Mz9;;$!um^0EiY
z5jjs5&<15*e9l`sS@me6=>eX}y|PRB^P2|$hv89kq6ECE?mB_*r!%_-;<8*VS}dYM
z@x*Z8)a(2aO7!7=xE*hI<hvJ{U%S*aD0`C7mkcUFNzt5sT;K)rM*FJ;RI-HC%B8!B
z=ia9~GkAE%2ra<=eTx#tM*_yp_e*JTq9**qb04w-FVX;cIaAJLf2KK12TwP*lOzzM
zEs4nG+iQKi$3!q)L(nkA@xOpPC!PNZ<b9@a4Wy(l5hdw2j%FP`+;PfM!#K;y@cPCd
zuP)nH`r)Zb_05ENRDqIiV60&WiwHA2hOtdpMK?LQdl)|Pm<r0yqg^R>6DMr+4x*5l
zZ32gWHRdczevC>K7mxA!f~yfpryYTZ3i*X6i~hkpw?Xv$9yRi;us^_)ScjAlNHM?C
z(YYWE$ej{jJN6p?g3}HMt+cC9;4_24Ud+>HxWpZoHf?JyqCr{#BBb{>Py8?nFP2)_
z*+TdwTdBjZXzOBJ+N9edv9@);f%`5LdVrdqSNr<zz?39uvHVvSLb=4fF<BSJN!&mC
zMDkCBC6ZnW28r6>ehFr=WBiQWu}k{`xPthM@+IC9WjkjPa(lP|@q^4_SgL;{a;QG@
z1f`2&+3Om}9ZH@U$~l+2`}vFjh$b)Bi>R*_6)O14U_0IMxomI#6@Hdnz=OfXsZ?)>
zsmov{YL-=s|3(gO*isHS6;OPrb|pjKKs8jkf)wnwtQ{s5PbJa;M}iqU1N%Tc63wVC
z^4@W-x@d-mDJk{psE${WKoTL5s%S~(mcd|w$$V|_&jCh$vcX50PZ%B_D^}9`e_WQa
zjkJy(QiY5|&e#LwQ6xXydCDOXrY1;uEc<rP2Xly>8TWQqwhL9BWHZwsWS#5|Jr0@s
zG)=@Y=wm^JpDvSIy~bR=5D3<>!CL^}Ke@%S-IWi{e6fsCG?I>))2B%zQbWCP9&M69
zYFtTkAxZ;ctKU{X8}9mO!iGcc(>OffVa^i$ScJFLx9yrpC0`;nqi`2ra8q!s*T6kl
z%;mb>>FQ!mea3eGUVsJ!6ZJFMx*h<UHuG^C=c59e$h@d?Jn}WNbXEd$YiE8A?x4E$
z9TWD-ouTedhcxu_wtvS^X{MB|i+~WdWe3}wCkZ4G6T#YD-s?LR=NaIA$vJL|_v_wL
zf9QuH$>!J<l_#$oeSTc?OYr^g8VV7w;f>GN>ph7-ACxNG*DK7jVx-oWBqDO<B+A2`
zt7}XochzmYcXVtF>E3_wb>y&)7<SknBC*5Q-tt_-51nxpZi3B)Ti~mYL2eVJ=7p7C
zR3!PGFEU}e&9JApmzJjIFs40HO)o^R;6XOZ%lAgvQw@8PshVC3kHtwGhlZ@q>{UM5
zqflZRQ9$R*Y|nkgp50LOq=tP5Ii_>yMxp^o%Of8DdsyD5?$4>jvbWB;eMwG+zxSBT
z@nw$h<4Wy^4z-8iT`D&@vut+%Z1c_HIHUW5^>TdS8GlO7tSFo3=5d@+LoX)?>a2Hg
zb_y$`*WKRzBQi~Z^YS8(QY$1j{59*-Q5_Idqip9hz=9BYT6KM7(mRR?ZNLDqX0t57
z>2N1Z$SBR)C#ZmyOMf+COSFs1tICX&Zf|5cc!Gt;5S6h~5ZG#_SdS0&dcXhKvnUL%
zjm<(=CUWa|d~Mi68N0tVxR|cF&;hL*sFJ%4E6dJJj%a+~;q_jh_;lJLfA<Z5GgH6I
z{SNgC<Og8P-cJbvB<S<ZXVt?GEdY;uKp;fY?-}jL#-Njj=F36m=NR-=^?4<-=mQwr
z?L0A_P5>``2>%xhKDOe`Ha&-?xiwL*;T60=5vP@<3-AxyY)U+S|JK4^6~Mt+1`+?F
z!kfnlCdOnoGnflCAdm%xl0JXhMDicF6bU@_|2NhOaInNBQamI&3x&FT)O#VJw{VdI
zr(f~jUM0C_;3geA61$v64p8Lkz<-BFg}exn5HR|KR#1noqQ>Nr<AEZ1dd|1{YyW=T
zy_oTjU?eHr|H5Zq{E(Bsl)Ehs-+eGDx%YZRT83QeirKEa-uh(HA-9859?VGmzd&(c
zy@%m=x<R)PK3WL+@Po%F5Q&-{aRlEJhHytOFz_h;zmV)j-T3~Hscb8iV$lO0hM0}^
zw0<PC#r!XTu4-sL@qP$O!<Q?XfCJ_)_x?Zg&uUS63l&^^W1+qe23(>2uK^_|X<k-M
zg+53J!=(QAFu5M=4Vc7l!8b%C_FrR2_dngUguD!}LSM{a%Pi^LNJYYG{Qttu$BwJ7
z%cvf>LzwAEJvG!z4MN^VJ$xPw?wRWNcTY2M(=XMJlw_wQ)8m)E*ZO@tT<P&Jwv1F`
z0bAtuPlz2%5b}j{n*u@jY3P9mDx4isfMsgtqq>)U*W*vI#Qv<|z?FMb90)OG1=s%<
z+J8*T+kQxIqc?ZZbdkMDgCa@UiNsv<=xF~7iU%wC+!5XW-T#d%r)_KIyhw6-*Rv{)
zBO6$tg!K`{sD6BlJCrd)4-$tGiSiNe%QNT$NG~Thw=av{bWl)G5;K2m@XSxI=DFIe
zb2IS7f3EvAuNR_Z|KAB*+&#ar7REFqM#B{vWhi<oLh?2fZE3}yYVcw><iFLP10d6e
zEm_7tAhMWx_0A`<e}@mv>of|iMf?9IZG{38tmZnz*aP}iU1J`6;o)6D(Npz$Jb0MF
zY>>6-nr5p%gP9tjnitONgirUF&kHVaA)HcD*KUy74_opYr6=b;G4d%QKoI`9j&Zs6
z+!)2g&M_g11sACU>y6cKZ$X_?;f7g6IUtXWL2{In_}`v^g2f{8Bd6dm2B>?ldJd7n
zTI*@I4?E-lE2ax*JiZyc|K@)I#UJ@_vzO!keG-vsOo!a>DR#&jB%ok!_To$D@+97Z
zBZTkeVb(+rxdEtnr)@FP)zu{`S<+4T@2_AcqCvHEBe}{l!@H-aXaRufAY|K-xA}&s
zUQ+`HWrkt|y?vtMZN#F}&h5}sDG@&KLsc$-WVUiJn{~nh3hWqU{H{!MotdApf>s{&
z%m-Tl->f#hI@>?(e83Jq+)Y2Oy_#%Lbs^7~k`PwBm?P#;uSXNq#t1caDLTYq=it~#
z9sF)<4Pp}3(|4P{R}>XOQgOc!11il${J^?5MnKI-$rBK%XAkWXX&lBn=BcqIu$Ux=
z6(uJpKVMEFP_@q~xPFp?Y4|+8>*jL)HvL-xY)-XCWNJLPAlhP57&rS%PR`nq#T3c0
zTZZq+<QeE`=iA);d1b{EKqi*fDC^hi$G18)aw{ny{~d4k6N@T?vriB6R)H7o{Q5C>
zjjf2;+qY5-yOQ@Gmcys<M+5~l$O;;y^&&&7NLAFFot*R*vD?a*fSUn|Ri^_|+_j%~
zc_zk^4lWkobEuhif`+QDu0(xSRu-rIOuyEcr@n=9b7biDyyh4Q%9W7cBq1AaV!=T*
zbhOA&%YdNZq7d;SBu_-)*|SZ)pL~(G5o^`o3KCK3ia0%qh#*Eq1<_IN#H4D|C_n*3
zFVHsar~opO*@PuNc>ERjw56pb(d?T)q_xwwpaLEb`bWG4RX{^D({(Ip0AXxqel8)Q
zvDITQyb(dHur4kL9>fXY{^?Q1n?RNPbB*2wap;zu!EwJuRGaC!Tl+fu`IaxsRPR;w
z^lXs*O0^+z(66R|4^r7|V4qShondde<qL`}Okpfo_EgNFl<8>XuCA}g=v0_?|LkAM
zF4%>fX`4^IDs}xFI)e>iuIR4Wk?D|N%9?uywZr5ilvUBNk8ziV;y2**MNGLVK&ds3
z`z^#paF>}ir0^i>$L^4pP-(fWIZjI4bTr1+W=f4~s768%DacS^geXEEso(xKhzyNH
z;bS{;ffd{|ro3X^R8x1YGp5vfmtOK}48!qd=$mKIlMV^2XL>*&+iP|nwDUoX)axim
z1V^5vyxdPtXHuU@0|&bXH)!n9oCgp_#0x_Zpow~`{T$`R!<UD&khBMTKoqXA;Qrzx
z30}Fse*I$6!EP#OVB_T6O7~cYj_PV?X&DBp1SXcUJu?O0gBH{G)9eBrvYoBWh6bQ2
za^I&*Cw=;KY!J^ud~C(3Rl)@sMCdSq4gxshUbf)J^P#t79*MxNaC{(~rIc4xM2oh7
z4IzY`0%PV}`X38rM?q0>a_#gIW#O~K;Z}oQiE5_e@bar($uD)6wvgvro1125J}o>^
ztmH6v4cBnn>B-qRHt^jJW&Vs|@PE<d%=pX{^v66zp`-?ecW7vrc4}}f9S%9y8F8QU
z@Ed@|vsaO27m!0jyOt<(Fc&^9A&|8L@@F50zt$MIW2@;_%?m#4e=a9{oTeQ$NDKt9
z+-hs2U%!6+8Gk_@LWqlL(m?zsukKP17(JXI=7!UdU6+!PA%h)2<~6m(Z2PaQK}Q7~
z!b9-UQ3J8`dB*!7JRJA$zs^WVaBBq$OvH6_G~0IC0K!NGAFd&1;pF1FDgj70{8{O^
zM1G@__?J5GhXLV$R@e=6@}c~&8)%Z5#FMzi&U_%Ep}%|Vb&3lA5k_*!&d#n(3Tq8{
zY<P6;v(><9{_p(%>L)j3tI#pv-`6u&yn=><4lv=|Jwdj+hV9$IpnH3gadHYI_HoRU
z7pbD<{(!pGpRitIn*#Y#5XaSHWi?q^PzgM?8=~)pn#JNMeI5D`%F7N4!%<z$o-fnE
zR3ZCgpzyc<%z-2aN4LtDSb;be<a?Dj2|SW-o&>2+A9uB3py?|#P=-VYzJWk%!)%*G
z3GW}nN4^Q+{XN3oyjK&1!q*Qq@-|f=^klGNND$=z*Q>>uWvAQYo0{Tf-nmYV)byo2
zDznwa$L0B7?rlqkD<qu|b_k-qpA%rs$Q=l?Kn<VaakPE+$3(>kOBO}FL}ds=0%Ld8
zc&&TCABKGTed1xqJ4C9TfZQ1akb$1Tk#P{`==>?flx0xV(BMT9un0*8|Dv{(imzI#
zp|5`}8GwT#tI?VO1f3qT@?A;|nqgg9{MfyI^z7pKUQx`W0;m*{F%|`($CYXZKXo>c
zJY?D|M$$(Cq$v&rPwgKbfipAU5fJ@lLyVfQEG#iGUqIy@QK=|xqL%~iAop-{=%4s}
zmUR1j_*)2m1mD%wU5AWCQ|Z3~rvD#X>X<HTU7p0QeJvgkf`Da{UH+G!2=(n=un6k-
zn74#93`5^U{bK-b3Xq;vEa#z3yxLHC@OPZqp#S_ug7huuQxgD>=0n<mq4%Z#eLTyZ
zbDA$2bn@5`E)Oc+B4=^r;m4d+V)PGb;UW^-V-r`1M-x78kos7JhPp{y6R2}=xq^`6
z-#O)SeaO`#6ALJ@m3W-+IIgvv7Kep$<1aIU)Ud>ZT^4y4srBmpbMxr6)Hlom%D8lB
ze`F_quA?@c@k{P@NXUTo&<CzsF@qolT<dY8^c;@%{yoXdr6dZXW4wPXPt^?dz;H+a
zh4FI7U(3H)i{8wK;mtfj*CG~Tm%!UQy}#}oF0Uk6=}c&z{eNcfRTcKxt-zL&;?b)5
zd&-bMR^_V<J&W38AnFWmN--jx%?~%M@>R$Gu+S~ZL9K0;_%004AH@GLob%+9BfB~L
zvTxgKmL6CfN(VkBO9-C>$npTC+tjC?hpo~rz<pIU6a@5(%%u|4k9@}19Rr7a%Vdz$
zlE6^fHsB+HE3Xp}EEW^n(*u_Z${vO9PE^=C;ya021cHrI!jK`b4dF!G?0)VVpj8eH
zz2Iu0x(A#omAD1z{*eAMJSjy+VVmng*;n08lC^Ae^+c{KLOvel5CP^peL{|BLJ)da
z<d$WT18A}}%GV^@g8!rkPbmLM%zzUYJM(7a-;I98hs3X`#YOTiSRf%57Aj$PY|s%Y
zjB=;=RjY~g9zl*T<MZg4ouqWu-iESj?(0IQ0>I28P%XrVLlGqw%h@8o>HC3wFBS;E
z2$13M@eL`Bgq)O$UPO&Og-iYtg#@mzWUpgZU8T`6+R>J(Ae9h3A~Tu|Z2f3Bo{4*v
zZ8U9?_BxaY3jMpEY#+z>+Xx_}A?Y<5a57n@a0aD->2PB!D3P`7ZQxMN!+<uW72x2F
zCIFNebFf-fP}}{NXQ#YA={@Ff^uLl926a$leFQPWSrbjgQ_aPNmSM`OtA(<n7>dW^
z1ND>wjp}5^*M!nPhmx}W-BIEn=z4;rA4!;_DE`!g1xf1-%M*~?6Xe)?_H`fI)1OLm
zOlb`{<0*VxuNO?EPYExk81I0Gtx*2+`*(pqdo5Db4{O{u^VJcylbfQNyE}X#V5L%u
zzf1a8+|z}S*OF*rll7n}s}&tEXSH;a?V=-Pz<WX=9}`tos2kl)^b>)qR#CwEsw77;
z&W_O(QJWfP)yg^LH$Xjf`S{kujDVD4<;{iY!^nut12Txp&{z()zahS|XOlB)){jo4
zlcgHuK{a68(;it3QyTg5aa=6<omT~a<4VF^2YZd7MhNZiO$Rhm_l&~aIkk}3X!&zM
z6%w<+ATN3s_36@mkC%j~gYZ9V_>dyV6ZVce94k%fLtezuTMzo~ar@iiyEmwmUImXa
zgnzTP?ePTXHsG@Tt5-`d)B}+a^Bs+5D!3p-KD_I}bWKKHVWD?TRu*%Cw+?7|gXE7x
z842rjpxBT|$X-%tJUA%XL7S2WvEFkhloyzMn6knF4~X(1e5HUb80E13;6E8S3YJCV
zTfT+aO<@3r7Nf|OIv1BtM*V}agBjlMt^K^|;j@Um=LrjwX^^${X!m<a-2#<T&d@e3
z{_<tPM&p_mP7BF~9I1^RLRU%wbKPD$VT@PF_qnn0VsV?%o*Gd-23!krgys-ia5E#=
z_tNgPjWbm@K?thRiVkqX=a5MBkJt;88qMk&hP7OUg6!<MeW)#=l#tjd@4Nz+lX<|V
zS<0pCIAQE~LFS7tTNd^4-W@;B65pw@g}&gw(}6Y{WKgzAfCiOq%m>&I5_mH(*xkJ)
z(=V>Bu3#a?T<T*2D7{#GEWdqq6I@dY<}kk8g`YxyXaRPHi01*IBCo8Fs2@-LwgvPb
z%-}X+$RV&Lgk1g%7I+@6VQ}?4BZ2v7_+kcpN>BHb2$*hPEh*5t^l9nLzD`H3+97-d
zokk?Gvv(dlNZ<!N4140uayDc;7F)mUQ*AW~k*YgY@b@G2uP=z7D3ifebbo(kwA8jZ
zk^)LytVo&EBQ7q^UiL6=bRAd+#oU*PT-p^wJ8m(yptCe4pD)O0Z=X)aZ_97VRZ8=!
zvYmEE39w8y)UL7^3P~{ON#M3d-$M!D)X8uLMb)NJ>U4KOFGp__50G+!Rz?aRaB^}&
z;EKq-Gp1}Bf`#4>{Iq<I2SKdc?f}Gs5X4BF$S$P=@!y5m;StHPByS#8<<;S`5d^tM
z0<|Mn!24o|(IQ<OrNGO5E0{jUTqK|q1uobZ@n*6X7;+&~bN)t-wS~ZL>v^2=6wnG5
zXUG0w9jZnEGLTZ}4bjIyTUrD`8=B|O$ocH^2IK)s8@tEik^6r#dU(lwDJUptfR)sM
z3mDOU{0F;vCb~5?5A{l}Wu}^ECq<6&fYDrAxH+C>iDYra8@GoTFTiW&=1JJ1D=I1~
zpE&2#_V7hc0`qm6G&nwV$!_)6Cl)Z&{qTE*PR`D|PFze`=%>FpOH*m@pd&1Kf&Wws
z0-F^3)i0|KRN~P(22k?eaV$K5sMd=U2x!JHzNbk2_5MCRh-osYjcr>@ffi3|d)YD4
z)EgTcN<fqNs(+{O02QY{P=}xpO;XzxpeFWIGg0C`hCb~q*agLpMM(N^vW<7=nrUzV
z`HdJh4Ou~%U}NS~KPD<2zqAt+Ly>*;YFMlQ8}fg<xcW^|1vxoplmQ?C1gm4<1B0JO
zrtNetbOdB!jE;|+EWiT^VS&Iqtyfx=Csj@nw!QeBF6GVh=ROZ7XCY`5xVGG%`a}Z+
zp7B}|@<f2!aP<uM5Fbh-$a5^|b=A}kbK=BM0t)DdD!7@=$3{o@`tLovqg!iteGyG3
zG3DErlp}^5B@H2(Q9$nxABh9C7oAEaR};M5wRTXZsM4mLzRU;Qw3bY&R2E7&F>P{4
zpDmz?1TYO%-Ng(BloeSoFL8C(`~UdMmz5P222t+x@t=uX;MU1zIR%e3#y@}<a{{x4
zm6B34r(StY%B3{3i;uOJ*TpI1Y9qkxTI2W?B#|7%>vj*%{fHH%Aq)T8p<ej^cc__$
z4{L@_Ng;>g)9FjkzdE}Qfr4zezw$OMA-|?#?ro;Mdi9Dq%3JWrST{E;^tr~lP8L1y
z@VFcfkr+3A>$Ly&rgdyyuEO)@TYOKUEo*~YHZ`q7GL3ff`akea5R=|l={v4(H-axE
zKH{tH4m<UsFX$Fr6nm^aeppMV>UUq3H2+;ju$G2yiRgDF4N&)N!G2N+5O++5Y70(q
zeETr>*on@O9d;HCO2?61gI_>P&>=+i9Ti1ku$ATiN$r`UrOGa7MA746ZU>}yeeesu
zm(CwUdr=1=jO;8d%aaQIihtFj@AKJyFLbn_;)VJ`r`jiz;Coj69<zWkNP02Fs|WnB
z%UimWooR|lx%nE~OT2an(40%d!^2|Soe8{#1_VY10B8fVV1VmEa%&S6ER^QWQC<+k
zVu!Nj2Tr`2ZGT?)w0tW*1)(Mjihx}19xp7ol}1_uaNevy*E^bN3476k1|8_lU{VdK
zKeH22z$4fXVTA4X%ja4q?XAKlyZxBBqLOGV1EM)MDQdT+M_=t}dvy2jFHY24Ye3B1
zNh$82H$iIL4g?b+3Sjl1Hxu2`p>OUH-l4^ShYK1&-5?%w^^qqjcNShlR5W5jM}I7&
zLZfjogFCa%PbG>5cN=ph`g04+L%YN>b4yus?-gS>c&hu#uA_zeOz^4>occnBX1}9?
zf?N4-o$&%b%_b2Z+c3Xk*rmPi$mSC`Ub9o6!sPnb_t!NOwGVgk3Mn}$E)c@9k(tq)
z2B${uwcqVEL@7nm)9DxJLPtFv4#Gq0Xmu(Y&jG)b_W=)2G=k-r@PG1bKn5;S@Uiy1
z0Mc;<itWUp4=(OJ&S-AM!|^{R1|*5;)>0=Eyx%w8Cbx;{d37pet*--Py@xu)(!#uA
z4-O8fAkW0ZryE?i29(7&zTUd6(E{8{4e~WY0c5BgCifpbB7F|N>s*=;BFDh--P2a8
zi1{5*Nq(K^!5(TIJGe(GcR-O82PjdK!AF1Z_^n@B3GSdlsUr2rp!1vA^YnZrNDWWg
z5u8@eOzEe+9NP`ry~_~@AU$P>gI?86Y+M`)$haC{dJ?kdOHbq)@s4*7aLsHEFRvjE
zFY3-+ftQ{)m}+wyrZJymsaXs9ku?_=-?)QDgOzVx`{Lu{1tA|uJqYe2{Al3AfPzdV
zoxz)|+Vz3(KWKz%@ttR3%OE$gZ}<LMRHF=KF^VT%W(VrDQRqVYT^-ZbNa9XShkN1G
zAWMh)Y?ar4a?wTFeZ~=s_f7slS6|;XW9&Evk!uY2^R)nZX+7;q-QMl?tfx=>YW?m0
zz;UlBiZ=94sB3<mdQ5fA*l#+tyDE$%rc~9My52><f1ivtq?W7;7Uz0jYwtL5Rj2b<
zRkvrc20Xq~Zl%8wqG>&*lV}wh`Zz$Af8h9%C6L*SUqI39UeD>Z*XrAQ1IIH1A}~d>
zlKS-)4w}=8kiL$~Xolk)6T}Ugz@_KCWpq0kXrDfnso&-Fd<zTcuGDh2q!tk#Ue%ot
zdQ|^tenB5VVw~l#i^%TNNqHT?;JR*t?x&OiNwZ0%<)C12ZO5DPR@B|lr`eZi$~(S@
zJk{){yHmJd^YZ?1Z(SkAX7sB;vW(bdudGtXY|PeGRer5WH$1)#*P*cLOc|~w<c@+n
zGrPU{EF&Y&+%d*1IO0geY|V6!wK^x~-l;M4BJ9O31@o|olW#_$>71e+jF(!Pu4#p2
z_lg#a!I^P{=W{RBdMYU-wZFWzA+d9vhJGtF)+ui8@}$G>V_^~V`{6kEKEEZBj6XvX
zwsO#6s-u*mp?!;G@c=&D*0R$!AZsxAH%PZB|8(_&m&A!X%FMMI$4?n7c<*mE;;=Mt
z`ad>^xfk-z?tK84DZI$4m+(H>R=l#tYvvKLBuRR9<-4%^($Er<)YFJ!P4X`h{Afy$
zTrUu#b<1Mr#@csioxvZLq3_C>nDOc`NiCmV`^n3?J~5i*+gBp)wvMpGUZuUYb$;M{
z57%N|;~E5jvZU5oLw_{jo*?a0nSe%kTOthl<}y{x86Ugd;jRj9(I2ZP<zV75b(hM4
z)!-7iaM}jdgJ$@hR=Rdk_g{0J$e5k%fOWS`5_2|F6r`529-nGL_;sYO?+ndcGY*OZ
zen=9i^lQ_zV*b#1t+&EL@Ip-B@>`>fzOHT~Ab%fiT#`~yD2r4Yx+Kg2obR;u#n1T6
z&O6GRW(hxa#n@MQ{*v#H-dl2oHttI`B+*@M;*`>9Sar0*lTIOFJMXS8U6y1H7$wJN
z=kAi%ojbp4J7A^R+}yZkB?b28h5<QUa$qsYmYwN+wWCts0-S+(DD!@TzCNgibUQoi
z|5QHM-O4a}))PTwc5<Jupl}(U2;{A`Wu4+p`GhK*Bu(TA<7#pvXwrSzFR?q*>;eT4
z4Z6EOn}%;LR=_X)K0{cu2JtQN+HKr9)rmqg`&>pNlsWjhv-6=doC)3!?6V?wb8;Ad
zT1rtG#sXrOO2)5ASerzVNvY|~_0XIazxS)OL*Z+=EaUMV%JRi&j%%(p3e*p7ex8kO
zPVSdFQensSUEKU11%-stsL|VeA!cqPb&ktNO7cV-Bu|XPFd<-%sAsQb7rX?#VbM8%
zp1|cu*p(!uHPD2|O~%K@uKRlN1uJ<`w<HF9@76uM6J&B?hj3$pgun6y1#NTg-2hdd
z%na*Vglm|&@s`yeCi|n=304YNSz9Y>IBPYm^RCnwAdlDsOOnNv7b%Sp;{}GP<8$>T
z`)Rvx1g=UDQJ0tsV|*MwLy@SnbKfK_aPQ6AuHFSQSc^cgCrKQ66U4N}yYlk=F;@#m
zZ9Wu}pl^X}gC@C+Gq(O2xn78)ZO$<!LFtJcJN-b1oXK`IR>1N|YOIU-FKY7H>Bg_j
zXo+Tf;R+?VNmNbD!|n*P?jA~+?y{>%=9<~q5m_XnH2ge$Y=vRhho_!5Mf%0X8jtLc
z3JXS%?xnCc3g$siMUJS$-foN5O}Ao+7U{LUd8Ucz(&a-UP2I_DhL!p5%iYl=*E3v7
z5-mZMF}-WzV9?lfYVbUk>QTGt%@?kRTkyFGiHaw|^{w_ocbn-n!E1iaux5IV)ADec
z<gpzVpNF~d_}x1(-TcYG>i<f_?E)d9s-tHtPy5*O;IsGfG`@_H-6}?$zqrpRAHLYh
z7x{USRVlLf{eLmBB+q~!C(kl&F(TgwjP&_&<KtPsu4+0oOL#6}T1)y4IbtqmY3R`u
zb%LEuLu<b99k*5vBX6cd@Rv}9lTB?OlW*Ljxa-dOR!^g7&c9pt6`PL?q5mvr$K~xg
zqK=lD#8L$ifGL4a)+6L@f&gigwBC{DOhLeh6G=X^+o2ya6w4A(h8IdKt$wcK<9+qq
z5&Pq%O#4v7_hdvwFnq9I$fagxZXAny9bMTJRBsaX2gOj^?R;`qbfb(ni3;8?WgTHr
zpP%N-<El_!Q!-M{6Wpl(Xua1E`7uiT*8ukAA>U}W6pLgwSK1A?Y!EFSU_1C~Dfa|j
zK%Z0x=i3!(;zDOiAV{WNuY!^C1THTdvar@=H`JfCc*5I`d31H04l-|rBJA(0OM}_9
zDLoeq#bsQLmuHDA#zsv+jdZs-8Kf~V-#&2Vvf*?A_Ye_LC2uM=J?dJw&O2YSKR<k=
zYBQk}L_woiMJk>G5Pu`Ez+I4sQRnP+jyMyrmQL^a-t7^SiL+6`jN8mHF25>=yI={S
zrg?>#_t0-t|ImWPcoLRiK}ZG3A(5HuLtsc$Y+*|b_aV}LUXdK6y!bnD@E>eNJOh|t
zN$P=xg`a>*a!%D~pB+!Gp1>h;IOG*_N616&qMrGe8H)Vd*EiAqUdZ_(FDg{Q5#_HL
zhwz^LBUkTHWhPT?1aa7r%<2z$Xl628nU4~YDV4}tq~EjDiIaxRIR_VOB9Quxb=qHC
zjcUJ~Kvc@VfYGl;`;4WmT$HZ#!)KDV7O_6^8FIYm(k-#N!W81UNGSqx#A6YL3G?g!
zBG#iOyuuo)Afw)W714VoTvo16=f0jQJYDZ5zP=l`Yx7qSn3>PNV3sP#tTW-&7g12<
zJxG$dTgmXP^yoCC?WE8-<3Ck2Sa1K%Cbr3oVlNT5=XO||m;aBpQ3gsl{{uH^Z6zP5
z9m!v$IW%-71AKDVdMr^pYbQ*5wzFCn!p@f};D65*@4b4B5EhC!h-JQtPkm2B)l{K}
z`ByJN4_>Aeh2cU96|Wpa2Bg@ky14N>w|!?893@5Vg~DXWnKKlO&)#hV9-kL1(-L#P
zJ?@bQS^tZ&T9lMFzYmAYUc{V#W_ncquG|F{&`nf~2(yih6ITfy7c*m@fk;CmA6^%;
zoh`T2<VOCZv%*Pr_Db-+;W<PyGL|=@b`-&M&`y&M-!^k&11%!1v$v@pj+xFN1sDib
zs|8Z1Vi{rRn#2_qgW6E+{%!iyY=I&>oU-{<!I&<H)}t#lyBi#=EiZk{&Y6C1slDQN
zda>&sf0sBm2&lf$E;#Aa60FC2RrVAj%Q;@e_A5-Ex&qXdJV&{%m^e}z2g_+om-{<&
zAKt!edXJ2r5W(y$G3`SDZJE99QUk7M<S(F05waM?&COjaEPP_8R+t_MD(GSGk%*Fg
zzGK5z`P66;oC64D*Jf~9R`zXcS-yE$Qe5m?CO`i%RDpif`I6*Q%YyS~Uy+v^hST3q
z2G-f)yC7t5XDs2y8=MkTt@INg?s^IW?fzH(NRr!90>Maev)613*ha*BGqcEJK9X8!
zv+XsPuw;fw=bLgz#0|Csjsv~JeHColZ3Khc=_NRHD}c$*Lvv<E4w0Mnt!s4aBZv8w
zrpwI>oGL?f)#8h=CB^Dt5(We2C0vuZOP75{b3a7)6*AkqR91JpOH#Kt|9(@f=7PZ@
zQB(qE+vaAc5TXU$Z{NNdDzg&ng4}*z3qthgv)3LATJS5MD7vVpQb{j(Q2oe478}s;
zMUC1)5ASJ{@(DhHlpRDZq#V@7xDnZ#&o9|UAaTe>rs)}G=Id@N@_D|I<%sD$XRZa0
z4~q)}M3_CBiZcXX^}lBl{tAn^AJj8@H6M4zL|As$KELX`$G~IhTUKzipKd10FTyvk
z9o^hXIlEPnb_zIzVyT-ffh}bCv@`PeG&s>)%C@$85_$cF>0^}bY;A2Z*ZPu$wmyQ!
z?pG`fCdiCA7Q%V&CJ=$Nm1QQ6sD^L`@B-(`pL>j-;1S`w$rUo9jZ)263rs&a<h)q;
z6DVN0<}T4>Qu>6&LE<!Gq-7)HVjf5AVUA*KEJL(Rv-SD=$j>sfGKS)R=R@UH1$}ZF
z8U2mK;k0zumk*}Ugfasm41@M>@R}%tfM3huh!N;+Zc^C`cz|N+{TSZ@<u*O6o!x6s
z(iAV-I_+0B=5O_TRu$6Gj~Ax-i%$j<R`~x~bD^e60v?SI?dO`Vr>CYu&}?QKj=M{&
zX0DY4E~^@~(B=^YCartlaUKaaT>9JoRQ_$c2cVSC=4NVM4x1*Rzt*{81{N_D<add_
zw@{gVas^f{pxZpRFCqE1%Dj-au^GF=Br?u5G;i4*0lpOVv!L6D%QSR_$zT0!jpxI%
zg?$bZ*FX7-9drvoW_(vi^VPIgI19UsGz|@9go(pOzjAon0Tgn4g(U`jzqq(wOA}Fa
z5^l7F=qP3<ZwlVx1FKn}+mu+q+uon55qKb^?pPj9AXKW`z|ky99-)s)izkTE>Ux?2
zwh@t;Xsu~CRb@F3?CrbFR8PDRAX`ne5VQ>2wiy7e1_J?s|7W_Mz@4CS<B9zn4*{Cr
zGc&19ZdnTcReskO&jTE1$DqtG#lF72n3FYeil*xd9Cw>(nCgxpgU{4W>Du_JMxtiR
zOp0~5X_l?^G+l=HVg$SW+ywD+4S}_nj&ExJ<pKn#`us^2=p=ElTsOf34YossL_}ob
zYQIcU)o+ow`+Eb6_2h8~kz#Is{>BcCu+5peyAWu!Ff=oxmst7&f{>o;b<_6`qF)a?
znrQR(k=8l=mf0~Ix7-=y-!Ld<>hLPKmIkKqwRhI>G~QiSV0uiJo12$6Cdcacn)SAa
zq;yer^)Wfb{q_`LB_ZNCy-K*?BOoC|9wB50*3Qht#KanN{#ha@ZMqEqo}SLztfsCG
z_4a;xOFGYbCnzqE$2>l_valX!0l|LcLJPa$qK@iEH9IzoWbCpVdmnte6Hwh;pRSM-
ze!{U7((3a;ZhY+3?TGi}%*;WlNr&p3(Bb|5vp-?+(LRSaRvjH36sQzyVs^iS7+>rB
z!DZfvq!o+BqY?hfDUmEG(?2yfgK?*Q&l)~{;OPr%>zZ933GSPED?6=_e3y+`>KmrL
z$c*o}z~}fLTF|SaW=99#c%sahh%viFhv#Q>=Q0ck3>U^d)Nx>cOMP6LvIk{o&1$l~
ze*MY;Q+Q{d?z#4d54UjUI#>1_xUzN_Yi|#l6>e*ty=exduOwd%v^@KaOXa;li7&H3
z7_tknogrvQs+1va_d9DA&TbZg<rNT&&ddKWd^Ph_Sj7IDX~vrq86VmqVkw)y`eY;K
zNTJzVLgH3JQsWV2NTYuU=;5PGZnX{AK}NF;uC=HFy*5MM^Oti2zkmOB)7AyPSAy#S
zqHZ~QJTL|5zHB%NLfRcP35f&X*o_V5e2*^rqdgGmPCQ=G+aO^_dkgVz*;3rJugmz&
zZSVKoqe}F9H-isvZkjvlNh#HPZrLZwJbw^uX2NLjUg3tzYbhn(R}eGd=~)&_D`AvV
zAj+37X$8(#&69_Vj~k~a(0-0)ACI!d%v@7Ow*SJXdp?;-wKI0Xzo!sway)t$jsijW
z#UwSWE=V%Z0{_&|$jM#E>^FANFXgORe6;9|t1D%=%z=+D>`~k7sX*ljq<7oX{njK?
zK=EdKpT`y^oeTC1^n?RnG5o8#5KO^ze9BHjKtDaRv=d<i4kFQ>*~Cu?3F!dIikl@Z
zR{{>N^3ONrWbh&dRwwMiY|LFy4n_-hi<Lf{pP9LMi;{6Y%e`gMw2{eUJ>`C{{*)r7
zY~!0qsBa6ZKz4C)ad}aZjgoT7<mjkiTdx6%!)Ys0Y#)REN**YpWsS*T-6a53ra}Ma
zUK((M!uDrhSg|)>8u<@w>?Z$C*$OkY1{$A1s4|7W*xu4o`?aQ~W?Z|16i^R75)PW{
zeLdjUM^K%MYHvjSy2`NRjnch1|6j+;MM=F}cd5z;W?6+%=D^KjW!5IO42hpUS#k)&
z=q;?Q#K(FOTTR-?4mi7sqQtPkM|g8~0HM07EH1`dSWvfKc$z(=n3fWGl}Tu_J2WFj
z@p@=@n7_58geN8GhrO{`tNW?8F{RRWpXe)5IXOA4d7sr-?{v?z_Yw+~bQdPa%0ow^
zg<1v&Jp|V|&QC+TK90`K`Mpt8RIC;7s0Ey<>e_LtpV18Vxhu=?#r(oZM!x<+?ZFm@
z{@bk`Ho1!7xw)GZfXrUJ*m4I2cEl1rp&_w+xQ3xnW-NEDk>9^-DjS!jG?Y9Azt`dP
zoq5nSVWmDJuQDp~$U8Y5$e`@8nm0!tsVqVsjpd*elh&K^*PJTiUPpge0$aEBmk|#k
z0*wOmCcFO)8!`VkY@}$M>UXh`@_^g#(RyE%gt(B*?NsN0=yK$(yu83`Q-x${-wT;T
z$21+Q+L!wt8b7Y=#D(i-Y{v^9KKgNXR=w_g#3BDjn!{7X<Y#2DY^h-{A#D3)68|?k
z52pEX?)6^ZY;t|<_+3{E^nUGe|84Wd_w&;~M-P>I+y#-7mR0!!xlTnJM*+V1`(^@7
zEv*=I9t?$hakcBt+x=^ZJ{LjvaU|oWfH%3Z5*Rb+Kg8|b>J?LP?_j3)O*h1xLDp$w
zTV6mfzUvVCyDHPiIC9f9ZlD?sVse${Hy^BaucxZl1u!%7$G;v8w_7ako1J}35KL3r
z^hrjj>8%pdXgxx++RGUNdxl4dx(?7GUpD#}0<_Np=KYTla}ZKA&W5Q-z;U@?DSvHa
z&PCqX$lE!WU4QxL3&4BA1Soj)CK9?&1(1Lbz=)c`Dd-O1yIGp1ml@K&8Nev4e?+*g
zlY%G#Flg&r{+uHjf%8EHqYRsVOO|{#G3bE>-gA-PA0l`dmW7!dQb_9<=+4R>9=GIe
zhLYg-bx+#^1Mfsr(TDu3$i)@WKhWwSVXXW4WY!oM%&~?#Sxix&q?0g6P1g84sUuJb
zatMpVI^MinL`e1i1{KND)0U3zsEnQunVXdA)*=+dlN%v%hJNR(#o%>z`5{4!9;cu^
zW&P#16F^Rc&z3rzazb9Eo<jI!I!;;9;CI~YX3=*{9T{Q>?S2`v5-Bo0V!nITJSYYn
zLyHRdz#%RLy4UI?UQzS$P3^kiN3fu#F9kdVRJ63l{~)kX$5SRFWQjfQ)LDSl>SQMl
zOF)gL+bg|ypey}-udlR>SGT5!gE)*9U+=BiDQb#D>AA?lQvr$WnwpbQq`4<`o&I%C
zoaX7N`K#%355IN}+DmnO7Nq~AfUm*tKP<T9%o&=Y-+gsDsuDNpEVVviQ*wB;o8he+
zg9>(6q2B^H24GVRP^CxCVK9K-&7#kY0;uTsZ)rtL8uM>XopkP!Xg;NfPcS*xy&?sM
zKT$386Ikcn9T`JPhr%s?+Sz`lBi&hkw7aw-A7Z)OoKy0qUWk?4)xE{NX^YRK+cZz{
zG1!gk$Q#iJ+*ik3I^pNV5T)APM(_LGp}TABpTIbwN1w3C)!(nRqRCW8w8!;Is&)sJ
zy2yk}CG`mrh+hO~Rpe6uKW0WPSj%+p;N{{34xgXV>jM1Uw4J$|z^8t5$>XL-7ipOD
z(+>jS!|9EQ&thbqdHJl;5*mS?XDKgA>v!J`sr#882u!YkQxP^ZLX5{8JRweALv+9}
zlGHdD+*OZg!U_GnuTEKEffs?`c6o|4em4T)G9RY*1U1q)h0ksmxONMS41?xvf(C+G
zd1u)Mu1o82X;fo-rHjBjtN^>6`3d|%g7nia3U_~|_Ow`}`|;JEKUX~BgUfCStDl2Z
z*!rpTM@AUespnx9c@bvxGa>j)kGa!@Hv7r_LgXL`075`MnTSd@dVPuc16rjRJLCC@
zdMCB^!5dg&YUsYL(Q<kxF1_@6gtMLG*5Q~G4ibwILn1=Rs|!*Q{eDMQS9qP(*xS8R
z%c*DW!eMBA`g-homSy+OjM5!RG2?1pamW6*3?5lpfnbiTxm8d=HbDLuzGSooSN>Jj
zBJ^F_zx!($?5Ep8e!4vHBVj%xxCVb`21EHd<DcO9u(o7LOV^oF9=s&B0^YxW%ZbwR
z$DfHS>0kV$K<OuT8irg072x&%^HxUGmNqeUea;qta<(T2n-A;~no?_xzzf8>>i~EM
zlQX6N>huG8+}-cjT(}07>_vu1FoO;WpM2$=dAU?gU7F;^uti~XVjR3o8y=YJKu%!N
z`5mZWvcJg1N&?sg{y*wYCqu72)Iu}R@$Krwh$5enk5)J8k18yEdc2M_i#Cd>tsXCk
zh_-ULAQLk<aJh#VnS+1fuF<dmzEKfre5NJG?c;%YcyCSU^WT+0eo}b=^$PyEucsGZ
zwIC7_wcRp||4he0&F|HJ5R1^2Xyn5!B8`{z%2))rZe`gH=i~lCC6VGOmIII-)4{TU
z4TScn$w;8g3fD&knZFJh)Qsqi>0K42<Q4U0v3}?Gn#bmUeRm>QZ$9#=AEr?z<39u{
zv-WCG!fq@9@Irz)UD{DkC<#YF!}?japJdsxuw1BJl`|*?cC>*%&O@7K(UCBob#bGb
zH9c#20K7)m$8T)ZP}4z&!<UnzG$&pq6`@D+>?922I@VJCa?$SllyH3KWrXFA^);C^
z@WNJPslKM~I})ptM2u^w4;w<y<30)B?qjcIX#ep6@ALjEkU3m>V+TXG<8=MzE{KTa
zUSXRuDj%1I_K!Y)?KUj-72YJ|K%agnw@)z$Nn(26qEu%0*=z_d-RggI0ApMSAeZd(
z47RCY#Q4JgLv(BBRQa#-pSwRiO*+iS?U6@+H2qBs4`gS=24b&8cYK$?rI~!0)OfCy
zOH|sAgl%DdP#}C%f&kA8JS;!@)Pi=3L(Cdmugo?qQ&oQG@`ZK8%ROhE4`hxgm<4sM
z8kvl!Qw5~>$JO@k0imS2LZZTtzNO1OpW()D8X9v9rs!*aBNV~V=PU?A#i^WnM=03?
zFgRF?xsveK){QQPIRmVc4>|Yvzp_@uRG#-Ef<UOYCxA^ho2s^cx>IK+TCXqm5+_ZS
zbv~7`@a3JhiXx!Fy2IKlINV~j#{e3Qi#hZdi;!qiH`QbR>D4>PiHMB}M{&&Pt48jW
zhY8MG_ZGxcw;+3ZtYlQ8S**LtX8R2#G^d+>AhYEfj$iTm&HeD5E@L3PFh51SImx_|
zlJo_H7K(G=b)StsWr#>kRn^SY#KeR_K)LPjB5e}js$K>@aL>SsMAaPo{af?+UW|hM
zk1z`R@2zFsSCuG{m=38s@iwJp=Qt7V!?y<A+$lI|YNm`0DQvu2Q$grTk}vbolq_WW
z<w4zHnBWa06=rN(7xyQ_nd&L{Hf4s1@x(MiC6bE9R&+g-7pNhPmm6j$5g<AzMH$f%
zvhKYNz#a+Wu%>N<d&8w1{PmZv-qF3ppfDnuo8)rp>=0nl(1#>Vn@`zx?*2Mos6+ih
zLPg4Kd`7j9F&i&xw01R4P3+&T#s!`pKXZM{Ut04v>gr-_<{R>J&KO<j%lw}Z4+{w1
z1oPn%pxtaEI?-BxolY22)}Y)KkYM;cDi*Bu^-B#F+q{Sma31hGp~X8|7Vf4n`g-uT
z%YNb$yK3=ONzSWQ<h}x>Z9w<Dnuu92TS&)!CL*6&=#&O@-dCAd!6F~RBO-Xw!e_2)
zz;VpWN*qQGPeeLNmX;f)U+_MQSpP11av*kcu-dZ=5Hu-k;NFUW!}UgrILu~a*8Rb-
z#HZejcZUxLY}y70)XNsC^vsDjlmKKX#z0HKPS4?zU^7;EHaT+j`7~)-GBSmhI^b=c
zFI6pzX5}C1DywR6a+2I7?hc)nY_LBN{?Omss3@6Ms)v-#u({X_iKm^zck3?4Po+OA
zu+D{cyb*z9L_?X^A@c}!lL*oP4Ky<<rCvN#nYfSM&MAs3eO7sQ5zG6|eA3N#Wgb3D
zu??d8#Ff{%qkGJX;B@)<VoZeu_4!?Goj+XQ2^&alT-V8I@WC1M7sj2#i4Tx6S$iHl
z*kY)>RQ*@Ya*zsnIGfHjd0y65SL0z#fWUp6{#e~*n%Rk^d+EK1dp10s2{*=|`0Orz
z`yQH<=<zW?BWNxpRZ1{k{4#6#(ny!k$jAQ~$vL`)kGi-%BJ}YGfJ||I^nj^Zx9`bB
zF?{a%R6L~eVTM%SvZ40NCF|zrU~<7jQ`Oel#*_w=`cX-?>Ef;h!gVkH8fVz^)>Pun
z>rgKX@Kl`de3FvWhrrBp0i%N88Y}a#n$_M~7a;xM-wA>Uc!M%6<;{i3AX*9f*8b?7
zn>PH3;6xaV0wm;Bs5Jqmvh=xWeWQ6ZW=@h+Eg)m1IKw#T+~>_>Xkh?Ci0nF=!NI?N
zQRl3y`*ZxkgXk)I52nNYw|~4qqs`pL%dw5kpOI9Dh2YjE*yh;>f}0qIvh3!R<6pD0
zXPc?MydKB7{@4Ay-U%l!-IUD-)>>GZ?aj<oPoohrjRfP@irA}BB6E}{-*I@qdfS{1
zgJ$#>OZEyG&Zoqc(K0X1G_x*aaWP}0PA25cgr$%#S_kHcM{u%|jF7g88HjK(r1k<i
z5A(17;PQ*+x23TBi`s0-CQX)9+duE=>B-Jdu0E%2EN!~jETs56`Q69BnUCt>8}!oF
z8!u%g-%I-2-9xMKU_Xev3|k6qqR%ahp8tLnl_BTNx!7T!ji1}yxA&>VoLEyt?3Hx2
zG&N6_(r-gfk7h0?EaqVdEsM@>vzu(XX}6m^mkeod-@-bTQ%hKzk(M(B<DH;O97?&6
zI-o?p6+f)3#d|k@j__+v(l-S0mS(c3r<V2uH<nJ0j{9t8Zi>n!_2bKZhY}*viA>*T
zMNpTq)1F|m+g2p7AX>I2uWQtWfG^&d%Y{!P%(Wcl<s??4z|>?uB|#-7986J)m(`)F
zl;r*x@g&Tp!M~e@YP`l~>Q@H!O>Da$kapx{pF`DDUs-wB+O9?h8rEN~xB6b5n95K!
zDu13VMBH5Z7`m4^oCa^SaeAAPqxx@OLO1HYVM%n;o|PCL4@R7ypK>e~4QTD2Ru<a%
zHVz%XvN8*4B)MFvO8iOBmy^%S-}7<L&8XEy4sk<ydog=)Pr1|^;lOx3TBxKfb<o?x
z{&VioDd(7@iR8W;Xmu-d<irJI>gmOIJ$;xak^!lq;E%!8N6@20Z<XYHj9DIjSh<hz
zGg|Lk57fd*lK&b*`=rwkTmM<j2g#N-j&#30hNhbXv+!rxFMK*Z<_vvjYT5~8ntGZY
zh8<6d%-mQ5l5Aa0{g3z>ab?YZXV-3on?A;yqrKbi@PrT@688#;MZ(fD5WQx1-|I!s
zY5^8uEoqg6w=!Uu>ET|nP9J{fyYkl7p4=A;jn@PmFLVL#Z-;9P0IRrXF7Z=v&2ZCf
z9K}A#6@jSRk9uC(yw{UxP`D$=Dge^{64A!L7J4s*u`2;CpH8Cno~DkEkV?{qkk*xH
z-vH0;yz-US8pnJ=ef_!E0-=C4=cuoaDW8R%T?-v3YdSkX2day-+N5_c)PfQzuZEl8
zgdvL&*T?RD0NPCVS8g@X)Xe!dVG7r+G<#W`q~YpnvwWEH`sh>R#MxDji0}3LW8TBP
zEE&98*mZ^9{9wyO{kh>!fS4%>RbGnQ&uMFFx)~1)M}#&dQ8$SJ4&A{(O^t9suv&2-
zP1I9%*Al$MvF#$gXzuGLpxYGcR;5&FUS6IMmg9k$P=vqW?0V^4H0g+GL8G8H>&zhq
zN`uy)l^&%{cLsr{txJeb@U<}FX|}E!ebt`g48Pr;&L982DiCB22>}3)`MPCpn%grZ
zBI|OlL`>8j#dxRW*}b^=iOa1mEh(YJ66@W1j>&Y|kugsQ;%Yqy9&x8$vX(;*%AbvC
z9}PIRE1zC-{l<=urkA=<Sm2J}Z}7&2_ae4=O`hMXfc-+p?S6Ylv;0i)vGh$2uT@LQ
z(csa>$nD*$N)Zzei`IwBYE+u`ICnIU*1Q0Aa0c>!-pNi;(8{s;gJjp?ReqdLHAW9V
zRsejqE@j$07ZfB7{LwCU_dath2KX2y@b$33W>B0+zt-26BCaE4OnNal7xR;x2qslz
zJk)IDZmPcBHs&7Z!QBZ1Cohy0>sQz=LGN=d^Uo-+3H_MYGoeuaFPQfD)wQ*TXz%`#
zRt&>Bm)jw0y-Cg=8$}=;@bJ~1M486ZdcY2201m$dfH7x%rD}lt-CZ9bC{WsaWuEKU
z>wZA;H!OLf^KS2}&1T<*+Xfl%bB2H=Q{}Y!3t3tnTR_*^45>GS%3p&<ZtkCoc)7Wy
zBsw;TzT_QdD|e;?2uPZZjjd5Df}oubbpas&!d8gt)|u<8h6eFI2sw<YH2nvSfbUwo
z@KYc1LuC~E%W+ljUdgMUR)#*)R^--?kbV&?(AvdqRrrC(LnyQ&o2p3+fIa^gLJ*~2
z)hRzXHs1H*7Nqcq7n`m|dlAnUoXTM>`n_Iod;FsvZBxo|+lu^_x|p0Bs4f047!wiN
z>9MxWp$baC<EIDxD4NlIat>Ko0ExQsZXE}ONvTe?6{k_s?}(L=?jIle7CzGnjlSC`
zl*9hMrCqdYkK6~$B;lts4By`3@bSla321Jz`4MdpAZ_O&4d85{?<6j&2Yd<?@!t;Z
zVYslG)u44bnjR)7OXT=5PlR`cgXqO%aA%0ARt8;iGB3Yh-?3BrsIay${g1ahSq#gl
z)w~GWHPTCabZ&+dC)4YBp9qHghx?2<_7jpB!ZyC__He=&e!r}}#(vTfv7Y0T_EzTg
zr!kL_&;ABREpurxkY({!T36AIUcZ*^*M1MisG5U#{wnTanw_Q#?%+>!wXc*0Yz0mU
zO5SXQ_^c8WZ6F=C*z#xOKBUV!udWad?tZ?<uMk%^`XmBW_S85JX0|C3KyC7VRr7eA
zecr)}zr!)dnnVfKw60zDJzP#Pl4!WE%^<#7sNKZ8y9g{rmH(%?_YP~i3BpAKD0~8n
z3eu#Bp^4Ih^d?vcRY3)$Di}H_0YWdLqJoqFVjwgLA|N15q=P6$Lg>BM&?7Z;?nb}w
zob#OXoPY1V`wu0*ve}*8oq6Y-8DPdrNhKm1t;~jYI6eDX?~~+;HwNQE->ug@HScxL
zP7{2x&Kp%nx)4&<Wh`-J2><t?0}p`N*E8`!{O0hfC#6XIjuIhoe4;sXM{#VV_QTjY
zh4fcV$;H6pvaz94oPG9IPxWo@K!Lw4^v%!UBi(x*s*7U`p6(y}$U#iZ=#kw&sS^14
zmEC1u_XQ1+_a7*0)5qSt4!sJT+R^x9o_;bjtj|Y!Lyt8+_V6c~|FKE_Sh|c<4uE0v
z@#SN~CHn=M?=HO~%LZRJ|Mn4RRum!(8WUw(q*zy;EBh?7P$voq*5B5YN<krPeylzo
z=r)~0d7}KJhv}prD<uzoaLW5HBvOinVn=hB)K98f2l(!4$syJhW%3ZK=_5T*%_V7K
z3x~yFq_gyqLj42E(f}>`f(EegnvL%m`$&Gtn|g;bLH?GcTf>}JdDr1apJTEClg*p2
z{+06D$K3_+0|spVVhzByqo>Gcc<%JO)_T828+;AJSLZfsM6bdkn!fr0O5kGS+oocU
zMaS<qj3%og#x;EfR}f~rtK$c#r$b2pXiJLNgJ$#PHLF6<E5^%!{&R6!RbTKh&i^-a
zU8CZt-%nk(gM;VU%!i{fl8Yvl?6yw@RzreAE*86<u=97z$RF-#T&18ultz0FnE$w2
zfCa>DH|(+3YjUKRR%+cF<Hh4uT{`ve5WZqO2ZkR2LACC)ym5@*ehll;+*I7sA0}PO
z*nT-PT%Ogw^Y?K!Pujs$@bmk(TzVZSBJDdotU|-Z#O1PW1FJ5m_f#=#-sgC6O62dv
zg@jPJ1FtsH>@RQY%O3I88peh{EgGbcOeuKzgJuU0BEk6iX@=ZqBT&zn)t^PzuEP^}
z?QxP_x_LKjqB<$aPqYdE`_f#@POb!HWW$8%@qrIC-X5L;rtB-t2LN}=E8hOl&Y%Bn
zzztT@iVtKfd3yBO>M(HbX+~4~M@tB$r%;aPY>*pPfTd6&$A#pLOhND61^+LkOKjM`
zc**$5F3T)0GsW!Ti+2AABPY33pYl_2O}c&?f#@6IRq_l=00>+B%0A^;U|@b_^QcN1
znSkM#wgi^TN6Dj3Ez{rFMCHc~t``xJ&QgJ8=;M~j&3LrBEz7)?JX`Eb*5QW|c@+vM
zLRoxc9^Odo%)I3vXlfqzRbQruM?%U7ejQi;wROnb<|pq?XR&LNVcmYD&V|DWb=$Mi
z`wGcnVJk(A@4gK%1&5H8W^ccI>XEoy<SEF@c`x-9zsL={oqzfta-2&Hl)vfZ6>;&k
zrYcUi7j$s<6{co9>+AhRt+_Vh<?yD{Y(f_MQ~F2y=1ijn5ns8NAt^8zzg*m9yB`-*
zJ`=Zu=#vbuySa{ym$**i=NI~UNFJ(XBcA(=Ldn3q!~7F2k5Q?VcACDgMS$F&m5O<_
zq56kX)oG+|nQW3q1AY6U!j0pW&8Iia;sLFGGgYOk?stbGY>WhEjMpaUXRE}=2ChSE
zeqm6l@+oQSk*+8%Rvm%EL!}>Qo8Q-`4Rqi74KMoz%*gn^GB$m+Vf2&|rn<9sd)~oi
z&z7L^n(<m9^-)4ycZ8{kufj#VQ^Dx3OH$v?Jw(oV5gi6k-lpOYZfkSqkN#YOFkJeg
zY;^59?g)Sc9NGU@q~cS7ubT_V7CL~~%8@P*q@d9K>8_gBZC+8fM*7G$S%yFcnXh4I
zGpA0>($s(FHky`CZi>21!0b($ZRJ;(qLxzczfq;%7iQ|Xj)Qy;x_CKkCT4iK319^C
zR8wcLDBA}szwN}P0C0`tsO0#Rovko1Vo99`5u~agIdF&V{RA#pp75VOD(mU8*5nN{
z7(>v96_sJ-bfa%Ku1~l2Mr~_%hv`AeALON-oEpq^IR|!x7L-hM8lM)9TDbX1A<z3(
zf<50E5#jp5LdCkG9Jb~XiAs^Mg<E}Jy!;7;uQY2iI4bS&?<K8OVHI?*AY+%VTsiRJ
zkM36hBwvjXKH~)nRzY1DAgb@$uo7mra!5&1eZH_29ONSvVMEFFS-TKoW!2j)LmI!a
z(-vB2C#MZX3vqXq=(^TZzt*0IAkS)C-#yN-+?L$RG8&~y6njfy1g=?ST-g>O+_2X|
zm4~M6lBKl&;{rfKHb365`F#gaWF{vkTjt((_?ViTkFa_A`|qsHv^7kvHUssiwYJiv
z2*F+W=;xcjX~(7NrgO)xz1Ic1LNmd9c#G?|S+MmhMiJU814>06PTGX;uQcmQI_{qr
zC?9xuH^KYh)bk6`mPCQ~nyd}ax3YADZ#`8>_+SHDvCVzhe9`-)ApgP3cXo5h9R>ST
zD~Fp^OoC@$%Lbi?04l1yH<x038D+<#y?hbV!$a*dp+OPM7ky3K`4#Ia+^y#No4b)T
zhz1nrw*w5IJT>2b_JaT*?;1LnbL=O^y{%%b1Kp<12Ma}hvRxf{K@j180W_?U_p+{|
zhk#8}PxaMhWN;>LZY3ZS<3Bi(6YNT~UNV~n$^mJ31Z0dL0L>E-2rz_&l(lF$uExR6
zPOf7FHYy?f&$AN1LGhN1$GAD4)8@^G2B>al*=vDF^Xbf;CV`MX07A5#s{bO_v&#36
zA3zvniLjW;0Z!H{*xA(+E`C>UyneXvKGuFHnqP-A0<Vfdkpo+{lwf;AKt<`B<~PjC
z<?m3_tWT`|be!Y*G2p~3v2Ki@dB#_Zw{QZDlr6~lRwYJ9P_l>&m=!8`kEg%~cTA4c
zjyhm*MHXnkBO!-6fV#KYlONC{sGq#w;_iXCPXi8MjgipSr5Frr=tJX^jB~W&<8mpr
z``5nQP3%#ydLL+=D;}T(V#I$xB_CZak{sc3rRyHQa@i8#D{(5Wai+ghXya-%Q_pSZ
zR+v{bFr&`NrQ494U=`VP#F3MXO+X3IyyWoP^Wxdpv1N)Yhwj;&MI(z}<KD%`-!i;v
zw&%7~{i~<!%#c_pgqS$61IYeQfhGzAQDmmD9gFg{g`E6=A0zpT^E*;}%ND($qKk`*
zOYcnsH1k9Dwe6|-J<0k|rXvu}wr;zAT7>N;^*7&w746RRfgq;fX7N|VA<^&q^Ak1p
zE-s_d^_t%6(=>q$HAe+0Yd6<T<NL3|a>|R{8tH13$cC)Lzq6=+#9XsE*xgL;ioH&^
zOL9YrOAZ~}ChY;9&6tQy@~9j6N#oYnR)P9A!+5=8E+M>Z3h!*@=QC4y3V{9pxeb_V
zT?}3+;Y+JK-ryb+*IqH2K}QgV5<}0ePAddvZDK_rDaA$mUKc<KS>Jl@vxlBM&q^ib
z@OulfG&<`>mc=je!%XjhFt}Jv=yzC~64IUh<-L_d!&E;&tSY5mm)fx~@C*E@R!&zT
zrCH|7a&Qz+4J)ELB7IqJ-UvG6v9WqeHb^aQm*4$<M~JtsgvtMHiR@S#Z=VvOA>Jb0
z{kq64>=WU0#1z7{TfDcaV{zP?I9JBGu^C@b&-<lE0HZPD=KSy29ET=IL0?Tvcv(dU
z(@IQI8)lQ~=|A&+w?Z}bKGh9XB_$o{Zf>an?!*CXxZ2-WO5d7%$q4KgZ>kna|2M)&
z>DUc=#MhCLL?5?2Mq{emv{nek5r4zr^`>!|3+BO?WeDv;c-B*Q{O>1L8o@<P5Cme6
z>V^rnliI2E4z}{V%m??4&)S!?Y|RaVG0xr0qRr98fiZK&r=0oD>k%uogId+Lx&#@7
zbzbg2?Ab+ck`GF})^id=?pWV?dE~jE=$ZUG7`6Nl{~*dQb-|2NV39Tp^Fj?dCiF}5
z%oj@{acKUVYV+xbi2QR3A&j8Xvgbv{`jgjALvUDi-q+VXmD>j}^J|P%++Z65JouFG
zpN~X?OrkZXSyFGE90f5pBjLTA=M4ZhAP8tN^g$J>ckkbqDQm~rJju?=Isq0PcI$Y_
zyLazgLAr;GL7fasM;DiXh=>g2AV5O4#Gi0+)PkiSaLlHXmIMGi0zm7m9g*xJ`3Z@M
zu5<+If9k)ufows6m!AOj30!GGV~|(JP|UP3HUp#$0446DmDzw{^6<jCjWy>g$~vH^
zUeY(<U<JfvhNzw?PwA^gkFi}!*)Y@4s7D2T?i+V@b+t4HKwL1_b*&sg@#^OLXN5+h
z$dZ!w{JZ+}>F7G}&Hh&!$YMW*2C{PMcc|anq!u#>V#^nm0yvUCoZ8NvX)ek4n-LxD
z?`PLHiq{c5LPsj`ADBSTmPs<!mtcBJOH1=x9`i(BTS!N|bNg6{QW1{dPrSK5;xciN
zC2Llk-nvwsr9zUv>P?p{!RQL$X(n4=3n>iO91PGmM*-yI@_#@Ut<<PUs`mp)JDgxr
z6xx*Y)(sCPn4<$4@m$7t0W?Tc$6<?IrStjy2v{R7AhW57>gc=SFy_cgmjL9#tP5Sp
zLqy$`8H)g?i<p}AJM{U5g==lKDGW&(0M>7RrZIvX|3AWY*0z))l8Vg!<5{RgwUz-j
z6tsi7#pkdsi+5?~Z=Jg%Bbpfu&MG&}lcUmrzm`v>K3)l6g!~^<s}CW71|i=xa(>RZ
z+y919Vl;?y352qSEdz-{kvIF5l~Dk|cwAt-<O?=#(wm=;DlA+7;0u7hfDGa5FOeBB
z2L-1BTNXn97ToA$dLFFXnx}%#2IOa)D&+{L&>bp3N+?VZAfvg2{d~mpa$>#qMC~ph
z(G(eCdD5)o@nU_`3$5UFrJp_tL-L-GRjEv^3*I+Ak`xyav6n0oa9@hp0}O|*lt0-`
zUTFqC+TLDT5NWwbaW<J5hRD)@Uc6li;CoVqsrm;R92yiR2z(ZP{LywIKl+vBQNYWg
z3l_+e@k_7@w*F!7(XpWOtxMrx^E0FbaBXh*{w*jX0u7m6T3GHedkl=c=BuKVAPKI2
z7m^mS)fdp31lgpcnjnDaw9obApD#t9KYzy6KD|!+8BisMf{P%-2g-nScx0Vmcpb;^
z-)h+R0wC6R_1Wg@#~HHmpIWz%5%E5(_Vr55Ss+)kVF=$hrdqrUr>x8Hi>}waq~O58
zq^@B9U?Yk9FCK0m(7xMG|ML2VO=|nU?(Q=jcPVuBZD!29J;UY7a;4JO*GDfiYFv7D
z+>)u4*MoSm7HXwBKK4Y6;j)(l=X#sy6P+_0e1FUeit&0aBf4R${Xi*n*<Q=Eb%(-W
z_dX%mId5xpxGRUO{>RIxk^9ZaQvA8gpMD{w&u|3&&pR`N1!_=C`ZUQ)<=<1J$X!N7
z+Z~m(F&FZ8cHT;l`W*Sr*{yh%xg&^dz~EAf!8|C`zFvNg5^VCPMVr}KHIrJZ_WMCF
zf%mv!ZKI;Jsh#7{z&A4(D&zEuuF`cW4usqU;nZNWN>@-h_T|}VD&2oQ`DZwa6%xO-
z{9|$-czYmihaweKU6;p?sRJ7pdX?fW|J%?)&biJeJ?9x)0DYMQ`d{C{e8v=pG$K&y
zbAfHTXU}k4{$D?tvgL|@m_ad8X5_;!@`q4-!2dSywUuAia^t^GobNQ^bQ%8gX2_z+
z|I!(brvK%BX4q8WQH9r)D3|#c4K{L<zx-?fb?tw2{7kfx!uF?9XAk{qrYXTx|NQzZ
zE{7$OE-=m=Z^sL3ubXPm{*NI#6P<<t6r;0l59Mtw&ZN!Zf}Vt1i<iPC>uwrZ103f+
zRt2zegQeXX7i-O55H5fqr2kkNlyT-rmHr!;vU?Ypm^b@<%Inf1TwgR@bo_RZrB3`K
z>m1>KuEB*){s^NIbw3q-_~B(jD*Cw5&D@!DFv}M^`uJ}Hmv{Pr{W_#I*~kjO9?+Vs
zfTqgt%_?M3dE)4F1D6%L&QsRX_JV9KfJ-RBLL-S<`v=#?jVAB9d~dv4n0ovh9huZu
z^sO3z1oRm%6UGn63rwF7o|gN~Ou5Sj+9fpF<_SL4AElwNcV00XP~xs`bzuU#1=m#{
z4j;o)MR<6rx}-cQx^*e!xGKfkrP!h6_nV)I)6FtX{0NpzvgDz{H#m9R_%a@PwR_Gr
zhY?OLIE1ZWX|gg)&NCtz*xF`?B}KF-8vAvqw$Is1Xd%PbHCLn^k40$RnUqTzmi1b0
z^bQx`=g5WzNH+A0Cx7_STDAc?n&3BkcAg0gkmR0Qwqj%AKtK0VrSxAA`1(smAb4xg
zf4(~gofwH_50dB2u^Nrt$DNJViABy<zYi=;_kKL0qvwtyuJev#ONm#onySBPn`43w
zq|BoR!;QuML`S`16aetb^flmL4K-Ej@?m(U?S{E^z0mOb9Y*%JYVXIiEF~R>b=c6e
z^S7`~7Y2G_?)?R~a36skzs8rk>~FLF?9E<tnf68g0eTu$f|M^w9>Ti)^m$p!#ZIOY
z8>1DI-Y`weOT=fL{9n0P(mv(J8Cf{t3aZ*MU_@|4Aji!i5MGQ=3ns6>2}G|%f97DZ
zH6*zJ1dh?&8Or+_vECq26xuv_gZa$-dBgXCIzOJ}t82$u3oI8iKOX+5l^Cdf`lV4m
zfG`290XCQ=dl0Cj^yC>^p_0Im&za_osG7%yJjR}z{50klwJnaBMpqXi1s2-_&{LZH
zh4J*04E%^1x)i3^rKfg`E-Ca^wylq^kKOy^2bN$b5=3jH0ZQ13m4I^+Z2e>S-WLzO
zthKdUa^JqK^3^s3R%vrt67ens7Hk;&x|SB^xPP<^VvEf@s*YC4WIPTo;%pBOQa7Q0
z)+BiQ^qKf#UDF3rRD-AJAD9LC++g|XTgBh!xIz+QyME@(wXC3TU-V4v1f7N5Pux~}
z{_OL2CYw8{IOYfR=7p_0puRLtM*|@hV}p=tUfl7zPMG#1kSm#Ct<AR8>&j@>UO=ht
zPRXa^-O6iFR?$13C=h4s9@OQAkf90*6|W)0pyCuP*_S;O0)|mx$P{{ZGwP2FQ+h&h
z&ct$1Iz@KGHhOQJl`$w=n#_*$%)z0%@3}lf-nV=_ana~fp}gw^1fo_W2@X7PAPXf7
zB%-T@=GEX8?+0wL$Ut@&0&l6F)-Lvf9k3o>s;9ZZd`&yApsS(wXq*&$4>K@;_$tFf
z-*fN&{my+}wZ-Bu=7^)jhdb|1jU7m3219Hfm%OWVAq3=x_da+UW+b}ShYh8s#hiOn
zxr9kX6v`U(yze?W<wt(&H%raeT-IVfIJ*+@-G62Ph2;GQ)DeerR#v^2N6vZ?hY!%y
zXfbh7(esoDmN7CDdZQSl`)(uCFLI>MI;1XDp2#Xs1iq`+%z`@XXTUY^bS%<V(;=RP
z6U0~tl4l&%9a4IujTG(gh(BINC=}g^E?qxMuh!78H%f+iG3=&GCB?+6)$+OL^Dk0=
zZAp{zH*#}?RxRL&_<>f!zJC51TN2wHTyDF3@Pti*-k5pkMmUp%{{@siWCLd~VV=XH
zlJhH(oO8)_^zFR_@)QTXzL>ih<mbi23YC-EX}9N&G<TyhBNmpHfrz#0Cvr+(Gcp7O
zv!Fb{A?~-c*0+AuGtcdXdYTFPo%Y&T;L$<np_poL(u;GaoX`P$wrtv*ag}Y;$x=n@
zMvfiz$KUjcgB*pvKFNJO?*qPk_%K;?{N%A|H6pUVS|IZ%su<Oohq5XEs8MlPd`Gr)
zgRBI!uB-4pZHFA76D?<OsnLUfoPyhGqA}b>jXb{lxwX+jm5JHjDQnn5lV`08pS`cA
zt86Rsffx_y{I4JN`9^o|Ugl~FuYJ<d-R+JhdOHd<%cBfIjU)Mc35rI9gPR67m_oZz
z#?40OJxq}SG1S5`_;KB~EaVOiW*}n(b(2536&)Z74_4zFGAiGLc@o@o-%C7v{HW_Z
zkh7HmuCUP?EU)wLPIoT%%3S-{W#btQ+j|L9JcM~}7yKL?KK?R{K>Yo|GGH*GzYovt
zGOK^JC=zA7ZXsGS7fAb|i~&-s=H@m$K*f&BmL@sLmuf1@q6D%O<_7zcKNL00z}SK&
z>lJeB@fetGJ_t}ThuVWGP2wAOtr6Nu$T6uRPPl4oV}BWbGe7X4g@G2eZ|u~vA-CiN
znCTTir&g_fM11uP`gA^YpjAo@L@BjP_kgl@Ff4eCHf4`gb2JVf`noFC0~8-tejl%9
z-<{nXzu`AY(-hW|W2C?3AE@(j6TmS{ZOO_NSK~23X-=C9Jwn^nDO(oj&&#^_&4t4V
z?A<L9$1>Qa(D~s)<bH<cB2=)M_joX!U~x$a_h6`oDNwa}IDV^nZ_R4ElK3+hb*rjN
zcGGh$hT3L|7}<lm?Esm1aIn0T=<P<Rigw6)1`3TvdMSw!hYSAL5}@p?%3}n2Lmk+(
zdtl6&fZl#lxy^JW-Wo)S-)bNK`Gh61QT(?OX{ajJ`CYN8${Hl3y(F2K6lK4fSFuuF
zirbXO(=d9-dMBfd%bx)@3q3q!H@8caWxzPY`y^Vgc@Basos~J$o(1B5Y}(dc7V6XT
zwH2&_rlv>CjjsCX7?|8#L?l#8X;1J<3!iOydKq%#4XrC@uv+SOybVG;Gy5sQ<Lb2*
zkP0valV>uNQh<yT<vJA!C`V;C(D)@sj30NFQ~#F>OW2&{Huj&N%gVe<Hy8J|WH&)N
zfs}}!k8bphUoDyr)E5L{cSL$t+8gs3@fHc)o#=oUG{oWe57&x%N~&F(C3bwGgNez$
zE9uoKJNcpT7{Nb0u06)lY1Hw{;5nDmpM@(Q?XQOdh_Lu79k1VS3JLsn+jtz#D8;90
zqcL-TcyLIea|`V`W@|R2PddsMkKkNSOj=oeh^SC3ooXL6zU=uhukg{f7bw)a1tOyi
zdmH76TcW*-7hyxIc|o$44+5{VhWZc>cb9|?w|5ptTdrN6evMoi0davb^hq=t%)rdT
ze_m{m5sANK2lBd-9&}nwH$_vyIj(wvH6-TtqJBcn(Q(lnEvUJB57dCE8C{#i4(S>U
zaxl$dka^;+hL^tK!UEufH};Prxg=nj6Pra>hpR0ZKZnqBTS+`6a2wAt3X8o)m1N9d
z_cVDvYz*NdN%^GVVl|@_RLJ00@^GYBikt=8*lO;X^oFIiG{+8?K5yo;ylQDI%bR!B
zkoGMNxxr2m&Lh3OLk0-;X}__+%yte5_op!>8|8Tc26<I8dxsH%J0KxVSIrOAC)P}M
zd4q!w@-s4K?!yx2v8K`>Q5Wf>?kNZalKP2DaZ1y?W7~3fiKv<tUGZ*td*8)3;$Ca1
z?ZR45yd5coe921P%8SEPMImQbvVxS54u!m=M*RLbv3E$@whk4U36E6cYv0UwGyT{C
z+dwS^<L{Z8E`rM6EBa`HL{T|NQXtOwV>cgUNRicAXz|g~_+Heevc~~;+;wv@H60QI
z2?e+|TsIf<fMDsvT?KjaKxkL~Wh<cS*n;fA@o2HR3@0Sm_*QnB3^jpG2DaN8+*dGT
z#j@LVzd=9h(JImQ0B>6)D;{o7js1FvyGEk0>|62QUx&s3#!<>8J3f|kFyEV<jh#J}
zJF|(*5vqsVA555wDl5%G$F?NR*OwiiF^L6uw%xh-%-QnL|65sMq13WFs5U%H(%Lnd
zoR9_issiKpirE*EK?p~_0<rMWM5-td0FZ1m6;i-QpNT~>X?9w~TuS!ddEFO*|9yK8
z>Fho!Vwu<xTki#V!ww{iQU%h%)ioY>V32Elo{^SGr_z71+3&6L?J{&ig)RI+HDJt_
zfwFu9cy+yIc_Vknrc*djFr*5ar4LKv^`4IY3M0r_s~rY5Wdh67Yw`shT#9Jc;eS&2
z^Hs;*%Q8iQ8wR}!TmfGn6)^c6ub)`FfdEuZWE^RaLqC70*uJY_;&7cg?5C%6E*{Uj
zGR9<M3pU00#DQ)1tK8h&iD4VWOJegWX!rHlMV~7xo`L@)YqOuuF#_l*zVI-yS)OcB
zaL6cg_3Dm!i}YPfi7ApJo9fmb_(dOpYNfay_2rM1j~pm8o}gk81wp42oMrj8a5sOJ
z6G&DQs_9eh`xf&8+ALoxZE4MPJ<p|e#FujR8Wv%K!#6(aCT@MiS+Xk7-PWNHKZLUi
zY9)iNIRc!!ludy~0;`jxuo?M+0AYM=n5pNlXK-iqvNIpolONTlHlD#)(|+xFQSQT6
zVQWCo=y-H;bVH}J2R*%a_;Uo++ua@hZbKLTrf~{2{d6f{MoSuX^cfIDvrjM!-t8Jb
z4%PyMFYvy@DhWH_7gLq^!RB7=zP{|j*V#&1n%nMnzqdZ5r%!3Yt8wB0WjT@q8)Pib
zP+!@@5@wo9C^fun`PFGabR;r2D{IG1n8ktNg}NxM&r&r{V)wAJp~2Mi6U>M!X5YS4
zxsGvD-UVqlBHOQ~OtGd+>ku(rCkgYOME4gMm5NSQUCT6qMTVp^wd0LvPga5O)K}n*
z`ba|VQ$i2m;C8=Mzp!%{@%U4+h^XkE-<x_GwANW(8U;_wNb^#ckqZk>D_WhlefgF(
zS6S#c*+<ta)8oa=chhEg|Gc-h)~<Vmq*Z>c+C|u%74q{n#ch?J?<!sN8XYi$_0>p{
z74W*-=?(O{_gt^9e`R^JMAbAiA40nks^T|>@;@k-fPVWy_vcv<T9E`<i0Au{U8s1r
zZ%vq4Ed@=33S13lo{8kJue%uQBOeLGR9U&rS8?@s9XPw%W-r1dSU|*1A9PtfD;LDT
z6xTit`fx7n=Sr0OODJYQ(s!w#ag3hpgs>J+nE2sFf;<6L;Ih(0b*i~*fb#`UCg#RL
ziNm^JKOwRxFbAatb5ug0@GcgWj<VXZr1?;^N;W~j?537Nhx~zO<<K|J1j2C97ZIvI
zU5*N;6^GojUkN6dgy7!Xy(;Mc9l7igQ<B<vHRIna0;RpEn&iobaBJTj*gW&p<|)^U
z)PHVfhuZz7t~blj8*0UMN#)8C{XzWL-;s!t=d7%iFybdY@yIJpyr6h<IQ7h*PpodI
z)-f<5^@2MuGl`rK#n=3uen<#Jm(%L~dPaPa(sbM_{#ESrzpq!jPkG0%){R3Ed-IQ*
z1bg_YslTt_W1w^lZ$4Zypgn(+*qD#^;Q9NB%}em5at)?cU}WWtGVg@uQQ0ng@<QH8
z?{L~bR|H9ehekocC2Ry~L$@|kWuuOpqf*!Oy$2gs4?F+9IaLTeUlT3pc<r@!g1|tU
zBI&7Kp~!4mnEqWaoB!>i^TaA-k<Qqi>_9dhOYWJa6|H-;>*r7Y?M_zv`0><BttVTC
zO*mo+|71^&x$yw=g3>tB>KN4ARunOPgQ&P!gwx;uHt|vhPgc7~OTb?$Oyd+b`}b(s
zFPc9e2H`u#kH0Ki9`wK#VxvtzM67ZB{CA{S|IEgV;yVVzg>oP7)qOOZ`S+9AKc8HW
zmWf$ubEnUbNp`$Dy^MYNcNW!dQU(Km@YzvR_YUzWA+O1w?{7bB_>Jcv5PH`*e)8Qi
z@0!?oMG-9>H$8%jxqaACY~K@bm9PSk6`RV@op0j#!z^Rtj#FK49hQWF<pFugkN7jl
z8*JuK<XF6uDO1JJ%u?smYSh<G!W!6NYM*$1_<k{`vP_kODARy1Ys41Ows|i;%iCX`
zPaxMfKe%l$7Y@v@t9X!E?&tkWYeia$>treq?1GZwcX;y7%JIzF&b_6rQhM&wperFp
zao??zihNs#g?##}D&)--^b)i6ihqkl3KuGqK#gYh58#XT!_!?H-?^vns|!)N<!DV5
zBv0~^qGT@#`)*oX3Als5O$T8~&N0}A!c$^wD-!cdbq{~_JXpJ)7P8qOUr~okPN;x8
z{SwwUYUz8nkE4cI&7}i-9$>1<txKxVBr$t&;;_OP*%g#5pjh+=rrO!6*5p-VS>S$F
z4yhX#g5FCU;DTF2M*xp8XK%d(Qfp<DCC?m^yqAS1{GK_k46nul1L864z%Gt;!~%>!
z(1iU<J=O21;4y7aEH1Btuc_HE?z^ZmgWs~n$YDNp<5CCX3!%M2Z$OJ6B;}Q<R_KRW
z{sMX}<mK_+@m4jEMCmmGoFpW;!tG8m#1;o}n3uwej=wEjtbL(jWw8L5hp}pWUq<oo
z5(d1{p7D1lvFQ^`R+?ZOno`%3qsodlCfMXvC$TPxE3H)qXi+sKNUem7`9!)uo7eTu
zr^-gg_l7<8OSC2q);KR~JZ+n`xwE3P(v5vV1+jXcRyPy|jGhP-VMJVnZ+;DEKph=+
z-Ru5Gd~p550ULNA2pu?(?<F62!p{(g{or(hKyqrMSPksh<V`PS@#|tRWnmvh_h<ZO
zz<?LB#=b8a5?D^iBfx!|=Mg8{8KUD9iu}sI-_q|PGSQ22&9xcp?P}iNwD8;nr`T0s
z#j{I3`hkMRy&qWau-ll_gcjaFgX~&MaiA-0S`P8jwjla?4+MQ4a#c7D6rKYYg#%WA
z>aFw+Bbwmay#WMW(4(#oh3in=g$z6#EniVw6F5f=iOL!&p1%v{ZT&p>QRm+F_950a
zd$u*(ZT;}9`?4f*3rCz3cNN*&{tSU|-=)+DaND{cCrcKJqtA4f%53gQn4uczrGW&o
zyeX_x1fdL)U?T9hh28H(=+AVQ?FZwRI#(1gMmHhgHE}pKzTYo^hXi<sB%nKb1C^!^
zNpGu0p|p^o;0dKbADPYM*n9qkmt0r$l`?10dYXroaNnk3foqw!5DtXrpXs(XZ9F%{
zDtsW2D1(mAWeVVZomLWZ_c}`jlCaww31-Pi0#@D^kUmNLaFChPKP{6JX1ajgdxz=K
z+_hI}AUqaxFJC7p(g2Tu>L>BWuO%f_D!yAz2q*;26S%*M2DwN}OHGQTiaySFWx5g`
zvkJ8=$S-?T_iRK{HYt{-gsYB%6k~?m-bCKSVr1)@rSi#6&-RzvV7kbf%gA_5DG$@x
zi4Z!^=cLLTCHMC*4|))OteI#<9qxGerRCgS8&bT(k4}7ywS>?1MCdJ-p|8$k)m#<8
z7zU}tU`IRD5S;9%^OX}%<k~^3{WH?uk>Gro7G6%KC@z_3ZjBK2RS;vLPZFIkSHZgJ
zN^>R+Y^Ii!c(}H$Ph)*SjgW%$Od3$QS9Nmr8y4A!Kz=w32O~ar_s$(j0^VY&rQqVt
z@a=E7Ogg!<w_^dXL7`*4(jrBhP?oP%F5mdHa93kwdR_e_gz*}}vxw;3)!8$>=R-h#
z?T{joe!&w9m2VkOoJqu*AJ(1D!fa&GZ4Pt6!=$*6R7MVa-4n^y$Z|8O7&CpM1e|~&
zlUm#zdo5{b<DG{9fysNc-e2AmEZmCj-d-2pIi0xycs7Umht4|&pUdtjlTSlfO6&T^
zJu<Owzb1m!#&baa8KU+jZNI~(@ET!JQPX7OXMCFwf;ZxRXDE{7uETfwlLQPi&|**M
zDG6{k0seG}d-Ww1f4(uzV-Utr1oan6A#!N~ddfvhpZN?G?uxO1y@LSoQ7a9XeLIry
zR&!A05%^P-9zz#W-<xaNSPg<0s(P2d4!Nj~etB4rDZ{(XSLWPOCR@$s51D=44IL}r
zd^Lsz(e?97(c7+3#Ix<<w*;ImL2L)@iXx;6BqK>#n9~fj$C4Jy${kmRh+8ScaZe9Q
z^Hov3xtSL6JqGHLJx9*__lTE=MZ0z{s-l|mUBjH@#8)MDs(IL4KyC;Z{l@p00CJV7
zg>&W11kYI26c$;;rsF>!c4WI^Hb2PIsJ052?8(2khFr<>{lN0>aRs;aU*$|fR(Lok
z4&co!h$YUh>nqg@$^Ly}hl@XEWo07JIszBB*2I-2(EZpjM9o@(J~y89Lup-!R0WK$
zCuvA4J5K47qXR{@Bwj|^HmRa8Xq+igyI)+hyf8i8+FnCL<8l`z%O9)OgZ?ocMDOO-
z2j?sYkScmGq>kV^)f9b^1}(KK&;uX`jAN=CQBW%(x(#^>+P1j2)}2h`Ui86!k??yz
zAO}pG!=uRHn0aLLdL<1ONjELkvUJ=q&OU@3HO16Tb!&w@Z#LT;)rdU7&Q2&}W=Nt6
z2?4V!GuDTMwe4Wv9eHMqI<8cWNv)<cyQ|N=ZyuQSMQM17UT}_Z9)SHMrPgDSp_}gE
zNIlCMhw6`54>aYL<iNQj`<9Q+oQBmor;hMx*(x<!F%g3#D>$9)Ioc2d)gwfLg7FbZ
zSLy7JO1v2xXZNG&OcifCYEO>1)%>%T<U0I8wAw4l!lHEjdf3+2Q=~FF%i(nrQ&lms
zFHg>}i@?JeUr`@Q+9R<j(7epM)-g`89}+Mlb_f6`Jw&j1xVv8m9{N$?C&gcX1RCab
zP)vK$1TgKVyeY0~h?S-Sq9HoLZ|%+ll>RPi_+2dRW?YL(O6FY%MDuaF*5+Hwq+HQz
z_awajNRXJw$%XTMekK-tdkqi1Hp=CbT1zz|RBwlJ)u<WR7CC*9Sqri`Z{gaDw!GYD
z-E#ENIZ@FlvVZTQ?*h_N?v7B3tUzDuD3x{f`~uQ-`!*6q<g}DMY|Om`@Cx3P!KuZo
zkT>{LyWa_z_~I&H<bgC5pT`$hx2tO|tY%LKXE{&@*UtSwn98*9ZP)@v6sKA#T}YiT
z!;?{}0+1fu%|na<U4k0*7AoFbsZfR0M$=YBL_~0p!HC0_;8@}~9`XE|71#mkG#&7%
zD`+bTFYUZ_og;^tPqw^%t9`kqrkNYXW8N+;!MF6w$H%RiobVp++!YiQc+ka>xGgJ*
zrprR9k&~<2`K_NHph0Jxh{NkI2+zO6iWCJh%tupC_aPgT{EAO0%m&%#EA0%n`HwYt
zjPYJRQR%%ATQ4p*{VgPUsbyRLv)6*$<ARSXimsJNlmz>J1QY<4$kwGAXfr;**ucQe
z(NfCF&YM(4){An#1yVp(`|mvjL@V8$h;F~VhdoDnf%DCMdC?{&Ys4|1GI;61w9AdW
zSJlSIiDpar^~tVYGEoU&myO=ZeVa;^Fm}d(8r7-IX*$C|sYTvE-vils2OTtiqiCnE
zdN=floMWG1tz9-@&a++JM&^)*33FD0BfSAEy}cylx`4SILKfxjF3du&{0L#xIJNo`
zG_n-8YpP4(eNgZ6zOQcubUEIEdk74v)ugk9iMie8p9vBMHRMU!mj<u;-~Vjfq4em-
zs}%A~KS(d)0rG(__K6=p0>K*HotrXOV2<3t|Gh|=qH%!ndDj#d7xxi6!4^qs>1)%;
zb;#*mz<9fjdZW4X2`G)qS?rCfZ%6BnM{WbVj*aH;B85eOS(q2t_ZIAsD1iIjT2BR=
z$N`V<Qr-HA2StchT|yaMXAjG7pOud$yOSTjJ!sodx|d*IiM=X6R?${cg(5yasp?SW
zG#t4r{ls`ec6uWl$J5ui?9x(kr4My*3h*BkU0#7o_Mv0CQ4~}K>?JA(tswGle~pTc
zt6ilXv->D}mTOU-^Z+iBIp<mYQ?-2Bwk6LL`EyPGCX2oHy}m?X&b~GuB9vVz^AfvP
z=G_I$eOtO0Gt2wCb%fM4siNybE#bxZhBhCXlb;_5?+gT#)80IQh1tQ0S_*ygC|?Qo
zyFfqHYFp0(=%R@S5D^z$*?zSV8)uN{*)>!>jTQbFNQ(0H&QOH^dlTWOgp2{~Csk{r
z6q`aTNT8a7I}&z3mD-==x7K|AYE!ayN%p3|h)7qG;&RjE$_ylE^6KMXu>e1~kVHGO
zmW*4R$$=3e2LW${&!#*?if<(4z~EJv6TUdb@efZk+v_IEEPI?RT&%4AZ8lCsKQ27{
zE=l@zlz<zXRWLm2KU1YrkDBC8Q>penS353g%vq@j4&z;Pa+Lewv*Qe6ev7%KdTxXy
z5|$y+<et5K7RX}0g5F33&^#U3w;ohPO5T<=F(l4(_n;5E$OmBpt@@)ThK8^~e6<LC
z)k|rp^y8ncfMS|S92?2`*<mk3N+is4!n0UsoMX$x-^8m(GV0v(;%)XZE6=$_XpKL)
zMVsH^Q0E*oMy{BlVwa}fWcCL6`j-K0_7+GzBRn0Nw1V|*$Lq-!`G(Y4Y2?A&-QCSZ
zn&USzz>44vt)GZSyDwyAt~ercdUj+YW&l;tH)F2=WP7(f^+Me_?e?ceqZ)wr21ZbD
zAql12&JT+=p4e$t?ZJZZ*Uic^`09>_tzE#QR~B?8xN>!xH(Iv}Zqmc{tocB>!bd#E
z2NKeinpl^pCCxb^IPU^SwSk%Q12Mez2v7CAwxRa+(>Z{&Jc~7*!sa-GVLSOkUF*U5
zJq*y#@0OY}eL@a|^2CSDwE;7Mq5-7?;1KmWeP%B}4UKm~2(t_pBu8Od^Yim<X(fUt
z@b~7|Llt%!G18sQ7LI*+OwtRp@6_pNAgTEP(Tb63Wf0}+=YJb$G`RMP|9DOF)hJgz
zdK>wB%2A0>nnu`>9#Sm!;m+bzM$N4B>7YvXZkU&$Nx`F^4ZtQy1YzEN>rbmenYx>>
zMB#ZOLYXY|r-1UM`APtt=>T?>DZQkpDl!i0@1^WNTDCMgNCnY=$IbG<Mp{ip*#bwg
z2X_d0vYF?umCh_*YTYA0=vi}x0^}KZpJav4rk90H{KBynDt@ne4|-!O>s<<8DcHD|
zgtC!E9`W3PCP$h<+(Z-ON~!oV{x>j*Rr+*06O1K!LD5t&>+@bUx9rEKo{O0W7l}eH
zgEFZAlyyXJ#dqJobLWmi`xH(64Zvm9=;T1#AQ}P6#1km`1ez)gcAYS&ImT<75-A+~
zZR~`EFfjfjwG*)LT+AUdj@7)Q4+%sgY1(`A0o;u+a@WElK+m#K%QpM_o)=*y&fWM|
z;-!QNuMGZp;^aPLEpp%%1OmkXP>h^(#ek{RTx6blVAzCd0mHDy$_NCCh5#e`d+6CO
zJD0pPqT%#*R*~B5bGR`&?BbjfTcn-T9b4R@vqA+6ywqxLWd{%wh}v1=@oE8kvzsM^
z1G_@jjc)`6@EqD2-*YC*cmitve5LEpR(0}%zUeya=M)};kv-9apGTYaijPRC$A!FE
zRSjkvatoV?_wAb#)QL0_U*60LIzk}RFR~QE7S>iB4xi&f-Of=!y)us!)+5h^7b$Lp
z!+e~`u9eiVubXEWaj!(wqwgj4&84CrZjS)5kuGTX1L#;SVXG}~cv*<E65M;kIgUTF
z>8{5c0%zaHtpi<)MY)}$SN$JA>Is^lQ2+`-_+B@El`C7*SxhPx_=B89skXQ;3c?eC
z>jhF+M|leLsa_=Fp17})kqLMt1TuH>qy+?m1_3P+VQU$p#Wwzg?|(jMXZUk@nNf7l
z?9wdZ*1XHd&$`KFq}dB+|9mb2a3e@+;%YV7gSd3FIpdPIIihRj-gFOeep4<;Rdgw&
zg{aA%xiL7sI6R-Q?z>)zv{~_L<`wbI|Ml4g_rNv$QaSKH|G|_1hzC)lC)+n`tZm*-
zL;iU~6uq|epI?AeKbkTce;e?wiU0d6i^!jWq5O3l_pO;TD7aX8b*%T<N5~t?FohI-
z=NG4_S;}Mt0%S3mb8-x~i}9c9<@+|x-mob!-TwTnE|s(w(1(1zsPymi{$Kc^Hc_Td
z2a%8lHFqtiu@z=$rMARM_xVm+`lJ)WtiWCugj+5=G)hFS9Ear~3Fej{t}vG5q*g#5
zT3-R+Y3vF%N3w`nU|DS5l28V+K5I*J`fH-k$q7o#;kJ{y#pbUNhhfzu0IrzwyuF#i
z)^;J)$Ubk~^ARGlr>`$1HdGVxk}tm<n9nu9Yt<%?D!`qE3sjQ=)7vpY{LFVJce<9X
zV<}_=kRjX?5C<Pm@KgW1$U@%=GMRZBb@H}4CFA^|ad3rx{Go1V2<WmnU|a05NDE8I
zZEh&qv%z^Ki@K7KMw&i=OrZ>Fbs%-9kkmi<dpt(<cnM`zb!VxCefDXnQ^$bzf-1(j
z!2{#mvAE+v2ibK2Mvddc7^<H?<;igr$i!XwRA_xLaChl#D3RQV!)#cjD5a_zC<OVH
zMM_BGJ;hycx04LK_v8vMr2ycH$(NRvF1}*?g>SpzyzA1kw?zqZMG81>W8K{zGSh?@
zmX5()>Yg311HPv<p#QdE24>KwU`0aeX}5(r-}bUeiQZ-`w<J>4ssW)(1!;s63GzGD
zM`Ovv));pekP0mh_xJ^YmW3I%9>aT!Kl&T4PgGQt7|EktezQ{y_>>p#Eci(tSCa)p
z(*+zM0=mEzZJJaWNI6B(7V+-R!YsR0nASGb%~WIey(F90#5NyzDL6MZf#=f|zoZ1$
z)>A#$IDs1R=u><Fr)LRW>GOt2L>YgeQmc6mj#mH=Xs-E>xyBaX1|H-|^;znd=!S-d
zQ*eE_0pDqM!*RJ<s~PBl2^#-dhVyNno)OpEo256>I~Lc`jO>s2K|)ZFO#M9CBJOTy
zN5?AZ1>qO~Pj7~>sns}*RUg=R9Gb3$0u{EmIYuZyW}_NMDW0-**-635?w`<lD)=n{
zo@(Q!%e7V>gZ7!+!uEl1dfE^pHn~$0(*mQ9-QbfGr9VzqQbB~-!SdV%GpPYzQzH4m
zvvYGd1r9D1>@mFfb)4~x3xJ21zk-}{w>yQ2!y(%7V*csvX+d?LfG6nNk68u)|Bx$=
zpQK@AMr^4)LUw~FYMHO84<Qit24m3I6oD0$04y#SAShmyi|eai`;h>pI1AegR!FS6
z|0!H8ddFIrrChRL8BSkZTG}E{uEvFdgFgRUn^5X4LRmd?V&pmLe<Z>IWnk9o%ewwA
zpjDkfdp$j_d1Xg|q=9z|@b^F9-eEjuHD^UD9;AXK_`|_)5B?b1fWj&HU4h8G#8rVj
zy^&h-kf7vPs0uYa<@^<qVvoh%6E9vLV|mf>`p7L^h)-WnZsPe?o-<7JiVbV3W-M;^
z385o}s@;}?L<@mzg`*(R#-Sm_Wa2GfmVxbIa5+gxqV}%OC=K-u;4Gmmg=u08LukGA
z15;BC#8xOk?;;KL<Y%IQN3+V{OH}RnW;Ilv*ubXnkA#`h?p(TnIm{~wnR`5fhS5RU
z3`ld(#5sP5f1%4rc{92J?&CY(nGr~;uEF7;=zgVApc!4xp9tcAj0hm>-vHbe3UT>9
z@uU-VGI8W~kX>ym5$kIKzW%pS2P(#JHNM-UCFTYOh1F3eL3MedU<+lJ`15TDWej*7
z^cMXKb>4&mkce({lNv3Hz`y4}FRWph*d6@FY8uy-;0c9Rb<fZ*7}&m&JsBmiK!a2|
z^UpEp{1*_l^p=unF&}J4X@vF4HfSDQ>8L+>1Z20N1K9N&;EpO}rGMcmSyU9G($l42
zpxDc#JM46_7BWz5&%DS*Jp#==%0r{kuB-}765W7M5QH&L@bY^5?SwD{a=f0=0%C$B
zB%$Buu0E3HJa}(l)()0l0YJA%S|G_5kjZzKO}?9wgH+a(TzwuDff^4^SM1oWo=CxM
zwI4VmjwscK)BxW^BkK)V_bn9Y+q+leSYiA(w)%VMQ?V>ev>~eE?t%a>hk*r9h(NGu
zqme5@S};=(U*nXQm1SasSO;b^0qy>{T8diZXE+D!=39U%zr#^jNaUJF?^xQ?!IWPG
zJBR`tMBEJC7T#z9Q@#jd4W?2kWf}6pTB3N$m4}9gY_)eWxEd|6TiNh&ouUJtQYjTs
zRkj#RwJ4Mu+kce|m@Yn<Q3GJ<teB<IK!nA>UYo$qNP8pz2QXbafP~+<Z&EmO3PhfF
z-C(yMpMEPct^vq$yuhMR(y=ftg_aTU_bKUK?}K!&w?@BH{cHsEuH)!D!-Sb8z^+@I
zo0D?}0ciO|aZZWtjO`?r_Rti4ajgX2w>NoPjAT;TGs*~iQ9#b3qcx$R3OM8?CxGRf
zwBYdO{U2+LsIA`NINVf(;z|ctK2@SwN-Xr;%-{rT^43iK7U%A}pN(9(2CK-bu0Et%
z<v-DW9)$7qftedJH8mwBn7qZsK;Oo1h{M6k2?T4>kp(PsS@W8h!JeIc??HL;E8Jrs
z(`~Par=xNy`|1BbTM*R@^qy9^hCvao6}<rUw=m$lQ>1QS`r)HcBXp}2J}9(W21kp<
zZD4WB=)=+ar!mk2OL(N#;3a#*Ni(80=aBY_xL^{jU3*#Mb$Mc8Q{X3@rKZo7=w)$q
z^%eWO;>5pRI!I8ZaNJ$-W&=N^3t2jy3qPiQiZsZ}%)j>+NE)d(&WHB+qVqJ77KXM(
z9u^5p>oE?#j=xnBKJ8h*+-cK}aV)~s`^x`yc>_f^HpV@&Q`SVc(;v-qfA-|4wqf}C
z1A_w+OwP6MmfTNyWY2ij<q#cbwBUxD-oFFigb^7md$@DeLo93@8mFnC-uYm8xF3(J
zb6ha%9M?vBYKO-!#f&(JN8e2&m{eVZ(`lD0+6U&~I+U?6qmvdf-DycRyceNBQ*k$F
zRj$G^|NCIKkB;~A;0l*+ycL=ccsw?q6gyj)6SJT$;(*1C3G^h5Y3Pcbue|l~lW6qD
zU4xP8DlFT-chU|E`?upNOX3GLV$DOyzoe~~(Rayev>a2f2Qz*8Q*vlsv(FcJ$ctrd
zVJ#eh;u;iN3A8z&pLPr?+VAOCbptM`DXVp+12syO<m$^|n}f4^v?n}6+LgV>{H=u<
zKGZlXMe)_Xl=-8%&gHD(miv6OuEN_8l3R)PTlb*OHL-QZ`hw1pcz%t*{i67TYFs??
z5Z#O%MgM&%@$YRbvMP&jAKy9O=C<U%PsJ&e=t$J5?sclB%!pbf@Jj*(JU=1kE4d~0
z@D&%9T+6r+sW&(}>TKy2Q&Q?U<u4^?6hB(cn~V!#9?G|o1pZaX7v_H-UW)$Owve&K
z=P~-*i4Q71_QXZJ<S|-L*GV;<scm*7wpdxERjxuiA?i-Kp5@ZxqdV{+{g0pjlPfMC
zouncw-Oh(14oYycKu<<_@L&P_DzmRV=>T35XTOlAmtBz>dJrnGd{oL}Ee9|CbIk*{
zfEahXH|43g#+ZdI{h0q`%MDSkVt4mzHd4^FPsd!p^8)(51F?aqCheuaATwJku>7u{
z*j;x0`xW@!5cKW*Lu@K8;<~O`CtJ7tVyWX;8hT~kGf(GyWiFb~O*a-9gu7URH1vBc
zZYc!5GNnz2#}q$Ml$G(_{bt!|d3Gq8#rkm<;KZ5!5#DkEX0hh+ZBW1?);omE#t}Yn
z!G4Uf`hN<s6kIg6JbLFdE*j0Ft5hxdQ*J=Uiv}GKH(P10+dbM#C)PpGR&hTvP9qFd
zn(x5uf3Jwnf0wof%;YH!zqfz;0Et5C8N}fP&@$SCd4!CUvpGfb3ujYca8*son_o|k
zit$FfV{k&<1MMT(1e2d*@1(`*C6(ZrIYpQX0m+6Dv|bmTtRUfI-B8?-9lS4V{iQqi
zf|)Pb7k$(A{eOlN{mO0s_I;&{S`(1x;G_e1b81`Vs5bZ+6mSB#o90(7>0>FZ0Cu;(
z0#uU?HNQMB0k8Okl&2N4H-g8ZlcU|$KX_*0P;0z@`4@m`6VnxYc+<97R%s|xjblAn
zfD=3-3I&t_Ac`tZQSPxgSqAQ7;GdNCn@YcPm9l?*!RewW(Y@MlG!sO#imm3?LcziQ
e|7A-RI^xamKff>Tl7pi31AasAdY-z~v;PNVrgIzs

diff --git a/reference/plot-17.png b/reference/plot-17.png
deleted file mode 100644
index 113f104f85c1a1f086a770e492ad0c025452e8db..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 93183
zcmc$GcT`i+*JS_^6#)w(N>d13ng}RerG+A0q*sGTlirJ{AVrW+0|EjOq$5ZTO_1J{
zDue_S=@39VF=TGk-#7EES!@29HFMXJgvWdD-gn=*=j^l3zDDZYQ>8m~`4j{Kp@XR@
z=|LdWz7PnN&57gS2qrmu1AH8}(NI+apP(JwMTdZolTXx)Js=Ra2=YIQxVg=T5Xcn>
zOzF0P@0(R@K;WB)86?u)LM9R_hjLCs-FQ>Ys?<VdPERRhMGSlNHo^Tey+Bu~i@CY+
z*yw24kHh}MVe^Y-<LzyaZ;u<^zG7&_qf?S|cJfZVjboB6HA+!~2c|Kc%nHX3uE+66
zBz}ZVYT`BLcLHVSzY;PD^>{yq?%5v$_N8?$C;trd|K*QFrf6wMXzSwQVrpjZoNzqq
z!L{lS_4WRKBS@XeNepHhIk>=(f>Pq#oGmOYEYY>Uv-H^6IYVgacN3P^zCW#svqx6Z
z)D)8=dfE|MP+h%C10Or@#gkZ?Vi|m(CLRz0MNYUnIy!Q^l0FZCd_((=P(T<}gj3QV
z!lg9yn8W)Kd>fp4*31^Ra2Tv-&Rr0yJENQ!yTrl8jfF;1ig3>1vm`1LxC?S~b7AMG
zEtER??p+d93AO;=A&`y_SC;=XTcO0ePcQ!aC5n;!*FTL<kI|Rn@7K3#x--;&zlL%u
z|GzjnQJD>LUSBfZ{K`^U?ttzO0dDgzHN-HKrE)6uYm|?g^hZ~M3k<{A7j|v|PlKGO
z3qosa`1AG1hN83V;iEz58rbry>kR7Y=s~ecae{XUuY-A_pLzyb;g;iAv4~0r6+|?+
zJ(oT-vbzsnOZcF>!1oYuVO>oSe^Y?i_q*~~pCHw@s^+sEb^Ww@I?|B-TgKH6HKo2j
zf#a*}T>Kn{Of7wI`fnv?ZS7EftE2D5!R%b2EGa2*?Sfb0_PUfZXuy{xisl5@7-<G|
zQ%A=YW2#_|l)uNDmq-~-uC7^3UhIk0V`F2z$7AA2=e*b<kaE}JPqcq$8loIUDe~_t
zBmb#}e_wS_d>fR1zbnS)wEX`&`G+n|%jc1?G4GeC*&OhS=-qr^2;}I;s;Z3hZtRJ*
zz^?l>5kK);larJ5(KpYjry?7<Pq!Um?=o|pmGH?LYi2w4DXhA>IzxB<c6WdV$qR+z
zy2O<`hX=Fe$Pp<`1F`yD@t(@crUzab!sP%PxkY~Lw_R9EQ&Yn3`@6u3Mc^AbSAJ~z
z>dR+^Tv=v)*VlK%dAcoYcda^gB7%$VIbs1nowyy_w6+y?#aHr&8lN*Sn!o_<?n{FP
z3IC8)Oh7l+xe^$~G3*YP6L0vGrEDiPt+^h*y;1d&E9v8dYrBs}E<qr(EYuc^N`{7`
z_hF{M7<)`iO}B`a5h))V^YZ2+`L-YvPToErTKh5!3JUzL_sH@&14F1he(h34Rdsc;
z*tvM{0Omx@?K855=H`t;|4!iDj*Ewnb$CI^9Y`5d{e7qZeSK-%ARnFW)^^Dp3+}!Y
z!nNfClUAxIL6r3gnq4ni`b4MO2`+5eZ#_+3wpI<kzkG&%ZJQuNjE7{n*+dRxVI)TV
z4PU~$Erst0IyWY8QsS9Ew$$RgF{NMO8fbwi+#BNL8K+}Q86ZW7X@Yeg>K6OQ^3I4Q
z^tv2`UrWK*i>E-xQ>69mS>+|W!97HAAC8NJ5`VlpHOZ0XqxZVzJhr&iEMygV<G`pg
zq&!e{WC8-=lh!Uye=Kt3&;<{26i=-C{TP-d@`wjq&PaW#;q2o_k8XBh2a^p%9GXzl
zuoCzfrvpqn!=kRHiT0v!LtkI8=x2>LN!@nrlSa>Ig=?+u38xUskwS?#REM_U)i+Xl
zv4QK2d-O!3<LY{<%Q528M{ixzehaxEe!3_8D8nzV*4p7N7UCL~d&l03F24cY(qc9~
zC!Y1b)p+GL7=@1KgPg0J-F6inSSgKq<skx;?P|fDq3`(MBy^b?Pr?#4Ez@a}=Teo5
zdN!5}KGWlLH;c2gzYiY-9vO=hP^ejMtz^$)bciHVv|Nh%yB?LLH5*C?jSRQ!-Vp8s
z(_LA3p1KraZ?|(kB#Wg5Oj_X4a_qMgXagqb3DA8My|^#exwuZbtU-~a?PsE=$5{h6
zK9y)BU+d;h=!U_)o6yso{Km${2e)8Z7>wT(v8}(qfBN#$3U5O9U@hSybVsdzvM+6T
zX?yqZ@UQ|j+bLoiG1w!ud{RVuN5^mfE^^IdRdhU{1~p$|FkDDtFLDB}&R5_9#zw^A
zyNeObzsTF%M~?eo2w#eKWqV^>DgNUnv#Jx^d9X|RTL+~w@c)YO*)1&bGV2YM=6a6@
z7sQc$_h4GxcJ63Y0JnLK)u*t*nvWk(`**&&azf>|#J+Mg6B|?b_lr<|rf9a1adRKy
z2hz8ShrRXOB`F_$Q47i^Uv?@8kN^CB!)rnzAh)Ssj}D@)Yrot#N#D4AzAocKXZA3+
zOk+dNkM&7?d!C|ncHe7=P-dUj^SElE{q8;J3j_leXTHZoj*5_|w&xZ(+1c4wlQel0
zK0NxG)0Q)EUC+zQt9cP~I!`P8zFgV@ecQ$`a6{t^*YK2g-7}=Z#=M;4vToH_8AGhz
z{2EE`@n_8AUNfFzTU7Jgk4`DOx%2iz&kMIwxOd#3Ou&P7bC|ZFQKP@<)AL|9-l}OX
zQkJ0vU{cSm_XDpZ0~>%vl!-KKwCV~aGxc=5Vxc*5Qdtvx^Xk>Br>JHGO7wKIxXFTY
zH}lsB?&QIX*a$($leg8rqcq$Zl6H6X%^RdPNf+eLM!m57MX@9jn+%=6sW+CEFzvTj
z+i9&ofd5(D64@z__(ViV_l%6j*QbcCeXB(LQHGiDUPgF$I8CeQ@4?K>%(>nak!i81
zEnqwh>WA~eN(Y8K34Y^h($I5}et2P+HfnfjAn1nq8%-Y4{UJAmMY3lzI;5tq?qK}#
zPljiz96=xNqh03nGGu(@E1!ICtwB!MLo>J>ijdd^Jc+yx*<colG525z;n>NKWz*)6
z)YNOgTgunP9gcAJWfmKIs)1$iipsNf7<&X6gIA+Z$-B*IWv(0uHXfM}7e1th_;#!n
z?H?!mAa6k=dMaP9paPG_AK;Iy=^hZq`jWw@S0^fst{GR@Wr<(CeEBjHnozShI~M{v
zKN>tyac_Bf`7jqH?UOY=HT8QY(;3nI^8&&J+v#8;EhDpHYGOg4vAuh@%@iAyfoz<?
zlQ@_1K1rhqJ7C>pU3Qt51MWEUA>3%qHH7u1c=|lPplNLZOI)DwADB4|@z2i;INY#q
zsGs_II7Irb123KaeR#$>*dguQLqC}xbZfz_OC{`am6EPYe)*`jcjmTb=;m`E(p66&
z2G%jbdP7lPZx+nCKP3h&KkkkQdNntG12@;%@67RPTy@iZPZ@rn3kKf}^0Kovsm!@&
zI+?eR;+NZ<Pmm(Kk#dTb=2h>AN8rBJ5KUp4M=xI^<qY-pJ%Y$VZ`P=r(U)df@NK5h
z(q-7ft{m18KHa^rC6(pg!{6-e<P?M2PgH+rgT-R^Y6)m4HnrQXl&u)s`hs>)yz*K9
zD(=^>Un%KnX{}^8>dzcbizV{FF2KF*r8AlySvA{T_e(?3aL;kK_pJ(on^^vy7!0nS
zoAMaUFw@oD#8D`(`aHajoFi<>S4^x<tC>_N1c^ycnS)ais4FK7?%(%rS@@_RD?sR2
zh#21=sp+-n;4vM8?szuuK5))x?Pb#FB+XISQ-%`HhG+Zx`$J(L9%x1?*?8WY!sO0O
zYN#g9owr29%6fZRa3~j{(WXJuI6XhVx<2(^*V$&3<g<)i*Gua}Hhs-D_n#mAuotUs
z27q}@7vv!^mfRt84=gPegyPlSEZ{j3wk2WLE3@7WFZslIC3kdmSoNn#C?GrwkV0~s
zpU;b6vhH|XN6OwGnnf^%f7jgu4=GOU98e(rP5=-A<+_L0<nj2FNxUd*wyzEMZba=M
zHRnqRTNZp9DrNcks@n1gC}=@Ng`UTJ->r>=`rf0xW;=$O)ys0b-dw4d41YD>RH=5^
z!3{oq`kOf3Gu=@o5c;9Uwv%3NtFM$KH}K|`egvy~*ZT~a{gF|<1;NXBY(=X`kan7+
z?T7_OyF*(zo&V0L5jU)pBc#FR!-LO}Iz1MCa~I@Dzqb%5<4X9_+h?^LcAv>0aJS&6
zLR1hqJK$X8=ligUqlH#c)-A#1rXP3)sQEH}DJm+ech85(4fp|0q$W;qso}vjN$nv#
zWf{7L?%UC_6|uLcC-@r%6VQmN85ti>5izYU4k-5~{J>v=*~iMtsb~O?(~UUNNBlbY
z&E(_CoOGVIOmh**d-QcuQiOHe*Z9Dru=wy@i79cMAR8A^u^v)2?(dIF$qE_J$VJf2
zrXho1Lm=bXt0!y$nCGyvyj=faNI^|cI3)mwot80K*!}e@q+-B2wl0BJVLoYC0m@+^
z#(dt0Ib6YLI_O+|eZ6p+te?C0OJT)Nt5bDbCa@GWJ#FnlX&=&~&++PM=t@~yh*g$)
zMGM+@^T%fmk;WUALlJ$eznaS0Uch&W!2e1(451=uR*c@K7t`_Kn^cDyF*UBuwRLs3
znmNiU(~)w^_=OkN52Q8@!0eR2mV2aIlACL*@$lh0XhB0mAOdyvqAxjNGaj0c-)&no
zZVo24q-XE+rAas%jHRq@747!UXXSk%saKrm7L+;=&;DJyj2tArx3%5da5*~X-|9-7
z4>P{lw7`{{N;hjDnZCwcj)spF8*8gD9p`pko2){QzjWKn$i1cFQuq7g#XIe|f}p9y
zp($*?ri{-~8(-WYpI*`ZlF6X88n2bHh=o4nmQA9c&g58$sSk`)jc#teNewXI{p-{?
z1Az-*ilvo@Xe6B{w~gRq3L1O)@LkKs#>VE$kMjzV)c7Wa=<T(tLH2g+Tt-Glw26s{
zJ7M#3*>^newsVv0#(^eJr8rVko`31prf*<1Dt~uX#(DDN649}R`&?uOO&axi{hc*I
z4#f7C%t0Tsz<JXP&)belE4x=zC72;f#mm(ZnT}{X+=c6o6^&+8wMgUyDZ{TbimHi%
z4E;ifB(rH7Ky>_+8z=7{hPK3xj#fPgvJ4Qfl)xWx)S2NkAND5G5mtnkJpCh8+{`D)
z*34~wb*Ty1!V9n%L!;3PSs_R92t*K?fqv-cahTLz(CG5f?O03UwZruh&^+Gy>DbtO
z&u%mGu9(iIR;Pqe&+Y8Yu?<%f-_4cqd<=B-G8}Yu?oP9S6&lws<5%0a7L4DL5<NTE
z&9hFw(;mZp;FQ}fAZNdtH_q3$5EgUeg1=y!u{DHZlt}tS9Kc+?gk>qFT|RvE$mr;(
zh;aqV=Vs8#%8KLc=6-(PDgobgHKBVo5#=!`=c8_HJW&VVZ4S~3nFN7rJCi6QBjZ|7
zT-<P5PXMVI{HgCqF5{lh7+8Rvvzr=(@p-SkxVQndGRzGlImJS|W343yQJ#DE6PD21
z)2lFUjgimMNpca^e;X@w!E{o8t65s#wAoH8WU+v_*`o^0sFUy-JKNaP%q^;&O3*jd
zVmQL7+|;3LQ5sG&5tW&*+w1X6L#ZQOJ*oE$2&*v?5^P`Gy+CG=cG7&tYww;o!Aw4@
zc`h>m?+0IGw*|H@(fP>8%?L3t`O#rG*6#YPN0=)o=5axL*tnf_-gA*HSOuJAC|0AH
zhTB~B+tcLa<kA<k7c(Ej<J9U0uw^ViS+n}>t~axPO{awJ7-$|jH#n-+boX+zR5bP)
zy}Er3!hFG0_<?BKdAYs1Ri$x9)mtfrnujU${Wl5Vp?7HP$G}Wp(%MrfGUvEK1%Y~@
z1|F;~6APVMOy-1<tqk6a@2W})hV?7c*r(1|1I{X{=eh6({5q0DIm0?Ntf}*Dt7xEB
z(zT3x&4Hh&FS2F~E{NQy3~$=;&R6qS87k{vrCIVkq*$}{*75Mz8m6P8i%>s&kG8zJ
z<hxeSKUs0*#F2;9?Muv&6ABNSa*0C+(+QY}@L_pT!ulN$)xMtN@oz@2%p=N3CmdUj
zjOQ*~kWxRK`NmD9<0P9DRTYtb?|xq#-A}(tD`E`B=(kJG;D)P4Kk0pMbM-=m9aY~e
zzw!#=ugZI+O>0N5Y%kM8LOCVCDw#C~j{8m<5?6cC2kSM?!5vh${=sr+$ul1nvJwEv
z<@t&UFn5)8D@Ab~m%)YHcFkKi!hUd{NDZe7C60JcH0U!fw`Nc7ymAZ4Xk3k1adfrZ
ztWQT$Dh52pakKd^9jCpaDqN?DFO41fJzdjXR^ni?)B9eWsda3`SoLQTsYgBvjZTFd
z%agod(H%S>6K;1dt5l(|3eiGuec!#ZzAoy}Z`G-VKsOoAnUdf+gL)LAZpU2iGWop2
zD$~+Vv7AEED7rmHdmh<C9T7L|U$?nRP7Gp%Ge}2bu@XG)nS|G<&rjfWwIk)gV&5i!
z+_S|IP)#~#3>A!b@&(ue3pG=^S0!Q1@gG}JWf2t#j>)CJ%O!}}yda+wHDh0Xf2W#8
zbhFCbeSTv8FsE*Phsq^|(gQo<Hqjsz_pthd71Ntrh}~NZsw3=>P`&nZbdXbd>-1ed
z+j5Y%?~#5raaLa3;_JdztPtg{YrikT+$$UDd4g!p;XipDZfPNTPh{OfjKK>949-Cy
zJm=tK=H>LWbYk`EH4059jBuc773!c3KhyMug?DV)k;nP6F%5@t00g2w^1R*(xDUQW
z(JyQB%VXtZ^m|4&a*=3S_EU;qyE<nhUX}Nb4DeozdDOAud$B%PK5YH~XW`0k0N^pt
zxjz|tn@#|xC&lj29#&n$qLj9$;L1@}U-WQ=xr0h{U@pvRL5<LWSi$MOyr#?FyC)p~
zJL!AX?s{L7hoPf2$W=tebjYL~b6#OylBMK$bNzA4m~pM^O9`V|g3aP%6Ob5J36ajM
z_z|&~GRQW8XsyN2WKynpHr0EfZH=eVNH6$hUbDXOQC0^hNVU#Wkcq)i>+9Y2!sdP%
zg0hD28rbBu0;8d5_+jvwZ<Yg2CVqE6i&4bf9UDR<Y-fDO6@Kk^6|l>E-{)lFrM>7K
z3u8GhgC8o%E3GVaHEwh+uB3qI6o-chKC(InC?Xul8kNa3oVmHVhosZDCrGFk^7_Mt
zv#P9{Rqw^KKO!gE5wUnoV>ebNr|;QNwKJwVAv5IYzys>g5p!uRg57_yAF!+rm~!hj
z1#d4eZs^i0*T7k+^N~vSBS_K1)uh;8Gt=_B!Co#!llE<C6C&YR2ZJ1be(gHq=CnDD
zjl?%v>5>^x81UyTAc(rKt&P8&3!t*xjJ|c$H%nuxZFFbYQF@EFKcEA*0WW3PwSo@<
z@-A3rnH3VcDah^x(bU8Te?&Hl04C}hQ*_dvE6r`4PlHst9(#`~u~~wWsm8)&^5^y$
z&9!OC%B7h2mubkgCgHetn~Q12%n#WayJ8-{$f|9U*&l1(eLTDs`6=z2Zl}+}#Sj6V
zc9dXF+N<+%o%!V3KGC__(IslZBXm>S<OvS1+)uUi?Q3vGWJ@1ALJ1JG<wlIYx@W4Y
zs%q3P?`R*21+I*jh2*2=5Q&2emEdrM3hPO3;q?AIZf=72v;`p4?EWB%QuBaUC>C!1
zlwRleB_(pgfHHF<Ll0qxD5x{8;UKZjW1)%J4a)dhM4tca_qp|iiFb%tFMf`$ar9Dz
z;RPwIgDQ^E;b{oA-pew{vjn=cy)A1%=Ia1W1_=$y{YzNdb|n`tC2cDNF@S?qty*JX
zsgT?M6{6Wdb{WDwFaRivXFoRgF;F4Ty=bUb!bjPJ?rT#<ogig?Hbr!T?I7kg3KK>i
z!-WvZAT>@REKh&?#w`1*i)$f$f(Du);1G^lfC`CMy5<cp(a{{v{~aQPk%=020wmd6
z+z+RS5`*az@*%>1NhAjM#V0<^>O+RzcGrg|Cj4|YG&EcWRdxXM0Hm|AHJo?dZSeuD
zv!2iiGS_L4Qn|*ivXM~+Xv`YdzMjbzJsrR;ft*M}#IB})l0o^(;^EaJeN1(PY(Q!}
zL~8Q9zR}(%oHQc%hZw4B+;&a$tUx&<1_tKbFMvOH&AC_ht<nbdH@O`hEO7XRx}N=e
zdqoPdXHs{_>CPaOSR4}aHAh+Xh4oFl=KmA3!*D%%j|*W3Hv%(YxvNm;G*MBE3c1IV
zU=NcPJQ4N+;lg?#^N+X`WlH@dLjl$fF~?OrRnpmHhdn|IJ$7Ko9PW60#TCBP6k^YG
z4UnDW7E0A|8h7&1Y{#qHvP}e4v`1W<@De+0K|}iXP;|hq&z?W*R9pGGC;Svw{DqAP
z+*_9c<$*ZT(`!ZL-`&e<cpSV%c2}3ZRQ}v<_XXwpu|zfK{0_EuEz(mOTesjQPN+qX
zr%&vGL@rf!f94qnU=?7l0!eAO(b1h6#6a?P7YdfZ&b+b;5xp<1$E&rj(;lZ;$g$s`
z5s^9(2CHzy=-W#gNXczHFu3&R9WkGx#+~FzQr7}BM7=U%idaVDm~byYt6?!Gntp9<
zjaY#!0RSlh(TxI773xrgas+PVD4-j`eL(Dw?}wMY-gYHmIA}`q^Yb~=`_&O)SKfES
zD+iU+PrMtz+9QogyCRhnPOb*V*G^vu-l?RS2=5fwom7!lPfBX;mEZZyV`#X(=lgQA
zpt91#L{&8c${(-B+RhyL-uYX(wL58-%%0d+PROliLt>CZrq!z!g(bzss|FoiT_vR@
zC5_jX&j3#q4JHfmmoD&8FHG>X1(@eqfFovjcuPe*U8XDh65<}N)@}AQY=Y{gxZ#i`
zt71qDr+9Q^77Zuxx{_c?t06#5i=Mqau`hKiBBaEsO_XQeyXjozyeY$uU>ScPUrfOj
z-+t#4=WJp=@4KswQ0Pn^kBx`l7w+hKPvi%cr_jw;U=&}$f-5$zoWBoSzh3!{H`Dh=
z=XGS?TW6NP)DJhY8@n*Kb)cFeWIE;nfSV>m+eQ%nsLEG1SQqd-x&XAz5xZyce&Q?g
zp*7E`?BGwmF9@N`9Yyw(m}$O%{HPZU7>>9B5*VDXEW1h)_k*3S=ivdlU#EYY)H}J=
z^z_iOoU#?lQ?R5FQ?Zd{zS;GJ?#27|D#tg8%v_yRM+7R!ZX_7#7o#cSQ<Mf)&Rkd5
z&Ck|$_U(&=KhWON%?K~wTvz9}r!|JJUnwm?G-Yb?2tu(lc*|tZxg$G}C64)U2l<2?
z+ISXQHXmgmk~Q??vs^)mAcM^fnpitpWaOIc`5IbPQzP}}N$W|4SDX<VWASOt`a>2e
z_RL-Gw<Z)Us^1>&CNW&NdAZ}uWR+uCyk{<SXLncL*wi$@>~p-%kNYJiRpvvCu*n9$
zEnSa=fgtRwCr>|AxM1tc)a&mf>d|Np;Ln2sKF0%yC9g9h<W3cZUXS-&-(ACqxVX6B
zR!rjO=H^_NFoEt?Zaxe?YW25=E+9&WO10N0tZdZYQ7*kZtJS&lg_i+)N>QxaLY_3^
zyQsRbb-QNo(6(wRsSGV91lFR?>F~WBgBXnfMxPWPZ*Sqp4i1-=!&!;SuNs-0pIup0
zV8WQPK=|&DjU=Xgbb|){u--7U9)d7CPG~GN=gLWtK3yrhc(=yVbcTU??y+n0b_|Cz
z$cb<1E|A@fN))AQ{IVeaNCzE--UGfyh*ibS2-H=EYy6ymLEL^;b+T;((bT_Vvnw2W
zvdiq5h@O0HC;Yc>yZBuxDp6dO`Sta*7OyPS@b1mT*MvE1;ayC1^x#d2C?|0HmZO{&
z)91$-eac>s=ouLqHQRL~aNoBZ33xQX^Ci$aLKL9NeuPoe{y+e}EsqB<t!jtr)0-kp
z^K1~s&wZsSgn_Fjb<nz%tj;TELn$56Kl>{CSDT$IIn2}Hseb;L7p_I@l~&}wC~W9(
zCo%%?b^1NMy*1nG5!TEJP@fQ^>&U5~d;_263P8;69ua}{u;!Wr5?b1EI8RFk#ErLQ
zXODfFLa#RwK0F;n8aJXz_lCYA#sCWka>>G?B1fxA)AaQ8^_wj<Cm^#!eR&rK?e(|H
zL>&rXk&QRRQ<@AaJ?FGLZm@=iiPDm^5Os8;I+gC56>+e`Rp!r^J)8@M8N}lH#r2cS
zMEv~LiK8{eI3_P)Fy>VS_=bvz*v#3}yA8;-l?BFCx7cx6kXm80LLZu~|3?cTMqZX?
z!alV~^EG_)>yL(dd`reYB$qSC@<GVki<-Lx(X^`45*$7i*o2BWeyicD*?N(K$-&Zx
z0cxvlklWof8{%(sUjG$M%glR@JRAq@jIVa5EM6b536rYR?HhP;lfg?3BCeVl^8Cv&
zU}WDE`et5z`8C(;qjvc*d-sanbtLa@{<LCmvd~KMb~At(d@zZcvcp%X;mVW@H_aLO
z@t+T^<$dd=-v@k7mYmBSpt16N?&4Y=RNB8Gts0O^>1goay{2t}gxBHMycvpa`*puQ
zlzUxJV1?7syx^dm{VqfD`b$nO(DyC}K*oyQMU6r2`iRUZX`0a1m(&NiCtX#SEV%MJ
zA~&^owBOmV`4_;3R-(`WH-)}d5d{(QH;#V*7mG$c$RP)p^CmM#0kH2U1gck@!mb&P
z;zfiIH+<fEqjAMS<*v8LRbIxkf3kwqFL_dO->miEwI06xYXoY#8{vI7_A`9<oN89G
z6!AMR`oKyF{i^z>N`Kx$zFb><Qr<u&Va#A6Ilc4BP^##Xh|0)SW^0h8K$QL#`gj^9
zF-Ec%_#W^%^x`H{Kk{TTAjY;MGR0!<>wFiDnxq*5rD@35Kj1ACp)<POQs%Ka<#wIK
z02|zu%Mf>Xg(d}!?mBEahWu*D;}egSu8B|`vqr6fM>|tRRxW@ceQ^I0cfa}C(s_jZ
z_h@bP#Tf}dM+0WB(mN+Lr+t>P%mrBd>1;yk>m|PG?Hhm#n6>)I`HLc~g4@xb*B#Ig
zaLE#1i@X(01gt48@8HWSIn!VPB`t6+Ti=ZGZBTkPU&N0qXXGY=7-ayjG=}M&IW{`s
zv>X4DEBiBXNjoHMqS@TgaHq;ilAuloyw$GL7MQPJufo)!=R81Nx{8iPQgeAaQ!Ze)
z!bws_U<E>hss7w%pZSKh_~;?JjG(piRsvV?OEhp$HFFq)3%<L{lDk>R=WSZ;bgMW7
z4Wg|Qh_-T|M2x=qFWQQnYQ@&n64uG3V?d*RLd3>HeFitT6i23rzOYD8gP0<=g8@FD
zbFZ6)vI)binZ3nTI3MGK!w#Fw7x3D(gaN8smgSpqN!|pl{KdL_dfGrmT`AqMqF+%i
zK2ll*Qx2+ZdwNIM0$u|Tos>Y!^MSfHJg_uw7{2M5*mV=&AZ0uSA&UGg{`|raWe4Kn
z&UzobvX|T8*UxT|EbmE&%G`nHL)HE}%l$Imt2aS(FDhEmzWM!mmP*FScaDIe9QX9}
zoZ^EeB_(OU0Y!&SWpLbiw9pWPuCLEi8|p)fGl04gJ~-sWnD^$Te3LUrniWd{j<$Ir
z>3*F(gRAR~RyL+JE?6TB22+-VS{lxu4jF>+Hp(q)g0l7cgJ)y`O6rwrvvR9&>hD;B
z0AAhQC#W6-gbFd*<G^~tvj1cbnrEzy*Ch~aH<ZK^peap@fw>@qc)R#lp2Bm&$Nc70
zH4j_W+z;_-y)IdBPd?ed)$U`~L%LiJ937r^*%jchmD8W(7{U`<8h?YnwH&Q;8yppt
z{Hv`v-<iwh@P^@8)r(Ug;Sc?A`_E!Z<<lgIhmG|<CL4L!LtxY`sc-*{N<7Z8&S2=5
z^$jfEdOz;P0C^3zMys+xXz*dQ!l?<pbMj7bTJHY%?FOV!|D3zjbijIJ$mM}s!u5+e
z^*9`E0g6Nk0e16kD>ryt#j6i;TVoBz-<3wmKr;kkrREoX(RDTHtF6l%3j0coWI!rf
zpZdvk>i<}g4)@3eei)6q4y=p@JYUq8zC4MtzQExge~;!nLxWn&ueS?P)9I_xkxCOE
z#8XNGulz^_=P`5l%?R0a+x-&9A5p{P5T%lp^PbX;jZIBW#lXUueVXB^3w2Z2!NhG2
zguF#pAGn53*3p2kq+V;tS=SznPSIL-Mo2H3Nl5zI={6BHy=zP|N2O@|p)KhH;CH<=
zOhg<ubrx=cszwnsXb{`Hf}cJa7WyB-iw^WbPPW<$>wO29``!u-laH#vi7-;q2o_+}
zz1wH@-%}q0U4o|bgwZ!z$JW*s=!BsBr`y(7*Zn}@5GTa%05`g3*%a9MXpW2=KogY&
z_dFqYxR_go5P&)?3}g&mGppTjhVNd2EvQXChVRu8LO_60-SdtmL%S&S^i>BxP$9Gb
zDDLg)sW~|3UX>eUae6vg$P_?d7FTCyIq3?~P}D+Zfbdwe9H=r#2lJkN=Ok$=ZdDUA
zt*Yb<iaBUr3j;&Leg`1^Vtfc{h5hMLgNq>fhsDHyjcT_sCzHoYf7D1|zV6Wg1x=%l
zv9VtjD*ZTX`VYJjtp4%$))(VaQc@w0%g-Aa7z7Su^MAV6?Oous0wnE7Uozv5Ddx-q
z{d_zww%ukFzU%qi4MP1XtNF%tzh6Js(~O@x=q%@HA}a!9K(f#ck_9@}^S!;jMxrX5
z>IuBsxJGkOzdL1X_pw#9!Ey+bJ3s+pZYEDv9pPtF$O;r|2e({WS}GW@yMEipXU{;K
z@ES<dfCk}bUBIs8$46gfeVTjv<QqV8xXRo~HZ$^Ms{`YF-YUAfmVWYLf~Svk#*r`S
zLsOM!3YhaJ^+<D(rCtPXGNb{Rl<es?8SWAVP@QwVmp@G1y@+S&cN$3vZrK1;8Lp*~
zCaepr`K2iRl{FKUJTs7YS;GQbo_Ey}SYfWq7@Y9Zc>M9!{f`c_P_#%;ZDvr#WIg&^
znrsy&&QCvmpl-u2ORUoVhGh?Sy2O0k8{wISD*Y;}AbJkyIJ6M+YCucJhG37mp%dw`
zH?5ihh_fc}D(>KMl2(MGXI1=yf&^<<*YzC)PS4WPvI#t<3oNR0mc;3hjO=%Haym_$
z<OZ^Pg{1A_I~g2zO+k4`hFd5F6)4{f>doHPf8cqUMWb<$Mj!B^znjnByZKa%GTs>N
zy(WQhNkoA<KBX;mcO+ig*6sa<Baa|qE1^Vx?S*~ePOoN$hS8X032T*hkJ0}E@d?0s
z2|pfXurL>ZicpZ@P!(bXgCPWTr~y%w)Tst#KBnlk%pE|fX*{c1T3r05XKc(56A%%}
zLYa%-d<b86_o*Y;f(7oCk0;as+_<}4rEn|b9?2zf8hE^x+R@}RpQ_$wp*1hIDTk-q
z3Dp_{fkyp9G<cR?a=dNHeDJL11XwH?h|EO5Qpg9CO9FVBj%&_PgL=~sLB+bbn19{`
zZlL>{cYE07grNSaxL<f-vC0h|?ZyJ9>3(pN&{mt{uA3mw=*J)F!uA;4f4hUyt%R1A
z7O015<PV3`)tnbL`M9_*dYW{_!I3AS;PP4vxNg>5yJ<m=CtX^}t!DbHY4Nf-;yN`Q
zfC`XRDskUv1AF@gAw;#BI^bwTZrr$GppJIQN!H{kTeVCd$0{_HZJlNew<e`E$S*$y
zjw&kjg|xF0+nqnOEyOC3?6!o2gv{~5g@9Kn_1JUnwi}w=^i4sn?@17xcoM#s&Va7k
ze&+R@^*W*Qjf{+#9|AA*fLuC&d%s5psSNqU{9q#L^|wxhNr6ccQAs}oqSUuzEUF_@
z5*54PiWWXwIvi)py?hyNqUrQ(C>gNc_(=q%3kXkXi&t5?yzJC2`!ep%JQyM*{I9(0
zt@dcT(Ujm){P(KrccTgx2N7Y%$x^^@#Xs?B(e~9T+iY_lhb^m4QXsEMfNbgShz<dA
zWG$EaqvZ^x_>&5iZrCLZ1+a_rCr*%3E#!jyho7<O{v^A}<2H1EbYu(?fL-Bp_?q0S
zot`k*K3LIGq*Z-uVbij|KKJ35iAHd$BIDm!8KCQhxjP8+7yiKvj<#5QQ2OU@A^%OZ
z2w*Pt=Jv35%J|5LW|XulKL(VNMvk0)7^<Bnu5<Osw&-iiJtSg45~_cYp_C};GF3|;
z>B2jhBfI2|_Gi(ZU6;h+niwD@isjpS`;iRh3Yp7U>|<i?3#ass10CXDQ72Hr-tJ2k
zn=LdGu7Ls~*6DD231r3Na#lbm=q#LqIY&;qfmk;nV0olKtIkZg^#-38FkXFA({w05
zdmKj_@R7*1()GfeoF8I9^yvW3TV7mz4NXc(8MKXwnIa-^>|Q58#T7_lslVfeZl5u`
z<Q1zo27(hDmW2RWTzjAVWpX#5`Yh~FPbNX9bbyiqO-6ZMK0aB3OX`5&+Ta!^7?9n=
z6cG;lL(P)p+2ize?%Dy3L}{XDDHM<&ynU<DXH^9p;A5A)j)Oi`K%ZiOPR!?)V*{O5
z2NMr{3d>qp!mW+8wE7viVxOO@0YFZqsH_DuF>%n#6({>E|E%hrlcs2I=cL0^4F`e2
zEVj^`>c^vnt_5W+>HYAK>pu9Yi{vY-0b!6ZYBU27N=Xc$3SwNR?C8L2CAa@34Zijj
z!Wa)kKRRO0_InuYGpS>xxG&DmJEr%4cU-=plJV~%+c-)QonvMdyBdT0FV_}{4UBy^
zU@iz`(JaGO)+01@{}I`z0PV})E;^Wi;Bj`U^asA7ES)gU-Oq~Z{s8~}D^_E^*@s*&
zAh#<fjJ{9ZTza{;rfpp~F69`{%uhROWdv`hJ|`~*viAb(fw|M{x>S&hZ%N6uycrUc
z<Ubm;<NT?HmYd1o^hA`c%9pu@J7gS3zG%NqYb<KyM|<@7bV=tldpI59C=T6xTQ3f!
z%1wL^Okpf9FVBc(1@q<0lh^1hp#qDcK$sW8q`z2s2m!#6yAzH&+=hvjheJ0-+OqtL
zi;z$qr7~qGVpDy28!30YnCGm%cSng}HIM+4J3QWY8pm1XFr2rp_SkJN#B;9>k*8LO
z@T^9q159mEKNCg^)KDTw&9Pbn99E;|pM~Or8mdW8Vawe2E&ztP(drXr6lwOhMF!XE
z*LTUufwwCliYGTzsGU>ZxETgKKB|8cwtHopcdHTiH8YKbAy48sceFGYx%bmbPiuVT
zM3&Y;Lc#*7(Y(M-z0JwVnd-au`lHEEjOb~OM@~+G{ew5z<IqSUiArUB@FP%}0Ts!U
zP%W`a4elLv_51hl(|cfz7>9OyKCg4HvAP5#uwQ@mWIzO>i1-OoJWp9klB77v66T7X
z-7u3PmIF8+F%a>VJb;d~M~eJ<0Pt1@$OD%!!nNcB!VyVravmR13MfHlnD%x}-PV8%
zhWp*|ThsrW=5WK1q{g<{m|R+|u@E#kC0|kIU23l1GqpW!n{XP!lNlQu6_V0;M2e?E
zZfEYMWxhhx7I1@pL*1TUYB;ZB`JuY`y^Jk12r#t4Uu!ODrc0Li8$7dahE9=;hr|hw
zZPDyY$j&k|X~_bJmCc0zL_@CKAt8Gb`4hey$5tRt&k=1#ehyK_Q{Z+%K|ysWTWA3=
zlg;>TEFe!?rX#II=t5hOxNR%dGrd_t4r>-HNG#a*;Ppa6fE8^A5p{dcZpO(kX@G8$
z@ICJm?Dk!sY>F*k9^_b4IsjY|yp=xPcQR~71$Gif!i|h<_oxlT1CqNA4V8<(m#^yo
ztPBe_8pQeR)vUXQG~BlUG@olVy2+<P?f_kft8-2cdthNWD)O8+t7>w<c-sKEeN;^m
z1XA8$Gs*NM0OEcE^)-WPJlo=4pZFj-?UfHYJaBe)R#igMyek0}v6v%g@uFXM;~ErM
zNH<rHzJGVEtu0x@mxevR3-ezt$UjHKRKok~3rk9z9m(l5kcZ*Dx<IsK8;_627ytpA
z(a2R0u`g2NILE(y*|_f$qNG<2?@xPh=Gai_+$9`P_65r$Jl^RQz<_nts3Fj#ou@xf
zwLNJ<xw3cNDvzkRe()}OFK|~BmcMsl&XOzs8__2R@aCRA!R#*$ruU&_EXKjMy2r#U
z2x14i2IwztOaBz1zYKdiPA_NEgdP-8l1$_g$20=DdGmDuda25|Y(IQssf{Ij@;^GN
zD9ZKEufN%b#3K7m%dI{ULHut^Gyf;4`=1<nZss{)KRxyw0xb0tS)*+C-}9yNO`1Mr
z_-(Me&4FUW{7e!@*=oPe+HomxSE>P646s9yfU*Q&rg^w@2U10rC`F4j`Ffr?X6e00
zGBP$snl$)sPH}4_Z{t^3$+8&07rP{F3yYij|A^!bzLwC<s#6*$Ptv)@MEFLLaf}9)
zpTF=5adaO6z;)tw-J|o_BOp)mgC!wGhKDOrK(M6b_?#-Y0Gy@?y9fj5R>W~YOsaeP
z(t#*m+v+xMq2W6%V*ixf@e5PGUiq1LO)LL;vR6MaRu4Ozovp}{+}Xbs?Xsh6RKoJD
z6H7AJLz1IV2arbPJgg4=3SeeHHmVZJpEXF*9U9NoGMEJH1$qBNN6lHS`?uW|1;5Gj
zEjc4_>D_iNY&mNT(8f;eZ~~83{riFfeg(L-t|++{X%F^x2*W`B5q$7*_+C02*!faT
zX2NkFY&&lw41h6&w4UVsePdaZ&)jjmv@tI?cUA>fSX><VBT@g}tx_W5K&opyGW#%X
z#gbf+Hvg`l!(8=l-rTu@<hHhT2h;C$)b~y>Wk=xy2y|=Ue)>>{`95UCK!!|87(Hht
znHsOIsj(D<4d-h&lTZowe&8;Hh-Z6uQ1A9v^7_LPJDRXtdtY6A+>HKe#!`O>L*5$n
zJ81-(;1wqAxyDKNCq4yoQ9hgW{*Ec^UsG}DGG?65b5#h~35)vB1t4i?B&c(IzJMA3
zBo~G51&;P&)yI!NJ9t1(vvE})f5L^$cPzvMK#<wJd04KQ(r;G(t1O1$?uUzVkrllU
z(aAZN!rKID_A-sWC<1HZJ6Q~hY&#cRNw*PwtK$K;j_$&Y)auJhOI@sa60&P)A$)#3
zb@&IHTrh<m$-Xkp#Lmb5=q=}}r3)Fl_IJ;oAbvb66w36i01>SB9>8-5A00rhUz9JJ
z8;pbe!RPjwDU<v6cb)Y0hl>yfAl=DSgW~qPxZ-|CAc^l#YzT`j!(s3wLZBMCa~J!Z
z4u~oY(k&+VBU^>Tg+#NKxe7PJ^=X*^g#qDyM(KK%tYx&_^!;5gdb$j_T0Jtm-G_Y$
zM|_kx5q97+9kl0oBx(+Ph}fjzR1&HZsMbE(T9KZHT}cht!}=)PPxYaA%V?ZBj@s=n
zJna9aKPj9(ZXwqgP$Fd97NH5@Ikq?Bhh#tp&3x*h>;V2#G13YE>yIEyyEsK;_93B=
za#tMEQ946ePzhLlF>ugttr_j902M9=#FE{NjF4|dWE`XG5wbNRMI21o4u3a2l&H$S
z-|%AhNoR7@7{D%%whs?~y%(o!c}lx}$(Zo_1cr=p%Q`^>9|eN0k0_Qy_O|41yUZ`I
zrM!=aeob&rwY;>tXkHo-a_ixst#H5S9%)?63v;WafO<W1bfiKqDh48Ynf&s%fc6Gx
z!R2G-6*4EbXAXg$2*iNTonSwQ;%(!)Chnz=x8sO@9>G)o&0;F?khfPf-4o1(m)aX|
zELk>x|4`f7Zp%ofhtIQ;Y3NU$WJR8q|6rEm<$#6KqDhfVQl8jLY6!J>r9=CcANRfF
zmDt<#5r`(+<us|rrBAUR2FUbUWF$8*N$T0;<|pH_*zYB$XP;VLSz;ikRn(G21rIAx
zz2rRwvrO7=F4Y^QV|p78HpVO0%|<VK$%D={uaWZ$YTLZr@DK&RZadYB*G@xXJ}hSy
z2j3ow%K?-Ec_sYq0*>Wcwz^Cf$$+N*cht2`Yig6G@28mr{_bZwfA-JA{Pb_*ftAr$
zY2KaQWZP8w(fMNF5AV~W)Oxvr@8#c}ax(m6luy`QbM1#$BB0_$kDQ&gl!3gVr+KHW
z26H%V)>tOV!!P@N1XIsMn)dm4H2&=;kcrfLtxUzn(!&5XRb&^VG?E#7xP%>_laP?0
zH8nN;qcKVDhfjZLW{Ylr%*>Vh5bg|w6U~6&2m6w8TB5^Fgv~g?G;#*S*uT*Zs>hWj
zSxj(h6(q*WTLf>rAuf1R2Ko6N;McC4$V;DKqYk=lCGy)<feY)AzKX-_;dl<%m#_s4
zsS`9EqL<Yl0Er;^RC;SLlCYV<acZz=OyNw{mlNxsj>m&vd;zVD6#*OjAXa63VAqp>
zK!9sx=PdHQhthKCdOy7ho*1>W+cfLdX|t*DIZo#LWF|w%E<^5)`(}Qoz)H|Y=3Sep
zh+X_ojk3>q0|}n#&==OJUVAmMV~!MDa|@QG{o?itk`H&v3iFyaV!-~HC&&r`LCXTJ
zID+>+ex%-<F2babf|(6yMmb(@Q<JwGvP=>C$i#Xbfdg!XK@AKHoX^+JYT^f?5o^FB
z?~|ZGI)6%!?t8Rcu?9l|0s{WI10Y?VCO-hZ{QgV^0%UDm0u#Lp9IQWOxhn6{$haI#
zrT=QSOgMSu@--|ikA_43_4p(ie!07lew5PA28A#$f4B6$KUi!&8hk_;mGP>dl6wQn
zWb;&hY%_`_7LE1!Yj4b_H;R#Bv%hOVO}WBc6`vqa+BuPs%~)cN_BI8Q*#$D!Z&-AH
z?dZ8<XDnL@Ng-g5o!!u{l<s%%Seo6PI|IlkOgc(}>}NI-kKH!7c1JOOCj0SPcVUU#
zfpfr6Pj&UQO9{FP?;Y}P8CT7Io>5EZR8lCbR7vgEmoMLko_OA8ch3DD94<G_fwX%5
z+^)5+7}-Zem)806u4Gi<t>O211{=1T^0R`avl8smYo>$Ib(m55Os>e6FKt*Z8goTK
zxI#Mv1o7VDN>;LE9`iQ|i{nL;%G8xq$qhnLBH2Z?w2DtXBgTV9FTD#<G}+;$!piA>
ze(&J^Uij>vrpsL3qQMCF8o!u9V|7mf9aqLSS?hbHvw*+Oj#9~j^0oJBb*V9-Z})zT
z>A`8%SE~CayqtgA<>wE!ubvXu*y+(ZWGTP1Jl$^KcSerUCvNVM<86GhVbCtuzcX|8
z+bBhJzb$RN_P?w0-2b^;O~2;TE(t|a(_CgXT0l8vW9OdI-*)3$`OIBL_^His-)CPA
zshB5MS!k`g>=VS^q)v?4+*nJ^WLb85`LFR+e;^M&`~_Y1w$a_Oh~P6P{hkVVJ<%2X
zPsg9asV#Cud*uU$d7d_nW;LA}L9)f#-;q<XxO*rM=_jrTmFq>AEDe?v(e2#)Hv}a{
z468TITZ)9DvL^r3w(qrPTELFH6IZBOmVr{<kBIG-7&j9iSl^rG=iGwCP;ErMytYLk
z#x?IwIuba~2R%vPE^pxWXJI5!#gCl#z8@$@nDg4X%3(X#<Guc)^`fDh<8<SNSNU0(
z`Bs70nRm6iA^N)2b2am=2oXi4?0Ws*z?uErx$|#AQr(WZz7Y@>&U8$hziJqyY}ib3
z<)R|u<#x>@k*$S6lG~Lt-?%vUC~wj#3WSOq?#K_{*h@}KgsKXVtWxx=-}RiE)uWbQ
zc0f~loia|XaT?a3k_%^kN&}G_GExXx))}#If2g2JI5(j|QDAQ1soj!KFq|tB*pIvu
zQW%ym6tSK#iO5-h9KxH$^LJL&A=frK_<Jh?Aw5bznIQScRX27`-P|^{4R%r>(vT@A
z=cNl5E`(37Ym1$pH9HHGckaCBj3zpq>)tD9T(J!M!4E0A`m5o6V6P%(zoFt)uQr@C
zF-RJ;xGTyL<quhO@GVvF;*)PWWI8r<y*S|(lkUe6A>lyAoMzoCYqhV!JQYJ0g!m6%
z(SbGer`ep#fAmtntm$xB08&(6D)OLw=0@+bC7tW1FlyUtz5XX4n!~ayb19bc(mkGY
zJxObjLdX<5?sFup$J-O)E9rX@y*kvt;&{=*lb$t}p!9)RQ4+#8Ltm@nJSnl4?D+fc
z(32m2oG)#j+3S9yG@PP)e+sprTYu`sB11#;Da%3QYKMo3T<88SsCL%a87;-+heT$G
z!Ol1~8qJvHZaI1A;Ifn_<1RXML2>9rmSx1;y~!I5^8*CZB8U6$D{46pxH%P;AMfA_
zJp7Me_<d6#`#+DyO51P}4x~ZHL(3qc4~@0Iu|~A2_4H64JX|S^5ScdP(rDqo#8>L8
zkq4)&z4<(y@&5Gn^en40e;0S_EF;~?`H~u(JB1oWC=gu7E%c}RWWW4<tmOh^S=sJY
zp}YwooRUj!NAhJKi@Dl%x+ous3V4Y6ST#rW?R>V{za`=Jlp6P0{4_-A=kWISHU{Ep
zsVSso=t})$1qexk8*SR7IOw;pfVgLLobsv6Jz;^27?3jt)?nx$qNllv!2Va|bS~X6
zu%W^dZX_cpxN?-1_LvaB_8%=knky^z!Ghb5Cf99WG;aTBEmGC-==Bn<<sEe>$aeDm
z8Z989ZQq&OJ)je7Z43&UOOF>87D#n~YZ)DoP5idJvNE{#n3tC~tK9?Dc)HD3L_}mg
z80~T9tnRAKQ>jnyA3xrhJ_|Xz#z_K2w4-R?(ntF<<zytdx#u(r4|&hl)zsGdYAGwX
zFF$4NtSENltPm~!&JWFVV0VswnGi=ap;fO{;+O`%y|>ckM$x815y7zSYR}b4dC9oF
zQF~0o)28zc*RWqK6z!$JdpTT8^`0o(EnU>54tw!~NHzAs_+7nGS+qD9kJBt`!^w_{
z;?l1!JfTKF)Uj?{^*HHbXeo$Dbs#M(D8#X>Zm|UDozvwULT(*S?^V>D)%@oS7!_`Y
zfu||VHzoMU?ZEEc3!YrqB3^E}1-2Qvm73L6$c@ZGd3l?ePaLa$qhtWK%a)X+Kpvce
zBtINY%eys<Q%YQ<eA*%nse^n0RH*9X$IE$m%tQ5fha?4J)+6VppU-8~+3LmS5Q0sk
z)@f#%ua6g3RqgOV6pllhr<*%(^dCs7<{Zl}Btb-vK>({zrlzf};N$^Sw*oyW9D5=T
z0ksP!ajh*$Q!4J8S%c#?KF2vZqnxj8@GuYGz?wVvR?zQ`bgt}-z01$1g1ou6a0YVq
zqOZ8Pcq62NDh6h)V`wOmQ&duT^ybYQpS8ZbAH7!YhZv!Oa%%ed$@8%R_kev9)FWoJ
zbN2P(WOQXmhlZZm<IAnzq$BS20F@GhSCwH^#)4V}X1+f?F8Gn5k<s4Kw-2_2TlRFi
z`;Ek1OqIE_q4WGL#LJiD@R4o>;D$+5bhHn}&9?7PPSb3B`Ve#-G(#DqN^N!ZANlP{
zj1q06`ua;P{h2GY>=)X1bZ*gx-r3CL{?*_JyqdQ?^sirx;s#Hd7eOIgS~z{YyeVQ$
z3KII#Nf223%g5I_IU#-`PDi|<A(I);{w{95F}L`tV<X?b?Vk>nhH$sHw=?u-DJZZ(
zK>j&<O<1_;oui9|VPogOoBsqRB~KgHPb~TH!ORR$XkDo<HBEj<8CqyzXl54ZIZ{%H
zs{kwUV*Ru?kkvYG&G)0#lyJv3^bLs)z&(G6s4W9|&8rhQ>VMIZlT#W7?l+C7H;>5-
z7Uq<b?-0CS{7r~6SNSI|@R7$LtdNabAUFb36nu7K@qZ4Z0+``FGKRVo`le4F7%>aQ
z@9CKr|D4v#mp@LcZv69yz!A0cGx-9*@~{11q;2@GJ1MzEQCs}iaB^<8Q;Pg`qu{6R
z8=I#o|GDZV-#rlP|LZ!&{0B4K|8-rIK0TW1%M2ir#Lbmls%2C;!G8_FvgVDZfP`Dy
z)nf4q6ReTvrf`h`;`NhABy!pfjUAsvMx4XMbEyvO`h!Bn@QsKqR+u-YF`%ZhC?hM2
zBv^yNI@z<UI{~joZN7#=p}_L~h;>Qn-C;rB{p(m#I3K1KGHqjDaVjf?d!y4ncRsUG
z)HAxPG@v&s<{h5>8P2TYQnPy|ldIc>)5g<L+t9E%!;kO7rEsyFA|cL8vZ|8<6FwI?
zIo0j1ZRjceTv!mHD(}OyMiv(qE~`U|C|W9tCx8F`?co%wbkb#ItYj1tPN5hSL}at=
zj3uV`$MZdPo^B$#K$>YFS%WnZ|I9uef}&{o-WE}oD(m-)0ytjE0K!^L2+a%{^5yH-
z4?+qGA>sY;U}hJO-K6<n=ljQrKkp*vroO4(lamHbG@nibPeFfov;O6d+P#Tn0rVF6
z?dWblIrir=-8VXlveD7O)LrlsOt~J*EHt6IX9A6_oyvsx_n-EQkzYDr)W0_sb-pV4
zR0y<ikMfH}{fR4azDyT`Z>80dUvQWFwZG=l{&bQ>n2jrpd&qby$)eut$ZQicDPo>@
zw(1G_)!VWBGo&A(`$Q>9wJ}^YI%|iz;vJWR1)~jlXz~5EU6=oikyWSdCdVYE@3h?Z
zMaNU2_FO41S5Y<ETEE742g5E?pRrT)YQ8+y({S;~J|8<l9%%i421-*npi>{Mu14e!
z2Ee2CW9hw2h$w~P6BYQ=2lHH2(X_p+kFoK_)$NY<D!>M#*qMCKk07`;Vx*=6JxX5&
z(zPf<Dj=)W<$6j-4=gN>vP&q)GqJeE$gv}L>uwD`DvtAF_^O_%-nOzv6=fq)D4%b$
zO%{z$LsCzOloiFfNQ2u}P%Qu;L3Zpem}=*umx)vlAXYa+86ozR`%Q&-Vc;No$6k}C
z#q+eIUx+_7uUolh(l52V?8Tzy0zQ^6_Foqi*@$6t8${thc%sW2_@^roZ6A)c+<?$Q
zg!c$U{=YzN@M3&K81_gM)OYmPdj16ctQ53UFPvK7aG4ik@ZrWv`zsFBt4zxAbHim+
zUHBr)CmrFYSSiNX7+*f}SlMVBBA0|AXW~tPuMjXQHJhMU1V5LAMMRo8TKoYDcqS-v
z9=w?L*U$$w6yNAs*$M(!DaNX!cAmX|k@H@6rsrszcHrmLymytH$Otf7HdNr2st`Le
z>cRN0e!1mEr3B@T`eQHe8ITW68=4Y7a$Y#7f@pO*?H11DPfR!)z2e6VJ=Rc+61~K(
z=aY2y>?v~G6nh?B%*{@{ArM*z8K8PHQ%hgy0Rbz^OIJ&UY*Y!$j<mnn=w}6<#$Fti
z)Bt-nWc(V7O&4lpFdxE6US8s%9LsNRDQwZpT~{9IqI-XTVU))?uDyX#{#38v)#8u&
zp;BSwl!Q;Eyd^zU=_k8k1{Y@tQp(k0vn!r!+N|EQpbUkAR+f~k(L++Gd_B5eUu9Vx
zPzQZ8!c4C7Ped0-Kmsp*v4iL=UCSP=3*_);x`;&@)o*$Q=<jMu=TA)9_|QkO>l!nB
ziaFY4crbscHh-0qGw0hEQe23Gj&bdi?<cYb_gElOY?TG<EmT}8G5dq>G*S+a*OSuH
z(lAOtzb2f5@xE(p#BP52AAG%aSd?Gb2KdfU(jXup-5>}kNS81)l2XzkDIqO6h$5jN
zh)9=ogLH?4A`Q}wbV~Q0@%MeZd+lDk^A|9@b>4HHxbORU4i;S|T*}+KE`mz1cyLE0
zW*j)pFS7#!^nhtt{C+2A_=kp-Wg-ZymVvk}6OvL@jd)}?l55etp^7cVh5Pp%mR6Z-
zJsKnOn75#aqSW!in3KE4tz|d-d<hkp0l5nqB(z!hEkR|tg7KBu2Ay^Xz*2|5x{Jvh
z;A#?XJ=U}*x#Y;Rp%tdR9!Zz4W_v`_`n_3#(frIO^^I`u%PqtfXH!w6vAC9YSz%vJ
zT1X4;d)PKxN;1%&^{VFZ=@`FDG~^x|^>{}QMxcYZ0XrXHzIep()9u+@(D3o@Tq;J3
z_`p@T!tK|tSJC1KiIcP5!P+mO=7{_m7<(CwYXpCbtEKt*g>p;ZcJg*gSUr?>=Y7tN
zMaZLtLWb_XaGmmWKhJa<Y;-uWWbnFuSIGT>j|a1fA9wm?iASCYVPWBx1P3VGh06pl
zN(~ym_h0%<veTa$9j2B;&O~%8=@3b)jo;(HsLs_$FuX+|g+P>jz?enn54gUZ(mqlH
z<=A$Q>507TK_WskX6X7P`?<bg!G;CFJBx+Ovja6w9v(@tj}{>c1^GJfghk%6Qc8_U
zzq*6GOZANiCMEGY+$Z0T2I=9mq|T@{`qWov<OpqVoZ{Wq{0X*BV_f95@7LZj*xAKj
zE2GN~3jS19devNy)-GqBl$i+)oVBVm%jqOK+^0Ah(q!r|2wF;s$nT!LA{fe-Y@L<$
z)N4wQ7joI@AbKYw_ac`FjBCwR<qAPpj;mvN;Z_6eri_hIeY6;(mdr1LoR*ra=XOsy
zI`nijsBez24C%^2LOh7gvdC73?Eq>FY*ll<VM9;d55vefBMdByYHChzATNrRVPJEr
z5u!|Q<w2{Q&yLkTkD$!#Y}%&N#c&w|_jvYSfL(qzGd(Sa2f4Vqt|ocTz6$szO2SgS
zJykuoedHZL2mzhZu8DIihDoU?iR>3ojRiNdHxDJ(Bc{urSj*q*2a#VQC>0xm=-Jq;
z50`&@iu8$;8`ORF%x>k+pEH=0wx_ZG*&pk-7ZjB&Z+jravB+X7m!AT?A5F_D%Pi{(
zWQurTHR!M1a((lI8)6IEIWr5lGJ|OV|7gg=*pF2rvF1`sD-?g#5&rY%&##ehZ4*}e
z;vRhcxq`BsS8ZTQcCNdnki6L0$g81x8rBEj{Qx=2zV<a&1K)?rU@*xv{KySKPk*y=
zzjm2=+vFuVls(3sAc5N>PzXaFmg575j%$pvD1nq<CO+<@ysRF}aHy6g{ngzC7CvKn
z5H<Eb<`Hq|W9+^pDkY}}g|DKNP^NA_9LzKeiq5Xqm3V*+(EZk;>nXp3k*GbtW(1DA
z`{oz(3k;&QT`C%~7h#sPN-D7%;~yQ<gjl3CP?w89#lxVDz~?LInjE}33lKKO#PeB;
zKxvtomNwwGG3e6A`GRLl`JeR?*6(*LBx&XI<5oydrt=8jpkqu8yZr%r!f^Cu;2fBH
z4fA=B?I<8tayLXFR%kpnB_(2U(Htmn-FJ_-z{F76R#ZR){fLTIcH!#g@UgR3QKW@G
zI-Ou7?g%DGy;hEv;YQ=l&e?|N;-C-6yY>h;g}R6-p=TJ#=$}l!zGQlBi@I+R8YUt^
z6iV11&$k)_LNzL`o;rO*Hmn;eRns0z$gVy7tZO`B=c&pV=9CisS|6^zh!qBPkKddN
zo+itvpS<N5$m1K@>XS#5fu;?AS&3_z?siRLx9IC?i97!JHqqzZKVVU9K$ItX@%k$D
zPG54}F}JH9^EZ&}WV%%X#=_VT$R^~OBgyTb9;KqV&aNHm!QXJdM`SKULi-s3I8vk^
zeq2Yae*N@IVzl~~s<NZQ*+Yc2{pyV}4La3*G_{(PIk$lwU+@mgLclw$bOB%yq=vzR
zK$!h8dg@#v?eD(TPqSU*>@bJE<#ll4E@;?r|KjXs!d^xU<X2y{-MuRae#ic4E2e4+
zdN01q;M%XsEI9I*M$yLO17e19^{e>QBg_^G8_O>&1CAHEqv4F+rQ;)Zo{;E!;50Xk
z4Bg)4H@!9v{%bdeDbw_&Y>TdP@E5tmY;*>rg9EpAYe>I?`Y!?asp9tH>)qv8$LnuT
z_QBYj-$6z5q&g|l8TMr7oUo@p7qfXKewi2tWQ~3s{Q7mh<bOgCh}5r)LM|K3)WYu3
zGaBr$KQWGld)(aT(fJZE)9$h<jJ?pGCtwLm*vd5IzWhS^<n6{Lk&w@NH2iZ`Wq3He
zHY@1j<~4ZJPf>xQ#`xN8ib~N#Gia@{<|5v;(Ve3j?#1}`M+iG@-p0w5|IIg#)sSG<
zK4I(6Au5<iG%NxYP}bs5A*z+h!ou<awwFC4Z2nkvy>oRMH>ZT}$yKa04;C*T5-Utl
z<-tu{<%=uYdOo>m?!bkabTymZj+s7x2yrAyFy}Q`(4I2oEu{4C(P{rZx<E>P8%J>9
zuyoCjBU6yDcr!5K>l(Rhf0z-@Xx;X>sGnTwiu6<i%7NBCu|X88&A@nNV)~-gwXtG>
z2zl}P%NxxTM>7WW)uU0K(-)M{%Y4mv^JByjvS3^VYhr|niQj&(-H2zHWgfW(^Mijp
zgfw`(nO<5E4FBTofA=6KNB^Db&15S03nnFQ0_EZp>1|+a_;df|Obbjp+UzrA=Z={f
z02k6oB0A%b*@>N(`uA%G6Dp==e^G8S-hkF3dbQ2-F^7o)6@e`x$#`|%4b361?`!I#
zekcC#dEvQPe~qBn;~77{zRV}ji3!1tA|Bko={Txq`0E>Rj<aW0Ma%KQ4ZwvNUTx3q
zIT)uMPw*jsGtUUb@7Y<2<sd={3K%5wjsX8Xouq&2Og#I`?(Am5&#Tv|xH@TFb9}$o
z_9f1a(Xbvdg}d%W-%m(P<dXp6E-|RPukZK+t+;pdOnj~uPyqf0S&l1YI+Clg1MmX5
zdOq17?2sx5LmTt$CX(0|MMOlXfqqg10~#6{dZs3RG$P|+y`cykPagC!DQQb|={?`=
zMZ@%b8ChmJ2$89(y~*gOV(ZAiJa!WC`a)f^Bw17Q{?$$pmiyM52nEaWde7Zb&@F}^
z61qAZmVBb97~1_3bDoa-%ck6aNukR)4HA>Bo%dKjR@)IgaCUwUq({*(Nq~&RlKpVU
z7<-wZ=d*5VJ<)|nDo(Q`HZ)mow3p%zd$`JNEX>TbX;JU0x%l_u>J#i2I;8CY?~i)W
zzIyNkO&uMgp`n?AZIudYj7;A$)hoTai`3)bl2xH6>W214P3tiqoQgcvhV2<arCbHX
z58@u&Mp!Z;3xK<mgi2Y4x<)G8<}aGv7JtkcveNAc9n@@}F%d4VINca^sfOs4P3)fZ
zqw&I^fACO>0u3iA<KL(mO0edtUZV31(rasLyJt;p8{<Quecd}A;I0=utTVo6Dx30y
zyyf7!+P~>*Ui%k9sJ6D&6;#!V^`**JO*Hvk%ovxPoIoE_Qm(Ht$PM=}>iINsfZz0i
zUEcFLvIn-(dkCzHMUVO$Y}30CMQ~JnV&askYi5W6_)d>_f~uj;8wh80C*pJ1va+(Q
zjEz(1fNxp_c&>l!OngQ<Ko`BwUi^H6WyC#(HYnyl)AqV7mP1#HSCh=0)y+a@Vx<~m
ztqV^b4d^Ip@mky=(}a~;KC551Z1?qmQ>_1&%*w`a1!5E)ytpzh&uF*yqy}Gvy`122
zn8ljMM&XFW9k|@(sC41K&;JvD_@De0m3jWZ;0{33fU^Mr2nk>0;7BlUt0VfDes6qL
zee>v7`F-R1Z>WPb`u`Wus43x0(y?Ak#aknN>vpXUqkHesznCHKTDjxTF*RuBq=Y(`
z$ekP;dejd}&MC6PJc4*LnIAd297$R@@=v^@j`C+E9{q?wc@xBlXv91V)B>73$)EZ*
zHA3i8MgZpIO=S9aRnn2Ur!h2p`dLWklffgL62U+IkYu<VH|o?Ibf_bVn&)odNJ~hu
z6_=P8l5$E-cv@X*`eFv3JaVb~K4V5}*Y21oZ253h4P?k_NB6%CGt!0uYY}q(0mE1t
zgbeukA~Y>b9@Y#Q;|upWM6x_IT1jT>VxINHH0rcWFBK{06&V>GW>#mek)q%Q4X=>J
z4gaIfOSVU~m1jf$G9AW@0gzhFuk<A+;?Fm0A1U^fr<95k5_e$kX0)y6p9>(G@}<1T
z`pFdDZRnEz7IhEDJ;jnB9-QYDiRrVmQMmdCM3n2zKXDE=6T?Vr1d*+1uvzZRtB`L0
zN<l*rxj>g&<;bs0ReNVumd|1v+M<9{3ku|aGh!sOg^Jn`#2s)ac%o#$H93qly!%6e
zZVAAdz#ZJ;-1`QEh}hWWq(9$!03(^aw(_w~rV4Qn2Q|QN`^-F{@-9$uJN&egv}FTy
z*wSrpyaBeaHaa@G^Q}+wNC4;~LNenh2uUHL#c2C60Kh6l|KS6x|KT61r)a5EwNoS*
z^ue!Wwo}9-?Y+IzO8f#22j%6-Z9v(0;@ZBKotgm8pj7S5&eyN~`0WY{!Vf*R0Mg}V
zZ_f$QKYHZun;|BU@9nnmu%|rRODrYV@07bIbuzcW<lL%RB<J~ya{q%dvy_|frJux$
zyOt<l?#(}jkhTf&Wv6-<SN=<lC$nf^RX$+8!zcIpxqgm&y~sn2ofA-(@@4zv0)|@d
z);?k>q=`)Kw!eCMeCBqDfZwVIW3A4?!4dFul)rf{z?J!U*IB4oV>ByjT<{3=#Sxnn
zz+;lvzA>|)4kOvH4>CJ4n#OoN{75g5zu>DouA5Ej)stbdNmWI~S5H5GYy0RIj7RBG
z;gGN*l)_{oZ)L?=htnV|EPRr;;U^$p-acuQ!h|+9jRT`EFxe2RJ@Hr@Ngtq-Fr+9R
zQ}#X|<d}_0NbuI+Y9hd2`o<&13W$PVwbIM3yXZ~D`6ktz%KTZ`*`Pt7D`qq(T(P!T
zYy9aA@of=22GJNQqOJh_3*kNGj$~0z0AwF>@T6r7a~(=`;njG(cL5Ezj$w4r7&-IN
z%UMT}|K6WE2u7Q`e89h(iM4|QK{et~VKoZWt>Cg3M&@&9Ws4ab8`I{=_}!XcSJ+NE
zzjF&c7_9~lJ*~uVdFVr;Vc6Y$N6{nJM&8H8#E1;8sCuj@=aF|TF}ukQ!g&;xBSo){
zJwN{gi}h}t6umPh5*Gz%U_xj&Z%+sN@Rlmc`>Kw&?mn-m6fvWSC7|1m3g^CPsq?*-
zps@Od2MfmnmBqJ_2SMBSA&GK94Y!bPYVBTpV?P*@rw)d<_Bt{BQ?}(#s*)!yfqu1n
zm2BYhpHkefVU3pw;tGoEt|BSwda<7|=bu4)XTI;CLhRM+_Q%>7EngsdqFdB3bu*`o
zTuNM15h&O0b4Y`OYfA1dbodvz5X-nRENylck^J=YFd#z)u~8#KP0-a0?<S9G1M)U%
zNiilBklsIo?3pS9HVjZH;;lPg_j^PZU1}Z6gl~6*PDa1Bl5*TFNR<jaj=1Tgl4$5}
zf=WSs11B-p9%zmvDdw>?-kfatGy^t;azEg3w7{0uz!WC&aNK6DFiA885fQ7gay;Qv
zMTuC+&pbUh^-W)kuFj*#rk=HZ^4%){9hb;k;kzKVpiAYq9r`37ABIysb)^QdP*aM|
zZ7}2U1>{9KXUyi7fDErQth*A$rs$A`?`kV4EB7`uH1NQX{i{@awTwWIiAoxWVN!_L
zdc$I@tv*tGP#xDp#I?6GQ-8t++)!@0FqkD8&A`H$B9`xIzs+yb1$h3}N-_4sZ{gwL
zqOY%VfTX;cWb`A#lJl(u`^VeRgr8ec&>tV(yYV6B8rvDKC)6*$V6s`R*=k}2{&LOE
z6vLhmX7Bpvb~4b-4rktC=4H}y%Mnh6vbQdOb`XXtS73Xz^+i)Bzcvx?7pK2e`QPoZ
zK|GLAR#FKO3%>5H@@e$Z82qJRjdBEx8Q;Hyq&B;u;qnr+*97z`S5LVIlv|RUyJo*k
zON?(`+f()8IL6SWIOL@yX*e-;FQ*hu+H)K~-fFjDSfa)4iOzylWAq2C$JNbd^E%Kb
z<Nx6Lx5;oD++v>V8@c{<c|gIqLHz@FI-P}OF25tE9kASA3$hBi3d6UEI+z~3cYX9`
zE;i%$t0-j|ltD&>zDF?!BI%}&zkGEMtt;bt5`C@GN0iAcv^_|#Rlk=p7~B4`4*P4n
z^O_GTPX!nPAd*q;ZmqQGD<dvC1!3BgbSLRwY%a=z4UE*hAb}>;{?C!WRro$m^Xy96
zQ=SV?_A?j>6`8|1$k38Kt(0hwLES)8kGAv|-AZ#!O5`JaJCGw~Al1IeFslu`V%;*d
z(Oh|4^a?-=WRDc;IGSG=ed^T!4(K)}c-E)D_g<{MB6bF&x{-|$D5z=X1)OJ2a|l<s
zGjkXmczUPM4GW=_t7v#OG?s9?Cd2V%YXgd&b_i1-)kxXo7S>AVgyC(UPm~3S-BRB|
z?owhDFDQVa$XMh32R9)|ojPM2Y{@=f1U40r!(a6LCeG3bktKer`_NC$`sK-~yrG1B
zrop`)LXgjt<_0r`-G1<XF2mquwII%~lNi!F@(d`m75I;s$$<SZuG*KxIE_qcIv>)8
zlg`UX-En!@c53}N09y(v!DiC;68%>|s2ppCjmw<6(CrrTi>W#EraP@DImHM*DrG<*
z(CCp?UmVT}kW|pbt6x>DQgn}^m)nxcKctS6Mzwu!bUwc&g?sLY_IZ<$;$8QhY!)q<
z^%ALne%ogLKaOG$!cJqKI)o$1H<Y<)N*3vYkOEefmW^zjx>Zxjk!|h#h1d+L%p$5F
zLSa^}K4bh8!hSv?trSw=ju)y<3dUOmya3eAn0?f)@+U-!d#tAIeS0c`kiNH%$?HOu
z^C#_bhvtmM8=d(pp?`SD8bwk@AS#quLuc-&CXs5EU%Iy!!4*i>s({e=;@Ov74Q&BQ
zwe~BPfB0|i|FRRR^e6E}_>l32Q$U~?85oGFk1VB6ohA1sq@x=(M<|BQ^_I232otw(
z++~-xv$yB-l`tuu5xzYE=%`@NO4;v)#l^Hv*0dRZ7ssGFetCI$!B@32;$$u$AU@w}
zGMOm>5Z;lqb1N$=!@rA-F4B`<ZSlIT4bVkU@~zyjccr#J*_jcRFj+O#&l!#<=UfeI
z=~25GEcAUv@RUz=MD0`b-0<$rxzlLm2W*b&d=XiZ=9?X$-qq376--7<*QgsAiT~U@
zf<t)i6t_7e-905z`==Xy1R;RMX6pT-U^p|vC;qWfHy3z(j2{uuxY?Xlc3m~=m}}0D
z(U;=##s8p7cr=oM4M)Fy<?3Wk?OS1A>j`L4{MG!AyG?>sV$UBDOHTq2VH>6QW*|n?
z9DVN1ru?rKp!$LeiUrRY<?iNYI#z51ArBAT-ct$o%}T}W0kHaAa4;xv`5sEL;3ndr
ziZGh>*%e2JCvo!+f`1|a(4a?`P{h=>oV)uY$n<b+5af(i!Wx2f&4f#8H{>>ALrYdP
zuJ;c6QsMfS@9QNO7aGIz7iT=@Fd(61Rh1;6!4<><DYEf80=W)8spQpz#J+gU#+?GG
zYlXzN<Rl9JlJeO<F1@9q=pfSP2J)Q{-Y2fw^h9yiAlq9xIAcZgx&?9d<YPI1YYb%{
z)Q>&lM{+?nPEG+W>u!#lWB3LJPvWu}h2szO<wBn8_d^P~#0}`60Bt%Iq)5#kt3o<J
zR_hKm!66VBMp{{0-^{EiqzWIYbSFzbsW#TDQ_q*YkMOy;6@vaTp$=<_RcCQ!_ZmMw
z{wk&5grXV48F<|db7EU;lf7}!cRiMHUJ(LX$19VxgUy?=@8w)*Qo;~O1+Ji-3gG&4
zH{6KN%=R{1p%mdd4fLARhkE3sA;m_o@--R*YM4c*++AAbMfSHu4B+n+L4W#47%5x|
z)UZ`2tF9UwpkYu3FI0-6fw2&{<Hf<~;^HC!+Y1GASsx^f^124vQy?oV@K|A(taG>b
z(XG$RTc^tbSpJ;%=H~tQguI8#7wV0$7KEGX$w`xRYURgE8~=&AxYM7@rW!!(=-wpc
zGV%ON$AI&X?tL$@FK|8bhVxb6VVx!^*z8bqp91LwLv}4qh=4NIagG#BZ5<ue4Ld=8
z{sxk-&CMxT(0z^v8qch-(^~GC{%FaecCVjNuA9uZE&kct>mYO%SLOiv7FyJ2XYE~F
z?dbUyf~XpJa4|@Vp2#Mw<9(DmAeAeoMrT4@ngNyHJTzKdTx=ZBNooCb0>xA1Hly;k
zjJQP<l>4>}o}c4(Q3eSq(-Tte`E?j+3XN%3*?uvi#Ckv$B{9}ViDzeLS5Z)k4uSK5
z$z70-{ZjX<G!XPa>L9HuyjwH-I*BP*2->=h-w>x&^|k|r_cDeFweLW{bQ$HB-0#+X
zqyrPms+NE!SwR5>^;T9oq+f}NY2=laeob~@NLH&%6frj6ZtVr*nor37-qTJQ=VsG%
zf8=Ss>uutWJT1r#oo972>1_ym_CI*&0n-G-i41QBol%!<`R;587H~~(;ftSx0+2ub
zGA5_07ac)J@pNrl>r=+vh!7iRqW_Y?Wau9rMB=uh@(ymYvHPV<`(=Sg(>@zxSY4g$
zTa!17iD7-UPgDooW2+_p)DQkd;Y5hgcPPIShJDmNC5P5ckZvxTzuy^OON<M6O!z%-
z!hwFZc+%$L`y7UU^!cy;c4q_F`ss1{0Y4`+TPBQ4tf8{RXpcZ>Q~@AOS$3Jb-W36K
zxCU0b^~J?IT_CoqcAvZU@6A~!F}c)IxbV$cE@r#ey#MX@9`pMeHE>XeElUU{%n06I
zFO&VwnFPjb@IAq(zGgluX{h&x5z?(kg-scyC@7Epyc1qCyF)Yx0`!Qg`#p7s<J%6x
z?WXap?;Wqaze4;I*d|sxm!-k)>9vdCEg(ajFmd_ukw0pa^dVXpHCa*c-ICQh8nSZ7
z<hdPRerI>l>})gz+Lk9|_fEXX$!OZjKSaI}B}cge7FIwa4*&w_UEanYXa!<N=Y>ms
z2q{D>tpK8+PswK?><{vous>MzORaS69!=;%Bat`o!#Bgmw<o%^G<lWj12j5vi&H~!
zpvIofzq^I?O>3Q7<N{3&76=jSiBDCJygf;)U<%m#*%=V}@1PGX!#B~!Rjo9EULeU~
z#cmOU!eNH!rld##hz{+P4T%GNh7uxcXE-1UysDgqQ`@_JSf!7~X$j}0_U1~^YA{>!
zpkw%{^!N@ovBrCp$cmJld^OL`oiI^VW7GgP{mjtN5Oi*;ai)<cCzfreLoC0nmqE0j
z?k@#cC#>K>dtPE=AY9)SuTqm&okgDsvX)hAcvc<;q-V=KccQL{!>I1WK4%H<{@vs=
zWV)zt==L_Z7*VqO?GQKi?T6V-T<#yQ%ns@`@a4@IlubutP4A0HauIG<LG;m+>A&F+
z_RDLwj5h-C76T}l?%==@`V{oV2^vRFTZ;orXHVShWE&G{026app|f6#3JqB#1X8=s
z1!o(2B0bFclGx`CRKhNny+c{bS3sI0gRm^lmq1Gs5*NR+hWxEu4Kp6AAqb2-CZ11i
z_YvbeR@bK&h<hKdC8N5|7Sq-_lo&l_zIXoq@H)7uw^ZHbn2=VFI6iaXVA8%sSW8oy
zW?p`O3$-dWu5i!&BOj68Nf~0Pk&0~zlM}$<QQj*}j|=M4x4_(UhYPb!#{;DOJ>*5x
zMefP8KmbXuFRU8srG(suqvU~SfB#fMN)cDDWMyS>k>!hvi>y!y>~tNp+n`%!H)ag0
znto2zGd-quqrkPP3vqW@za_zKxfgiUp7h)xe@j^*=!1U)=dhNhTX?zx)}CUT%$)OQ
zCp+ryCZNwcTB9x+IeE#Twc)9fO7OzVg1(E<r&dX@sL1H)Db=m(eyMv6@JCEh;UVn7
zcv#LlrA8R@oRG)Dw>*j@pjL?^NSkMlJR*VaLrvJG7)Yjh^H@ZTl-_(+Kjj1S@0;%k
zXy6bN_>i}l0Zd8*?$TJga39hYh*j}CLnj&iYp+RP%4u;XX}8@c7u3Cg#%p{^vUe>S
z4&(Z5`c^f`27pI<3aUC6LFI0VJQc}5->Z8Y+>OIVV$2GC{`O+HaaLA#l=0jZl98@0
z!q{FnzyrXJ;<M`GM{7y0)%5}pz~6J)og2Qvwy@j%0e(Y(+i)m+ak48&9r0By{Ep~1
zB4P}3`b9<rsYm%@Jl}_(x`D^PQv?VDfBFFwE4;#TNDRPogEutF#_VXp2<DQxWR8_v
zX?j)rQJvB20qg41_+>l0{SKtJLfZFT_#fUTp0JTbf<lazw?a)VRH`=wcRWMc17Awq
zJx>D6RdGpYgNd=RL|RD>2*NwcgF3XdJ~TI99vFw!&otWwJWsX~CmQM-#aU0}iN$H=
z;JcOi#ld-%duoqw?Pd4{wY+Gmd2_Fyzn==vZAAnr<~HPrU6qQqrv}YoI1?kMdFeJg
zcy}X1*oy(|Nu#Z)`7jgc5-1_{(UO#3O~5&^QUw$L{W(M`0jk?brg!=RBN?Mt_PzXp
z$PY)vYj6IHHWCZ^7#}a(zO>K$9j=`3?X5e-`+iE?=mtATkDkbRYrG5%`WO?l95y;;
z27-Y6d})NH`6CNG@r$nTGBa&|lNnMZMQ7uVk?3W#Xkn4bh4J;%yE@GyH=q%RPt58~
zBX#A?L8Rohmz|at$-fXz5a3p}gCZUQet&<keE~)T4MIr(Xd|uTdZW%uYWFS9jvDFg
zdRws@k0~Er8#SKH-4(lu|6AK-q+Q=XbBPw&HB6fqSC#>qd2fP-(r^fj^!vz;gUILl
zouEY)E~KEO^z#Go_G$U69`(;+*iSA1Ng?#LsOZ2=(&|@S2|&2)0Kzqq`sE9G1P9*M
z9yM|s+QD$<MD8?u+QBaGv@8jo<-rJ(mbZZ=xr^I?4zY7^JOFKMzI=5xi<>LaD>Lu;
z0iZ-~^-+^xHX8glw*FNf;bs&vXI|eUVq#);wmG$rl&F@*+sX^+z%D{}5fW6mb_pwc
zdVVx_g9D}Lp>|%>C<-Y$Jh?fCsaH~;E-$6ps-_;zs~$MQs~on*9vqJFrIQ<ZVDWp~
zw}1ysD`-hfO1jkkuvm6ca3KsD0vqS#k4;dpF(-(S*4#YkrYd=SuLtH-O=<O4UaF+w
z&NJtZ38JN~;x40xM=izo6lBCdreLdxQR&R?4ah}w<v9+&X5iphKh%@FiOY_s+oWW&
zjLq*W7e(*`&FR<wu9i$8B3~@;wUS%eqw2<8dh8Ba4AkyTlEZw-T4Y*#9Hi#vUJC`b
z{E%e^R4hogW4ZLrUrbeP>qEMs-Q@rp^KH(^Xxq^PkpeM;dpbVt7ZJ5)>rw4@Xr~+E
zALnMVPJK*Lp9~KLPBjs_T@WELv4`LE-3vArAk<t977&0a^`&%nR_S&-y}J*T8E)#&
zYK*#T$=mAo3F!IQ!dgQ^7`Y)+r@g;-k-M-{81fzLQcwhUB1!&Bhz*3ohg%nSOki9(
zqkq9Wp#nE|JEA1KnSrtTt(hdh)IZcL00#Ax-$g8o0-!bxNUGsbHJTI%jKdkDiBfZO
zH%=fGOdux=u+-Nn^xb}7{&Y03uY{&;_axvR<H-bdo_y?caZGw6Z36@dUJeim{DN01
zfm#<c0XJQ0(w2VBgw7lBrFZ-s8Z@pC1qX)s|JGbRU-4xZoU0sh!7mB*cvYR1^~Va*
zg-{o5r=+A5&7e+K1FKaRO^^D8WYnim-&$7`(3s%M1VPKRlI?GN8~AV|uQso2?6+SL
z{Petr9b8p!9u{BF4)OIBeQ<kP50U{xC=!zL@wupH{7aR{L?f?_52?YCbGwr4jFM%-
znon~#WT72AV?S9$>{LA4Gobu!{xDAp2SvaqS(%x>cCO4or}b8#*@wDppjK$Iy>)Xe
z&GjcHBO?wV4Y3OiUY-D#1Zt*?s)0ha)VH$ZkBc8PALP6hj+u?pI-=t?<WDAi`QrJ|
z<W&amvu{=(SQx>uf@Xa;tUoQ5JnBniN?WLIp&NVD8rHvvAbnZVc$~n;!je_cVnALH
z>HhrrpLS^Ie$Ph8==+6bBSRswmKi{0-sl}nP5by?JLYx`0%w@t)BZVjUQ^=-M#UuW
z;Zb21r$F}-S(WduC+J-@3@|xb5x2EnP=tD3Pa~|FSzhi!@fyRpWUL<K7C(YEg!7PI
z$~e~q-Pq(5LlMU>E_bfq);>?n#yuAg5i15ADy<4b#Na5N4$ETyQtLmxi94BB<ahH<
z1yI&%OA=KK-<|p|lL!ZBC9RNins3`1Y;AxrdTIf!n-^MRYL{kx<vV#XV&Xx1z&X4Y
z$*!rdD=T~13JTFKJyuc4Kn-$V2yb&LS%&K(Kk>&3i1P4lVhY@@jg2T9h+t3y3kErw
z%PYxxa-bvYL#$_4664E)W>A#SZ}>Q-DK_%v@oJkLP7C%t9;94<6d?NZ4A92-Aj2T_
z1Ji$ajccbE<(tKTV+vu{SZ=6UprH{yuI+2)QR7v7Z5OWRvT8~rt|*`pPom^0L#!FV
zm_fl$CWQeB0zN?<a#-V9GbY^{zBT{cV2GF)9}Z-PagH4`RIoxSaM~oGOgnDE!9zV@
zrsVl(s{66@;j?<}kq`>qKj2!)xtL#D)p%Jmuk*h$#k>h&9}<9S(4>^<uI!+sxk!J2
zD&eXQel$h(7$r{bFoJ3h2hwGFL5wDT{0PtDFK0=^U*nNB5L3&BCaB^9RzM|@AV%_l
zcGmCP3^WAW(*3MkL}0XWZ*JViQ}yoq!)|Q(2s-7KRR-dY0thJy^TKI?Lt|)h*z_bV
z#%`(;dvKAQhl^=Uj2WLmh;Pp4{90o=_PMOjODB*g<PHrk1=jCOfnH0F)uOm8I1}*s
z`m60C#zZEH;1Kq^lLPwQehlBIw%bD`iG2zpw-E^JAV4W-fwBvUQ{|5UUFyDbzx1b;
z88td@c}Wwz1})*x{68_P1~sS^K;=vbjfdGUlndCKBK)~#tsghVn^YBZ;FO`DDU@dC
zpDdbGR=7u?=xMk6fbq&0nJuQXjPXS0pTU0;N5p=}KQNCmgT04ysz!`so{EQ25?=c3
z-3PnY_|?bc(|9)(@vDDg9dO)>|M81T>PbQWFxqZR%AOky+<9Hd{|}&O^MG{JnIf3#
zM8eWZ=wD_^Mcw-UEzAD@r$xVpfE>U&f5WZbmAM>XSv)l}j-Yoqg$&RO2m;CTV=Myd
znsF2kH8eb&!!lsus#JY!UOi>xU=Yo|e{mI&Eb0jdwLQN=N4va1c^*gX^H$#X%X+;O
zupCeO)Og^<fwNB-85)8^6cWCtqd;ty)EF9Z@`Eka&}+FH(3u_{$61cmQx(FR6*YBt
z^Ut3H)wh*(oQ;j0-OrhC)`Sqfu<$E4Kpfrg0kqmU400Bt)-MCxIuIC1A?k`%6=N>0
zfTr@jlamt`^qyAyT%>x+udFw!?kRx#asDr%IHTUQksUyYKxLL&1(1HvFA?);YH1B>
z&jT4lBWBAZVo^Q6`=tdqL+V(WSu$)Y@W1kiEoTO=fgNbjG$C~~3-E<<400y#i~_!g
zE8H{Zw|Mh}&=|9EWaOzEnU|K9R<pHh5-mRELv$mI*@}t^9<<Z)#?TBk?EojwU>gNh
zRf$Otkz{KI%znw4uq5ypOhiGHOs%{UC$jyoT6@w~-9*66xQiqcj#KB<9|30wU=3q1
zxrn2_d^uTuN9TV^j(Sh1eF$L<I8dNeT6R;@b&Q0gw1&&xd}|^kh7G+Vq%)3=i8*vy
zkdRy9Ywi{B(7BT1+GK^~Z$!u-q9v!mh6PW$vumD{cNaBm(VssbF#=I8(Jk~ouszWw
zeL`2YjDtfFNel5mz^uBd$v>et%6s|bLXs@10dQWR)Le3UM{Mw8a<WloehC^fU*gqr
zcT_!!Rsk}DYG?-?+30i9M+113yTFg-)YYBSLd_5o5Uze&h?H{R$}FEhcYyPfL6`mk
z{$SXGdZQS3=39eupi>9}YFXowl9I?;#zg?dszANa04Nvo2a0>kL?&5TIXSsdGl~EV
zi1<v_$Y`d28(bEYC-uW>?m$5C$mejnx3D`&jScadwW0stP%#8EL~jaSzK*u`pG^;K
zHOZ?(kEynSRp2r@=Nu3pC8r~wv{#a^oKq{>!;<pWbY`|Qev{Me1SRqX99Fhnl8@~Y
ztd^;L-}W#tav0_py}clO2GU3d78VRBE&suV2w*M6&fi%STH#eEB-}Nrj_L_S)Xss1
z2;%4U6XjecT(i4Be<}s&iW1#=b45*A02bKbm>B@;d>E&9K~b!mTO)z?XjsRE;eSpG
z7P5WdRWmv=LhVPFmyhil@$Ul09p!~@yh;F15`=~lHQcrIjmVUQ36ZR&*HC35d-&m5
zb#I!nP-Y2y7AbE^*%<F4?0}vDeZJ1$Bh~Q2@$pOMga2yNh&3!xmWw|wE)IB+2h5o$
zZ4>H4YJdX#2Tjd?P$=BmbY|~FaN`lbHSc%e(|VZTD=4#kPf8w^tGmXEAE#DlJn7v1
zPc;XQOB8wyqB!0Mg#$W9<dla66(k*KYCl-Wt@tlI9`9L8kpod@3w5BFW!!IhAmVQ_
z^OQqbFvhSaz=UC&fe>I*8_3gyGk$=~#=UZCa?%jMNtGTeWkDgno8>*+KC?j7#~l0O
z_pFZ2W?<Xo?OJ}s2;J;vc51<0J>183JVwD%&uzCeq<f93*7pIif;xLfTo;^&1zo!r
z{iB6vgpKCami>9`b4rEM3qmMA;RRTmJ$<%5jaA%#m|&`#-@au$n25T30UUq^4=GWJ
znK*%hQ^!oPKXU%TN!xmfOC5-*z-J=@6y!1Mn)n^@zH5KUfP9r-M<j)b90l6igBN{w
z@Qu-LbZCDezVkof!v(Z5&?WOvCUYoh!tlyO-xyi03z!)FI4sdGHI}Da^?Ky&+3C6|
zrV%EG5YKHaUXu<q9?)%`Z@DBDDlCCcpTDs{AymW~)MkN#FfwxTehl?Nzc*}9h=Q~j
z9l~<t`4^+#R-O?G1$lYGMhJ;p#zxyNwP6nUq%2_kBw>5F^{(R_^)K3~W@X<VXz$@6
zw=Al|GmHYARzEv<{|xp*jC`QbSs4855OB_k^Plz%Odle0ePvKbc?QLVcHf6o@qrj6
z<1w@KMD}N^=p@#bM(CPA;WM-eiRH!yz*qN~!^om3&2I|3m;dQvGQ2($F$t_67=w+k
zv>_dLBNNcE#Wwc&L|jM+Bg5vz2X}~;z&L~*Gj!3E9hXV^Q0jZzzlo_~_&-%Ji4-@f
zjiQl}8BKJO*XO@h2(Wcu&%ez`Lwb#U`k6j@4;@3c2yR#Od(v*ZA8<I$e;q1EXMg)o
zDHUPq-q(=2y>A(WV<&E>&HZ?9o~S%DNmx)cb&f7oznr=Lhw-<1#o7KYi}?7N{$CBA
z<@H-lt9-kWe;aO<kp}Db*WY<hBz=qF4N-=2G<pf@M?}IQN5BzDQ5^Ymw%Pt;Qp;XX
zK}PTosIPRHu#HhahVcL8w+<lGL!&ufE*g$a@m^l)slTF6|92+okZ}ck;{}}5enD7`
z_{>Z6B9mr+1yq8~Bp=~ByL#Y$k<qS{=Uen7U&`JuyMOf{TR1ZYXtH@Qf$kaNb)Ijb
zyYOwYAEwgq0cp@YAYx!3!#xB4DJs~@ko7b7;~5*EvHT_(6{Do+O^fuuVIqMhk0JhC
z!m!cXoo=eCjH^I5yEE>^v(16lpRAb&uK;~Eqt%my<*8RNE1qi<^!VNG;IN}{PNdi|
zY%Y~#Y7Df{4BnZpB|)|ZVcm4D9{9zN-}&SOdN{nh@Rfit@$b65>t9WbUHPc_CvF}H
z;c^epeDAO0v62Bh{`Yr@38B>E$E|tsr)OU)b-?x%aFib0nYD8*8rDi#bc6?LCPv*b
zKiy09{rlr3I4jybXkp^}ef3~8OPCpo(WC=8;B@Sc5ExraZpA&EYHGaq%AQ1zm1vyz
zH{;woqLf;fxsjUrl9Y!h&RU?&)!L=Asn>Z1HV?mCWEUSgMRb$R(8{43)<>M!=WqMS
zqtkcW=dCj}?}g)icrY<}+3Z~Es+^i(UUO!>##3O{8(^k;pY%|g&Yg?e-|c}?a_svV
z0DKIbkwlmETH2E=m9><uhqH>uXo@q&-#M2!^b))3-s=YnZ=CKuikY++5x5XzqeiA&
z;!X9ZzqN^Y&+OG!i@?nvT1L;4Nr$A+6G=7J!>=z`M_$N3xBf~fisQH9MIpvNGl<PV
zT{j0&`1#kh|2OfZ-59_z#Xz-j@U6T&Hd{0z4FqUkoV{HH`1z@|_4LMLRz4b9PaQ`o
z<&7nEeEF<7md%&)G!N_^fN@GJ1zOBGfcC*Y_j(yBmk!RjF=)|&0Qc$$2}8SHUb%Kj
zNpi0V_=9_smh}fY`)lLVQLb*P%<4(P?pf=vR7?V@4NpFT`3FTuV^GC!P{BR9_V;=S
z@Gjg~<LCn;BNeI~?#2l+IGjYc8VC;w(cwNZnoL79YGFymmHw0QsU<OWd_{wq-9TXx
z05ilY1uSX-sPHs-;;OI>^cd54m*}P-T$J1zE1CObOTfYMV#+u_Mub+1;}P|EB6u!c
z(2<ILj}=vE$ElQ8Fifo}>fbzv=8HKnJ>Bq_W0c8z57ap;f+Ft@{mUTD4NqHqc3H3@
z1aWI_rGYI68)OKCEsYXn;4Ql+f;pm?V0Fenx!ur429(H!Xb&^2$dF^3GpRMGeX^=v
zE_J$JJhlV9lyRl+P$?@ZRb}6gBmL{$==1sj1Q()w`>()$doQf_%0~d&Anw>%<M&q|
z?P;v{Z6UGgrMv3*7sW%(4dMgMmq+$b-&u&UfQ7OhFvpSQVYxWRIuXlll#ux43l{SC
zWNKdj#O4v^X7p72U5j#189VmIkk)uiE5#GYY_4w-QTviT+<)cMI6Kq&s(CIoDs{;b
z=zA_mgp~3et}agIhBYpBOee~p>$jem0ebA`_vRze<z5VEs-f|!nds05b?J3=A{7!R
zvnhWY8jYP~oc#gEHOiLcEFI}J+eUV$^lmPzZ)giw8eSYh670}@vI$L$upz|Nh2u?I
z5@`=i6?U6amo?=od`|~%0wX~4K-e6>v%mM7&W8&`7GXrdl>%qJXMf!eqm{eeAbo4_
z!*icgGUDiZ(<T<U&B!57PDF~@S=g!;N&J7)c;@%#xVwKBidEQQv+#%)t~U#P4q@NK
zsBdk{76(ot#{v~w7KJRrJ#QErxajLY2~5ybgj!vBRdIS#fc4IR5Y+ZLTJPX5SRbs?
zUE4M78^`@;bE6*g8J5yZa@L=56&*1=`__YVI_I_h*XewSMi-rSp0EXTzEL2Oqa|Fi
z*#Fiv`X#xI^Y2YG>PSVrcbn-Ktfd%QI}VC(1ZWi9Zlpu-SWxDBExT|f>C=m?-2Sym
z=j_CNl=YXsMI0W5duREu$Fa^}>cfqiz4om%(zU`D;@%5M0UW(IKHEd<@gBD4vMtbj
zeFaZ|03HPfzeSVJyVyrBV0OINdQNW~>{XJ_CtAGb|4>*cyWKEKEDh5j*RBiQu;?{+
zFAw^8xHviO31(kR`NM$Sf4p`@!lv55I8%nhu$c3iJL2VMy{Z46c_FMUQapA{uwegH
z(HZ4WW=|RRd8HyR>%-xL;7q0_g6UP2c6<?uk@zOLON-xPODek`EM^eLPG+~;Y4FcI
zKOLNQn9&|Ux~9bxmqQabSS=+lF%glb%zJDqwIP_lpw63j>$cQCUbcipKLWji=pjYK
z<EJJ`RroFU=4sPR-W=N}*Z5d$EL_u!V{Wxl@0PsLU*+r*$r0``uoai+ZF%#tibv=K
ze}mr2eB86bT$~^S_Y=BP|7RMPZ*4OrG0|i1h>YTJ(5t==&A#q+>|jI<F9U+>nYZ-}
zp`9UHFM-^Inwn8*%ctlx5y4x-G9JR}bSwLQeQu=ts}{$mm@=X?YjjI>p*zjr5=v-S
zUy`j{>mu0#ae0I4$b?=_!wS>=h6dZ1J#=B!P#5$n?(c0uv|Ti;N_}})b9+qcX~eJq
z;V0&lP|A?45#E<roAatz5|~YI&kXLEZO}AhOw(F54p<QT?zAS_4Htjo<g}RalwbV!
z6u0rk2S7JCJ^2elLog+Wtw*SK18UE=99mj6iqgFLW<vfw%bT@n-P9C5g?3L38@HTh
zBem%r+Y~&ovD~=G%Lu-rkiFA@<aKDizlDeQN0r|FHs&?C;mNOpcYjO<6gFWdeNbFB
z4py|Rf`nU$=DjDse5)b(H15#JES?HGHvg+956xxzCC?imZ&tTfy?-4zv@t7}KvrhA
zd$9%+wg`@hva9x=w0Nj!9^RWLp093s_$+c^Dz4$@R9tgME3*fe#e}EzZPepL(<Xq0
z`y?-Z9EWB*%iu%4>`mV&^1mktd9CA1(;dcGYd>kA;v$@Blf0v%EAQn?&U%KKZT&~!
zSDNiN-X9IE8at~p*4}dO_gU<3%5Y*SYp}HlS|b(L1cZAe(LI7IcJVKN#!KSI1+CW>
zCc6tlp0ISZ?V~n4N8ND6nkH<1k<G2v7>QGY?A1MO;zdJzl}={6N7;p~l1Vi$tL9>7
ztnQY09n+{jj0jqLJ*~92@@_hMLI=ziyz>a(D>M9q^ec{iB3MPjpy$G7YSumOe`}S2
zad3^v+<>9;ct^M=-s7<Gygx4U&qM1oqIpbH+C2`|K~ER9i?i?XoeA?#eqURVZB0yS
z%uIvs=EK$xvM4eh5PVwlhCWf_ui9GQU4PSJ^hRw&Vn?NG(Rk;c{)KAiOCqC>RM@jJ
zyp>iPwtfsA%RCI*Ga?4eZ-~UB6GCSKmVd6}xPQS3IO`QL*sr`+D`HoB<@{gl_a<F^
zVPnjrRQmn)jO%N@qV3F?sW@gUCl$@pi<X`~3nrxJgU~t!yQKBd{kQRZ*A|sT!w1Lx
zaj-6KGy*0U)q8QK{vvHl3+O)BbNzuq$!}+0hprG4R}tr-OOseh2_XX?d?yXJ+!m$m
zdhGVtk=aW|UfEYD?D*zZd}{Qv9*ASRwD`&{K9|%;N*&V|9kHZ2jVbvN9_nAo6;tci
znLgL9vmpI<tBl>!%{0R}0%et+YAJk2@)Q1qdze<dWhF3{isQojY%YDQHCGuz<|>`n
z7Dp^rX(hrfXMI(YSR1@q6qpsZCL1cA-%BG+{|+y<wDw6mwYHuz;{U9!4%&8d&2BUg
z;kff-((}L9N@3|B-R;YdktTU7zBy!VGYnYVkXGd1KNmKlw}K9*$o}0AW#@*^c)@Ys
zeJW%+P316q(LkV7tUse{x_dV9r{&%V*Q;lyBCP3SbQI_>@Q1l;t|FpeJMKtEN8AoJ
z2&8QJ&{7tc(k=+vXY2qjG*NmH`5WQH|N4<H58Jhga3imKSUWbW>gM9KNB(j0`z881
zdO`2K6w-&yqG8ABbPD^%#`OQK3=rKrnx56XynpAMm7b{f^AJvv`t(I-$FQWTzB3K(
z@9^!^76EB20?v?u?C<e%!+G5L7lbrJOq$lM8X7N}iAhdOc+b8X@P;nb%S>K&#T%;}
z&bnulSm@;xDLi`ZYSXLhpk{iJH~v3^qp)|3A0KH&+B!fO+C*%C{oe!!?@`{%{U=_D
zoJnzIgu00?=fFvcSHKdFCYAoXS?BD~_dKMNwmads93q^vZ&Xbs%XUmGFinH@Dh+(l
z#=e&|o7UEtF`P6mtFWQ9f1U0N?{mX+eoFlknpQp)n-EM|SNe6Tk^Z{Tk>vgiQGZjr
z;{VS1jgM#3?33?|>p%fs-G81PydCe4pS7&GD%9nJnoKxn+ea*~ZXJGAa{U>`!M9_+
z^2<gfHtVNBAHSyBZ@$1eG>Zp+u%vz3j@y~}gC04P6ZwB!9@y>zOVRBnQe;@TWudR(
z?YiGNYm~CqC27rFA%4r$*l!>7`~E63Y<t3PmVPzk8!Fj5nG>=GUh@%fEtKMF?Z8yj
zHhP3h<zCXV>R(!@yb5tvg|EDQA&B+K)Nfk<`f;55>&Dy?hU+2f{T<OHOKPk^3=Z<D
zGoD|#w|?D<rgIy(qq6<Evd<TT_s82fL>RzZlHTPNeIiEf*z51=>gt39C=@q{^Jxq>
zhQ)1)d|>1v;U?@^I+-BWo&bcv9w>8Nf-niFPk)2ZT0+@V#Q}tMgtI{RYzu*i&BVQ%
z;a%bOi{u^P+y$IVznwN{-kW&T{xHH}>UX|&0XIhXj~^3$TV-8M;}at5rJx6<VwJ<{
zv)Dh9J2gvpazs;}4m|r}?mrE^&{S%idVJux^sQe))E3AK<<Y1;#?4AlQEzOkvZguc
zQP2f~YidGNs_2P{N3o|^XQ6~qbo8pixVK7=gKQfTH_`!HFtnZ@ZL&e2*TpUcU|R~-
zBl7M~Re?Q13b@gLRBx8g_jHeZh(_z)c$HtA<^~>=HFEcv&$hcgm7t^8H=yxRM}CbE
z+309%YqL=zs8h9Xtg5IuOmXXC$Z)w;iZMLp;uhD{H8g2a$+Qb!C93UOnQF<4YyII%
zJo)C++Vw%;hq%dxh2O96Yw$m@6I`hZQ)x{JMn^t)eoNZ^*Rz)PmTD0_CweS`@7c&X
zsM@oXAP)A&m_UCPqQ<`@I(2yCz60~8UAZQvKBc5=G3=F~^!i=fkNdv;@{ZjUr)ZR)
z{G?+8u!3fec;!_ef1~gABlD6gZWOyO`G^fdxn=E>)lLbo+_3$M2uGe-59u3MFJoE9
zXx2X4R5|yl0eyioKAuTD+gx$$B#F?>S96{Pe@XdUo$mhB45_)wjHu48BxuxR<<z?Q
zt(0e)bWh7dMWD#tDeE|Om3xh&-+)7#;^ACLuRTALUaU5ClCILx4o3i$wD%Qz7QxUn
z&x=okZw$A;w9aWC2dJ}WJv??5n@HMKZ|yJ^Uwc)3ReU?|+E&T7s6f<-<>7Z7QyOOh
zb%l$^WyY|GR^~r_SgNhetNg4#&hG6ovufO37qoxCT6@7=#^U*@>;_YX<83|CHtx+a
zR-AEm0E-J-3DtvS7*7XtJYm=Su0sOWvkRWI?Og#XzFSjs$X)a3sMg1G#^0F(rc!c$
z$063+`|3=(m`yhWBY7J?rDj*EIv7Mk{T6?plAC%rc|Z6=q9dngji1f76&YA!uKL?M
zq*3O%*hK{M;Q!doJEn4Cc-O05xO)NAdiH{An9u5$E*`vXJE<Z1w6@S-yn7)e+w)oT
z3~VQ8ui~muTmVG2`7zISxpY8Y6`S=ajdb6;Q4+S}dZ&a^(*JS`yn;T^M?RY~%L<J>
zYll|QaoR7rXqX1jr7Ti=?p`l!ZP(@3-?NeZPl`2mE3NW+4{V+If!|B|*{jA`yI+4y
z>pLUq=9o4Xz&>HAUt_#MzUX6XYujn)z6f~F{C5?<=g>+WE-pa_>)our--?e)#(dcJ
zkl(znfKG~TJJ-*1;o$6bV;Lyv^Srsfs=xAFMD2U6qvW^2)g<J9d5SwksV0PbhieaD
z6&1*J*;hNYtHA}L6WdR9bxqP^yodBJU#YE&Zyb52wlSvxzNIS?XoTDO-!a!^D<frF
z$l@Mi{qT6iOYT~^a=`tfnGNSSV;jv9(eKA#{ztClgZF&?$3O3Q-EE3HlW0UJwd9wd
zJo{7+jWPT%`d_CY?Pf$Kg`YjFSS~usG5@9F*c|xImgL*`rhPgyn3se45@fEy9MPf=
zIx^q>pT@f`AF`4v6pjo0cKzwQLuxRG6YuA47HSa8d`xs7qeTxkxBDwsPNa47N07Vo
z|Gd(zsCzf)0%pZFs(0Szxz-0q9B@g8&~_KmHaw=Yu%q~67Uo3Q=F9GYC4e^5M34_~
z8<8RneEV#Z?9!64T&+Mjek%J&v=nXtyNFo?-qwHTMI6FgG!$pEJ7XG1r9(WXRrfF-
z!KsoHfzQQDAM^W``|`<w<YTn~29`F`_mFtor;aE1pfw;@z;eTLTRg2>4AJ>7x8Iuu
zN}0aTD!ffPs^@hOnS7vFw{#mgDa13pC!UZr^F8f<ART4r;j}{vuF6`PK3!4nc=hqd
zun&GJ8<Qd~LtVu=VTqWEzJ5TZgU@SCoDTw5>2URC?`N;**)Hm3W#|sFL{<O0WAg<o
zd=%tIS?J$e@z7K=2fy-X^^48NCrsdyC;a}-qeNW&QzCyAi}F&kSE`STnmkOE&`AD^
zupWmV&eOmyN>+_%7OuStcN#*$gCaO;xX*uk8@5l{M|RU)pXL-}fWVS_&^vpy)M`oc
zG2MunOjg9+d;s;Sw-Wq+&V+qiaqpVE+haxh&Baobf=kU*JSdlW{A^Pc`J+WDYoxx)
zL9*OwuIa4j3y)K%bs{RYl;(Jp^T}!z>`>D{*?;_&$Vg5lh<`1%1xRyu-hiZRC(od9
z-#&IIM|N1}D|haN^LgIeuPY5p9)|z!UYa>5#-`g8oJmZ}zLBhkh5}A%Q`DRv<DCKe
zOZ1kPm%sL^Du^lPRXhz5EUILle7ko2*W9kityuXb2=EHHvUuFSH4?376m(jx_i7ys
z)4W4X$sx`t+J-KorfP@G3t2qwn)%oMN_n~vp>u$-@vI+c817EOsv<O0UMTx)Qwu5F
zlC{4hV}F$l%2=COIIRE#`&BULE|I%ymOw29fNpEn2jZR;P)y-QNqZlSBw@NZp*KxH
zY^7KQ{znztIE@ry%|GdcFAm(g*g<#~GNQ!XdZGCKB%A^${ilQ16)yls^~b37BGvDB
z;)zy{s(4OW*&vomE!aU6lgFs;@SE4MnSN|?p%)lG`Syf@zQIXvNQfnPv}!rhCZ^{z
z_;Ylyc-f)EKxtQ{=^KTX*^?YX8zhd6u8U=w+zk5hVcnzQVs`2WCleLu|8Ak;<8AzS
zMYEp|(<+-vwu4YD(gT}e0Da`tqDmmj;4Cn3Kf-@zpxr__SfQH+i}~%zsZHzrNQmK$
zbNJ%X_~hpJByJSE&hAAln*k5br)hlcTz<8&qLi1KKmW}Z#3aj5kZz6T&lnE0EB^qP
z6rj)QoE<8D_p@g?4n9+Z(<;~^r9-&${Ch{O(?(4m=?xxR4yUPKGB|y2m?*|%dnj8x
zw%Pb-=5b~l*gz&?V>pKp>h10AI*M7vPx`n64W-i?@YvX_z|^lYY_4DVdo*ViW3eFK
z-^HR9z%S|#%djohJl+KxLUe(7EgkZU9(xuKgez{P&-K}}s(|vWfyFQpuv{Mkr9i~-
z1AA+u6HKN=<U$L<ymryiSHNVH%pXpMf8kRw=o~RG&d;BeG<o{e(zb}N@Wi{;PVb<N
zW5A6d_nO^fT&K8tFRr;H%rhJ)UefzDp3k~^bb(z5KmT6&QxNmBdll?0M|M7|p#L7<
z-f9|alBAu=_0#K{A5;>nsl4*G$>TnC+1dXGTYnuE)%!+`qcfzmgmfq%AuZjAv>+wj
z($dlmN~eS{(jnd5Fi1;FcPR)%O80sAe9wE%_5QBkpUgGG-uu~4-1oZIy_UuOl|m7=
z4-LZ#g+r?eFCX8&(a7W$<v+4-cp|V~6*W+N&A7U{T3!D(QnrBQCnmY+C)0c)*dEKv
zZI9HJmEcxy?rFdJSNyW>PF*1u6q$$bJO``2x-EZ$z3T%i(&qXXMPJc&U|KnyAzJoC
z_!pUJ*F4#b)GdwycfHB*>F47R@oK>g-{8AUCA=$YvMVaKiRDlQX66dUpvSa1QS$I`
z%?GAT^YoRV@dNX8miqwcwC}I{`lW_k>3*=hD;K`%2v$ftbm$;!;2$W<8aDb|{_Dk^
zcj@0cUV;ooxZj!q7jEQE`a!2Ys!+9W$G;ko#6F(g5XN@+UhYVYIHp8YNa(dqJl3Hn
zFEPiWzAs=}3ah^wi|1xFeGj01+m&{zfY*4Ui6Gp~7tA?x+g}_2B?Ablfa6opDZYO&
zSzb`NX`dtJp8KzerH!QMcLHe{Z;9r&V;<1MhfTr9I*cFOU@=VoNuGGznbN5npXkYx
zJmDy}#T&}e4>x#%-XvKr>euLHKID<*3}1v_T3-`V+1T4}K2Qj=3Jb&NVgTR{4UUt~
z+z93sM{Q|84vFa%<0t4R+kY>$G-{dYE5n&BZad2fglr}#$NefmoDn1M{;=b`OkNQ6
z2!)yk;KN@@C9B+pl;FEUbrGxl;bAMQ<P;RQIy#%9$(fl4(44o2AIjWW^6XU<*uacm
zQZ+QR2Q2gFZFV4Ts=E{BtOwGSVw}$d3!dc|*V!`q&t$^wA6+*Vc(Pp@-DQNdi%WQ4
zeHUmxD^23)U%KXCy${^D`jcffApUvNkR(^5*9V{Y8SC|%_dBo(3JLzepYCi7&w}LY
zl3Eimeq+<M94VD<#;Tr))oMsRwPd<$<%YkIa}Sf0$~SF=za0!02Io*tfg?+&_v1rs
zM=wgm*#!#6*sdbo-Ooxp@B5YV?CfXx4h~@b?CtzyGi>!B;cNira8$-+(_a68q+^K4
z^=?h-`#HBS(_YDHFKlmDUB+~2Fi^N-Y2u7ABFt125*n_n=&C4$hQeNW<289_6gupB
zBetqsnrC#knDRmdBCiA3FoGic?w$OW_jN)|uDe{eC(U?U1N-i+O<%9*iO;;?uuJ}e
ziWCUEL;~y<j36x&Pu|ei+nuAKK~a#PG%AQYLk)IEGesMwQCXp;gRidx4M%Pmbj1<P
z+^CeuS#|bE^X^ll5M4xe0OOL=i3?`xO=<dX?P#6_PoCNn>6Y2Z<@&v0iF6y`u!M!o
zjrkpE=T1)*N5hx=shnk}_zFcLmQJE$dmMe!x|YbY(Ajze^lpu}aHSCqqdp|7vQ(~n
z|GeP6ZSSC|UteZC--H$zE#A+3D2x26hAte*3)Xjy-Sy+F{%`jvk0zGH*~FAVL;<Ey
zDu9Cw;V-i2f42)6vwb_JF_WpWYniWR4R{te5BBa=-mk+|iI#sALDOL~*v<0b($ukA
ztm`)f!p87`k69u*PR~J`S34dhUM27XBJ7K%AaTD8H!IQ;wTpkrC@J*k$X{1Q+xNN-
zm2T@sczzEAv3NZ<5-8e+y1KeR;~jF9PJu=K4<VkDfH>6{B=HPxZ%NvUO}pOnIRDC2
zF>_70Dh39;{`O+lG<LYLvlFaNJ_BZ6iY%bn(5ad$v$S<4b{eAhks?de1~a!rC1?6;
zB6a*ELdaKah2ZIhP-`K`?Q+fR#R=IAntAv+I4bb70FTZWMBYc+G0c<|GkVbpDuhLr
z%&D(XX*dgwd37`qT6x66$$O`Cvzq=$^;qUl0o!5%F$xUKljGahP(WNR?1c7Z3DC!u
z`l!8ka~}#oK4L2%O9U#ZzIp!lciuued4r(jGH0AOSpWeZ1}vZ{h(iukk65)U;p%E?
zzmOiPP*t_HE>g|!i;TE^@4a)m$)``KJbpMS^tk8L1KJ7Zs=Vb92iS(ryKB<Rdl9E%
zsQQ?P56mMGKq$tWpo|CG0R9txf>!Y?-li7RMo7Y<l6UMtIW0fGd+N#ox_}R0QtPY{
zG=j#t*J)6><u_+{U)Ao%b&bpx&a%uL^lH^ou-+GaDy!sxfIWgi-S^ckuY$8x0S>QK
z&#9|T{xB;z;3J8G7;?Uch)qsbT3q>z@k>onUt=_w5|CNbILbGFmJ0^PP*f{G)?@|w
z;`S~NJoTlo&^pu5d@6sdLk~wyY465@pZZTVCCYh(C>n9OL7Ur_w0QaPIfl_FBbq|d
zfpwX8_lkDdK7$eeU0&c~YG5Zm1#{WBLUtqf^6&D8t*)6GJfv3WkP&vNlL^hzJdmPj
zNdEej7s(Tx3Dhz$)51D-WEBPJDP$gET?Ap;9M!;b{iEzKuE6&DmraT4DHXCak0G(B
zM#wMAw``Rya}QKEojoOF_lQb=+A!UyNBki8mS=u*aWwdgO6mp3d#+!jEiMY#&Q{in
z95hd9iRY{2Te>eK_>Fu_`amgIyz+wuQ)VPR<L@2F!04M;q)75)k`5L?EdG)u%TS6-
zC0Er>T%cABP8CeIDVCb_^dW1BBeo}U6J|BLLnf(TA+%b31zD&+_;9@n%;6&f4v<ig
z0{x#m_kk#pR1)!g(t+N2VO@gUt!LVB&xIw4$z|D_KT$KZktkG7VlAx6Xg1||oNCfk
z(+4t*;<{7%H*}j=vO90PYrE^o&O8(P+mw2Ty{X*A82LwWb6z-^=h^iFI$h#K#dZ~U
z{7<f0KgBuwrzWjFK!b67_9v?bSZV7+?$@2q7J<yoKj1D~DjZ-9V&oeEaq;^Z5X|9_
z82KET)Z1~Od<@82Lwl79;&fhJL<h14AdH|-zHB-Cf%OIydi}m9rj*s)H3Ew#>eYwp
z(9pnupb_(vR@UoFaFpY)iw{!|OIC~(B6942k-?gU2$gz^X;Jli7U2yvK0=v_jm`E7
zetpsEDtTX;o_BU=5UhZ!$Gc6JbGH^n*?978F3;59^(SVK9G0l)K8u%6PoD<=T}ykZ
zz49xyU|Fq$@m||J&T?Uzu}wy-U397H0eGYpU6eZ<Dd6o~NKTBxUAYxGUvzE{CuGbh
zNi0@V7`C;wP1DiT46!>H{pJq?3}We^DN%`fXJNqYtIw-zH?Z*)meA3eb!u3Klj%z|
zGRAJA;{0A#c0s1FV=CnE7~g}@jvqFEa#C%WxpL*-zz<h_y|oARgWgEmINKFWZWGd;
z+6oATsO?$_%!6S1u|bSbX)BbZcjIAhOU-P^gonrmT3Gs}U@aVQy<lNUwkg9yz5AXx
zkA%Yc*>;K3YAox)yZgMSc^C*B?1B_r7TKLC&k*@cU||U&l9NadxS&pMm6jEbvE(}K
z$_rPiF4Mf48-1I)t~;#$A82hxt3_fVcZDwlB4)5;hx}Lme36Q03Qhv6YulvG&nO~7
zrf1Ude+9#F?#A)Y?j(*v`6Pd70?z~gu<sVg7M7MSV*0E!Nt**9@!0nSdnOO(Fx{cv
z7Q?SX$H~^8%~RYAEW3SyVkRj***sc*VpCG5sjDu#XzpwMrcP*>^ZYBU#c@@_ngh-j
znJ`|1YGqpVl>;kHGUgrPy#N@7hme~6XEX{2jOM8<@+d?DhH{`D5Ich9nbbm*@R+Q~
z@%OH!Tf6&ST$5lR8Iw1*?kRdPFDr|cs`C_v^Ug0SI*Ge}L4R9q?>bFg(SRc8()5-t
zVlX1c`@L(iDWPOodH=|O*ZjopeR{~J7zJLdW93{;9pyYGo`*|=*Fdta`Dq4;0jVqP
zPt^z;c~xRAVUFHk?_V1&0pfxUQD$C$My^dkrbIlaJ;O4rBf(WEotweIWS_P%`3YpR
z_`x~Zv6nkq${laAX{55p*V>V-%TzG)4%ic+iT_<8=&?y-+-h9_<|-{oJzJ1m%?<vs
zIPx6`2f)?czlgsdd3|b=f3Bq~d!etP(8YlzGj#3yf$kwwmO^4R=2gLovOb>8g&jqd
zBX0&TRs`^KX}!rnn(gMmVnWVGwuJbmYcE4H&<#<<-WXqt4k)a$z)j#WkL(QL*a~8#
z7dgYGHi5aEpGyv6oK~pp_^Dp9QRDxJel?x{(fH%}HGECt*%y)?rH5CMBddaGAn<WS
z)HpMSSzTj8<Q<ksuz9)$P<B+#t8W71HJ+!BL5X$qbjVakf^YAbi@Xs>03VhJZ+{bT
z3YRo~$@wV3MvxMQG=wDPx=xsmb=M{ozw|2AYC6`kW&4ZuQP^X!>9h%&-6^aZvG2bA
z@1hG#q=NTss~;g{mbwbVir;5q9xha7B|1^8cYDN?chdP&v*q}jr!Bej&zcCFCIAQS
z%~k$T4EK1fExFs~@njV>21N|T{zp{Y<ZAa6x&U>fP>dbcvzIUs=BP_<CO204f|>XB
zAK108z6*!Fv$D1x(EOQ5T2@lRvX3zexQ)N_h0(@O<vwbGJL0U(u;wkDyJ!pHrMaI~
z!$v|S=Gh&vD>M?ljmx(fy%*T{W)>eA0EN46`+T{q=Q$)a=HsOH5bQk-h2{oW06LKi
ze4lvetEAx;L%gL=GqPM<#y^o3dT^Fa!D^%-(d@QH;B3PwJ6(b=ZO70(0nNx1%@>-@
z>vGQrKjsoV9PiJh&I<t0z+1X>$f2S>FPcA~rLYp8I`Q$<Qi`=HSj)rTy|T^)C-R)i
z@1wZel4l89U#D-|P{J0YL!@nR85t#0Q#p}8E9dc!JjHV)%dbDQmSC^MS3f@IQe>NV
zltEfS8!4!(M>qr68XbRH5ZC$iVutHE6W@CU15p$ovV31Fszrj;J~&DiiAu2Gv_k`^
znZvGHI^$9F1S@Bb%~o|R#pdU0Wu==Ata`jC37%|whET6a4mr1yI+v~>QNvqtg$M=_
zPhI?d18CmIY-huR;nf})`eE8Jh-i3bpjKeT_}%xn{>Fs)ia)tv|CrV?WNRx1!xk_8
zGzpls!-Mz!M;>^Xl@ui$qQ9wYQwYlON}W|n;hcFoS-i!*CAFkM+pk(E3uB!ryg5H-
zr^m?S_Po#Jrjqu#%qUYK2EL^K;{uF)DD_I~TcCEpJH$MscZ6D~G$1j?2nX;=bnqdx
zVP;1!R*4n=vEWUZ99pdUu4eHem{#3k{|Hx{<~L2^KXY@*cdB=FQ<Z)ern(@#Al`J)
z2r6>FA$IhJol3?^A2y*o6f3oie3<BMK8We|{%<VMVh{4*Td<C;HP{SXTQTRX{Lw<x
zmh0oil_^-z_IcXb9IQjVbzQLBzYkco_ngc-TrVa;dPYn$Wjv_mJ?$Ego(W)I^nc5g
zxUnpgA_0y4+yFM}bLqpcHR%dj@9mYrg9C|Rcb%juLbQ^qfaQ_a%<22xmb#stomMJg
zx8KN=AN!xV0w$HsTTt5Zz5J6b2<*|>6oEcj9mpw>fphdi!oI){(0ZbjDe*K*ggTa*
z>%COI!RFs2)}ds^0EYE@{OLT<U{wZqNqi*6fT(>9s8F{LRhP`s<OV>E@(BQ~ZjJ3A
z?o|RIwRD8_t!j1TEC*umBVf^$o+4(>(f&@AB;NyW+)(4e3tbz{GTkW>K&4}J9OCLT
z`@Q8+ZC&P^s1`ipldJYZ-6ox33M9Uojh3SslmQ@9_EPrq^Lx6)C1}XO#+DA-f2w9Z
zD!alo@XxzGtN&R<YjC}*&RU`8SLR1$+*-!*X${C7v11l|4B6bsFv9F%WlqsRj%;7$
zxtJ5Eo6xa8s^6<VQH9q!5{JAr+c1BU2|cHm=x+PhLJZ*zXcLj=ri$`ml*W%7MK>>Z
zv30ik;1!&I=-Y!mGrYN0#$zDu;esLUUtH~bCK(MkY>zu&2Tsy3Q*o^0^!=x}H@7WN
zf%$JnpSHtWsS;XSFg?0Gcj+xtvzL46PjMOkRh_)|st>VA=x?X9<6qwRQYd#6B-YpQ
zK0aGakz;=_kTI?A>V5l0Pz0DG%^iJQ0~=KIcT3F68}k&iyS2h--2KmB2^1Z@CqC_*
ziaz13eDhAZQe9rF<U=%vTtod)!ovEmU4(TJ8WO$q528iG0+zmjnbTign#2?OTd2&{
zKX0;ZWStS->e8F=0$@5@#v>Wy!19q+)#q`(E>LB_fZ^Lp<AEqp4f*Ezl2cG_T{|ID
ztzAnv_SLRig4&15>8sR{^%q|KjZlCkkeGO(i~~yBW9I=K8Q^6{8Qs(!4Wyrw1Ma``
z=OkY8ZK4w@6m?jYwfMeo0jUh$$fIMwj(catUv#-vWo)a|APDa0rswnew85Ed49XjX
z&P7%YTj$pm5E%&QK^W?x$Nu$}`BbObRIYB97=h5mztu)d;w9@3LOIwTadG{$a5?rZ
z<CQJOvH{HS3!aJBQ$KWdU1R2ats@0^^vr}-+O`s<c0~4o&Yel>SsRzUDCZx_WjxQJ
zSb3kiy`woSZVv-ChBb1pb!)tQ#yn7WZN*H8va#W15Ts(Q8-v^+Cs16c*4#STO;;@_
zZan#4QQ2`&oU6_LGdb6NgVAdDN4_$1wx{hguL_g7Fjmddh?IXf?xX$WJg7?Mq%sOM
ziEr`dl;4keulUoZD;nkC>Cme?qoL&NVHp4M$00(*F5FUWj}pHCjQRbSSK|Gk&XgyB
zu?+O6Za*$II+3Z<T$My_rsremm#cKuP4g8_!Ns){zuoy2u()S6H+wN{ViUoyz^^kI
zDQ!tsfxhvlgw?X5A|0Qbqdv>U*d9>tQPL{*W#bxe1EzrQEO)NQ5T|~4Oq06&79XMp
zGz|=VW7J;2R&SMMN3U;4EJ*h4-hHmtE%dhgS-ZzHW$vvB+77<1-&`y{SeXG)s}4Zj
zoCl>Y_jP&?;FFM?=J*36_JQpSxPIof3?4Lq*SmctDxD0Hb#3NT<Nz=uq5+Av1~|zo
z%gYalLi{>97zj;WUITZ(CjuCTt5g5c{SJ=6a|7f0dplsPOygD^;TRq*B$PB5{p9NJ
zh+bZ|&Ir#q{SM>ENtD3!E%%{=;`*-sDj>&fq2;@;yTo66i^BQ@snzcrxF0Anf}BM&
zaZWJ9UOY$Re{7QzFqB35AB=nlv_3g!PKva4tY20-&j{m6eHk|d8vi8Y@B5=A5ZPUe
zMT52XfnlI!oq2rY3X=i<8d*{8nsVz71Ey~!8~<<Kk*@Qa41M}HQ#WORmAS$m<uiR=
zx6V(7Y&uFcwe=m)pO^2{AcTQ5(SBYdr8zh04Px%&WlQbo%F6kPUdlF0bbcZyJ5x43
zpmQi2;(fkv@Q`!uC6K^sTD1y<AK_<?V_C+^=$uc^Y&%gOTA}8W49^YyXu{(+o4H+E
zs&+UJtHynZ42pFM8K_<_%cN%Hd|O4a)J<J=d*I)INPv&QQ!}cF=IV(eUe_@vBiAx&
zZ$j(xB^H-f@hw@x&6P02z@nRRbyh<9_5{~1-RzqSKV9@{8y0PM52>I&I)kOTAhLdh
zhw;-gj59NZ>GlLC%EfB7AGe()R|_!{Peu_>Cqo2guo%huPT#K5D#jXcp89A>RULe9
z<pY@(<})FupIvyC4Q9OT&?oJwZD>UN`*yo}KZm|0b2(e&B!3PYIz0i^NxEC~Uu6%q
zP*|K2)DJX{T&c?LwR{HMeVagLTmDl6mXP;a8Vyp%tA<;rB3xu&hxU%u>WR#6_ox0w
zJ{mIm?JOX&c435`cvnR|&ad?wtuua~u%$>?=0v04QP`!KzJT1iws;prR=Z~e%V_%p
zhY)axLJpr>NE92Hc%f$=EUttx-=Csq34Ql1;*S2~Jt<-NVf+&EF%E5hexXxnfit|;
z@c&0l>ST-CNs<-PBLvU??q%NsAeY#6PUeX9>fCiOou?<x$cW^=aND$1+>7#XkaY3B
zM8>p_fi0p=#Lf7iyrmZX5Jn#glF{H1LjmuIPb<3}jkK+VHOPSlFu{rF1GR@&8mkGL
zXuzD@9GPc6_OHX$*8JZ2Q;kg<ld}ap3|9Gy>N+EJ9SnsPW_FN8I&;8e()fhiQD4kL
zrSJd=!TH(Bt_!=Km>j*9;UWir7u8E}lf+rJp%cG;Ts58*wptSTEMJ@6W@SJC_Zw{n
z*sjcyl-~3@iQ+V}nO(}~47M=*sauo5TW&Cz&40WUIvpYH?66mBP^lfBXaD@)-(pGn
zC7noW@t>yN<zo@5St1JCt|+?&hAq1(c)zGpAKRxk8wXla40j;oG*48k=E~BD|NiwP
zAN{{hgm@_&-@+pM(<nV6ud1d-E|D@-raKj$_anvcdS?R9x$5?>k=s$~)z=l5NI8Xg
z%E#VpzD_XLfh~vGMMu29j)pbfM=~!QiVkHtxSOJNf11APAVF4^s#4-Ke-2ONi8^nY
zIf5r~o^Orm;+lG=Exet2n6)pZMJkXR=)Ls4p`*gbjHWxt%1C}v2#Jqz^ndD}wY?3J
ztKuHCq!|jd<PgpmJRL&C`nMD*_sG;%Lu2}Hu=9RH?G{~GBVHrH$-A=4Z}p|;qkdld
zrUmar7VGWfA20oMQ$8lc^p!fX(bj7Iuj69;^OKmg+bFdiz2eUw(G`y6?qb;7^Iw*~
zZCz6&+!OQ+;=XnNw1Bbo{mbRpt9{<e`EPhyrnUTzN|e`F)0b4lAB`Np%nZN}nHr9B
zxYMDOPh=?}=LBMSXE>J41XjYmKiDl(thPh`$1QxFAGYPXnj{rd^5bNBFLJx;j@p7U
zDIFHL^){^a#h33hA@m->a$H~Yu6mnezEXsFQayX0Q}<!6$(;C*PQ~QjK}EqRbw%xB
zXtwx8(F%Jr7u81c8{w(X56RDCQJK3i!TU;)D76&Xw0IookonRZRz0IH{)>wglZ}7Y
zyi)Q90XHfoqnrw(oW|zUx#H?MTzkm;AbVxu+JwM*FzpX2yqymrzN5GW2^Bsj&JSHn
zm0}k?Qdg8jjD7Ld9pSSO9OMYDT|vEx<-X^0WG-t97lol*jtU(9x;23PCWB@CN_IAm
zUKde}KkS3+;IE5dz(}f>C>06zoGdTVO{Ou<a@s?>6&$*24jch9RMzyv3}3<(d7E9&
zvC!WkUf9-|BxO~%w|)gGgYnzLo8Be|8plBHd)j;#dHN#|I*BS@#<5F9{4uk%b2ySM
z7F`Z$qAoJmN+`&msF$P%%f&+Zm-fuJ=Tw)~E@h*~{!OReuug)Yzizf~Y^FN(vP8cQ
z%^OLYp_NU|y~&{bzL)Q_$1!`>`zM+IyD<GTsnM_QyNj^_6!1hd6{?wj<p!<v?dNa@
z!r2ib%Vyf%(~n#OStIem2S_u)P4^XWJT8o-(I2J9vbd#O<Sy9xB=6~IChj~IL**g<
zSROBac-73a-jgSwv`}$t>(d)WI^N@@3;Zw$bj4eH6C(y)443D*!*-7E?4=*f+<O{g
zwksSj74w<)h<tp2XcUn}M?G{Lxh2@^tRD+cBeh2PDd0Hvqk-!nvW074o`u$yBV$IU
zWs6x9%$TY_=2}x$(k{8JFC>colyG#2e9L`}IUds+aTdj;u)Q$GWeyMe`%*-qEIO}k
zy3~n<_?g<?NuG10EJvE}!Cz_ThGL09%{Sg+XJ)*6<ZX^_J%6Q--^#oD0I+dwz<?7u
zx~vOZ%!qW0FCtK#3u8(fanb%szJ3{nwVU4h7M9^>6VPGeb)F$}@ur=EGmpp*!zV`G
zq8;6bW<vO4<I!~+zIw|E*D9f9n*n?f_GVf3m){pAiiAe-DIJH%n&mAx9b7Ab%5Hib
zW;V;Is62p|?s|rF0KpR&9j16TA{(#~a0$y;9fma*<pyTTJ}`p%HVKBFpLcfamUe`$
z&$-iwQ-&IIjJv0~{$k6R3DkQpT0TdqZIP37W`RUw?1iEnUAY3zy6@-?n2ISp+klTs
z1rxvczxfJ%?t<Yp!_I+NT=$;XNX1{(1_O28067RU>m>j^8fGW%*p8Ju1FM_mC`LYf
zptR_p&0BbQ&tVm;Nghn)@w6^GJi`L)Lz;slY`Lb|=7+OAv3Q2KnaxD;kF6)(BVyT<
z=oJ1)Pb^no6fAII#9>;BiaJzIS#>~mU}M+3LbBNmIeS>6hCJ5`ER3TUvd!-SHHCeh
z-57r}eh<}zI&*E2mf1Pqy|2mvr_|jZudhlB_GZw3XL0^Yd}~^hE}!3qRaE!$a-Qp9
zos}K@q&T|Z?%xMR!Eo)?K(a@y@kdl5d^~1;?orqKRxOWLoJbLhx%;bx(ju?v7HLji
zrZaY%dQ!R6cF0<@@e)*6#g=p=Xv68lgiAWHsyf&gM3uKY5!DYMY*F2E#P+%<3e%Xb
zzXtQE6cBt)H8(fs!qa{S8lrOQmSc*>xo`vs)bV~hf|jeY(+E(AdJgQXa8lt~3OcoJ
ze{j7gehH6iuyB0SEMJS08!&*F%Q*NO_vFpArZx9VDZlGDK3?wD_8*`*OmtaR2rK`S
z7)mozuY#4Uw2xHLdUn?<U5@r#3Ns3{Wb~<>5hzwvAQw)rm{R9WmDXft@U!Q2VgQ_0
z9Xe^uE*g94Uiaf9V@Uy#r7TH5L8OSp{k}kAoLBzSj5LY9<16vJNfuc5z^$jJr!;`Z
zlk65`p)^)y%AG_UvGSUlNtz0xd5O*1HokweUVZ)u-al)W3bU)=Caa{5s_OW?OD$E|
zsXO@iU85eLa8OW{u@HPuG`#HWb+F0v|CDcFARN7%`=p^1CbC!V62$10)X8?)0`X@h
zY4xnGR(Jgt^_V$9E5%=QvlwP|@h;N;1-IpVI{y>en$sg5gV3UeS+Y0G*XL^avX7n#
z5y4MlM98^4)L7qkP13bRBindl#$h+XnPKr1-wgMqy5$sj8krawxA+ttW*LEa-8th<
z#{WB`q}C3497fEh`|b^Fo(eqose~sG-xD)tKAZqX>vtk*WRC``e~U-CqzFzJVswz}
zaor#Ib%?$)p(&}&<1PJJubaf4>9fQR%h+d<#!~_ADuEU3Y3g}Twq%=rhG=oLe~cuX
zRS(;Lo!$P#)l@04;O(9(7v(cGnIf3jn6ilXz@iDu8r=*M0YG{WGU4|~c=)#irom*t
zZE^Db${!C*OYOG(lqPwGVR+7@sDoBJeO~8V33Soi!CkHd=mFX6udw&S@>~FGPwu!_
z$AXkJHa3bEvB<HlSGMjJlMnT$Nq&!>K66l+zGTW1MtsZT>5yR-++b<`T7*UH(VoS8
zWylxtBWYQu4|32#^7Sz@_Yarf;rslVA5`hSJhWiC&_gJnYzUDH%mo8JyJ!Hthrvtp
zpNWAW6@p{|Uv+~<K3xPLayD3%F~)VH8~_iRveMF9e~@pOVm|AJYU+Cp4}x6aj%q#)
z_CYp#oT<#yt0$C#kWH6Es&_82gPVWF=&}0t3ku;)p(|R829UHF!~?KoEkRjKaK3uz
zjpItYuQK?whxf6A$o@iIS);0~zsXEbu<yZ@w8oUH{H$yHw8557w%Sf5fJmSJhe=i#
z`8yhBFw5?PoE(_SMV!7oF5j6fKS?2ylq#zUtPGM1{oc<0yb}=*lDyAsgObd3&(*?%
z{tpr6o!aOxb4$z2KR#SK4Q2!JG3+#LRiJv$U@VZId{Cl`kOkAq=vk*1D{#pRMWQ2^
zlX+RDTUX5V_z{E?5P6S5V{9sDpL;?1e&O1`^eYe#knz6sO|A#<SN7@YW~<D>a$D?R
z_tx!pK_+#B_qja;6ALSLB#~5aHSgO&heoZJ)*d#6FZ`*xf+@{-iN_j;O!}F}_-y3J
zMv~t9<@V`jo{QXp>e3pJ$f7ec{u=^D>uGInd%6%2YUK*WOTC8EyX%832vc75OWt0y
z8awZI{p;Br8Lo4G2MxXbl2_`lrCOphY5aRv6F%5P>OUIgMa)!)w`O-H`f)Hb3+4gT
zY=eE``eGwCbZTk}l|*?=SC;#qV{!xFwT<;)sW%<G%CHr)wvK4N#n+iLl6NCu6&7ys
zXYjiN%mN!gYX;O>T|{-vygLa1TzL2o?wdu1nogNMqu~gZ)YrSIG;cu000)(s*;DwT
zmJrmhAHV7f9J`sPbaU(O1k@<<&Sv%pDTLi@HNQ(avgW@{Pg$9r6~L3QxgCy~7y8_a
zih~9ys`;Rz%Jxvl4SpcoMe*Y<)N`O%^tAX@gZD?Mc&MACfggXd%lx*%?mE~7Jq{D-
zrJxYD0^W{0AX>|&y#7!C2rb5+8nh}5XXfjD?Y!t-vaW9BzBP~S_}1{D<g(VDgIs{V
z<IBVq-eKbJO(oQPhK*a(g)?!2WjT`7BagFDOG`_dZ$$^MWBXn|eGKHt4Zo4+4zkzU
zUH{-h#@+n0r<bE^Vx{Wv_V3a`rorBQ!jB)4pD>nzTAh*k`x|g0LG&Kdm{C6mVud<W
zD_E~|2JSI{^n|+t-tCQi1aX<p!ms{<I|lMvC62)^Rh=g6!I6abkxnm|m@f3td}Z1(
zqy7)G2CU7Tx)E0kx+FO(TiX-#H_U)-?i~<B&<eP8r#!s8Y)EKO|Kie8A390fmal63
zOnFJ5yLF$5gQJ0?>395cy`A!;?mZQ$p1!^a4J=+x1Z+!10>YukXw)hDoI!F;gSf&G
zS|ttRls$M?C1`_R{c^4am|Ft;y?18u58dbMUoEsQ*T&X3d*?lUi65=3o8sAg_vUOi
z5P9Ff#+n9MWKLgoHdc)!{vSvigZe4|iGya(86N4<QP5F)x3Mxd_Bd>kot)cpq;}c&
zeCZO7s!&w$R_yXO4d6`%+35ivi`qTkND_`^)FCz&7H@RzlDZ5LlOI9PM#Zjo^F7w9
z?7ZvxHawX!T$g~jMmsc7Aktq|;kip|3sRXYjd3&geUZ7&n$hDWMW8iX1pfvQo%`o`
z?`Fwow}IG3ENc5hr|V6=TCuGD$AY?yN0i&FBN_^Oik2cp#tZf6jbuRPeey-#W`W@M
zN<d*o`e(Obuclx=G(}=x4X7%4?+&FlKjbC48~Z`Es04Ql=TgP4n#)|fy;`@ZUH+yr
zp0e!4A1r_|=Yk)BYx`HJeQ*X-u<zhJQ}Ki@CX)V;sV966we|5VmB#fS9B%=(hA+GH
zXE>+Vx0x(?8S?XJ2~w$nvK`6$0<?2~9n9Vuu*JQ7qk(=iG_N-t-G-*J6PM{%1qwvp
zfMA<e>b?NPc3~N|PYM7BGrmis<>&QwTEfWY*x{4C8%HRJGhRluj86Ie5hzfe=NK3q
zZ21V}`kf4v->Irrm6w*<GCRigZ3tq80hZI)EA%$pR_f7a`$MlbQ%8bUU<J|c3y?mi
z<jlV>Ko@W5i@bcm7uWm>n#WW2VQQU-(>oN#&K3~!G=X0Gv~+;J=>T$xOWx!R1W?v=
zJlDe+my;j9xphqS3glwTK<dR_9O)WJPdXp}`)7{^q-hskp8Ebh^{VK`cANZC#KoWo
zus6Y4!=is)e~+o#JkGXfcU}<{H0apn%B);=SjOsIbTCKqG=YPF`iK0X{c`IiM<1p0
z&>L&qBB}?DdHkwU(<st7p7L0bMM6ioY4x#?17#ktwsHOcBYpBR3?8>u8TPYJxSA*?
zzw<`!xPN@DT9jxL)(H@Pw|f|usTe3lJU|~j^xz0tTwJvC>!4XZlsk0=WP%U@<aezk
z2Y!<AEn3hw=X(q8)*v#JY*YU~EEYEz>=zIHP2Yb-ys@~U5HLOY!5G499$jts)jXTO
z^=z&8o?r{(9ex9%FP7%s3b0jBLCZd$;|pL3*#;yUI~c!S)3iB`gu<Q1>oc{CycZ${
zCTJARG6$2`C`uErc;f87$4`6hHEm_2a(#Z3*=BWis!$@EW8Lz6Y-lKbL8ilPyY}mN
zZAE!`yCn1J_nM1ifaQ6w<JMvUcvKmhoG*+Ea%oykup<xr->en1QtJf<0ZfP1<lH*>
z{q{z{cZ^KnQqk+Tni3{hO8UEIF;Y7s=-F?a;+h%oE0p=qw-~agIpUR69&^=`4n}Qe
zDJP&lYL@myy<7*cR2dB7_n;hnDqrU^SC`ysjB_hqzWmK3QQWW>oewBkSo!n!zJhU_
zI3kaX7tRdqXwrDB)$k7u0wkUuR%r9@)4OlIk8c?SiU>ShWCKeWgbdiwpJ_dnpKCG3
zzdiKro=+8Q-I$#m$=cQHO)@N0-{mH5k*&TQ%Y%((d=4b5FT1~<3a;`0vK$7;E@9Lo
zo8nbG4T+5TXSA2^5ElgF<1gyZ9`>rgFTbN#L<07gA2@GZ(K+qbyTfT7thk$jCi5Cl
zix3;C-wT$&t>jshumJSXW*M7X_pD;zJN#*^4@`s=ND>}7&)<iFm8;k;m*ZM6Ns@)R
z`K=r?ub$1V6F3D#;g?&PMzDkg=XSLnjDqcBz#OC?0ICY1<OYHj4sV=SJy&ZF+2%F<
z)fKu9K+dY(^8V*ZR7o-zryPWE-ul|79}=D6^-5LI%m1#>Q>W#cqLHm9O)-5<6Nk0h
z&hA3o^>O5+a)pw575b+CqfHouf)nEeG;UlnyiM~cS1VRv=B)veYZf{>Wc6+L=iQ+o
z9ng}}UP3k#Za{7Yl-VxM0fPYN2g||-US0h#uv|iPTV{A#_<M%!?gD_pw`KauE#MKN
zEpRe};Al*@%4%Ok1vILgXsL}f`wLuu^JmrdJUaC@h<FQD51#YpC)8V!jC}7S=X4xT
z2yw)CAKu(hj^KHDquKn*u%}GbzhBl4F#wVlSJ7XQpCGn#s-f?nP|j_|6s4O#<NJxQ
z5P5h#ax~22tJQ$1^G6r)I#8TBuO0$jU4i2~L>qY^UN-vFQfZ}x($F$j?<?Ux@c-na
zaM?kqRTZL(3JV2<XKS>=w4)3KuJu=cG-_%VXI48iF#M-EB0BXKwFd9pjJGX%U07gX
zNppvj2M;W+eOrr>#DlIF0Gh8%aW<T0@HIIoeU>q*`WL@lk;nHE|LvLmeTRfYO!cJ=
zOM(8agBCJqS9m5hFNe2G^nY_ykW6Hw%L@xZ%QIpzx8CDuJsR!+8ZriQC;x<(mh&<~
z?-kW}z;C?@{|n#y?Y+o{zr@Tl!}zW)a<wV~uGOFDF?jFi=%tQ%_0C0Z4L8)b%jb@u
zL<9`O)dmwzTeLVXWoj{`*Q)=_7|aNGt?ibK&0pWowVYq%BfR}rE6(|?Q<2z+bh9G*
zTeEW3Q8~o_(EepWw~ff^5kz%YcXwM^6p`Jk&#=oVO5IX)cRPA1f+U3ZRZsorrfm`C
z%gMJ49|=`nmE+Wo+04_goFBwGd_tY)u}<)rI6L+EBHL~)wxVok^ecZ`IP+i&IZl1#
z3QK*84ylhsP&XZjiXS{UU$4uJ9E>l2iSl^<j5g=a#$Ok~meZIs7*Y&5$u1crx%cZD
z5=IbamXbUkcZGk@CQ)#**%~tP-mBiK;C?!aEO4HP7J>5t=3bgAdY{NE4kVch4<1yl
z*O_M%(xt1o4PSvz?g@H!|B_=uuwAZ(Mv=6W`W2<Z|81tVOe=fr6k!=OAzOrHt7pA@
z7l^E~G3DyquF@0N0z{g8X@K@O5&w&3`FHuftu#bP7(&<`j8Gs+SJ;7{=KSxZDvtD%
zGXY5m(^2t3+YDY7kBjvduJHPU|6IJ2Dh6!?1%B&T+3i!v?^unRjuR)J2e-%$vmCah
zO-yU0cMpCIv^i3y54UTvH+;-5VW!^N&+vb^0IK|Mx2ealdvqr}a5J|F{Vi2I{EBq>
zXW%a>siGjDb{A2SC4OrNDao<*BmIJ&CYWoX>AU@=U1}5j-@C4NA~be-s=GWP8D|!>
z|5K!T{$Qm?q=ncTvD3L;E|(1U!T;nwzclqC=g<39d)`w9RmFZhvnYMxtr6Mcd?l<|
zVjv}SmB#PTbZp0z5tCCSd7647`ri$-k&)=+2F-T(``hd4&RexO2iYyR3Qq7SLEyjw
z&9P_(5p8YXzTD3<JpR567Z>um&G6uld`u+<ylr`U`UG#s2Hil6{@ls}1deK{Z~P8U
zdrx1v@6I}<j9TKLQF9u@NhxAl-T-Gi@`=SoY=~o%enH;*UKBFQ<;QVOh86Bv=_c{n
zb#=VAptKM9&r_{<^z{0m<KUe}SQwoKHv@e3Fr;|%EO`U8ih3LBrJk}l{vFHS@@QM8
z1J*VVj=jnb*49u6HYw-o(7qTyuh_~b|CPqOC?S7DRE%EO{d%XM`1RaHNvmA+mcN7t
z9Vxcq`l(myE_MR8VFmv^5LAOeLc8M9Bg8|2bjobK9R)Ctz}9+Up@^b!tJsoydHbia
z!G1}I{&A()!NBlv8ys<(KS#-2%EiHPfc&3s-cEdZ>f@df#OX4Q6z@j{W}v?p=s13m
z6Ui>w@`Le|%dmoEO`KRZnmjlyjWU%my|}y_7Zl`BAyJfq1YpPJW6Y6K)o8uHyM_cr
zMPWx#3D-x8?pB!(5~F;6B|wS$;3=^$kSGCE1|Qtgp%ByFs8;~v84iE@WOT*}RmA|&
zV}!fgS$~ylg@MM&+7&$L{E^FM((=mz`ItFPq5x1ooo)9OhCF3q!2-Xn;#WcU8bk>6
z^a0@+RU;!j2&u3e>&urfZQr^qBESvJ%uxXkOEnVNLp4cQz7D2w&641c8^8Nw$Otq5
zq{i03yu#NHrJ^`v%~8M*b4efQq~F$~Da3<5k4vC7900m~jj*XjJnDwy-Z}wm3xDY0
z#@N{eV1Yz>mG>_zN>LNWg=eQWq~<sEO-sldr&y2&yntBQ*esZ*Z2EW;y0VgDmm_h~
zlVbCw<O?qCi^zX0G9m52vP(|Mzbq`Xj{n}?4o1r!8UTV-U-wmbpu+;?HWARuX+eh1
zKv(~c=XOwvcsS}PU$yx3bLxwkkpgl~*hKL)66A;F`FEr)&O4M<9|)(?u?z$}i37+k
z<REXLSwOLtZGs0V6n<9+V~-4{fUE*dKR%^n+*G>f!6u~qkP9;4TH#!LRXQ9O4`iAz
zO(5z}gy^42lP<Tp)c*LSBuxi*AbqA_#|j27ij-^palmW0tw+GC$rjh)V6o}=U?N`$
zag62l&_unkZ?P1HPj}UW^b$l5>Rk&U^%hn;YGirwVgxsq7I;)IB}9-Bg2;TjzU6Ws
ze6@aom_Z(}G`aQJNg!~Z;mvox<pA2a%^sF0wul0vdMTiJ2=_F3!w{Gf?yc>CK;E`#
zw|Y=*zY%<mF}Q((g2KHQ(e&-#Kdh}{IsrW;G8S6Duy&Gn23MPDm%K-RI|^D29ZWqI
zfC&gGpRIN#$6t`krIFqFk^*mm{Poc_U?5Kk+EO3^#+<Y9;O-uL4Wka7>1Z>tX~bHM
zDBS^|cw}t@_yxT{GRUr)8xQQ5wf|qc`%@kDOz#n1&$)GRSV)tB26gZ%3+8Rg3qOZK
z9}%ay92$wQ&ivr}B4m(yB#*Ms*TfJj;q70TtELbRNFy|^V1fuj0cnlgpK<X@d8TuR
z$)e{+d8JFTV<Sa*6;beRSh@}Sf9^Sxj7kPIy=0{E^^(~}pWO~ai=V0o7%AzV{h;IP
z1{$gDh~vuqrERMnGRA<~1omyn2+AQUqyp&z3nI(t%L2I4jYJYA{Kuy}ud9P<dR||^
z)Rl@ia|}cNr%<|+@X0mP2dj9x^;de-Yqdbhle>XI;*U`9pNZTW|E5gft+?%d<!w+D
z)QVJZPFjN;u0Y)2<mL<CIDTC1(N})po`bX#G#|)7jAAMeR-n76hbUf}Gu!fLv<lCP
zZgT!Jx1SU(&iQiP{DC1mt_}0%=}|V&Q6<Kd+gp@xpkYqAXot_T<hTCe{6sTppV4?P
zH{PA;qvoQNzeFa?NeFj@;N>j+$e^<7_*H;p6~OEIWAwFAWUX9%$Q2FF>Xkj?o=%8a
zBemwyWx%jtnA(3Pu|TtSOOe$;gc!a(e(-F@nca{!XEM8o^MgbBNn&yPqtzb?`_#}?
zaWuHct@b$YYr8O#fv#-CKl}ZTmoC1zv@^Cp_WZ_&`@&^IA|W@_@KA^@q!n+q03zi5
zmJ(wBnfLZJTL1&(0P2AaN2RKb<w19-^)(>d*(e^{^A&K~UrU{@Q$juzgeXHVo=$Gr
zYfB%z5faR=EX%$_Ll(e+CuIAkLQ0{B9E^3b&q-91(+v0@d`EUD7}$MHQl4#UwO6*k
z|LCSm9%5;>&ixt9$YsNb1lPy>>V~1^SoKCu<@mimjRicYx`4YRa8t@0oUwnW>@qxP
zqz?9T=AJ-FBcC+N)5az#CHX)|kP9W({WB+#J<#D(KovtEezng=8FE7oKUcoG3eFz<
z3}FcQ7c2<;9QcpE`?BL2_&q`hd{|hTnWJp2I-6?8Lw}p5*La&6+z0hRd>4~NzEJ!7
z4-zo4xraBJ=ZzImpzIY@^zd)|=JpZeo3uLwvcNPLUp15ev}sG9Fquu4{=Mc4$50MZ
zQiBbf&g;GpYjfx0yu-uQ=068VjV#>XM$ov1@n7{q0+Mz<dFv8<1wJo1TJFb_&oFOz
zen#v?(KWo?AK*bf;(6|}xBZHyMsh|&zzH)rh8531d8c#3m9eYA_=$T1q!CSmq4esP
zG2|SsTe-&z`8>aOEU1k-#|G&Qo+pLqXD`IkUD5io`nKVeXp^(^WriqkQ4vGT_pWXq
zO&6}lRYp{-nKqM6%PkrqkuFHri+2M5sYiR67^~%DzQ8V3j{hURYdD=p?mKO50Z9%2
zN5*^sakOf{Cb0%6hj9R~7`1t>Y$#38IcQ?L6;mRxcqEXw=#qsxw%}=m2tNb{MFg|a
z0Gp9_P>*KN&0=7AO-w|z$Uqy}+|Spuadab^^cl<p1^HqE?RJkqNQ2(NsW9o|Y*IUN
zHLg6DwV)ZSE&s%mC?a>n@=~4YuZtSItc?jFW%__e(-QPgWc=~*PrWTt{DlGn$8k-F
zx>%7^;^1=Q%|b41O%&I-5KeS>?CS=9vTVEA{IS#fu7ZgdCsI6QXq~!J`;P)xjFpBT
zYvQevY166RLsqdWBhmwSpF<%&IMXg@CS}<RRYSom>Fq-a<--3~sC>$(r|<W$Ar)OO
zjAegbSwf5)+!xS|p9O4J-ib3HX+JS#&tZ2$aUjCP!~{l^zL0%nN8$izv5<~b#^}h%
zUCY8A^;o*!lKr_7Zs<lparwKc7L#<zFATK!SO-AbSxW;C$pZ>chln8~5{k!*g^V<2
znwR!}6{#1v&Blh2UYxuO#pU=Z_!or0v9Ma7BNs4S4Yapj36v8Y_g|wF6H$74@?CC1
z=fk_n*XW948AU+ID~)K?CXei$77HxLF*u0H*^~4UTKVvJ`q8*__JTCTFgz#rF!a|w
zM?FmRCnt_3+qP<&_M)AG_21^orI{aCdC><J(`V~*TD8lV2ALws`vZ412^9^QGccxo
z?bW>>C)*RD5&brmVu4=We?FRficoj)4A($g3yKMi+aSF|JWApL1o+?3S4+npEwc-z
zzkL#yEDc2cV}6c~DxZ6Zpalfdz9b2t2@qysK`_Q$sZRNydWifQ&jZJ<uX_w|wPQM!
z(>gw{vx(5rs_=1=Lh4c>;Pp=R&y_GQz>y-tg#I3w*%~@%r2-B&UnM_>+h$6v1vloT
z<ME617yH^-JPUZDy6-RG;%l9_w-NVD#On+{t&vIl`b{H5xO~ZP(0J;_9s2aTbpw(B
z@>0p0o|=k;=+C?zhvHD4^$AnViYZ4wlIw$m-w4-iggwq<`W?&ytG|+7URUlW(N?}i
zpi~ifIHOgjdujs8b3~b=Td16t0NS%mN=}wNW?pgvs*qPKr6hC^IMCu9;TtZO{lj6@
ze)FhQdEzYqwNs#pjZ6D7+AMB&Ry&hL;s{$zj4n<OObD3R$X*&Vb}@tn;+)VWqstUE
zxzo1uD@KoukCs>HH+ZEzs&BLMucUul+N>;)4O>@ANhQ)_&Jj&8=h3D%+b0s0kJX7s
z)El83`dBSMjj;nzA?=t?@up9{7${K@`K6U=OP+Cd^^esBmSYv~k>HdByPh3!K!=bW
zP5&aqARU3O(@IK-O&fDs-r}e#i_#RbB<JpDCddVWIYH}>mitkrYq3?PLY3QVzJLEs
z8X%4o(6x5q^D)<z9kc{ZjpfIsr;83vbAoi`jDl3%CKP+88UK`cYTr9`X5`zsIqzqZ
zqIE`0R<y&kVxLj$6w9)7=dCU#_RY{`DJkY|^kWAksW$C=vTm&KAwp!Mzd~#g^?Uy3
zDA3(cu2=U^?Z~txB59hnKI<?(W6GC7vN<C?;gO$QT?CK6`<_E#cs6TQgAgBNI3MI5
z*XB1uhs!M1(kP(LLzk;`%aJc&?ZU;Z2Cmy5&14RiEQ<5y5ZU(mH9ABB1eEnL^MyMK
z_?Jn=#6MC#_qqY~&hRxCYX1+ZdqY5)I1$MWxj($V3c9h;E&Z&ewB<B1b-Yd50Zi3_
z5hk4#HR1F5gV*M9km5LoaV>){4JX6y@uVE_{y%iB>`31;sWeO=*(`yu`5w!r*7m#f
zlJn~sik2horC}ASW_F0mui?JK&Y!LL@B^q1A>08fg0ecucgoA_ax3a0BHA-X>Wu#!
zxLV@Z*4JlPkfKSMIap3m3Z#!1bQ(XV{peYJ?8?B#wx_Fs8ej()wQ*ttEq!}#{y1<{
z2rksaK!YUl>({mrF`uh|;%QCIvs>%4W&e9$e8D?0D;a>E`ZJ#$yGDn-{jP>KZnc#!
zEbuE+bIFw(oYe~7v)?T1>kGU80*1(PmUbx3>$6!W4)okwB-33l4l^)pc&y!9aZ9v4
znM);YzVrI`29@)>99K0bJZn9i_koCi=G|#1m<}rMW2mgi%|^Vc+@WE_%Iut)J0U+m
z<<zol{wYoLgjkXmNQZkCV`?nT=}_Qfkf<1dnu@EN=A*YsplM7I;Il};^tD^LL=vkz
zI32?{BW%dIWk~~A*>M9b+9!Q);cHhYo>5tGgNysU2M`U?YhUrdaTS^vVI3fPae_Q8
z^6GZ8u?zSl+aG(ik<6uUn4#lQw~is|#KuMk2EOf|dZ8~+GBdw()o;P0@&k~#&E8M~
zcD}U`pF=$wc79%7^odtnEb!TQoyzY(ZI!|xMbE9leb)WV=*?zB$xMCyx6Y{YdEzua
zuL-RY>hr-B?-uUA45xd`ZP)WxS6&c!p=$2_nO9q8jo>flv;8G+4Ff<xv9!0hkDEd4
zec(!HfB(+owTA!Zsn^0owhN3m=7imK9cqBj-7JysZJbhSX^|)}$2%k%hCS-@-<SGb
zV_4B()Ec|k<h<1bz77c(&z1FGK`2EVq0#<RwYT4=*&Q(n&3-d2=5>)IuMv`s$8;#D
zjrHzZz>Pt`${+p4{k-fRxjbt(u*n-mxgqwt<)A)P@Ki_!${~Rehmav<bBsH%5)H|%
z%zx#9G9C7Z16ey-xggxJGlj;TxK#2eHM`>e5xfs#i0t@yy3iPT`k<rWwcQ=DdHeoJ
z9saa_n>+ME@a?TQj<|crCtih1QEhc?Y~;hPvFG^zd+i={DpQ9vpVDJ|K;Qk{q06Ih
z-@f5WiTToIM?^$WmX(zaLVqKBR6iUGvU~4NVPT7Gw10p%NvMK?!ceM1yJw|!3Tp8c
zc0keG+#ClIY+{MEScp2@8yfYkYfYjBiQ|+Ie486OR4d#!@baZt8CPCGfifEBTszOe
zM0q0?#Hflz_GO*81;7+SiM$MKInpLLqp-hQ^=6Q+{p@~CVi;p&TP~v))T3BI<0RDX
zbJjT8d4*i9te~vy(Bb>@Q&~~gt<CqrKUXB$&d@WK;w!2aZu-w%N83WYK##}t%clzQ
zJu%kMiEZrrxRr{+ojOQ)ov`f7?|FI8^jDSMyt&S4vlhz(lJkF)Q4Jye2F-5B0jYJQ
zy;XP&IMgDAd8!%Z^>>YkeH3W;vDe=#3rNTQD{+L$6Ea4)u9WR7#=Rf@H5GfErXV&n
ztt2*S!GqsR|INMiE<T3b&`K_(huz(m3j2!UtpfjbT!T-Qn2j~iQlv`U>0^FWcnj6Z
zB@Uz`@}&{$f3JhRnpnr<cXz-DIV|6^Rc;wnE#^LiV*uM=TR_Mq(Qs`JYkepO9=J}|
zgHrzD#*v7gu5NS3_wV$NB(zGkod7rAg{yR6AiKE6p?N!BoAEj0LzqW`M;0Q>(S}B1
zmMQeaA&DEuJyR5Sh0@z82;&aGhy8EWu$}uOZ4fae?&=d_+UnejQl*it89Rt<#7fo%
zYI`+QA6oJ#wQzhh9r%EV2k_$jsDRjh(tz1Q4slCQ!^@Gz{Or5EjfWSmDFb39_hOMe
zo=m&oksw(`ywq~Pnnyx(6{~4u^?=wvhuH7B0Ae4Y(Z7UU(%+0XEz-RK6?tM_ukHFZ
zlIaCyCq<xj3U7!B1Ygt01$Z)K5A(I;TQae;^MMi^2^3aRoX4JwIPO`}$C=+avITq4
zABcqzWsD8fXExmr_?oc*y&g^XGbI$@Z;v9Do*8x(9ug9=0u)$Ex0}sB^78N~A|Dz+
zKml1lEjc;a%M_1b;~7K?DIm<pK;wKajd2YrP=Y)l845Ih56}+n0J<{7GaL*-OU<7{
zS;cbOl27D9t~tz1){o}@fSYlqI#E?rfJf@G@JNfH*8<0rcZt)_#=YGPT%Z*_`$-i5
z5d2fVetj}3t+Sp3iVwJ@6GQ++x?cby+FEuoiL9~60*tl3!dIP=O#sg5f{GwRz`8s5
zq1z^sS6B=+2Bx03cWsq-hX;5JYZK!?@6lJKDQf7!+#T$A#-j_Mh06u7##-!RP#yr9
zuq84k20XaeQX>XD9(O;JrrG5Wn-u%Mrq?rgJ&Ek?WFV{d_3|h3y)=%Ni#C6^bLzB<
zRVnjawpbm-m~Hax`t5$~iNBLQN$l;1pxc(8(v#1LJ?Q%T`)6r{%KTj{j}b!YBQ=r*
z_O3iZ;%oCy+yPIfwW{hU2^37Iqv?s`{0ny!CwPuq7G&SP9kJIgNh$~eg^SG42I=oe
zaz&mg1V1>VpvY+0ut$)5Yp>MQRQ3JX3zW;K;tXU6K%~4M)Bq;;B^XNz3rQ)(p0@*C
z0KErz5y%cq9o83X0us7i1Gn<a#N=elMtydO1Se^!^3>Kt!3GENZYr~wY@-awtVD(_
z5;tF}`U~y@9|THqjcV}OgZNf%I)dFAwDZP}-s%_`byRhT@B@d5zBRip&Yz(;6q*i}
z8D@o`P{Mt4U0GH(oEkBD`ac9`n)bBQxboQ(ZU#xvp<~fv9a4XuyaB4nB8G=Tj;j(=
zvsI@2z&q574KJVzLLI^zE=5bsNdmdYunF2`F{pi>;|I*^9s#KYx3N99Ymh|ZQHu~y
z0~zqbVhV_qnR)eb`nyHnyzFcfO%WOVlAY(-!|lzVOgn*?^#6yo_YR7p{kBG%tda%E
zNeqCJ6eK5c00TjUK@iCxAd)2KF%bpC0FnhHjY!T}k({%VktjLmaQFDWzjMw#_0_Gq
zRbT&Ml?BZ7^mISZ-fOMBR%r$1p{+W&S7-WrQk9mR6%c4=wfB=r%*|c9u5FtF&oCTs
z`A&>^AaviN@K@PSy1xE`ye7W;(p%6qz`wLk-1VFWqK05BqbOrM&5ewV=;Tu^iC%oW
zNfT7BmsVWHpk-ggDJxsW!B#umQC>DKZ|E14|2G`~8tl{~GON-Wlrr9hY`Wn=d_x`l
zJ+4H3|8uToAVOesk%*Ngv@-`s_O_OmmU`G29zYtWldok|v!RNxekFJhYD{DKczMh3
zpyTnfftXhgy`SP-J3NS|IW;rW?z;CUH7Ux%yY4lkn`-L%D9SVey9C~dt%%)3uqcJH
znp%}QDvt8n;T;f#g}9ns#8C)<VQ@W1+Enk3o*G*t+*CnT(2|zxX;-)x6n*+MMobMU
z7PT_*Sw9z@YlK7NiE<jJo(}glE{*q6c1x75*D>3&d<+Xyr?0e*N+Ccrok|B6ZAg9e
z&@%ulva+9j2sz1gTK4ov3JWATH7+jgEO^ui^6Qm6omx%50IM<?x<U=TOn_0{(xn^=
zSH(3<{~q@UQ!V`Eb#!z-U_<a%KKn1c+O!-ZMqGePRfG8(%Z2OPmW%v~N(S^{H{0LY
z-(%-`sN5}O^q~%3^#(MkEQl;^m?+ZY7h+C=F;|$B0K@q!Qn~-;s>7S$U}i;VfS#~p
zmS6I#^YHQMA%@q@x2x8gE~{iuw+Dd6=vQCt;diYJ?HtnTtQ++ITrb=uOU5N>{O3rE
zz>!h^IsR|`@96+5W-kRSDk*V^I)LO3mO6d;%a8!b?8h$QnASUe^JFTvg!A-Bn;2wU
zp8%w|7ThLDi16_cO!ps_n85l97+?Ihc$MOI?dyO~SjO??TQ!2dn|>2E78W;0ap^rk
zIhbSF)<KL{!@eT{nZsY_pOC*huN2D08Xp;X5;;Nq0!Ku^;-YpRjbNImnr_uDH*RH3
z022Qz{o@sC&1-xIOpIj2xHIDJhBGPxrhf>~aiB(OZmQMJlok~9<~Yo>t+c-PVht6g
zW*l0Yu0DD?1Y?@5CyrZzs=>T>%liGa<lqGfC6p8-)YQrRIg-ex3alnJh}vN6xhJri
z!RQr&3F0q3F)sTlZ5h-DVy-J3h+$8fIv8~3h<B#LL@1al=_9y#QBXyhJ8<5nMb`qQ
zLy5qeTMcW9wQ5fepKc)s1K9xUi?>f;2CJdzc4WJ8=sbDW1TFjjU$Wxx_luiwE3iSr
z^-DpOJbBbKWODOToj5oJav{U0X;xpO9oUx0b~)Ej7G)36>;>0U-h#HOmDo-afjH{}
zn}#=Mvwi`608o?$;whAWcn?b5t2n^l|9}~D5|V31iLysWPyzHO2)mGSs)3OttZcUU
zJ-_GQ#`2Kd#kT8>y+42M1YLhYTmj+Q@4y4A4c|`5VxRJcFPV;s!{^UbVFvhOlKEZ{
zmQj2nO1H&eqLpOKC1KGAJ68iV)+@^>VoU^r+n_aimw{Z;$vhLb3_E!Flc?#_Wds*)
zz~1PEWD?^h+}J;-rY`4*pGM?R6oau4`+kp^wER8QImOpIRMmd}Pz|2k<rOg4X9iTm
z`mabXMu^L7BckG8VS{uj*ML~Jx5=>34y|l~W_Bb>&`7?C)kQ(!)n%!`a)`Y?JX#w9
ze$DN~8CQi#0ic5h;quvmIR_(u#N~%$(&5;4T#OyrMjn1v&A6B(Flwg|)zrMRkaA$`
zl42?eddlXB`q1OlVR2P09i3gqm#50%PWGBvYZg#zjE{DOhG`qNPt)3!@tS#DR|HC4
z^T38>+!bKIUEALPe7_is4dWrOLCO5HLA9<HJkWHy1KS^LzJpYkR)Ahw`2cp5w(nZQ
z;Gs|ajxPX9yR6IMX1~v`C9+0bUV&|IU!TlL1bTq|%<{D+-Nm-Qv?r>ERk2^?KPv?^
z+-LsHCj0x-o;&}?{~f--D(f8$y|D`CZOoTE^*3my+%D|e-?h8@BA?edZ?z{?Gkq&X
zwRq3?liJT_L4~JQ9&)@{mBmrDIg;o8{)vk{A)F_`GCJ?HcpWc4kLK7pg-KXkDwTUy
z&(hYn+0W~C<XE<b`xWX<THMl@oo)9Ldqh?J1UaYhc|f?;94asa(8!+p`D?pXbX4jh
z-0NTbX%Pkj!%+WnmJ@_=d;$WatSY|Z4%3tShex4!Cos5GH*sGH_xVh#cm!m9bOgBt
zf;fiv0v&r6PWyFhy8*4!Zxbh3#1{!6dvN@I25{}&yLat^u>e4_WAYGyX&O1#aPx$t
z06rqod*NVhO9Ib;66aG{j6Fh(h#8>SpNg@t(}O#i67>!IXVu4eH%Z>mmEr<xHXo;)
z4fT026m0RGN@+ggG~3^|;l}$P@E^j0u-TV9)L{#)hr;g(vYb9ghy*Rn4lX*&reTjy
zej=}?N1Y#>SY>!7S8c{n=SuB%F^a{eXn!3m!B{;pEKUJmMxv&PN*PCYqP4(tNg7=8
zIefqBVhT6tA)Bp=ltTq=0L79A$H6B0i<XzRrlxT1%(@$1OC(&a$WB#1s?bXFe|GyS
z95H;wrZ(_Q+4t?Q`W2;TsxKnjjvvNTqHUV+R;sY2a_D1IhB~ht76SNu>X)7}Jq@cX
zY_1Ks1@I04Jleb~`^DOM|Iq>{Ud&EWg&;7`wO9zEsE24Pc0AAKaXU}8>ZHcGLE%;s
z`F)wb%3|irHG**@9Rq!RTZ^w>zZNPTq^y0UxciQgJotv9AtuSl`Wvg0^9<nKb5_dL
zo#U^|Sk)1HCHx7;c&xI1p=(ND8@?$StqjqKst|q<_L0vYljv^T?WBHcjFy}l!j{d?
zXNX(FIClQmBoS=sRY$##8I?gnLF9-5J_Ldu(>6~O-q+UN5kQ7GpYB6UVOP(vH@CZn
z7H_F@SFXYrW>0Z)yC$3b$bxs_d;>CWzrHWGa{FAZ`I~R$^-7{;QC+m~(znhcwYNN%
zk38!prvx%It}}AD5bUrJ+E5?xerlt<$ftX7E=^Id9#{P&RsuJ0&cfPSx>b<OzhMLb
z7&#?t=zTH`1y4|yNa=VTPp}_15RBntK#EC83c&d3FcLgZx@vXm^t-)WAgh;^@Hhbf
zLHkcwe!P|2V7rCY&Q*f#kChSniOAI@NKm#Jn!Q_~+QS0&xMPt1aX(jR=xFcq=o7NS
zL$Xx<acQly$Gr>8?)zCHPl^}6nAF#qp=>#*8YpbPJbnN!kZ)W!qUS`c2PRJuX0Y?5
zc8hbI3H5*-n_yrd<JCZ{YU*X&Ld8KPR*6aSH{JG|$o%r9c?a~^cnIh3S<3n$T}bcY
zs`pU}@DH1vnU2)c^+mHBOx~%KGmN_AOS)lE8dkXUHs%|tx8v&Z;Jb$UCF>E7<%$-#
z1fS}&CX{S4A8fXNn!C^`%r<Ek{%co==>*|GC<oiYPs#Z;DGwR!-PuUi<pfK_d$puX
zmQwHD{q1#A)7M3MWzJ^(xFlk-8oQL=R;C&n?<PQPabk2qG;0DRt%haMhtm~Jvd6pa
z!rj7nxW!T@{prv%ISw!>>Nj+OXX>^d?~3A>o!FY@8Ek1@o(SjZBZc`W2`6k`p0Rbz
zJ{xtZkFAzgA3a{afwJ`{Y#WJS6%?&G(7^c2r`bj{Z+f6vfjvH*`w6z%8`C8C&r)eR
zU0h5}O{-_t@etIcVKMTNcb<}6-XxnGf1bA^gfQX$l4a&$2m6Vbdw5Gc)RAl_2={6=
zso=}}r{I7vJs$}D@yQ18%a@Wp1RT2X5npoVt-Bmlt_{;SZ>3^wFYM2BP~a40m9b%A
zS2C*(Shk1()e-aD5k#v0IEJW9I;{7F-EegpWfLY*JBYJAxtHZS(-5|tF_cFt8rye^
zJc}KucB^BFqe0Lw4lZqNz6;8fTTLlSp!Ojks1drl`s3v-S#m;r*~KN3+IN;w9!#EJ
zJ5EH_#M2iMz8D?#&fnTMr)&!85V;18!MTTHtYUU!o9R1Dk#TX}o`X|QFKiTWe(;Z3
zvDrA1<FKpJ(|K#z@#%Vf6_=XiV!jj4hW?Zj@*gn<l;Lvdn$#9imp%y2*{vCbFG{_A
zxbIyh+DUF!g1uUZbIkyXryB6q=o}c3xQPOV33bWnqVuTh(<)E{lxm@8f*Pta^(zLL
ztsl+BeY$kCF|LJ#9p&ej>$))X7I@wUolxL*2-~+LUD3FEL*%4#s|%T5cLu*z+*0k6
zea(KAY&%AXLQyF`#<x;LSeih#bmeI=OS%w|tx0ugfBAlrP1R!q<@p03F{#zrYpBb^
z3SMQBIQD@~VdB^i5XEn%8_=MP3W5>{(bjqJz};Y{vg2W?8a*v78C_UCj$7SXIfP8a
z20;}Eag0iYf%k46pwCxOA#wgaX>-0W9Y@Hq?pNfG`%^3{7wyDaOr^K#By6|Bv_^?K
z3y+(Rq?IxmqWb?dELo&u@1VY?CbKD=YUBHO;+x@$Fk+7D7t0WqLeb82NaG~36G6Za
ztpeHl70oS<GR?qFzqC1`!L+wYH5x{x;EmmewDCts^nwb?D8(`a5hR9$I-+WiRd2E?
z%O;vZQ9Wbty@aK7`S=6rGOHjrown@Y62{0Q)xj;A6w7G>26CuhR&F5Ddz0%1Yf2_4
z#YcTXWiBBDy{^2H2S!%DV$bhSPqRiIN3L&8C)XYCD<H)9)pe!kB$ay|l7?3+Oq4vc
z_3wLp=)m~h!Gnuvf=E1!_^$f6)!q|rA$d0IO!`f<UqFL+Li9ee5u}y?jTe1&N5|Gh
z6vOLxrn(Cha-cB_h8|tHJLBY;d+l=>!8PLzr>qMVB&Bbt?Q8YcEEf87cUD3Z5p?K%
zY=ZE|T`^Ito9q3eiw6(L5ea6T^J-3Uv6zuU8)&9KAX*o($J%Huts|NQ`E4-&PgRZ<
z;j?^|eCrOq59{+$g~7_wwlcswmX9|#?5+M5opHi8*82K!WvO0U#L$caH_3o(Esq}O
zt+&hiB!E6hU0wZ(8j23T8X^Raa8BWVi*4^gn^i<Bx3j-e2y$JS<AfVfhWa*7twHvW
z)qJMw!4nP7Cqu=nx_~T$v^ZOmy1jKWImjlLu72)8v?Y^E)#d;xhz$4Pd7ia~{eGEb
zg&Tq1cWj)Lzu544`CjFt#<vNoO7jdFFTVcCER79SK9Zk5w>nqu2fCF(Boo1yyV!Wq
zR_Aptd3-KNE)lJ)#K~ZIIN1qs+>S4t)*qYz?O!nr9s;}9VX^3utcwyZQw#xB)PaV+
zzW!}xgSQs&xdjM{i4d8+*(Ax#vLy*`L5QNQzz6mP*1(1B{X;~sG;h}lz<8a6+OiQ9
ztx@QlcW%-_gr<~FR$*ZwKZth@v?GxRcvk#<9Ya;RLrWTSUgHil4nsCBx@m`J;2}9Q
z{w)99pC>i*3!PgiN&&Q(=dap1ML4X-%+3*BNO-g9Fw=*h7VK94+t_67!i=&II*UZ`
z;U_<pD}6CbYFXTolA_^4#feurq$kF5<gCa%7)XBO`U`)5cgnVJ%-_Ibo`Tcy)GA2k
z>YaNO(ZdJ_=apcs#ap95(=#S?FT&|Jglx68wXv~`sNT6VKiiX-Pr<K9vaLaxE@3Q@
zd>d7C@e(NztnVH<4Nq%uGW0lIzFpXOnJzS1_-LUBy@|$SqaFK#IVelm>nCXI^HKKR
z-Cfn&J{|olHGY!<Q)e-FdE<ktIW9LbapVP*3^)jAN3>g*9TChrjD7X$l@4}~9M>!#
zM@&}w5RK`e4=-k*(P~?62vyh|Oe?H-R&n$3Rb=GJ!Gg#Hvb9Huk~Eu|s`(<fNM%j1
zmC!oz0*T!NM~{X(dPQ&r->3eh(Qv5c3J8=R7ILfoZJ!b~PxN;9*1Oe9YtseI1|K)3
z7bHdM=IvACogXx_JKG%Cq7RDa`vxxIIf|Llhs9px(-C{n^_u4tYf1`>=9T_(adYqB
z;S#v0dz>L-i%>7PvS3b*wLO3SJf7~%88iI9{O;Y}+sGfJ3C9PIH}2EX(TzZ+@gP);
z1{1?BkTp(<%ptTT!%GIVbaapK`Zym$V6eWcjXt_!dD2VLB<hi?>c~%3?qvVJn$~ZW
zh}>Dd!%?jY`}J5JA(t!E)J_J~j+r5rBscC~;6CM~WS^6eQpr7FY2VLzwd3x3ZMs$8
z52rEISJ?@Apkk>0#7174ylgeo7W$`#FcDSysBt}2?LnahV@$3tn*RDKGqqPEW+A@#
zNvBY}UaqNKe+5oH%Zsb*VJ)e^x)VSIXHNT!m2=tf<EUviZBnBrPfH#aIpvlSP^U=D
z-Wo-t(K6@?COK&gd-~&w!9}gq?mq-n`kHfRL+jTch*HM>wpd(n^C6x*`%E#)py&bq
zQOO(rplZ~W=zP=hSy+C~!1>I6xYbN&^D+O)Y8ObcMx}8o8fWu!b6@&K_g-0ZGA%O7
z^je8}H{jczD%pDLAVWI#mOI*DENfU&C@!?hZE2{`ki&$Su*z?&$IY9_qo?P&lf!fc
zReSw~xd5JIQrQW+21(>myl1>TE52v%%oAPF^aY|gK2U1WDnWsi01U=$v=W}{J=7Nd
z3H+bP*wVql!AYp~)!W19+AEuznw<R`xT)Q!s#7{D_Yo;8hPqeab348?ow%b5YF$5c
z@I0O#fz@Q8Lbj3uI~noSi&a;)n@8F_IOd4ak>!r7MLlOFId*t3?RZbmd~%@Ob2Q9$
z7%aF$Z(!-7oLJfPY^>k#FvQncgiK?TqvvDLZ{xfPVpEcuALKdt^kE+dluQnXO8BhI
z+xLBq#AK5Kv+!IsIb_iqwoOfG%w<b`yHl&)%853Y`60lH_t{N9tCtO{PA@m*^Z#hv
zQdY!i!|W+WxM4M0Bv^^<xNr#;9a4n%CTet%EeWq$;6?=7s;kr>f?_hdEuFs<F4>1@
z3)xS_=boQ!p})p7^3iMNddNUBDl*@*vq*FDB(=JQ-{>QACc%DX&t0D^eV2K9Q7%b#
zb%XoY_U|ZV@f%3-d{ywdz-jQz$+<n1_VROJT=z;`7V@~xIw?LI&#B<)8b2BMyx+F+
zkOtxPSHB(Yx%x{;(_X>xYk|dDVC}?6@ow8kx^vKCB;wM?O<y?WRdyktzLj1RFxE|6
zih_H`Le85A={V8-!k<j|$1a&B*IEs6;?AJpdYe14t*W0pcGgKE6~BT0%Hv+!%C-^>
zjfk`EV&U{Cbrw}^hD3e2XVJ!%wowu^$Y0T+eKmj(>g*RR*U;$CO5Ius{mMF}D_#B1
zTxep0(yC6W)MyyHPy~j^>l18{s4p9Sx@2$Z?~d0gWGv2jm&5$LAUQlYRo}${;@;y%
zKV-h+(f#J*01fq~SzAo^G5%~JceIc8eEZ6IVJh{ecfQ&WF6R%<e}spvGXNKZJ@<w1
zg~{X0(;6PoQe`|I$VYzIb7^{o4WuWwzI2<JE^!!q`yrV7!by=IJwqK;T$cdFQ1_|%
zd(O$1uJ8?*-hQ$OZ(UlDx11(oq)c7gM#0Q2^HfZ5u~ppVtJ+iH#P{dCkAv=72c~?L
zm&vqpIK0*wiCfEHHW~hocU>SyScKr)eHyw~{%H~J>*124HZns}s+xhHc3xZx$9KJr
zmWltM<MM4@wB4h7=`|{SEV#sAOa8U3QfCk023v?b(0hUG{JDy#9;Mx`2Xv1>qH>}-
zEpy;fw&CGW84m7EBk0{8nI~C!IxCNg)rQB0=v{<B)7@HI3<aG#+M24<&1z<7vuH9E
zQjO>S`HN7j^XOp4B?aujkP(4&WdU_a;WF1{dZ<!2^=w<>wn;RCb1*Xawb4i+fh&2U
ztaFNFySqu7-sQzMc({A|{fF_=Y$52<fM`VtH1dl>CEYNxM~xIi&cSoe!)e`7>QXHv
zCHDH;+WE~U$bcnB7_L2jANlTGlHTIGCt%E+3y>%Qq8P$r32e^=;fea-&9$=FXfyMg
z9Gc2ia_Wb&2abnpzh<H%Lq5lJ7;$WPD50KXdA56g#A1IRZjZX)FyAhRcIhGmTo(g;
z&DyvhI};-4ID0q7dk=gGg~Fx)O?wJb==a{NYHud8!jQe%?WED=dN7b?LEe1$kPOoA
zTP6Js|DeWV%jN5PwGIUH9#^Co(nf*x&od9TU--D`{Rak&uIcgL$Ce+J$H81{#qe2x
zP@L=oiEsNG3KoAGw>zj_f{10gF+W6=06`zRP@nwC!ZWUUO%26Hh*1AZ#I{3*Gb0RN
z!4jL&v2{mWT>P&-8|c9MNb$M+Lwe+-w2FPprzcWR|6E%eNboW*_t|LpIBnQ*?Y#8i
z&&?XOQ5g}BhE9~s)~(7@WAZp2@CtSMt=}etoD5_Vcn7B-75(@+68bCJ|2v4yFA{tt
z0P>9uf5pEdfI9;iM0aGVW9uzYlH`E5hA$I{0V}e3mU<P(hd$8NbwCcz=O5zTv#^u<
zh-E_9NN{3iTm8$Ln^&nLDG|=KKy2OBET1gE54RMMh?j*vkQ7bvGrA!W8>1Km$;?1C
zoHm9gG#Yu5E2*RncxVJ*k~Y6`r+gsxp>O8c0t+G2^wlo_J!S{&^so+kQgt02aU=nM
z8CU7LH(rR~&4=z4Oc<sE8d#Bzj*gWSz4q47cL%Njxw`{s3TPX|dTgCnx>*W36EbL|
zFC)<O`t3Kj8`)+9zR4j^G-4>IT|aZt30y2v&S{PT(|S0V9HC!Y4evL;>AozNRy}Vg
z<_+fl+3o;oKLm^QNX|8yzhZmb&;yu-*iTMR%TOTC<UWb~i?v!K(-2v2Gb>Q92d8`z
zPI;ulgWkYp*okL-n+z6>^D!^}?^7-RU-S*oFrSLJd3jk9u(E`L%+)uk(?OO03+6!S
zzhJTWjZ6D+9!s|*?Z&DLIPaYpb4N?l%Q0hoke$5dDfgVupFdytJ~50K5@Q08HM}1G
zn`bbo;FaFLIq3?^*FOihxb|4EJkWmEuzSRa(vB5Qa(NMk#eD#ORD+$ZwQ*s;`Wb-_
zzMvX<(8rAf!;B1#t?zn!t+$FA6dKnCh7O9_l?|6*-LqeCC!w8WFuOKIgWVk+f~l#m
zta6oca&)QE<EPb9HJUwlN>|$ko=eUxv?<`gLflco4w}IKl0u%!4lRC|$s1a40@?`?
z%(kWzK1M-gqk^Pb6fjpe5Gq9iP6`t>jGRwTV#pESk14x~;A(QRJ*WGRCbgRm>QVY!
z8dM;T@Dd0EH>U&M4qh?NJG(2^B+)rADEL-0y%*QrBA0k<$@Mb=q${TeTA~7)>G-0`
z9n~v`ssF<PrnyXj<`P&3A9}ecY$e=iTXH+|I6-1|TAD>5Byi(@6awX&P;+y$3)dH0
zsMtCh4`He}AJ4UY(b<6VlRjDr(%wReviXJ{?XFEJA+)jOwifU00)V9?yh|~r1k$WS
zMx|9}wa&K}LZI@G_*7i*qAV7FD7}L@rTpN#Q`V?^eaL5$^HBNYjzk$ir_a$K+j#$t
zry>W`2n$91j7IYE-5imJJ*=mjD|dHyvwTH@7Zw(bau)8yOT3x(KxVUx_j$rRyE~>d
z`ZLB!`2M-}u*~y84JT;c783?Y_J8U3nmU|}up82_bdn(^c-2!+t%J=*)ns-*A8w<H
zrq3)bXYw1we7(pvTY3SRef4ZW%|`9XKj#Sl&V3PkD)<}*X1d90F#ZdRPq6r@-6fY9
z$mCJ{MP<vse;ZFM7LBr*Fx0(6x)3QQrjsn-oS1KQgS5!{G9RC{zuD*(1*?RkR>m<`
zKtALXe0zr39QFK0MBkE`p1#U8mlL;ZH;DeSz05{sy&o1yPz98F_gf72;>=EKR(g7X
zRAO=H&TLXBKM+9p2pQwm$tIW(dB6Y37#`^RY|`$u2+sDKbidi*h4bh;tV9LOau`e2
z5aRuE+-T^K4&J<>NWt<`@T_@pKb}RuhhqEkcB(a`nIL115+vJ6GZIs<Lh3V(h>)EN
zYFC-V!JPs&Po5AWrsn1ba55UU4xddMPuix<We`?~-sFM1)~(D%H}4IUj=?u;<7*5B
zEuIWN4vr8cJ0}T?&vXBWtRq^z_P}+@?u@<Y9vm1zaKY~f5234e6H&rP{2Ruh_r6r-
zKt-)SXZ%bQAB0w^*t@;4u?5s&cGNxBoaQvm^Bwx?zAOVhv5MWJK$1#J2aFhq4OQ9&
zhe0dO7sYD?^ShBU@ykR2@CelBE;o|Z6ysHQOaD0+7o*Dcok+h%P@gvJrF5VF>z0oL
z<`OLTgKpk>{`;lM&F?0k!E}W8C22VTi+d^lp1`%HG6-8xeJ2`uMnE#c&v|G=TjVcg
z)Nc}vP0RS@XBO}2PrwImx$6%j?czm<h1S|J-Tv(-`9aN9Uj!{P<UTq4WQIq5Vi1~s
zM;Xy{6Tkp1?vEYbr{28}OLi=lUwP8wgJ$Q^Ovltyy|=Jv<In+TcX#(H&#zQrx}Ci}
zCSEU_r{s&PaSLF*#_Y_>|H2K0<fCIl5Rk;xD?NYo=#hkkRm50jH<DlF^V<KRq|;o_
z>o#XGa0UuXDkyLg&!Bcz3;*!pYpAn<gnAV9AoDje5?S-Jx5F`(zxxM8e!H-j`^bZu
zpuCcfcgOCscV!9F*GC;GFaLd%a*WJ)atk}0Qd0+`Q``=Yvy+oAga^t!+~eot^P=o~
zZe2VR=g|cexl0HPb!qj+!(jE9VJL1QWFP^9I`!PY>L>$3K|Pdy<-uBW)y%4^vDkZ!
zfCfb*m6y#t!cfpv6f+KcYbcX1b??pjC5B05S=MU$B<uJ;cID?deS%#m(5$C8q~<LB
z{CNYmO+F{WV;F#>;LGXi>ZbcY$pr`y3I~4TyC-n!Bs6Dk8*G>fSO>fq=F(jyYVo%V
zWTDo>u7`sK*ZNV9*GD?f?S!h>R&6)m$@=q)^elpQLUvrBF3D3w#r?62r)39cira`N
zKB#x3I5~at#ZSM=FDTF>7<0!#Lwq!0J_9W2^~!S2*VtNZC_c6c36Y49H8xmCmZ;1w
zq_1d-RMCa?O{OSU8VZUrREy$$`ucV82K8<pAY^CzsM<*IfG=mG3rl?x@}3sJNSd%;
z3DLIx4f8N|ce-lShZM8^_fHrjn$ac>=MfCn`^kUi&=g_5PQ{|>MV^G!gq>Kfe6)3^
zYWthC${VkSzJj~T5>WP=NG4A5Lye)A9OTj*(jzp?#D<qX%xtHpr;qMrQ!)oOc9eL<
zS#3WcqN(Y&DAXN5;nS)`i?l{$lR(3GF<5ryww%E0pQ@;YOUrCu4<CRK)v4SBqN`xa
zEx&IOkbBD7{VVs>p;USS)H0xh;rfo93Z#oFj#$uUy1}-M;8^0#+-Ckek(4)1H_mT{
z)Ta`Sj6!*0qc%OErn5RDq%347OL$7*N5Yowd}v6#UcuuL=Aw~JeKoec@^Yz~si%jM
z2wZp8IRRO}if?tn%vG=MKI<1&5r~iOf%3~uR-Q;dtH103902qU9X^AZ<Uz?<Zzmb<
zK9VZ81OW&eKl>)W{K|ev(RXVYc6Ed*%-86^^Jlh^m-o^*oTpHY22Vj@;gT{eUjNe6
z?1V_w=qOZ=u*%vH0L`A+*xGuS{x8k$2`>TC??&6KvZd1Z%-n#<_E4*KQe9V6tUKzM
zp2f{&Xp@2HXdHjBEWOLc-aaQCtLAGnLYtshT)a<&Oo1+;2vQSQ$|vJ4@r#gu0BvVE
zch;%2Q-oCplKUaW^ezFjn%Bu&@Za>I?M0Pq)!&cZMD_ToJlOlLgXdIvu9|Yy5}wwW
zyRn&%GvkXLJ1d;0pgDpb@x{OTEh-Epf>Uo50v|X>j8N-DUh?uo#Zc~i{=IY=*6ciE
z#%Ic572eIPSvSlwvCdNUxhD|4j(gn)!;=@!%~heM4t_ez-Oh*s-BSDOH7+5J6q4#+
zu_8bO92j+V&B#WVo8lIS9{QYW=16>3d_=k1b8J$Ql$10N)b$7yG{?!eN8x7l_4lL)
zLZ~2Q3Cu3^$-WfphQ-GlRR6X)_*ya5HudOkz5IqfqKhYt{G=Q+BXF-75MQsl=e1lx
zJ)?9nu{dy*jA)9H95(aKui`EQ4D>l+ScJ-db-qIcx;<+3B0Gu^e!%?eVd1638`8OZ
zH(N+ZXgvp@!Ems8h&76uzWTqAw`6QPqCK(dno|E|`Cmremce(`7-lz=0Nde11fe*1
z4cqqmh}%qTy~5;GL<LdR&Pdd=-*!gOP=W}RK6n{iY771A7T(TvXj?^2_p0Q4xI1}>
zGD{fAvS>VA?_DFx>g%b62}=~iZ?G!(w$at<FcX)Hq1zNlEavl_j(C{2(LOU?+J1o-
zrYWIsh4E5nqs_VA5daREQKrHEwELcw#zlkV-BOhlzb-~J_x_4@CBfu(T(I#f%Fh=A
zGBs4W`yi}+=EdNmjCVZt`-25~CGwh&#ZRZ2e7E+spgwzr9wNf<z?8AY<0>ByU!#Ym
z@H4(x7Utx%lWNi*!kr%Ujity{U*Y~*V=JBL-59o{J#I@Xb(WKWQzV~GPFUq+`TNfn
zJ<T{YM?lIjzmogS>Ufq8JKL<N1It%-75!3Icvy!@+A3d~isJEiFbqhte*2m71~<XS
z`PffG81+9PwC@tFE(Dme_4RcwxcP-I<}LSrxxb9=&B3}?`IqB;V#Ay?6vdoIib@`4
zQmfONlY2hE$V?eWT~woA;L|kg%6+8(^z?)1NYQHg6_SeeWBt_;pYmf05!SB-lC0{t
z)1LZ_XZs&8V9-AnUt{H_Rz=Ny?C<C1sxMzL`}LjFkiiz8VH85&%xR~ic<vZ{d^=?N
z{<iRez#MCf_T0C>TOHTfiT@)U2J;V+$iXGYfGCiz5u=KpOKnon#C`p4wecS<fcqsr
zhuz;ZcqJ8w{phg5oFze#KGr&OrPmVd*wZ;_3O$q=v0gJiYWkI+srYKCL7Q66iETPg
zIrQ0Dl9mZv8)4s>k<2UmMfldPQT{GfP$0K{TK^@KT!#N68vp0?{C)ZxF%zBcs$ULu
zJRZ2`^moVg>8loKUAi8sa=!Bk=CRf5(>WIDCjvSA(>UIc6W){;@rYbr4t+v(HbO7A
z?47CXa=(>9q-oqtWV`9R#_2Mlb~4G`XFaJ&GHU1y^Kia13aUkms2DoV<`h1itCd@Y
zgZMnRmAsmDG0H&&_CJM3RE_Cy-zw9XE3tsOY{VbNjZwr#CiCg#c+Z`|k=Rr<??=C4
z(^V9Elo)eS&)DTnlTJuXbA%$Uw+J2dc_NQ%2mW-sP27@bJ99>jPR?Ha$yYOFQ~!0}
z$%~1Vw>i}6M(DDt9+cbZW-O*w>~N}e{N)7H#z|3$-V1?&$>us8WuGvP_WI6B<M(+F
z$KTI!#id=28U2`xL3uSyPBqJLwe#*5T;hKm&yOBd8$0vsE0(h71MkDGk*5j4Jm~^?
zWLEAE?+<^`KWV7bg`eyv1p8IW04t|H-}uUwfk?lt^^3U98yfRhx5tu4zothH^z^*9
zW>}ojUx4X}+kns9QA&!?rHoy5&z?$o)Eo3YSki&?(rrpg5mx3lVWDt(GjccS$o;p<
zoNu$5f&ftfmL-TtKhaxR!NynKILh(Hd1mV{c_g{9O`)fsRx`X!tM}^cs%V8~uR6jO
zo$LAQW$8S<XN|9Gk3pL2@96ty+!{&Wi7W-l?0vXSL9*1u7|<2&^^^Ie(?A`eprlor
zedXo(ek2<t5#T#AtL=KB<2C(2l+5P4R5>k;^9Ln9sPFLvh;82+Z}DTgAe?muD&QLr
zlmFfobPSCF<+-*k)Zyq<b3DTrx;VlRfhBKw+QjWMn&JBKxrRN*{(J8R;)vg~)cjcd
z#nxgsW9YxyD1JuuM4SDi-v)U~W1jc5icHEGrC!hAlazVsj<bKGJ1E$^`}`h^76JtM
zL&!B??+RW^ahul_ULhSF|Gi5!*-I}t^Y*ovVhgTUk{Y_5GFF%U{G3UhH>5=R^Thyr
z;#^TYDPtFZl%2d_Z)|Y7PKg);#UnyIhqRA)e*R8RbYYC2Appe!B$XOVVaBKBm1og-
z#bv|(_)gX;?%~aPB6W_e%58Dy<DnRo{43z`sl|tT^%0)#?~?NxJwM;(3!=O}VCk<I
zlxNVS^7<G*9Nt_n;<`q&c_)ijAT9cy&K?s-#KuqU6(d!}iEbYMYhw@HrarXexwb8_
zq!)r_c$bxEB;kgi_e52pZ86K6LHC89-mdSIqUMUq@8!-)HmcmhU1l%pR11!|c3!@q
zBq!YtVeH%RS80@ws|L$o2c(9c4byl>>$fLKjn<&U>+05ddp&0|9Nj-M9RBv9@ZDH;
z0|(qU1BdY46*B$Xr=S!(jahY`na#HBPWcvglMu0+jx^UF-`;-#K-kAY1D`)cKA)Tv
z)fQ!hxcA$uiVfFvR`%O(EAW>=O68#tjC_f*;pUgX_aOCG^sMW0)$x%I0Cxxm=#BP`
z+SiR7oHdd*Y3Dd2K*q1$IP!?$jhj>1bL?DTL3@mbs-#27sL#+;Hx~}5D|%unYd<2~
zdi~GZ!)=oNtwqTS6CG6UZ#Nq)?}KmmBfN?ipY`3HeQnZj#xR{}rN*|Nq>!p=_q<)S
z)@Z)(t5S+O=r76^-dBjHdj0I8bvQQSzb_MdC4fEv^;#u*z&sS%#!M3hSr_N}=h(O?
z{{2IUr*{25E{GTcHb)QQw{cw0m^Sx~`kZ395l+i8SGG4n?vl<9R%Hg)*`s7_;e7y8
zLog0);v}2|GfJ{l0b?Y#Sv7+s+b+Kz&H2Bx+um_7BKvf>6bN8yoU{jmBq@c~btK%d
zR&^|gXcTzjX^=PZ@sxnGI3zw>d`51S^T*Ic**ezuSy%HTj+&b-j=k8xq7ipe<V{%E
zv(<_HmVJH8tmI^(!J?w}A6FoEGX@Go`^_TF)#+zmKkW?JDPBHra26B1;=;hsGWV=z
zjOZS174-jLm2w8L+xlgmR9D4ZHMpnMa+oD)$y@0;fbpj6r;`dvji8j-LHsBRM)U0;
zXbu#&+2Wnl0v=Lv>CS!}Q1JzsuiU!_&E{0t%}I2bgClUI&D7FK;4)$`>0CzAVwU$p
znZ`hJ-Lrms<Y?v~_NyP=OTN}+ZBYp;!RgmWWGSIiyBGW{kIG`jHIiXU;xe)n!Y}&<
zltMEp$zUW?MPE&c_X&Q`4~37{bn6?_Kon;cu^>5Ad>i;b!vYtvb79LaYPqbLfxM}g
zjU2Rz@$u*}P6!nlaZyx?&fBz{bQb_<ZSm}6k(V{&+bZ3>fvHs|#}doKin(n~ul`nN
zy4Tx}*<zM2g!alTP@8kD99svYbmA<#&n=_o-!fg5X0)hUWtUIraPAfkvTl2K8a^u%
zA{*qr=g}?fV`Uzv3BC6&5vHA>8lKMEYwPPp11u*Is#CB25!CL9(7E#PK*zI9FuV7@
zi}p);`bkG#TflY~|8%F4lv<6def693>W<s({wesdhCW1`FVA$r@=MZu6ey7B_ATWO
zwE?ae37LWz)mSh4Dg)Q?SMAO*3T$M=+gtDi!Lt?m(26AhRp>4i_kL)@)<)--q@AHe
z;&Haq#6(3{M-@4<$P-)2MdL!Kz{sGg1K+O8Z}{7te6k@1ahSCkJ5V>{yMS9`&S}&L
zi-_1%+IM}{(+C&XCnDV5?72^PnLXy};VQ{H<|}-_*F}PS0>Ruo|5l_v5IkA}l=N1$
z;`#UWG&CI2OnxnI+e$*VX}57z0Qf9{m5~Jpb?3lV9;b_DPd~e;=y`RCOYx!tPyaNT
ziHZ^Q(O_tuI~Rui&FHfDN-h}H>H!)5*xF)O>ydt+>xoA$y^2J(D0xOwjmW>-+8&+i
zX%L<gj0{-OvtXn1vQPNZr5lT5wvoe|&zRDNA{T9>mV23Pem!wf<xEWaS~d~Vk#@6=
z{8BM#b*70H``s}=y!vsTthVTW^UNq*QIXg)!B`IZN;XtJ#Y)2AbySbQ`}SyhycOoO
zqpiTEu9Jx3x30QyuV+Pl!}dxIe`Sd%Q&}8E1;?HQwHKo+$s*<7mpUh=fLr_jIn-rg
z2l&r_Kui3;{wMBg0vZtEd+22%K@z!2|BO6dux*&0<^oUG8j;t0{rVn@H~nhtPh#ZJ
ze0XmN#@yoD?=#F7LcoyMV=<Pa`_w#y;0Z{nF!{$$>|nO(!)4joR`36Qzo)BYBMiXk
z>!)DLz&xN@4AH?p;NT3MnVdWrp5J_w?EVBC5o1BZgx&;Y!zlnjVmMD*b?QS~x)h4y
zTn=jRLfVMm5E(wldN&KVvVz>-DR%aMyu(^f_23K8e^StdMIav$?*-?wJTIAeBb`1=
zL?mtQ6>?icy3kDLEHcHJOI2Mw%UFH|wA!PZz3f0}7b~EiddkF8Q}_-08pUvN`iO;j
z;jZD|Be7?kmRuUp=xxLfJEs8_wA(J8UZ9;FoN0^oP~f0iA>c0A8nCLblT5yDTc@96
zT<d_u&F_>$8oF?EuJM2kK-Ub-x!wryIB|z9Lha@KIW}NRYmJED7fZ)^{cRF(Z}s=v
zAEwDyi$&00D?dK~xk7?>zy|9}c{*kAFkf@-QmD;#suFa?xUU;`r1=;<D`JLbZPy|`
zf(~?O(UMk4*mgx(y3G$cWqGcv#st;Yh@nm&JIyuz|N1_X*#M4L)YTol<!dB|at8d!
zR}u%Hj0B-LslrJ2ntl%N&d47B!5bsSz^JgT089j-P?aagE+HhSNh0*&hxVC$^Tfix
zn>=odMm2uJw-tqTFzu42lkR<M)MuX3**FpFsH88()|cv+zBSDLl+4S^t3t+RqtnUn
z^(r8AXXxmT@*$;_nX5Dmo?ITz+)pm*H0EF&$wl(wI#T^3fW~KWdmqfOZoe7|A)*%V
zU*0|*X%kG(wd&7oUmC3fX^im?e*xI5o2^3}V?ZYvl9iQ3f&+s;|6JY<%@MzcjdAH&
za{&Pu-r(EMH_xBGB5FMVbwkNRuGHpbBOg1+;&PA(IRlcm^^&t?*2AUOl1$uw_*2Vn
z?H}?vPbWBsn>7yFR*~cW<d0*9*kan$JkWDD;ng|-<&z!k>N@%&3a(hkYx{@$)*(Gc
zU+O1zOK`Lz&Ook+89T5lzi)Tn8U_^`#x0IT0HKURj4yq^=&-<67=7kY>{K;7YLXtr
zjg&GFsu_IWz{@Foa4{|mO9{6;kYzZ#&4Yj^N+{gih3)#qKqJq`B5ws?gqt5LLFYBQ
zdlNchr;Rm7q<isBcS(}aBU96dWMjtrT|cK4%LqPy@2qEoiJY@4-jw?<LaMS5Lxwu4
zm6Eg30S#ehHG+FB(UX6cKF*kNcdcvN8$l>QWQv>VuW`GUf-6lq`ouWaAN);-MwW<@
z*z-ys!<`hiYy_)<l=&<84E#FbX9DS1l#zBYI`R5yG7QqUQT(wmz^5Nl@A6=d>r&f`
zae&O<BT?oh(TI(9IGUZ$H)I|gY#H^dv(`{bgZsS0!cyAf8ujL?g^b|lsRwkYUU{Dy
z8v@wEECZ~p7YRiQ@V#bR8t-y%D50wQ$L(BGU`lwaBY8BB*DBwLAsI^HEbN>p$m>o)
zgR)L7eP<u47+EF>$Jvvyui})U^DkzqK13ihcta_)KCfl^CeK+;O6=TJ=ad3=FKjsS
zEg$VEaXcw@y%CDBJ6jhMDPT600#k;^!9wA4Vob|}M!i&{7wtd7b(V#q{hCT%f*=2T
z{BScyL>~cybq5b4@RNz#DTQE11ZuK-)H=ovaqmHnyW#@#xCj)P22C)QY4aDpWXs1D
z<$rOGbLY-YOx)!rL{c5Ek!044Go)gX6-5KaGRYLnJ;`vs=Dp%P5wKacZE0&;YKT0^
z0e1UvgE7dZP7H!MRZ5A88$2%pYVk7ChOqgkWo5}GndqRm-K63kFi(6|hP^sW+=Y@=
zBELjnX`57_Trg$gZ96>DRipGm7hjAnkM`$eV6))Y;cmb6>ci#xkBWzF5V`ZGEpJ84
zl*(#Xc^{U*p6vdyo&1+|Yt%KR%)ZAK7OqZRpKp~dwA@&PiIo;WClRoNwRFgU_h<jL
z`al8Y{JGh?zo&;~*T53!m@KuAf<{TPZm?b%n{5o8_d;uGF5poi7*JlW#ulif1%Ug=
z^D<jlg3;+3q!7V?mUB5n1%^_?q8-Fz+xTZr{5whr_mKyz^wLBtmvM%lM<(}i(lUZ#
z8iX^fA%4H(r}Qq)#jMID=jb8`<D6g#CGfsP@b3)>wu6Bh!W;0N9uJM%4UrmNH!>P8
zgi*+NFe53Vqo)te;r)V-A+2sYb6JZ;R8|8VOTJ?FNDzTN&&tREmNHS`nBp41s)i9$
z0uLt%=h{0^^rzr%R56wiH&%yr>|DP)Ltv?()zl5f)~`3R?h(*m4@(b(VlQWH>X#_x
zjyR~~|2epwKt}cri~e!|@;)Kdqyn`LjJVKwROoAm_jZ~l4K7yZ1T9@e%s0Y1a<VG-
z=B~r+a4>7UZnzmf2HOUOp;B8Zy7v|ANbwQ`#bA?uS6kaP-T#1~ngF|p>7SQ&rWsw^
z^%QW*G3W_?K9$^mjcCK?ECYe{anuPzo$-eEfo4@MEM#640CEr?uvGtD$>E>(o-v7M
ziQ#1nuo6twH#l=?<Or)DAO?)r9{0S0E8037`X)aoR!RDGTO~hdtV})ps;+`*zCXBe
z)Xh4(;JNk)z77Bq>Kn~)s-xE3k|)njd179R+a!M{+-S4QH0k}3do_Z5_228@Z-E(y
z?HIfRM~KG0bSf$19FqxFw6E5AAMMWQA~!P@B{k6KUcxb@PLj>8znwahTmGTvu&U=;
zujLJvXC#OxmqO=%zZ}r*+}D-QJq^X0{^-);IKE=t_?F!sON_e>=F~2Re<<;l7_3BU
zr4SmV0CBtXj7#;DxfEK)R{sYmcTlK5z8b<*7yPjgS(VloZf3e!c74aq`A}c@ncs9F
zmBK0ikRZ?xuLZGt>i9TU2cLJ)y0F6YPo7#xzW7CG`Ui~Q{f&fWGy6EXE?ho|`%86B
zJ+xLha}J9Sk)2IMFkyt`xv(?cAW%r<XM`Vv0wwWf7y;s2JM^&gV>vIJLyvBEQ~0D(
z%rxhn;ASE%iXT6D9)Fcuu+^(v3Xzm|o;i0rSxae<06&K9MmK4Vc$*~Nfz&UYVkwQa
zeeA+*Dm<<Fp1XL6k>AVXxE3jRLC4>xE`@ttcp_2wrAdwO0a=4%8oy@4v6*|Wh`xT#
z21}XH>Ex!N+^ZUxLE|Nky$|Q>pfE`^$*#Whp5szo)?fsx%$gg)bPb3XvEFpXJ&%E-
zlYz;{i5*Hv>q25<Bysj$uK67k@#wI$(Lrn9I{b#9q~^%(rY2&p%Q$G)iwTruC+~@x
zA4#b1Qh&m~-mjovyMJAKJ4KKQeq3uYux6G-w00*ApL00>{aHmq%2Zy4uwXI`lG;J8
z*%PK0&Vj;~nl%)#B*F)JSC8@JDv1<|^EUC_dhtqr7rfQoPL^PX-`M0eo?S$)mI^Et
z(u_4dHxG}EJ<vpLJtfj($Imui`eY$4zA!zn2Oo}hT&0hjhlz;^$g(FcArv4SQuXHU
z@fLWW*Dm8>f8)lDRHC2ygGH3ZznPxM^z1Y=h0028*Mv{28>N}}*SL+n9X~h1Coz?8
zot@!x{l<-ZB0j=s={;Xta}k)n<oLt9NPrVYHwFhUZf7xHG5)a$jKc)3SdsSFiqZY+
zPkGg}0cXvkr-v`~DQ)g$l1xP^jFnd(Pnhe3@Ag5rh&<%)Z*bg|UZ7-RW)>K!@{z^O
z)KbPX&)Ln)3-JBybr+l7>YoZdTyFC9_1SVRTdv&C{PMSq*#-**;Z0}d+b4cy8xi0p
zLsD9EzUO8njTV1A>iKX#%BAwCOR4MuWoU5n@y^f?rMJYNwdEt*Q~K){yFdib1^R4!
z1g5<m?!&dHXbnRUY-tTxSv>v+Z_pk~HWk-ES3Pm@r%17+nCD)>FIGX9#d8%!w6(B3
z7iwpDx)G6!#C#5G$TzjwW|&rQPTfeJITr~Z=~Gg8Dt0c}dHT=m=BcOb1)gEpI5YD&
zT8Xtgz*RnLYj<S7skzy}gYh*YyAv^5b5uQU$5P_>Q@Hh^jA+j{;g<2$eD(!L+O#>M
zis4Lu7B@W`hPqZP)Bdlw?}e4BOohozZkDoKM*Z;WNu^Jl+mUShIJ?*8+g}#xml4bJ
z+(-N{ChTz4W6Dv&&ym(OR@FQJyY->}QWieWpVz;%XT882MIZ+oXf)DBfKWHa$auNq
zlt=ds419e3%b5@J!PLUln*2TIwQJW#30Dj^x^=T)5)=1{T41NmzKaUSIaOHgZTz9!
z>M`d4SO$$qcvEn>{=atHB?R`XI!2*KG3G889~VIrT!tGWpu*`4Y;pG1-BZDIbJwq%
z{W%4)Nu>zg`P+0`d$zC_TSY-WDF*^pmOp22y}P(LxA?Kg0|5F*AkJq1O;{|v=;zw^
zESsB~qJa$>ooYC)xsm`OI&<gv#@VXblrKbMg{TA7pE#RKOQfFwMo4B<VkMlC{*Ij*
z{}pS(f|5w<Dl;99QT*OV+SqcVCnAGiXG0d_idR0^I464cW2eDiO!bfvus+VjkC28S
z<RXr59uUQ5k#_W<$hrJQZIIi^E7auNj|th|Wv}D;yO!GW-2G5*5bJWeJwG}B-u>K9
zOxVkr+LexP&6sF<PLncu&wrS=9VQI>toj1Pxu{s239*0awVaHWYdzEl?ZXyj^_bT$
zwJ{$*J{__2`26(GF-xZHVZ4*e5cQnfc>sf!Kl~QBrlNoJBRDZ^hU51_X>xfiqy6x^
zQd5+dzkelJ`Livh9{PpbH(<Bhg4;OXL`TaFO2;2dERBV3>(keDbQZU%0N|kqGgB!T
zbUw$#Io_g@&__)o{PvIU%Q}gN!Kso%!H<mF_QZtnOg_KH@~5Szm;78}(L0%D#&2n}
zGqmJCebR0r7`=v`;+j}<SUx-&J=$_9GIHGaD1fQ?AeFqs5g%yTsvv6AQdMiIQXbtJ
zr%SDFu->$}ZV<E9(uJ?RrX<J3)-Shli{!h@k@d*)7dV4YIQD1($XWEinJ~<F4=U0L
zsCHT7rwO<VuTUVL$OAw4Q5t%!tfZcL<z!vW?MIp?<Y)N)S8L-vg6@~INoo&z-A|OA
zTjZK?MthC2o}7}V7t_>^x%PQ52b7H=M~;f}_*?dz<q<sUK4?*o&Vdp$vzcXk!W8Z2
z%?e=zhUP!s&*<~%=ePcYKcHAap_};LLW4B)nkWko9h;Sln7^#>zWw~CafbPZoAwfO
z%3kMRiazI)E~G(@$qL8TUVmHpcmM<K1WLQ*Zq$f?0y})mfES-Wrb*3*$@VABxn0=D
zFb_A@932&PbOM1X00Zb(X(iVW<~l}}uj%#;Ee*Ux=X@^tcKl)JT_&S7gP%uPtDRVR
zS&_;47iKzW5R&kgIc!V@g!{st>!P+O(YC-RUA|0d(A;>eaDNe|OY~sk#tvbCY3oJ!
z8otoH%l84*L+cXR-;+TJufX1CX#}Qo%Y(6Jdtu5_ECui~=%5r`;h~ORY`T|S?)d&x
zrk}rj!Jt(iU-kO_4R(X0rkGb%x6qZE`v)_!K2&n?@kRExr*ack^KJ9Liyqx?t{=tm
zH9vpt6<GEAN`?!8*HY=^+dEJ8BmWhFF>!-j4~B<2Z=x&ct67j;kZ|N?D1k*ne8$(W
z<l8;6IwtWJqqRZw%*Q+RvY9kt0(ix=)l`Til5si=H*sKZQ{p`hLyF2Z)G=Jo851Se
zbR|W^xg_)KGKFY#M9ej2(HMq~pGyfB1FaCUpZK;_MwV3e%4aVJvJh_WBS+GfxJm+d
z%xb{7;L2n(Y$;Q9SXsiC$Itv}!J=nq5uz}_Gl9!d+|9O5rn3^pPMWlfZ1iINA3C;R
zPa+r7=6qn-UeZfubXnCjp4$!hwBI3<Vx&ge@A#S-IE@$Y-H4-U*5p_@&tdXuF`{Le
zlt}%|sozB6FQZv}oBPI$V}>^QN+q10<|f`<AY4nS+b+lJ5YU_T*r{|Hk#v`?C5Fu&
zcWky~OUfCDf>6(hbM5p%P?{`EWBihdZ-)*+7-81j1hOfetxVL5m7?Alp~n*=G*lF0
zao+coImbc~N)^VAiay*bozG-;;oMhfREK(1(beB1C@}m>lQicZksKQpsla7U0lcdz
zdER`F$8)^OeD(E97UpP`WaN!l{qydWFY@qgSjtVriuNsrnCg9eFZRzJgmPq|bM+W|
zTC#*UrrDe7DKmPAE1wZZ9C`$k=4TLj<aTDo)<8}=OIbw85jl9vB(%?b(J6S0W=22(
z@}UZgovT91$q-*7+PX5e_bl%&N3`DSo%tk|oVVFkvz&`BsCn#|uxzCJ$V!lW;=_)+
zMa&`(qn1IHUD1h3R$JL;6t@)Fa4QBJ5@3WoOoJGJviX0(OFXYkwhL`35bf}S%x*(Y
zbZ2Od@d)$ns4q=TPiu3-$J5fqR6EjI?)H%fG|*;61kF?>cZyONQ|0#@+_*=e-lEm6
zzau2~yX<UC6gw9W#XnyQH7kfPY=GgOBv{-|eGHS!8#c?>y(b$c>e#v^Gk&6M&Qecr
zdM|T4=pGBCmrq{Mt)?zMbMjK*<-2eE?SEfP6tpi3WSc9=87Z6I+hsp5OIwRV;dxyO
zVcSfSZ`-`p46FCwBmO&}PoQ~H#RrYHj;=2A5^fyORbXKHxn}V2;6u)}Ye=|1FV5Iz
zVQbLV1-}}oOa~Cbh$1b>N+>jdy{ii}7yP~#UsX3bxgdFY|HJQBx-I^y1S|S8eEBMx
zOPSGmEB)y$3erNn6Eha^TNmhLc>p+m`?s>jox{dSzhS(+tKGuVIAgv(G&8kjAb)A9
zO7!$Bb}XtP$0HiVypQidcJnNt7^9rIkG7YFA-)C(8oB+}uay#yd(w`*Cc~6ei*HVB
z@7v<r4Q)`<R%KP?Oiyk{nTV}kBfdWsT8ZsiUQVozVmb@=<HGQikBQHXzDmRB$SFJR
zi9CnFM<Ei)jqk=QNyy2!tCx#}IGv1pmqIC|au==DT&ypWH%cBXItY|%92`kX>Q(_`
zS{^b5zU#8n+xxmSvG2IGU>y_(ynzIPcDDOKgq=A6H1^J>aUixmxaM6Yo6`UOy(hD%
z4agu(O=IHt>rbIshjMs1U3L$ThyX(Pz^ea&Dg8fM07>W9PaZ!$7JUraJeLo>eLALZ
zvg)t9N{J5QT-f*c>0HYau#BCxHF^an3GvVRB^?fajEuCz_y1GtYFnp@f+2a=6o2Kp
z&+}9Wz1K#k(hX!Ir74`;@S(t-Mvq_w^YYm%L?c0dQ&SugBzbx<z5!N>vCpj(z2>~_
zbr9bH+5lrgs~!e27qhXjZ4n~_-7$@M7EW%qL&hP;rawAet;-euEQf#6&MRC9jP@7%
zR;ANZ<mj$}zAcTG&W%y64@^3K^kweu=f?T`q-|Z4;vmD6w&Wsi_S6fUNrvZL>qWkN
zAr3cV-DQJHxz3@^%C>F^oe3_CgXm2oggtEzM@Geip!if<oU~+rZUz|eS^3qd-_z7|
zd;l%&h}hO2i&K~gORnl4XKN-m4>v2~RLx*|pu*zkiObZH+|qhiJqh_4!u>=k7a|s0
z{iq5#uU>U{qc0|`<`A2y?JM>)NsfAa_1VHjFuz3~M4m5W^%6oKNp106wEO%sn!mlZ
zv;HR08@jBs3P0rbD{r9r@|Qgf-tf!1KtuQ(@CRQrolUTC8!p$dje6wlc!gnpgRFuT
z*2=$tL+5XPO{b@jbCen4)GoE&f`?RrF^&x)tI>N*oGsYrB>_4RH`?kXuC53f>)V?&
zHxgXgpAXTn)|fHuU2jfUR}5iT)#A*ft~FU4DDXgD6Jo&O?K_}^#{wcsiNb}rau<;z
zL=x^=<Uzbs^YqR4(|s8_-DHMT$ilBY^M%j1-Z^h}yt?-(Bcr9IxtWvzRE~p^FkAMd
zOek62%j+<hv&;Tn?=tX!%YeWsfKV-t*iIS&%Gax_CtX8yfCcat?u+G(4@4mcKqiDg
z?yB`bzG~1og<n!q!=55;DH?T13epYXUt&w*-CKQx8vS;0zvoPwEp3jCf|@y1nzK7i
z#yb_Rr8edYAC*Q<y8nXTi5cJyhK)LhL=HiKRjYUD`+u5y>#(TWw{3Jl0TmDx5F~wQ
z1Zkv0rH1Y<rKGzZz@kG20Y@YxBqXF`NCD~Ym_b@<Xpr9b;PZR;`|jh|$M?tH$KK!i
z<9SAA&06=p)*aV%UgvdQayoSYjO_brw{E($`mv$1>-m>wl(cKzg_G)b1zgqp2IXb9
z8dG{l3Abvjs(Lk*xbe3x$o`}bO}|Na>HD3a(k4c_p7D*4-G71tR6#)C<B+Ve2=cPI
zB>%*<+XD|Ud?)h!SRpbDu~OMC>!cWvZRgch0x^FAOosN}rSWpkwcy`#a(l_X*@I{f
zAyDRK-(5*fYC7^xsPf%`fAUI{UNsE&7;FiJdWh>?K!RqRr~mkZ@On~I?0;cX9bjyL
z3JvGg&9hY$jjsWq>j<*)rGWm5^FuB!I$$p2K`4UaIdt>K0~}_+h&R&NCxMs(9xgwr
z$U)cvC~|2d1rm##A%F7oMt$qh)WDE?Gz=mQAZj8eeCyiuwO<r~<w|mJa3UYJf7&hk
z5Tck7lkvG<yW{Vo4j_;|*YIFEfB<(E<?#bipu0d`q1m)Z55S_g8f77mfc7&r^0yPN
ziyZxR?2whUyc^^P$-bhWn4C-t?hRxlK5cbG4W#537ILlSOZ~eo0fT{<bL(eXO%2cu
zkiX)iAmfD_3ehM0#gQQwRRd~T)=oqhtCcxp=ckMC%ka>bk998ae?cIGjVF1>AD>aC
z?jF6|ZCbha*FpmSJr)40#|%I=HG$Mq!CErGrp*h9VYXODAAn6<hdQV&<1!|9xeKp`
zU=!M~NBRfE>0T;JxUduM9?fmaypnstA4Yl*4ZRo#@CeRs|5f`39*PHPSEun)+9or|
zv2)OkTgjxVW!LfmMdgEJtcb#9C|VzsK?!}{f=#Z&9&FXxRyi33Wc^}d;m9^YoTbV#
zg^_n8QqwCcdEks3cTTJrJ2*5nReY@{R3J;`fupvtN{gO&1Y^nhL&02d5q$iAql(~i
z`NB>&3aY3lj|}QT!#u?Q>l<6FO&9zNE?g6tq`TeH0VM?A3p4_|`R_aI;J@~?A5}|~
z9QwjiA3g{`G-BcZy7{1hApRlFwWk)Fq#mM&EDy2CevtX^Vi00wEHP7OKLF{TSt@=1
z{o3IP4v3xI+&ww7cZ`;?bpqli^1UthfaA*zVmpt;cGdj=iHWK@@FD=KgOI&L3H8G2
z+FG}{1q1|?&uo%RB0?r}^J@1;jj)ACB0WTPX<!HIi6E9ESVy}>y>=1s2le%eNO2i$
zyvXmZO|AzyPxi_k@@g;6PqKxCgswnTCH99|!Odnr;QyYa2J4U6`pw5+AD{pVRb+X7
zw;FXpW2;|6YydPX*SP$~;cMP9jtefZZ0N828Ar=}=;iF<gG5FJP+g6Qwd4n@#K5<;
zRU;!wG{yBb9z8jufTc8DQ#TuW$A*EiRuHk%ouuK=_RJ6cd=tV@Vy5F?H#3&vp*Vyz
zy4f9?wJ~yJfrfm~sGN#S%v)kKoYoy7fzT*;2KVNysv`M80>R8rvA8?L>R_urt)1Bn
zq=Nu&;2;3wp4wF@MoKI`^ipl<-P76YcA=C_{YxXe_2Y}Vk3d`vH8eQLcLxG=M<mxy
zcl$1sR^fMGlgImOo0AQJ7f&JYlauR?ur<cIFv=g$tE2h00cZW^k3AiG03nOE&W<Ck
zfr{MgqW&eiHq_*3P{+Ukfk;fGzy}tm4Zx9XrjpsoSJ&2@$zrq9Qv>?uw_mc}o>who
zJIjR}J|TZ^aZauS!M(QiXH|sI>5T#nsVI{ifMA9og#<vpr!Ps=bM>b-HcMXgw)j6m
zK+5S+R{Gv4dwkdVotv_&I0g^zn(BtQZ+_^s3Vf(<>h}tD1SV%3a9@{FNk(lrrKWux
z0A08jvLJ+{bsPcp)2zHkQ*#=8ta@KNOS_QV?`bO_b0lr|)^+yTjPN4EF7(Izssmud
zcV*A<sU6q+aDCXb`@ySKof~KV0FvqXw#;elq$JId9$+0)x*>)%n)35A35Z&`c8cGa
zs2Ll^#ttI+e?|Z*ll3*DK>d}w%<wIxM+CGtXC~*2d@O*(RyyfrjIGr{v2}i4XOaCj
zjnVYk+fXf?R5l;$39)ZB+~Xg9_<-d=wP|NiBjKGhjeoh{@<Ifjn$-EX5uxdte<rF!
z;VUD)y9ZbFFLAl_SFf>!Ji^};Q{sB}Z7@B3$cfLxJrNcv%D>S{K^sCX)eV?>;*Gva
zfD9-Tfj79HWZkCDy9|)8F?<F$UGTuSwFrr%o$K}!nt3K|cC|)M!1SuOcF)wF6<`DF
zGWVs77;p8~Y)G;?S*w@7DjsArKc2)Yzvci_0eI8ln(+X4CI2>;;xgaM&PBQ8ByW_H
zmcA#evu5KtHghb5WV}PGbf0u?Ep59zTZY=wJ4hRZUMvRg!N4k@L4Qe4XtnHb|NOG+
zM%$m50!R=t*Yi&Yq=J?A?(Luhey`^EE;@~06>0_Q!a6R3EmeMsi-W0SJ*Ga~&giGG
zu+Tg^<;~BYdkVdMMR{b}oLv-x_{dU5dMPP)4JFNrA*qcLG3mT8g~Gg>(Oi|ec5hG!
zhg%h5KvZ5MAG<uKf23@ZZg*=rv}xvvns*{l`#?1NGg+#!xYYzApX^$wel+<ZoYiX#
zM@sFPKsv@?$IC%aPA05AxxL5ix4$q4|L3FB9L_ou*`GQ3!qW_~V2GM(4{X-)`OmJ(
zI2AP@#w1(2Kvk;DNubPp)CJ%m7&*kv&27{a5Nfpkin*EaU&@@EJ>B?9vWnB<?!6Wq
z`*?TWa`B&47vIHH;GFE!^*WG2)b(dL*!;WWK3U4}hQ3XCmtNjR<$QuzPN4F6ppjI+
zH^C}z9b97is`C)%p8N0v(mcJecZq<pxv4UIaJgNG_~1K70;Yvr-fA6W3DQmx>zQCB
zF$v0#T7tlI5!cL0&v;m;j;1Ex1F{<%(kD`lQIe_7mfT)pq4?>o5f2O4Odd!pU#v&`
z=0i`1Yc`d2Q|p+^EeBu{LM@wSKPA7`m*F*#*o-)T-I)~uz_8ba0paB8*X}n2Wcf5D
zymbI9h30!<;>6~fV^K2X$3w3>2RE=2E9L#hrQXhu{{?$wg4YYz9SV#Ueh7sRDIUXW
za8)Ed&Y!g9gyYy17x9mK(lQWj7boM5SMKomcfV!wQ|h*C5=p4;tR6qS@hbGZdSqs2
zU&hhB(|5j*fG6ujcO&k~x4C8auFh^5$0Y0{x)Xtiro$HT_!wGywejkfMdeiT#_kbO
zSBdJC?~ir3ud07qEha^DfkIiCd3e7*zm*w&Q}g2}>ULKxK5AN6<sY%(66+B^XJ{LB
z$@_;)=pUqnI>VKrzt`7io8G<Ib;)DA0?Aj?UA4b8UE+7}BWVn<dvKQp6O-m&h<g=%
z+Zb&5w;do7PX{<|{t#iEefHEo??6C*;)j?3&WTs$6bvuuRuJ62sLspbL7S>};aVZm
zi7iMfchw@n>Wmi+yPv+dyGmT?;PsZjm!kDbP%ZKOJ1~dWzXO{b#5Uup>)-b$L2i4>
z%LuiNn8toonhv|OP9%?CFDy)QnEbMHCHA{u&4{m$x=Ba6(%KT|X~S=t+UuD%ts|nZ
zOG%M1#RR1Xbo8cM-6f^mumh-SH9&4ZumCeMB@K3bC1=$&kqU@7(_h{4$L2hLC5uk}
zZXEnf{ASuiN+Pr)rQGWa*Zb+O^1RTsOCw|4r_W3}Uc0Y8s$Hfc?`_ZFFc2<w>P@Gw
z)jZ9@*1P!>rmxmEteOS^gyKFXp<5c8pW(68FFnx($Wh--;m@@NEqXi^M2l-J^~yS5
zwG+_dV51<Wh{P}7bF^V9u~C#=Spvf>aFL7}SSnYr-W-s8?I3uM2LUc{atdwcC5a}>
zZe3;(q%qgv2kFT;U4|p)I@Dx5TIO2sI)9b(GK4i3P-`!$Nj%aH#cg=$d(cDAEJdzi
z#=6-uWT+zqf`9wHAf#5B<~DJMEv02!s7FkTykTnN#jc7`NsFVrB(MS%C9%js32-3e
zE^{fd5FZf2nsuoKT+sUr6q%q63pK~9-w1N;?gcFa7(ursPU0@0oOWG*_oSIrf~yq+
zglmpyyBMBhCQ6t2y_Wtt{`H0=`m@A>TlLNnKQwGlcc?z~%x5T=A2)Vup>Wbtgyvho
zVfDQjA~ZN3OJ3a(Qk408>dS<zJ4E}-tJZ?NzkYWAgLPMCcmE6L=E(*pdr9p(N=8}>
z8B!)5S%Ex(YR8#kqXqTyQ!X57FB<xbv`R~YZ6xXeBkY2r>m!Ic0Nwz@+!F7g=~r8P
zAY<ornb$be`hVdG&1+g70@Y+16LMox5ibTeApJ6vyP5`o-1`Z&+a;W9azFzI`g5Et
z35XBS8-Q%gK7k%C=Wy8%H@+TqjC=vAXG+oMpK~gE>G^QGxRPG!Af;qR3Td|p3}f=C
z;}P%R1M@D6?Qu5u7>u5M!gecmA3(zbbc`KN!P_10Ji!I*;jqZlj?AvDji{pejW;Jq
zCngA4)n-1_!VQHffD+5FX7Y{)ziBexE9HtnLav7VX!1o)8ZIZp1*?+0Hh^Un&51NF
zRPu=|Lv=v_m;Qf%8(hm$g#QEF$ZecT!s0=IALiCoau;OC1At(s4GuyaIYs=k>OO<|
z#@>Va0%hfM>TGX~vk9B|r%?$Kp@Xsiygupu^menM?&j2|n_iQ0WPj#tZD!%KwR(DH
z`;+y4Z%4wvSe=k{dhR1#yK-%qgvfxApG9*CRfviIcc_A&zzTr2vzMUlcg}Bj8h--y
zs9ysh@-EEKrmEvj6})Y%iJ$#ZKS7WNxLRp>k8k@9V4ZVqPf-X37AI7dp>l|px$*%*
zIotaq)X6$YdqfNJncT$bc5KaXCgbYoJX5^JNUjDZhF6x({A2O22>Ou2KC5pEEX?v|
zB>Vz-z*GwkjqbZQ$m85?JH<`8B=UZp#$ygGGAxYKIIchCY-=~o!La6U2c|)Eq`iS|
zMrf_q9D8o1wr2(k_+B~A7Jy)tU^ODook%^VLp>x=ni0{YpKbGdW4vw!C<6d){o7Dw
z(^?-_PUpka_kbszkNf?B!Ms)<fb#$NLi(<3aCd{^&V@j7Hme$d@faMQI}i|LZSd@5
z&pvGh-TgUQWEMyf6avB0fLbh=E;%qg5PY0Qpc=5O_DR52Pw`4lCv&_~h{!PSxGPW9
z;eg1ySMpj^L8>y37Y8qJ7E(Y@g!})v0l9<Pl*ziIrKE<dTRM<y@19WmMgqWpbqXwz
zav)=_?F&fQR9&$PFag?}syaBoq5yW8RL{W2FB-VX0_c$rmC+sOV81%FdD)pP0Qsrs
zCWNqV<f@I#o}fAXB>k|N{*ITidw*zsav7FFU`d^^%(C(=J^eXJ5Cj5p4Wl>nqw@T4
zq-9c6M#isqF0L;;^C(HdZg8Q;=?cFM*bM!hu0g&t+}`bj_F%!)ieL9?QCg4&+$~~2
zJ|y_Q^*jE*p$|T0ocq=Qe=yR~*`FxHC8wZBbu<{Ag4_f&o-eK#Y|I}Fzy4KUvpuz3
zwll6E`i!%{w_JV8Is-HHe$5Y)wC+ZOTWbc}Y#m@ZIW$3dmy8_mrQYPE>kyDLXQ5q=
zzad0qSPdiyHrfszf|Yn=Z6CkR)Hz#-daVb%=PB~8649lF5&e@(hDI!Gb&2y&E?dAG
zr*LG@2l{K(4Jjq+peG+RIkyhX0qZhACm9ELQ~-gX0VPK43`RpBT)5u{fR^C6tT!o2
z*|Q%^&cFQLP<-6mCa|*~NEwZs#;zAq*E3<{TcX-aV?Bt#`KjF{JV#|L0}|ST>r4bn
z9@y)63Iv77Sh5CP3o3`y^8rMc5C{h(ZBU9KsxKdM&V@d>9C^_P&d;mDlynQ~_u>Q<
z2#UB2e4wa(!j!kyiuyrxkE&6az&R3J5L5)Q+q;cr7_~vf6C#UA2M+at;l*E5S9Zxi
z4m~cJoN^+BjN4+v@eGQzDNX<S-Jn&&YX3v)_y47Ti2#%IN*o9UQBm=&_r~@D5@g~8
zppXU6Z1z>~lMH}=XIV_Fcv$b)z}L&bW{`0qw@nGH0Mg=Xu@;%2z*l!Fv}CA_K5%_i
z|L*NLshEQs@1DfQsfw<AM~Q$*jV-$iPyg}-$T!A}Co?8n*#z#00<xA5gr<a>_%?_Q
z#x4TwlN;0-xBd|mqXxCku?hQaZ`u?}5bd$YT){;5N+y1?$zT)LL1PH8%+0tai_8L$
zCWu{|liz@`4uEHN1JgUP+~c$i$?u+g2?LcqAzdXm!`e|;RMN&^^eMy*JLfN^H@uud
z!gE%!I-3VMC=hikOvWps{S13nb{@%<b_n>{dGoA`0Du_>^?4{ill<C++`1W38QLBL
z@|j0DyC&_i$;#MSo|U=fBfg32GlrpAahU%Zl`3D>_${vhSa2~@;MdcW<@SNEU%Wa5
zn|Ywv{;<}?b9W|1i?y+$zQ<*a_6DPmv3r*NX#UZ|bUZX*%<f=={w3JP3C+CREe!>=
zA7a~0=@_w2>mxKsoczy`lRAfGmCFcx&rxjJ$yzr2>WarnBq($6@y@aR#^))oaoWEp
zvkhEAQ|UMsSZ7qDdi}OwWNtd-syf?`)y5CugW0dYpPylgHWSHlmltQ#`1c31JK<T&
z5INekY7>F#4!rpvw)L#sYPLMs^$Do;&Cy`*KfI~pzbzYr^TAtddc#GIjq(Vl_~det
zxSjP^Womfy-N9xJ-TUjsW#mo!&gGAylrNb^;2PSOyo^x%*Y<O^wS_B<6CF1$wTNXt
zC|)8;8V#)JtZhT*=1jvw+Bd=7z;TA_Tyi%9fRnL#e%wJKkR4geWGjQ1M?ykU*eFqj
z{W>Y&=<r`@<b=fGMB=J*Gn7N<a`W>1vV=~auGl^|K-lMGY$)m&6TO-WTE7;;$N@MD
zYh1|E+Zu+#lxh|fk<ulvZ!vc+*j#Kfy=BB>(4`CVBlNLV=}7lOx?_8$G4q4p#{$dC
zL*!v_w*(>c5X)Q;K^0Gli`(@s8cVWj8)qFHTiajCU8Mt(fBKML{E%sgs>X<Hp5gMo
z-!~w+^K(wB)xmAC#Na7S(?f`E*e`$Fms($9KHjuwTj<OpxwIs`<rt{OwV4ZrSleuf
zCsXn{W$O0Ff>F)c03&3N-BP^`yYYSoqG>Jd1y*91gW)3lA1@nhxcM?u`hV%#sFeXd
zboJZ5hBDzgo_nA?lQD>{LU<HpmT<1&Y*_;#nG>0Vu))3$or6iTp6{g9B%HEB*&len
zbMJ}AVwbsU@-izwR`?Pg571cbs^~2&I4arQPbc+g4{!8-F>?bV^z@&${6~=KySlq0
zVCd{RgkCCo!)_7Q{-4x)z$iQiYC+bJ8tire#IbLbQOGF$8$UgfInmYC#Q^j86WEV*
zbWBX@)?(%9uH&c!H>nD&p)}I2Wh+<2BNF}GTKO-8@bi>{=Ff2)v#HK%l<Dm%VEmpP
zp=_vOKWE#ZU%e@lxeV8yHvGO-tQi}@Mt9@aN^{irv<w`N{W0<&<!c0tF<B3=DqDYP
zylVZDXE@sXO%{59NBZAAoHAnoDIXho41f{>_vqBj6r~c#>j~aRexxEikJA#orNK7>
zGBY2fS3Vw?obbat2!Dm}ETt$=@qCo7zkZFNm~r~9k;SXS@ND`E2RG8&%!oWP@j8#C
zg>b7r<y^LcR80-4kD3}aj&6&CAzWtXGSjU6U>nThz=kC>HQ^-p2_b;d@dz81NR0~M
zR2&)a=-wZYr>8yRcNI>l|Bz4^o#E)^6FTn>ReEwZ2&VWhAs`7N0aT-gN8WW_nf81_
z4=8j}TYGb@Ht6;l?6Jx-iK~*4NH2&;n`yO|s#g!)>lLnZ>iuB|_=BE{D3|O_6^{<S
z#Tq8re-vxNtgH&L=v-#-scc#+0Ou_s!XTQ;$5dkFX0;-YbO-+ZK0)o(25JB_u-*&C
z`5)oiJ7Ll?TngAW1C|yys17l13@~{zwV4cTX(08H*Q6V2bz_5rZV59WVo4bs3kuXV
zz<)bVS1ld&U1cmbTNLLATt8K8xQLWlPCK6kYGZxt`lH?5AgNR*;38vmDS~WrTL&T?
z%=o8&PO#j{JB7EnH~O=H`5zoksU>?MMn7W*P#ehM;`|H(0W5l^c}gf*Oo36QD2W_{
z>;63<HgD^!nFb+fEL#)^fVzt&(B6~};ERr((Z4u6+I&>5a<;(H)(d60m;R?g{=E8N
zP{iV=r=O-0)db*<z5F+C4Ft(QeqGVf$+1fmdVV`Lnjy!3vdlKR@J3-6jhI0-))=}6
zr0*d>kL)?;$w2%I*TYzJ<m0+|578LT2!)$K`jg1LrkM54Nf?@f`0(Yu#zfNOH^&6c
zFB+upUv2H1nc3r}FZT+N%I6tAtCXOBC@R8lKd?TK3bm3^WaJuSJ%kCU@YA0r$vz^z
zEu7T_^%fK4^&TgKXu;z`2Pb&j+P4+9c-@SK9@kEJ+)wx_r*c+!!3lz>l&<FiL_FMA
z{xarT*r{s1<k7lEco{U^xQhEFda=7(nV;N(m2nJ2+k1e2s|d+}h-;geoY|?tassc*
zFpT?{OP$v{b5{t^O1H0`oxSVrrKNH=dU_Savi<8+Rxj)fNoLOyQn+s2`hfAFpg$Oo
z{@r3}Wi|WRy|+nZ`STlKH*{cxWgtPkH}(3WqWL8K72xn&%wrMz^lj5#%d>Yui+RFq
z!CrMNmC`g-XTE2lt4GY)CAuTw;OH2&TvLONOzt<1px}N4=#Ga>K|nxk57dX>qL(z=
zMBlIAv@du@X|_3Xf0zRVa)}_`WqI%7PY&OjP8}8MA9wV<8{#qIrg?OlRg~h2j&KTL
zthaxX(c5uP^sDt|VbPO;p5*$e<VV7X`%O2OQ`IcR1bFll_LJY+9oRcQ<So<Hzj2DG
zQ?U<xiOyPtlE|;T_#Ap_<l*}ea{f6M6(Br-Sn)Hx`9-CGF>cI*xmh5om-^9Z2;_-`
z0CTvJFbyEB-omCUc!$wk?8ElGaYBz+>;h59Qnx3)@Ad8=zDQ||Bzpn0JiQ($C12)U
z)Mfe@>p22}=x8o{i{z9ql@_dgdSz8cH_TJ=bb;%gbeG$7dQD4AKANT)r!g2E{gl-S
zYpFPwFnFI9XW^VMK3gu99lw}C*R{!;%pB1L1_tn7!MUn^je@%%(XIgH$L{2a=P+Ht
z_Wn9<u<5hNM8E%V0j?I*^%r_gm7?4y{c3EIPz7}xD_*ElM?FJNoHehg*Orai_76K{
zb6dkb3fNky!&$xr-MZG%z3G15F4Ok0Yk}{G88NoI{ZT%!q1&du!Fz5`sM)4|_pID<
zFoVuE2e}>_d_roWbkI|fyxS_yJ#}WyuUjC@P+A0Cf_slH!2*`+kE?+;(|(-@esBno
zcsVUf<m<68s&O=`eb)902ENLg;Wp<a%LQ@42<CD@pVVKz!8wlT+M`W*1D@xYob8&m
zEy)g()l-fYGzjNN!aA=DUO|A+HV8Gba!mc|vE)OZoY08C_B%NW2#C=H2eoA@^jtM>
zcLgAjWb?3Ku1Fe@k`3fjXCQgJ_Wd5vIGg&E-$0#He8Qop>V8NH$B2L`J3l}i(X?lw
z<yTy6{_&RN8GGEuX2W6)dC|qMrvypS7pBfJrQ2a9X5M2nQ1rHYTe+-lgHLn1+Q9TX
zs=BsR`uMvHAO}i!IhwLyu%*4ID!;lJckFicLv)$T^&S%<gQa5&68e&F_a8$blq@fR
z)&_8jiWKbNncMSisS_GeNLJ>K?op!Q@w3=(ChmvzplD9-S_hFOQGj>t@sG0QL0{!g
zLHCX#4&+)zoS$b}UY^U)59SLt5;VJ3{DQrSpLAgoxlilQjlkL$^XcDAe~*k)ddV4)
zn0Xl!zk}jJ-i!&UO741f{kaDZlpkbVnq(id=0D4TW;+{K{W-|t$A!&Ri1t0{s4KSm
z5TG*-BI`tsoCE!Wi5?l5MZwQ}>N1(Dy>+#<BRvn-nAqbw_B<R1>LXFv#uG~zN4O-{
ziXPBhd#IMBDc~J~$c~|@XDfQ*9Y<y#+ZXSl<s4J#IvbB<I)gdW$IQ8CzI=O!Z{Hjd
zOmG810KxyZ9+EL8pL~Vw?VoK{bS%2^gah8YQ8T#kYUaCD@7x$Z6D0)ll#Jg88LOJz
zz|H=ltg5#5lp3U=W-!^JYkgUfCs*Kd3N<#U8c*k=Np+w|IMQKRB7#sxPt8~y9u#kU
zLat@2CX4z{lvo*mwxAT9L*z>23K?YEi(WZn{%7b!@u!E`izWDv{!_CXbT>fPbM_OB
z;RytUma7;e(~D}`slv^h3TzNaFb`ENDhKF0c^{HM;Rqc-LZ($^iK7diC4_UmFRKG3
zXETz;#aY;6>-N{i>tO3(ee2V<ch`e+p-XMlEGn*_B}~=&?)LWX_Pu@F&iQ*lc4+#?
z*C$puZfCL`!tpJ0CPj72PSycuOL%%`0YT3}#V4FH?6b?yTEy*ejvWr@S&?u<I?bQ9
z>gCaML5LxZF7j`LNI($LWD~;*J=Wu&ddSXb8<E)uI_zl(NF-m_equt5U!h+}H!R)_
zewLl+^z)auF=u_LJc-McVH#Uw5|ice8ma18UZ$8Brwgsy*s@S}L*5w0#T5wT%}lUt
zYTiuMuRJyXeBbnrE;1k4Tfv*CKP~iXlT|L;F(os<aUc*>+A9JgEvbpEUpgFXX8IUS
z$#2Q^WK!<~+K%lDpWWrYVGR<dia=hmYC`69QkHIP)?Fa)Zejc;anCZ;>r+Qe)zs)+
z!p+m3*WV81k|~;vK!J=_>jQJILsqatC@~VhO|Jw5;irHsY{9ewRN<snYIl46*;kd>
zOu<5Mk_hq;1EBDqG4vPx0N9kI^kjiR`bD@(@QJiGr4If`aMMi}5qzp?V0nkAmv*Pn
z82|3lVF9O$5G8<MkEi5H0+8l$0;QjJ$L$t14{T%Chp3mhtNqjXK=;6rcG3R}FPJyZ
zV+E{8PIDcB@S_cQ1Djq(wO8j}MnB)?WR>~$LyhIDmuE|FewVqTepQuLW3a^aTt`nX
z89YnuaTal}_U|VMkIcX?yuY7={t3tD{`b#B8u?8Tug6Pm`q;x{=1z0m$x6XmkOI;c
zxS@Qda6sfax-I`Cb_yV8?G+nf^)MR-f{+h7#o7F9kVvMEZq-?uu$+zY;`#l<MN$^!
zkEs6s??T{fCm<V}UFk^hm?zfVvg`-Xw3zK+W=MK34urDy#a8<Lx$Qsl(=dZ!k?@Jo
z+l;0&H(#6AB8$R#Zl>M$nO9=m_Bp*M6C4Z5sXmcsf-$YqusKyPx63eWllI1jYGO{n
zN7qN2j2TweB+z5?VG%HPD2PqTWLV;=CFo*iNIj#Ud;RYJf`FfE!TAu%ZxG28u1sXr
zJVb%4x!UlXG{WC4cNRJtqJ{5-Prbtg=y{!Z@t-JW+N$sU;CU%G$Og4@a5&us{oM=$
zr|4T)%+DwtmoW(fukyLIo(xTH4v;z&kOwdmPu0LL*~iKVzAOz*3v71)BH`XJi9iUu
z&MC#d!(noqC}DP(Qq3MB7H7Sk1vVahe)}oMg`?2ReP`c&0f63zSQT9GH3Fz4(ArIT
z9a{H-JPaZyV@u+HUsN!g^%gGp3=^;z+2PL`$L;9zV7J)Wpi`Qzwnev@MT=5{Z-a%X
zurFMiJ09g{54C<dN5FOQd4UC55E1y1GwvK6sZ)72|C7}#0Ju-j60cwfTV2Z?>^-Ep
zwb8EfoD!1tor|f7U1&}7M^AQ~dxe*2r9$u1ffK^7^80}EgG|RsooiG}Cg8)5{q@|0
z5XKm@&-HeTlPZJ>r*7t?xXC$rH@;p>uy5XYzRi9EHZvRCJXHxu2@f^eK>O>SzM%jY
z`_1k>v^rHk4KuM{6MBE1%|uljs=Y|--C`OhXtuUw7VnXcRAPyGKwB5MWKwzNGt4_n
zd}!jQ10FCAM7j?QcW{yZr~JC8fV|AgWur|^L|&fHW>mW}a^-7F*Y*-}`qYd$c^reW
zK%E{pM^$XYz9p!IGSGutOIU9$<pR941bM*?-DOtcEd}=9`5s``R~DHyNYajs*S*Z|
z0cqxL;&g$jnQq=@_iaCCo@nN3`<exS#s{U}>iY0t{e?-eYSf>t37eQYGg0*%1!+Aw
z6`Ea@j%80RFSG)ou7aQSeK4DY&YL)2i-Y2rfsH6L_?{44HBMj>W%Frb2MFLwz?E6#
z<$1lF>64UMmVZ7kmcZ2iL`iYjzRrA$7}x@fMY}wUZmM^I68zI<;7-Szci~q&td>kT
zvO7<Uc5^+)KF`ohjs*mR*-Y62w)y!%C2k3HH{A!axV>Z`3Raw}{oq>jmIqX=>O=s_
zLEyJuFaC(A`7U(Cx<Y2qcH(k%Xdn8Cm+GTCxa&)vWI5=kqx0^a3ZBQ0d}B?9nVlKB
zX>U24A<*jpxi^*6*fMSO@H?3EUF`>+FtFVJ1Fd&L)tG?fbP6TA*fR6p&mz7>O{4S{
zhl|BPngl5nulNv130FUiSCpWRX#J=H#y~%*HP?~UvL35dcfHw`vNY=x-x#V20zSm{
zyH4=+@yz>E#dAOmY-^q0u7;x%6|;7Z^E9{-hKW+|ku&}0y})?7(qXwX!+fU0dH3XF
z=H|`*@ps+ta0GUH7nk2Y=IP4J<(si7B&JRoFMoAUlJeS+rce}@EnQuUa-w2DddERX
zVbn90+QnH}y#_K*NeJcoTwC&K#$G+Dl-Xpo=H^qRpl^>c#=0<Ys@8lPreVpcgp%YP
zqNyKxSmwiUK772uN|oOJ>II_q=wV+U*Hu1r1TKW<EfE5L6Es^Ml+<i^zb6Y#v<f_3
zdqU^l^cqhKc<6FUnqV2xq8+I^q2fQ`@Gep(p~{8bL%y%(;ju=SyX7T0dD$MUIhUgr
z;MSkrkLR0`yQKE}$@MO{euPpkaT%7MUU{tK5~9pnexA91Wb@UaxC#b4&Lh9D7<&Ef
zTw?G&gr~irVFB(jQBc{d787$3VW~7y|7Hozx$ErY^L(rlUvw%`hodik!h7f;!k}bS
zmSOPWz0_*EMec`04L?sJo3b7D3)(8)hN*RGTVt~L>Q=uJ3^F>lCWu6st#yqrHO-?%
z5#ro&cNkYk{Gv!85T2_<tN6}<_=44}{=q{e#l50vmH5F0$o1U+6Nm{2l#fwjiA#;$
z%P^(c8OzmP9jr(O>N6>~uD<GaI0THNAzF$1bTJpaaUJd5=&86X`&R2umY&R(Kk|k}
zSh9L+x|zOc`il4#s9W*uJig&xzYE(9ih}bA-^Y~aieEKo?^elunuy7sDIXMiXzVrw
zX}c2ikJXvrm;m(nbWe~YFs+}0ji^_^S8hqoaxg%TY$o($wcq06vz!aQqfjqzgi`4f
z!p^TnhTGdJgoTEJ{S*DgE8Ye93L*8=-7)cd$MSefpXjCT==RjQx5iT!R;{AZ(ez!B
ze`J;NZHo{pze~P*j`|9_PrHhFR2}_5(}71ql;1S|s~nUduNS);VKOv4>`{;jU+)X#
zSV>h<RK(;K7E1Pm63MTS%dp$8Xq^3&Iq!0KdRL$Haz&3+1UNzs3=C>ZgYEq<2gb1y
zsFo2j4O@|>bI<v+Iv>;h^7z+zuE|}5T$BnEoliApWb!Ru-Ep`v5YAmVJ{~1}2{zpN
z`ug(*KmMD+6p0?F3=){7+R$ut5)y7%aFM6n@|xbsfA*-FHIfRJDHVMN)Y9wfQ)XP-
zXuV;8X)Acf>m+}(On{YcW{hXHJRyTInqDMt<<x=|LrWWDbHh`d2%04_ML$OqhWnAr
zwx*=06*U2R3ygz~IM$0QaBsb3w9QRg%9DmlN5taYpQ7J*heAK%jZk|_xX*~S(M@F|
zzWhd-)g2EG(@XB2+{4foeKEHtCDY$P1&*Kp+6&WK_4EHM$SqO1E9E6I;`wF_#f5k|
z$G@!n@${U6q=5~p$aH6Oj^<vq)Q5xwU0GT%^?Zmw;co)w1{%0qQnMJd1c3n*^9nn3
zMlE~D-p<~B*Wa5n7Ze;uWLm*`7BG&^WhCCZ;N1-0iK#rlCm%&c9``|yv7Fn2f=@5d
z^>B#kAyvD@O+0QI$;@mLOy9SI@}&%mM?eL14&K3fu&Q6IuMCE90#g)~V4lqBVNhna
z*;J8)N^*)Qnu#{JxQDoisYtjwu>OmIre?bGSHRVYXyw8rKAsX8A54$$+vqPcg+aeG
zZ0iie^_4qFRR_{5QwytHVH&lqwsYUwL-nE1FwPSkqgz)y(9v3@Cn8zR7%A%BDgt{m
zKTL^til?bF$-4j-f$rZneG(533lr5`yp8JDa!UwVl~;i)P)n*73pG4nj~1$v>l{A5
z{yoz`OLcS^?RsDk0;(?t(t~zsK0wRMZTth((Fg&f*7Qw!laYqTHoJGj1j@=9i8||A
z@u;b(sPOV`8TeFsAy!lRnuo~jf|^dmEt9ph`Iyh^ANMaoV3d@ueR7y;(5rLYV!3lY
z8JID3ew@KY*`{5aSAvCB5BH~vvy~Bp&K8~vYCZD%`;JC?ibJ!N%CAH}b>tkkO071S
zoy65|+XNQg#!)eayy=3nCJ!(34nE6~KO9{=SgYAfs`IqF{*+mmRc<LZSg*Y3=!I$d
zzI@W&Z?q`P{Q0Jt8i%U5?52p<YYe(hDQ0u(+nC`eLDA4PllhpTX;O6IZcGvzRCMX1
zg-17U+upG-1Y&MZ9FAXvBoLW8>0Va@d)L`c{IB|J&EL6Rz?!F4@>BFVA3Z`&OvhQ>
zy6oqEB9Oc<6m?s;w$@hR-q+Q4QJ(p27gXPNMkRnnZ!F0I@A1|nsNe)D<Ok2`bLXE;
zTRUBPrd7I{=&o#W-(PAmA<zDFoC|y1x+<i2oW^9uhv@2AuRf>LgBy0Q6E>h+x?+`&
zU#=8HMOB=!lB^F=R5<0IAKLOe{VuYwc)?d{YwcCaRDJTm`ywQV|4S9#V;5v*Y3bC^
z(29qJTHBvi4rDeRF@O2ds8Gn4(Yn+1>2Ok2+AYu519pq0V&}Sp<VT}3HTG?e6m(K6
z7YB<Uo#_ST%tYrU;F)85i18}@_5F_fv(%*x+{CZbHlFkdy?gd&zMh5@W??NM2~4K&
zuCy0hUo~@+K%|pipp+S2zks;_-f_G#3%%DNH9tZbAiIv}wKw}?vZJ6%#F3klQUp_7
zL8+(1{PrK)i6ZU&--<1NSG2Aw*ZBj*N$S>Og3^_5dh6e#6ZO{mXO)-Uf%U_At*01l
z?l&u8BU*#@b+iP{U6CUkl?R#U3M{)GffC+;R%#{{)AqbLTTw~Ld#|ku4)%QVL_hOo
zZ=)AW`zj$60Uo;&W(HChpO%l;{!sd1aTYVu$?gNyPXb}?Obj|^W~s=4-c44ul<E0$
zR~vMBNg;tVKNGZ4ywqaKrY}&67#-SeT*r_>cPjr)$}iLEK6(g9A0)?@o;=uO1m4Ru
ze%E2~OOWZ%m#LZCkOWF9y28IgkBc)2yVpVP?Wwe{?-4f#;1dCLE5WbGrfyF4HM}7|
zcYhR6fiR67W`fSQ#YmMl|H5>v^N#H;P++>S#><ZzKYo2l8BVf)`lTdIxHZ0BO*3Ey
z<+va5#A-#&=@ECs6Cvp1Neg$vq-zj-bnr&tOP#@Q5<UuN&1#a!Y<1s$3GX!n_qyzT
z;5@Psh2N{K2H}t=&=_cq*UDg)ncvZ3Vs_F*xwY;Jh8{bIK?8M_j$J`z-lyf|<<16W
zc@118Uq|n@N5Si<f@k;P2@7bTsLeZyDvHZgW*Hjd=sLLEoN*k->vw1GEq|`pDWHGu
z(F1FUA#&n)`G72Lf49ij>F4T{u3~-W)bp`Y3;7Ks4Dx(Gd1d(Z=#@rwbsciM{qU)U
z10i3@=UIj`{)`M?Ch8uq$$s6P<83H<?+bx6_6z7+Z+X1{(8)tA$dIa-br(v-Xj_kP
zbU8f21egJmk$+S%S}}WQCXr0h(Ad=UJPPQOXut)W(Y0}?GaM)*6RH$zl5Wt0^~}|i
z58(hHlf#3<ahNYBgSHMi{wov4U8k{ul19nW;?vG^<PY-qI)i+3MNDV)NhMEjyzZP7
zG3(y!<Q`a?sM&aS>qvAHqJ9mL0C(=#dvX|m{=D>uGr8pLniAis-V>{1$CO3tM?FTh
z#c8iV^WO{PP}f##<_^yS``8<hY+HdW3~BfJ5Ilt)l#~OxsHn52a7$|)fGmW>b!38d
z;dLe<lOs1f`_T(ll9=2#S@g#iz)JT52{INXQOiq~d=oUmvvu$VOybfxNL#2aogsv6
z^)E(L8oi)UI<^SG!<_VZv<4q3VW`F}+3p^48+xQMNw@_=yGELx6SMe_<04(=PdWU{
zOr1Y;@1D20+Sgh6X_G*nwl+uHJ!5xiP%g9)d*~)%|JY^1`L1!#{+)DP9dc`2n6c*5
zL{Cf++eU;{!Ucm87*{WdYM7~ZTsP|u&!K}4Cybg3Ky^+S81GQ)19E<&Ul=XsGu=To
z0DjfH#J)j|NK~6K?W1J(z!Z}%O1$sSX5npUSR1VkJ%xS!k8T_2rw6+*%;YT+J}DzG
zW6v~fTDH6{(q##lU~WHYInGtLe>GzQp5LG1KKe>bSa^z3KOwM`YwE*&H}l?r0om)M
zXBtj>P3+Vf53=xWeXJom`*6b)ZnCpv+%X4FiCtHXp*@@Z>YFM_sZ$Yh*LXi<#okGB
z>_<!DLQ@-)GtzwYruUh+M5og<pN(q__gwG~A0IS4+3K6~*=|`|;DeQ?v>%3ll5?6g
zyx7LP>ct1{bugW&4Mn56hMt~xFhU|3WnyG#_$gB0R=6oJo}w}<mS+tBfFyS8etPel
z)5)iWnS_qT<c500k1iq3s!xRKcdN0$bI*FyIz|nwV2N^6*PeFDe$!RHF|FI>p~1@B
zXGE#%P&K#8x<gwE%s7qm079k5w<1$PmFosUbB@g_Mi%Hu#9gandu6hp-C~KU^RMC&
z6Eg2~>HUshsp!4jX&?}l2WXaj&-(HNDiBnQJNqeZ1Dxd}YB<(^f8h+qLW7vaNZ(I|
z^GaIOW}!7dEMbW&sNUz0);sdr@CWz`%;ZBIswY4M))Sc0fW;Q*zPlu+6^nUTN|Cjn
zyupG7w$}=p2qDsm-7n{Vv|h9s4^3}K4cGMUOG4EYYj|&G&Q!ZL!O3H-uktL=QaB&y
zTbc-Z8zEFml#i3H+`0}fAR7&Q&!A^k%|DpI62-Dg0}19h(KfZ>>ewr`%}uKRE_hVk
z{H_P!TS3o%5}ATu+$WhRcj)W-R^`dOrIC>&pE{z#4F25DDsTx)hz|=rVi|5e?7OQ4
zeKiNZfWTi|0w%r&fjyvPx}V2@K;Q2P+*WM!nf_v7J2o|9n}%Zv#Dt0K?B-P9i=azj
z?0r$=+)S)TP8UO5rPJF$qdzj>D>kpYo)986Y|5Mm|A`DhdJu@HL=P*k6M!mVQ6iqU
zZq*K_84TuxvHLZHjWw%6jX3fo6Im3)b2Iqki79NwQEK97rj#F3+-t{`v{TWeG_zgU
zwo5zIUHD!A3ZFj9hJQLqA^TnK{c$3H{lFP@CKUUX-Oak9w5$x&+giq&Y1=th%zgMu
z!IHwA-mB3)4SS0ycT`(wAi#eWCNi}t#Q%vRbny3QN7K@N#za*SXf{1R=3Of0rQ1&Q
zn-lj`rB9-E-2<)JpEUvpTG-o^I!}BTG4qqAl|Advq0btE@%<W6ZHwzE&#l{J?&8A*
z3Qs_xc3um7ZXRXqCiu)+&Bl8S$U?GUAyS}{VaAxd;p?Ld(te}acGIfBIVKL)B-Y-6
znudlo0)x_&C7qa>jm!k;O^x4S{T6CcwOjLImP-<}+1SAirBKkkb#<Oq`Qw&hLkQlE
zww~Uh2LFk#7bv5yq||WR_C9#O%ztsOy0U=TITYNhVVhJp2cQIvjdKk4;I&|n|66SH
za%f=BFb_=!<Mr?Hl#CcVp0EdH`rk1EJ~rNxy+o$4e-385MA5Jx{~ug>PJ^IvYp;`|
zNIAT1K3gt1<0E3C;~8be2HjMDzss`^zCk=hG5q{cI5HLc2Zxqe^_J<g_0K5iXU5xt
z-t=*$Fd#Sc)Z&A$`W|`~<2c3Y79E*@p#9AA>_^<+(Qr8;0Sz}bzKj!0A#8hFB(x0;
zkDGFyLLm6u)&OW7T-iebXm=YxyVYQWEL=a3<3Dl3BAwG`Cva;;@CM>B=P&=>{Uub$
z%6s!ssk1DCz~yiej*dswwRC@MifDiP?9Xj}<Uhz0Eq;qRxw4ybwBTlcq0eN|a#Kwh
zm=FeF;W@7;YF3(*a-D0D&*;7~JT|uJJax0##s9wzk*H1)`BC?1nPIFo=lmQ0@z(g;
z>!hX;d?evmAl_C4$?YC}6=Lo(|G(iI))rrdKyLhGrj2Ftc8(S%yw7j%b;vkj5+gu#
z{~hp(n-+6Q`k>~w+J8s6TPn2}U|-7rSBE+q7!Z(m8o(&B;y;OI!h6hv)pMK5=zTi}
zkVfa&vZ^=8Pd)4G-bK8nNg#{Xl=im+n-d6awFw9c_6U{U@V*Gjybd=NC^f4-=>ujA
z3zW`PnaCpfGrbX0orCpMGtbNMZVWU_Q!RX%0bhZlXv%pG6H^GxfMbXZn3cXa(t$R~
z1)%2OS$oxGzRi@O3=X{K$wHP@RnphwRmqEyR2iH>o^Sw(1gimAnPMnQoD$GmJ5Vy5
z#po%BYG{UK@=u@b!XGikb;twtWVJ{EwqRs*Z%hVrT{CJqY2uLSOcJ4dUZ0*mcY4h|
zy~^XD=Im<{qLoqU8U!NR62Z?y1SB5!#i69kXL5sh@C1;QeNqH0uNb)`qSp3qNdqmd
z(ptxck`lhr@o^t8`e%4cV-OQ{0L7XgD>9oQvr;7r1kIa?-y;VhkW@=P1pV7*Ro?)&
z!53K4+bE=W>aDAfs}V4<Yp;`-P3}5AC-UCtjj8Zh8T0{gO^b)<*<<gE=*&{oSuyyr
za+aojMtAt!z_}+t=!pUZd|o|kM-}n&11HP-@h#(jR`|j6Y=)1@CKI=|p_-@9fX@<{
zxeV%UCPdPTRXiE|Z4oW9L{lDcABRka5ApengS*~f2zYyR?R9}zGYg)QxYumS=lc!G
zeeZW9n<QvV3XBdDySEE0<5d7uoR4EM<&o2)-E2@k+s|15s-X<n{#PJDO)<<=LC?OH
zl(>CBz%(@}gJ=DqT}&OQq8L7kel!-KgI7Q`Iqy<axY7_gv#dOVx*g_eKiHSnYXxYY
z%Y*nDTWL5>a!z4kFNk~ZY{_*lf)rY6sLW$+Ze|G%cHksl*bF1U1mF<{du}iS6hkup
zI(p&ET~tsYAqtYT3*a{uKbZnYLuAsUks8!Gwt)9`U@)P4gu#_v+zqD!Q0K%<AMpT-
z7T2I!tA_2Z{dJ647$Fk~0Hj^PKFgoo=apcjyTmbpQTGmhF6e*QmfnfC@EwovuCj#N
zLOUPSqT8VvNGW;A)81&-6HTtak31SidC$~I4f05nRG&PZ=;6hd>i(n93>s>#{+!s<
zoU+4o{|pxzg$2AFr~3F2-QoJ22cF&+pA3`g=yc@Bvd94fy|nIipM7+2kbz%}ZKVra
z=;@1(YI;!jI%=Xvo{<=L?+HtsZD^a(ObIArp&Z{5LdI4$J+@Kla>xw4o3o-<M8LqQ
z0rrhKC2wgfu<@n30}8a@9(@vr;uh9OW)*gBZmH}coS+|2OXuRxsZsGC09WZVl?oGC
zZ1Kd!BCZal8&~btkAw<GZ~A*ycgf;qY=rpdXWoN>#@7v=cNlO2%>fQ?7N4olSy78Z
zVK5g=D{}lu-N9vi@%!Nk2w2J2uPlcmwqOyMkEWCO<N<(t{q-22y({4kBHTy@!mwFP
z8WUn-TY~_{&kh7u!z%98T6+o13A5qS)?x?I<7hRrSI?e<+YB+$vaZAn+K--oIXerJ
zNf$*PAwcE?9EUh0eF^jIgdWsUqsK}psT5EUhXc@L=vw+0g^v?)pz9PKdW1xeiuE0y
zPkAxKj+27kgxoZ(#DO&4ZvzkLBTp%J6_QB>EEvv+S&+dxc<@S<$7#CC0B(oRN8cX{
zLVLHbzG5;d0rRj06ni$T@h9rnTJbn8*5_n_0+zCbzp<m^hP{Rj?hg6YrtbAMe^8Bh
z26lf{9K3bPmmix5i9l4rtqf@vznc*~s_YNj7+P-~4SH+^WCrpdhP`e#2r&jzwI*j5
za+H^!A6b1uJb}7k<x&6!ZO~1gHx!ik-40G(UVYJEhH!cAkTin%w+%eK6K0q>HV6O;
zBDd;E+|Ip!O%jfB09uGnkpe`){D~g%nFP~kM&#t=BtVJEVKu)Bh}ByHJq>y*VnrEt
zWW_lNQ!)!XX-wON^M9L6INlAW@b2}b)#qMnAQbZoWcP`o&cea^a~(QEpA#uv2sA6B
z?g2?U2dLBsCjOIzkBDGg5REwkUh!#cracN~0X`&xf$bc}qT=@qyO@AU)I_Q#&I@QX
zR?(S(@I7j%fDdmPc4CHrVs4UjF={h_=Y7bO99w5U3+f{u$|11v!#9btv=Hf;Zm`;d
z>3W{d1S;r7767z+SO@DN5_CJ1W(I3N5QGYhsNV;V`CTUS1Zuk6leF3s696ithgf5S
zZxZIS>iL`qWv=)=M$cm9c!WfXYBqNRD7ckEU-g2G1#CnXervmKA(_lz{TX=G89X=A
zx^`d$1ZX0`^Xw&)BnL4w0)C%Y|8T`BaCUI@xyXC>m_TcOy(X-10OPqKGYrhzQ}BZH
zo1Wrse5!$`2$m9K0=fV(oBj^HfVsI3Sk!V$N_Lh(H7Aqb7;^YmB~jDC2sj;2X*?ic
zthKt|xsttcI-WkdJWFbQ@IBKL{B^3;{r13%i<<vLYrzBg-|A@OF^6|Doh4@CLkXBh
z6pnK;^(&T4p*<6y4to30W1K53>wT*69&a))5<FxE=|x2ja^dG1gLk1(@B^zD&w|Wt
z)IvSqa#gXDosiX<QwRTr0sP!x(0#%<M&UR;lNUt;VwIN%ezyYDp4peRvW+QT!S6mY
zP>;Z(Rw(=25~}CP(_82jbDarR6fs|8NL{Gs5etK=I(ieMA5PSe`z~BKP#|t9_<0Lq
z1I5@Q=rupIL48a`H*rr1niOyL<}g_Gr%xfu!l`Gz%CWLjb%!Fo>10EEM?mS*WY6~B
zA3@U6xCWAv5>hqSO0Kc6|FhD^<TwmuwNQdSCIw1VoCzKdUik;;nZolmxZ;TDHzBSS
zBB=NBbfEU&_|2c}GH=G1v-X!7TCYKem?UDgb|ajk|I;_Tkw+Y3%*jJgmHR7?_w$?l
z?S$H3_p={OK%=0?P*qep?J!wh(@BczPX&d^1TAkjp|-yS4bT(H3?ZucN^dRyuRG3L
z0S!sTa?8V#ALM^3S9lo_9D;sK4aknWnY`CJS#h$%`lw{p_kpDeF&`KG$Hw1EsJu@>
z^z(8RtzeLt7^cBnYl@9M9wjM%Wfz^3f$vh-=2kId)P5iP2!bWC-b<P~%aViCj7ovh
zF)U@iez@lhgl+m3c+s0~74nOSFWjcTm(JA(d)E(E@OrV&+dMP_>muFA%7{qyY~9W8
z+7nO^`Vm0II&-LVbeeK>{t$a{GLgR*rvmp&kUvW$|5cT}s1az|dl2h&xcJBr=|AJ~
zoURRCb76E9-rMf9+3Bu~{hjP3<%I1+qI-qZfXc>-UEME(z;=@#v;3&x1cgtDD0E?E
z?-DAp6!wy0UirO%V3EgQJZd^nwfA6CwBSGE0oPx<pw!%Ca3Xp5Dm(z8H8vnSj81aD
zfcnxLHhn%W;aYX-&OLcTaG09xez7ceubJu0M%6@~LZho^6W+UOnO~JTvou^+7qO0Q
z6FAh>=?AELIy3B&JEk6K3BC9Dnt8Inv}6+~;FDcRy&D0K?m^qwnXt^-h#U$jQH?bT
zyFynSROANRhDFyZH|k57&eBoEJGNVN@q4ay+m)eEzZmrNtp+JTLHT<J;43gmqyc#e
zu^I`smDqJJPlEZqvL+5_<+AH^g^^wt%S(xvHa(@uucuVC7oXdFY@w{nNB;lzSi0$h
zLXBRs5Pu-I?T5^V6K6?cU?35rKV}AQ%vCu?4P!<LRDbs<NtQLBzKu@QIvyeIw^BG)
zluRNIKV`b_p3-865;)R+DG!D)7r;HF2XB{iIewQZ=+Wpp?`DP-ylFf9+BCI93$K5;
z!dw2xgBr|<KcM4X?vc`iFUudflUTJ?84WJ_&8L?iJ@T2_No8G;x2vNxF)Xm(JdesO
z_PwzFdk+ZDwa-~XRB525i{>?$tvZ&YA)Wmi1E535jR#ySQf>$IVQiO^iAV(NnoECy
z=@0J1dlLi<3%YneW?C@CLnpabemHDULH!9N8UDR&8e`(OHNxaY)2;QEP`<lVfI9^a
zwrf~mAgC!#bx;Iu8ZHXV|I6bS@vBU!M6i_gzz?9E3X_bvhJ#O_jEqV`JYb9e_oV-S
d{-<7hE_KgxpvpJ=`6Fx#pt3;z>WTU5{{ud9QdIx|

diff --git a/reference/plot-18.png b/reference/plot-18.png
deleted file mode 100644
index 3ec2e587fbe97f110f359deb4ce2a8580667c5eb..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 86607
zcmdSBWmJ@3_%=F-7^EmANJxh?g0v_ILpRdW-7z#+AW|}PD-0kYAdG-8l%gW7L&p%(
zHFORzXQRLWS?9w!AKvrfUF+RzA>%X8^X$6!z3=<Fu1&PIrZO2ZJuw6VA%m(Y>Ovqy
z0T2kG^_5HD%6RI#P4MfIwVJXb_zlX@LsU5Ub=6bF*are(j5_}zNSNQUgFsjyP(=m(
zfb2C)V2GJt6MpMpvD?;3t?mwHM1h8gNJ$~{8BwESAx}&WG3#p*>Vwt6)!rKH@Lus}
zv+@8XrMEfNcVAkRKBytHBQbbWebs>YlW%phJ2WX{=jb#Mk&G4-UGL8Bp4$tN*<Q-X
z-jG~m^iJ<S%HB8%<M`XO|DFE`H9j67fRK_gGBVBv1qDqCKozc?lvPw<6LcIdY&*^V
zh!bVkNJI#cR_2L0p=)2U5d`-C{_W+crPY_k!E@zV)XK^VHcsls8?6%kl?O0p4jxgf
zuP@39dI{(HBgq&n7%zV5fZN*ht*)*z$4K3Py#4g8fE@C~@u#gwQqLMM)cYWdLuXpa
zSQx1Vl~RNA^76W;DL$pOV@vAD{@fLO2s{6_{-`#Ck+nn(JxJ+5u+2*o)1-uk;VdjT
zqb~?TShy5G5B@8-NO0v}N#-VQ{O_Z7KKJH@f1mG1=eLslcbVh$|NG5<bS~cyG-(M7
z4OLS5Zl$w$f4_T82$sfPk2@S18j?`Fb@eA7jJdzX)t>89ZLMa!2*n#MY}kN+AlDXC
zy!R|N1$3BTQKIwLm(#qqo;9J~Ei}Btc8pUq`%r&C5<==h;|EDD*5~!hKpg2T{%y=#
z?TJ%~8=UH&#a3=XA|--0^I1ePy7{*PPQJYNVF|YPe@quSe9&})%m@`n-@l(*k4xa#
zJ9w30XU|a3O*of?$`e?Y4i&5vM&>DIT>dEmYcWQ%i>{{n1`Py8$e>0r@|;lneGO;~
zTkrE*FRVuB{VqY?w#Gb52b1~;N~NW$;^MNpU~b_*FgS?0VNCZHqFXZ1>mzgJ*<ejg
zP2c%Di$UL^rwIqDCVj#8IyHDxR#Mqv6kuK@TXYP;-A<=n*ydE9TaY(=oW_LzhB)H7
z(o@?1Mq8DN$l||`+&WYN|4rFjMV)D)|C4W~PLP6h>VeUF3b(lFM{e8KP40DYmbO=3
zUhZ*j86H@R{+3&jGfv<_3HJ`<Xb#J@w!N)A*{0+ERp^7bXohj?Q5>)u<@N3Dl<e8U
z@7+VgLql@6{Kz3~SxStMd7czib55r)U^(Z7n+k^Sd2#iIK%CU&h?zL1M($SuGP&Kh
zB7FJ1PuJ>kiSX3jrjw+MP_a&#`^_({4qZfpZKk?JyImh5o<^KVbz#EMbtinHI17YR
zgbYdBCb?+V9jB;_96CQHmb{76Q1}_r<LEWZ6&j^mSL*|BuRtL8pVATuhn*a1z{Qnv
z2uEGq+-l4t9Y=JHj1C-=10zDz3U$Yi)1I>&D!aL@i|Q;8Y?sv3?4_Gq)b4=wkfd4|
zbWD@8!!Xa8qUyG0DUKw|o`po1{hRkUH3QN`jEpbB-l?~D_1Ml7q(*bNxIf-{GB64{
z8_7o8*eX=Jk6s?Msol%AFJf!+2ygfyaP~f=^EO;qMgjr};K>WtRl*gmY;f-;s1eV>
z=ipf~S6T*A9X|&t8x1QLiPg9!+~b8nT0cH`N=p}MR$VKpCtu_yyTcv_OIyp+Qkgn@
zvG#_Iy0<YPWZ)(Nr1jSfyv9%8(hV(rfL=qFuB`CLg&s@Rh6FY^&YZWYo&}7_>n;wL
zJlQ>7A6z0-P;gG}6!FY5WcJhkw*ep4SC_$EZ}%n1q7`1FPJ7+UHtHLvpMpm}-XOAg
zlJ1m;*$m5Q@~Dgx$!N73E*5Z2*F`*rK%Q{>oCeowXdENdVh=7tAlmGtzagY-a#{+w
zwtc%j@=S?z!e_tm57EUWw}1;HwUV=WWPlQH5suP`cWjwXCiI(??OW^f8s^Ys6#lyI
zDk`lYVJ<74?VavBvC4f5EXg(C9fe|DgO&zvP#r25drbSspF2jO4~67Qx0t)Cl-;#;
zboiX|NE+9Dip=M*c&(ynI6{760=HzPWM?p*?X=?@vORn!qkC=j3@qTztDrx0(92)G
zNXB=0J6gdZkO*C!O8sEe4?Fhn)tDl#SoxO9c9*RFO(+ciC3h&W!F1u+8&xj<qY!42
zn!rJTvtBRkFKc&KYmrXw6C1UHd!&qqF6@_OG8oFxdC5Q^{???R3So-cdGusq`O8k=
z##NtlFk@497jKCN20ZXSiVT%BRJLA;_72q|q;fCEqp-BKeHq73LnSZ0iHnD+OIvl<
z=0=&HktES`gx4VS?9RL>zP8l%X6(CZu~#=Q6F~0Y2DTWP7*HtB!jlwgMB7sA=>&Ol
z7+A4|NX)uk%-p=3)Bi&`hv;YY%c*5k`dynBUCm9EQCIEq^fvBxPCtC`IW)IpLi<YA
zq}=)|p<QVX#;W2y&W;KB{))Y+(f%M{EpfN$BdlyEYWSp0Zjh0v2RVqZ6H<ifEU80;
zZhh*NTmR7dRj|P~=ut*`5Tf?mC*_IV{xPqd9k(zwt$~|Qn@Y?-uk6=KlF28pPMH%L
z8JIWJiq~&vp57Z64Ian4lY3~0-NLvl<ODBJb&z`=_2))oEW&4B@8J>&+RQkArNA#W
zQVzpxzHs0Z^sUEg_BvAdzpGt_w08Y|PD|(iwNE9t=W_tb4?&YamoKxB`x!1-qI)j(
zGtA~vZI&XDB`xTQaP)7T&|qi7wxy-(7e-5rM^2bUPK)b3rrK8AlR^s&^(LDtXa5|>
z-ZX&!SVVJ3aGI_#RV1(B;LX<6mY1tGm-iLh63#8NN>Ea$v4*RNk@ZN@SnMv8xLZ)r
zlO2R+V5gw7UxAn9?y<-`+~&IEJX;yE-q2IPD`5BMI%GuNEjsd`Yh7)7@}OUqy24C~
z?R_<#-&>5ufirOZ>%xYCzCIK0GNH(+w!2a0Nt=~=)G)hISO>hLTaNk3GTI$#Uhn#6
zyY1kO=mJ_<nfLKur-IAA+y>8T5w0Wib$*%;fxNjJ=?IB@*FXf5|08PO*kCiTpoUos
z;gNYjJ&1;qDIb+MbR<Ys%x@kaXA68JcKc_%>vDpmzL`Alb(U5&j@u4rvpnhoi3@r!
zLr}rzCIk;^clh)r_uj$iI+8)N%y^(Nf3qXvRCJ+Xz8oHwvT>R@=Z02=dQ9e>u89b2
zd=fSdZsuH;Kajo^Nnuo>rN4Rc5`@K@DnM;uqTYvZsM1n?u-hIg)sIPl5AZJ&K!**C
zj4ZR%Ii(Lit$#2^ayH<M;r1%dZ^Vqzk3>xcoX-AykI9PJ(PNns=g^dIY!`{jh#GG|
zHQ?n6OYF1~v{m7*%V<V;*zbcvu*DlS;4-c4dX$_s+irfT&7Jx@`lSi!aA46;zuU1N
z#c(!uyuX>>z+fm#@26?&VE$)-lFg1QbNpV+sL!lL{r2tGyWGkt1_pS}63+2`tBu1Y
zssP~+FVP9lPw#AEJ2jYHw6b^iGHts^j>fn;vtRc8^a=bmn9K8+1_G)6c8!d~`Sf_N
z1^#IQgKxuVC5x#x-5|*3ST3!os95Zk-e_#|pQ@;;s_Ml^s2n#TU{EPD-5SKhiBk-=
zxw@*#YioD8pXV&?4e;?(lc|n}!H+VEFAOslP3HnXUJF)G%Y^6Iz*U{aaprQQmNcIy
zf0FMeL_<@I#Lz7Sx2#MpO6xtGd(X+$)Ov+Nnb$F8O+W%e6L~fax}NKlBf2*$m(T*`
zme?=NHp%jKrbsSmu=PnoRU~eUY9>PsZW^@kP@ujFRTUtEU#g8)%i<@pCHWwb#98YG
zUs?WTwhZ-!bbiy?SVw1P;?lCRvhKBu&;u2E+Q^yLfjbLLbFf<p?U-(wjL~{**sr;c
zB)+W7%nQ(hZr6AO6Ev<sIpM7G4_;-2w%n&E)YLtOo$&yD*MfieGVh=L`@FwB)!AqE
z`y<tQne1pR3clou3&q6{%?#%5A;!6D2Z)L_{0*k~N4ePJCfxRHeQ-LsHnN!PEH!Q}
zf(_{X3ia)=%!-4I<#xgzqG>c!o-Tumdf1aNwAH<I^guOcq9tpdAs|@B9j`Y6awyVx
zlI81z0|UE-n3gnkP85gi$jC@Q5uyhhrmd#d^&9Rl*AUjPAyETWU9$=w=ip#qV5n1V
z&_L4`Yvr8{&Ud91lk7iy(o)l#yYKFUk2O2n-eQC`%Lu)3-?}NPxx|qA&Q{;o9i>tP
z9j)0fS>JHHr$myuA>hh8*U90M@REd`5tSw`vV5|C6=Rv|tMvOT0TJ~F9f5*`bS54%
znzz0qC#g(OW)H3G<E~%p)?hQN3>fsy3sI@D2zG{f&WFqXsUEg#X^)t>`6elIsMF)q
z@Ij*v+7u^O&=_0q|6(E9YgimcCCLPV2tIrYgCOmDR6pb1DO|hrc|O&2NGXSlRzpK$
z9Q2CWz$2m8tKL;QaC=U^!m`Bz{!G~B$AD^3GlVknvRr6TBPOVsgD0jV*jK%sK(J9u
z7rH03dB2}OYEe|SK$Kl)n$cQnL9F}D>Zhifs!2f#%$d)QQh)e>#8V1wG}?ge?ApfI
zkX6$|ZZh7@>4pam7gV7d-h&U}a$L}ofgU{E_Z33B8pUpBK3$`pK77QH+p?4#y_I{Y
zY%68j<YVlc{)v@TM83?_Q$oVIGlJhTcfX#iXJv_Q%dSJdaPs!LzhC%|we4eDA2q5Y
z2xNrup%|%aBVr@c>d@E5W|j@~JZS1@G0&30@4ahxx^RIi`ZrwP+&rg0s@rdMXgx@i
z^qvB3{;=?EpVtVU(OTFj5mBt(wT05Yk|uALgg>4}hKsps_PPczDcOQw84=d$yC}Qy
z3&V>z3+$eMN{0tip4rwS>$eN4xqOGr(0BC?UJN7djN!B9_|D-f(;NN03B|KooxU#-
zdIP4ua$gD)Oz3dIt80;cqPF^hACLHD?{=hE5(n{ec7@bV$hlQ?Jr@cw*Qh?4JDyX0
zje<nne$LcL_`bCC7&yw|=h(1}jHY%>c#39m_+6Kw57bt4XDwJP)~+j+e>`>ljGlo(
zo@LnqcWu76sHmvUw%>EIG4M<10S5N?+C?MP(l@*7wjvq*Y);hF)Ot8MKKX`#tv~$r
z?d$W{Er;B|B5TXK_GuBzfDJDNBl3s8G8miOF%6W2<>YN2$6OcpR9rpWVirU<oQ<TS
zbRO{XNk=NWHx3jFN|{*gn^!%5{o00PH+UkDkT0o2->2d&T|;j88O&e`zpJBq5aQhW
zh~~qG?4Ysv)v%9G^7^2nWfO%%)QcyNZjK$@w!=|Wm`ndbriJLq%36wa9O!Dh?K~tu
z^KCv0f8X%up~FaRJOVKSA<!uDs5K9HWRD5Q@3$Y_<IJx!7<)H)d(GlxlENUTPgw@j
zl#loMoMB)a-221bRLJ>TQ+Q?(<MPA)g&gYH_Bz+_0|({~_3ZQnN#ex9=j2k{Hue~h
z$qF@k!6Zqd74yB@;~B}YXSdfy!pbxF;k!s^5@&g~e0W&j<i0Q%^Lp%AOz6Fn9I%2M
zuy!q#UI(~A)6Yxzk4NtjT4Hzao+?PH-_KTIf8^`uC(GkhjnXo&w(Tel{mBPaE!0he
zX0}w0+j#m<S=zRoDeN;0Xc-=+!0F#9)UIe%PuYcsQiUx@qpg%_g(&xxHj^q<TnrsB
zlYOCSC;Ui~^efzQrbrW$2JQWUf~I+HXldS^@c!|+;ueF~K3Y^iqfxG-WYtA3g`N~P
z=2|k^GZTH#B#Pd?EfJ?tBRN(V3~tHJ9d7M)WE3V4SJ~B9^%V2+-KVujL|yK0{5Lke
z4u2C1HkwHtzH+Q4dx$;Dx_X1tjQ%{CV~`C%{*nkB&4FXr_@3o*EZbn^Ownt5tHTG~
znRi!>z}`?Q(Ktg^w%C((P_2~_Lo1kD`ITA**1{{(-Cd-aCwD5?l_t2+X@5Cdk+s&Q
zJw9tdI({Str#GSGb@()D-Z$z8g;Yr8YH(1{?#WIQE&t%)U};&*S^hVz%%R|!+7h?T
z+RA_w-z6O-m6v8m3Q-3%b=7{_Rst%zCCUD4OF0gnXaw|i`_YY8f3}R841)$;=FM}^
z3mWKGiD&|wpv#;c&nqefYHIh%&1Hjj9pf0K#RYLhk*~=Z#JO`cySja+pI%nAZ;~L?
z%H+iCagE9|r}_JKxU9^#w^#(7it(hj90UfRyi<rl!EFV2a#FG8y*;Y7ZmUD(W1pOo
z*qu%ex9IOW546U~CZDbZR}aG0>rq2v?bIO$NMiYrm%eK*#5d?5LUW#eSp~KtqdZR0
z2o6#EZaO_3oh@+~9)YrCy7*<DLT0md9X@IDK?aOqn09FR*kW&v>DC&uubK7AvyFOO
znsNeDPP#M<8~uDwAXmb};ujk2U5=<rAI%?b!usX8Iy;Y5npx4+URKzwI~}{a@4o$I
zyW=vm+bS~nMVEzyU%6V=Ws|%P+aJEZzPALRU|YtPl$Dt#Rldi4s8*ekmJhkVx#Pj~
zPDs>~+V<P~Pm@P8T;B?G9`>&m8LU!BFtQy@_ff^(9JXn1Tu8iQvfv6_PrJu>Ek7p|
zU_b#6VnyS*^%Is(JCTgi{_5DJrF0f()nixONR>_APn<%;m8Q+FB$dlpY;E&}jo!X=
z3khhZpVt}QaSdB_nM*Y@+07?-?Hw(!cD(WD``lbPIXO`<U*Fh7L?WAVZ2}nMR!Ww&
zp_hBL#}zzH{r%*g3&p`y-S~~zKHO-5oV^+OY>t&>`e?t70|Y|ch<k@bB2oGI)=3C`
z{J~eDiP?0s@4t+GhS~zp7{|}gU&fx(7<PL60^xFbtf=xcP8#Yn5$fgVxA!u5<!N?V
z^bM{_mgUG(48?v0zwn-Z$JvLpgFj(c(*#RGSC5dcf=Wo>h_eeOvuKIiKGrlhpM5(i
z!ob#}o%+p=;}EjG>z{%}b4&$tH&=$w3YNaSsi_?$lYg_EL|6{nn=IMp&srSN$PMwi
zE%L#nrc^^Wj@{4K*DT~@t9{Do)htBkaZ}KqMB(IBOalsqx3IJfSN)b4(P&a-t(0(F
zVV-{&i<WX2tK$geTqR=)_4L_%x3h7#U(f1IbD{eejHE_L>`WmjP_#O9tT_}1w~3cr
zkPUWTjGpBkzv~`4Ro1le$tAd34wq|-$wxKAhd+WXC;Ea%{KC1}P5AWPRXKi}VLl@9
z4(e#%Le{pq$o44qGfZoo+6G<|wzJSf;?SENSeRkXwtS0Bdm2;t8K()YbHg<3t*fQp
z=~#RCND(0(I9xh3N=BQxOb;dWyJ~8mk`Z%YIKjB$4CgZn`t#YS*6BoYAuJO?FHxR$
zLQ&JPAgy-fz#!mk!Y9{nktTP4Aq^@><BQ8kPCbfRrK#Xn@uG!9oKn6xPFobqf6aWS
z-%g<8>qy<$4ZZL*!}%yK_B0QEZADVTEY~%L`6h`nq_w!-bpYVeUE<!fzn!S^t)Za-
zOzP?CaAkO9#@+$?R9+OP15n`{wKM@tRb6Ph@iZ1E^1QRTx1oAC_Zx~DmgY8EJwE{g
zznw$PtEHv`AvWzVJwgmPZ{TC0ad752G<g!%Z&c9b?vV*Dw0e%@AT{K(DOw}clJ40#
zXw%<Y{cDO7`FwuHJ_BR7Vq)7dFQ}3^p6aO5uq&KI&mNbwCK>p-8`EM6Yb#YQIJ*f#
znHw?19y{vYq0Gv?{-Yeid$(!;l;m(L%+KG)7J}OOMc(U`x%-fqiU=a~O?#TLwJ~t#
z*=44H>5H(IPi18;6$odj2$BYd50lDi_FG0562?0;pe|@t?_Ur7J|3RNE{fZMtX3xa
zCL-iDa%BST_4ICh$6OabJ4WGomfPTRYyHLTH6ekK7UkSU@gzk|V@$}^?$d!ul7yVJ
z)(sMaQ%vn>zuaS#qg}Q5=*eYh_(&nqA`j%rPloVQ4Bk~u?4%O9F_nO6^jaIr<zo*O
zwyYOwe6d}g`tap&AwN+pA%yM$_a-@TyG(}>D183Lya{wl6v;p%Xd!dBTTkTwg@{%i
zyS#j)a_!=<p`l^Lde)%^@R;vQ-@m`NaE9B!H)k{%qQ#-Z>kaGf-!bz^Tw1xpBkZo+
z{ExI;sf<^?M)Fvhl(B0}_6H^X=nApIWtXH4+-Y3RG;h`d78H4NgW+8<75>t$e!{V$
z((t~KTA26Lu3<gd7z?y0p|Z*MczEf1Oz4aks?R%EU?cZ}5+ECdsRAxfV@<o)S^&nq
z;nB1=RGi<7EbHy<eJsO-x2B7<Tw7Vu1)*oJA68&>{cauL86r*H>*qOOZl)$C@rE8=
zMsQ8bmax-W^SUX6YZtjDy}S5bh<EfL+_Y7(_<^4N#}k;8sN?THzU|6ust8PJ$C^PR
znbJP6d7;xkJ#wIn@jZ0Z$`%`HTA8v-bxe!*=W$J9^4)Q%js1{q$@{$*+G1Mk26EeF
zHfmfDH~tmWpx$-Lh)nLuTJ7#CNBN!mipllIt66z2T4ED_(DkFA5=MK;)gTcf${<q!
zLCrAPBriQ!2%+{1-5JViQi4~WJ%Y(Apb6kZ&;}!PJ|Io9k~(Y;SWv{r&HLP>ZJ{@<
zNn3@gm)HB_fBx+2Dy<s*nI5#DdhBeaM?mMSODnD(ZMS>m&vV<!T3yyzP3%L;Dm}7d
zW0>94;*}H|aRHm~-Z$-9#Uesn%kTlg4%xzfr0_VJ*-=Y9%Q(sG$XEZSkC&&nK`}zB
zhD-Cf)b-o4w+1<;Qmm_aWR*tq&J)m|`&C-@$&wmlJGL&;T-AQK`9s?s%hwrXkZX*@
zjU3J$a1Je%WA`beq0G!oCWH8dB!braA~elBX?0XdMKJ(a{NUT25fqd@#P3dDH(uIi
z*YMTgv_e1M>QQL*QI*Pt{3qw#*osYsmNI`%X&Eat&2y1iPUg|zACIANm*2xAJHipQ
zify#)jG61TPuopMQb^`K%(ewxq$h>fyM=bMI2dbjK}mBJdBa1#VU0@wj7_}x@qiNo
zc`l4>FdIX8yHLFS<Q9bc#P*<$v#Iejs;rH0xwbY&>!#kPC)}@D3M!8+Vo=hIma^90
z+KHLo-|yb7hQ=M_Batq+P$H7&R#qp^Qo_e8HVxeObvoSBc0VIpI(<{3_J1@e^)Oc-
z@pmIHK-#Xz0`3BLN6hV0c2URHPp;C7+A=puS_F*6Uyid4=xs5>7pU)De{Fk?4b_w{
z=6^BW_dZbL2t|2D6Ss@ZP{>}H%)xwf*_SUI(7XB2Xv*`p{5tLegw$sj5Fu;G)^$xD
zT*)ttpEAKLSl!}g-uv7VD3<Yk>{-Qw1qH9T^|i2<h(t+O@3k?tVk25!t=n-eR)fwN
zXI+I~uz_d%1&3)dNq_|<<@)!+%(>3u@db6%<b>d@PgO2zc0~?hl-2|fI~%^0us!-5
z$|8hnEi7Q)Q!!b`wv?Fr4qJDfF+rgvE0brhwMVsN?9hp;*NfwqR#PIcr(Jzs%F?bE
ztghuyh^Y9=&if(rWsjwYmioBswg;5H3QP~pWomz?^_s4_N7pDIkYq_f!bP0wQpSXT
z;KKM+BiQ!QcXn9<4$_qlYN`2Xg{VGpXa=Y7nYk9R!T9_b?d=!qxad=B%U1K7uV!_s
zn$UFBv>FwbK7RV=?N8nm<1EAJ&M?m2xVtSa4r;eWso~IyxlU%;@ZUVc?LpQ)>i%%`
zLHi)y?h*bPi4TojE3pvB+mLgnVJ%H~nBODxEMKqaNB>qCN7LRs|JqGx+91E#R=869
zB}HDpFHT8?n_8S2V`|8o6)Om?>!JYYrs6lF{PuNStD#gBUd=xyCw&{b@xxnKyuP%F
z|F>r`LKHdH>Z>QYYo8Z@k?m7E_By9&S~tJaop)P&t06eIQbW9NvsEi~*RK9lx0+cc
zcSMFaOyA9$<GVSjw8EW}qJ4{r^#cz#6MXN3MV{Jbi;g;nrl&4eGia9UPn)@Ul9UGh
z%j!ds(}93N=%hNgzCgXY=?WY>hSraq#hKwoKZWO>S@Jq^nX^S5y+NG8L#*(`^5=^)
z4^D%%ns>pb4Xiustlu9b4hT3lqul)4=IxJPUcx>`^N=_U-!a?s6-ELqKlw(`IY&BY
zd}m2=(mVZ$I+{TSSEZ`<<B*6CZ|_bUyz$K&wRILQ4wWn0`0byZF{adGl;P5VFhG@i
zL)EBP$ZAN8?56q|Kq~=Y0$4wmKA9V_onA@29M|oR>Oc7Z?zFvjsS;KK)jj@xcgU$|
zDs%K|O^hLi+W0<|Sx^1~HHcbDIuR89N5)T|EvFE34sm8p83sSxY13H*<OR1MphEye
z>)>@zlQD5TTftWc9r8EuX?uN{qer#JnU1mS3OVd$#=64cz!wrDzehP_-}Qn4C883<
z2}o54Wd1KVd_Om?$!dNKz|r*3%Fd#u`b^?mPAcy$CmNN|&;PNpols*r;wySMJV5UM
zX#pURYaB@v=0>|syG`Q$Q`jxK8?j)_AR}CViA>_R?m9bh0;hHRB7vD|vNRWIhb-`E
zeoC?D(u23LLW+dMtSPFDvsk?Xz|r~n`!^Sbk9I$Do%p=fbWWG8mlOkh(vk@zRo!t6
z5=_i?Or#X=b_3QL^7@9G!gI^96`AJ4qa&GIZ9WTTS?&R9m>+qLEt5W`rlzpwsCMhB
zpmRErDg0zBhf(idi;lav>xic(-W%}OeD~VbdgVFQ(?7DVTm#fHaU*wfe~)X}!0k6p
zkMGp_NM3p7WxwA-`Ga6v;EF|vV_|;&pO$X#ECP}dBaq3kn;$AHTjJjo<|h52-hNJ=
z#;Go;i>5xhccZM-f3ihehDkpAz%yC2*CRPA;N*MPc4ud2iTxwxjO%}}Es221zmMPs
zWH#9FD#Ff=7Apa3SzQO9mO51vcIdKi&_m9oK&A6=ywg@>l|3#Ep^+utR04n(jz21N
z?&Oly2W6I=Np<EPm<GRGgqE5v3<gsJ&*)A1lamry(KWO1a6&=m903CTz`@bcdKCcS
zrB*V6AZOwg+wIGywXRU*90JIBBf38*>uL>1zG(v4hVPzWR)$uln3L%_gK1jhJlGpu
z@l|MT?v}EoF~}EsTk5g7GbfMgjjOD`_a1nj_b)XyHHYY{5NZEk{!+Mu9?b4}`Twsp
z_+OGG{4d=MGz_hB$hxL=w;QuthaGVH1{@Lmhl<J2z<`+OcuqEKGBB#Jf0-g<R2>=x
z4I1mF0skv=N3GX}Q*%m1O6L_nnYAe%NKiM5yv%O&DRU0ULH>4@TL~}``XD%yglA4z
zn$U_OUBipT0jASuCz9cjIa)C#KZI$aiuvB0tioOjbPupxRzhdg=I-+p@=C|>5r_7|
zEe!EsuH(qa!f|LJ(G-1-S>kIcw!bgG&tvr}@;Y^-RRRK$Va~ezs+p<L*u~ZL<QXbl
zPqz*M%N$Jyw!xC4X#{d}dQqI+ZEu@H)t+2&^NihxA}!s8Tc{72Ed5+4YPpfb^dM_n
zg7akxIKykQfIlg(n+!rB#&MQ;!Pt7`NUQY0+Qr;p>=!G#Fmt$d)kv9p^U(qqZkXS^
z!8ZyK#op87VNL4rm)46t;)lm2vXA>9q3_1svijtY25sCM>dMmrV*0lNYNAOi5XkkZ
z!BOE3!+!*{xtCj8?3-*9H*%c3wHp5*;c|M1`mZ_{cC;!mI>vtQzj~UQm;Qm8Hp?*j
zffxo9Rce5|ik>Hg>`g}SuR<pV-@JrKKd5C(eMRi}q${smGk;7HR5f<7UdaG#Sc>e{
z>me3W(+kVL_m6Yq+>i`mAJ3Ueie|t+gYr>22$1^#4Y}V0@|g3~RZrBRS7x-knPs^L
zE}k(8*nVXYtWp5Zcx-&U-CiUkI`CC2>+;glJ<qhXEg(?wmD^t>2|FS_JsP5v9B?iE
zB!JpO47g}bW+%+rIKB#<XG=TU=?e=w*g%O*s^_UkRgIP#oh~(nVWo$udE1h-<ZiB7
zG6`FMyA9MqFCQ?HaMuF=hHZpizo_8XJlXd2-Jjpe%$OsEXFwjXkl~Ok`8HgVCMIs?
zmLNa2Iq>U?s|@+Q1Wy7ATBT3ali5Xs?RS@cr>TzliPY#|&)lIKxspB$u+%%o+oPef
zDx66fY))rk97;NIw<Nt6z3<SJrR_eJ+jl8$Qaau#uXpq9F2i=UR|zYg?74$>tJ=d`
zK5@bDgAI1*%d1~d_@za5M1Ycl^~EnS6y>pW!qz$p<QEhX4Glqy=dRQxx%j~I$$O#T
zjyP-J4^m1?OLMva#<mER5^N9~x!jWI>ojm1=zB<i3w3|S{cH~BQadipIDLykp=wJ?
zcmRCUPGG;U0CLJcF>&!+MyJQgQ4P`C0ksEkb$VG%dg0QyiQ0-xo$ied9)WMHq=w}A
zQH#CLl>e^0x7oj*!_u#|eJrk|HC0FR190n1AKw6csauj*ILI?>de)@j&0NcOaRJ%I
z&siKhmF=l|E<%e{;3ydM_@TwVi7^M`R_E~QRum`2<Vc~71=7$yysxW@vZ);}H)JA_
zo$I#qWWHLzXl4yXLirst#aoi^5%?0J@$w$|wU<Jq2`Rq%Q`?+SzI=^~a~Soe!|BU;
z!a9aug=CCFwM-gtQqb@?aY}D1`;%sfV*G^-p;>G>>JHvh%Z-<huW}irq45Z#K0qKi
zgH&4DTekOG^bL<RUtBYnQdPMza^V(Kjh%<IE&llU*!XA+2KcPldU+OP;&U-K4g6lb
z9@kL#>QM+I!*Az2jAvC=wZP_aOYTHspO1%8nPItel0F+0$k_6v&JJg95?{2l*;^T$
zO5P2Gt}HJHHfrOU33Pxn|BLyYZ;>V@CYb}lHy)1Hd9>`L+r$TIeM^KuI3jP;M`k>}
zle%I_-nLmUTjS*6v9V-XQCdo`6KlrpvDdNCt}bqh9S?5V`BQK#oi;H}_1^1`-+}f7
zKVL47Yw+KtcfQTGsq5+W`|oL}Qzo=?t(Wwq15#*T&`4k3RZv$fxjs{=M@^JTUdMh|
zZD<r9npFGKH|1-0bPY)+$+n|h{;WI2{D4KS-hc0t2Jh>KadcqhrCN_kI6Asb&OPgI
z-FIsmpJxQM1_aCke=q1z-^~#6_r0@eG!VN{T5diPV!R%AAalb{ZqBlYtj8s}?VdL$
zswl$iZyUL#W4}m~o?|ix=e%MCLJ>Ii;hwc|U|5qPHa}uZ17zfi!(A<4_yHEwh5e2h
zBf$;$uHt~Cb^vE!)DhQ%I_UxF`q{~o!iCQZ2>5rE^JoWhKk#j<X@0L9ZcWF)z(Bga
zldMgM6N)`c{qf^R4xmC#;T!<f8u1$kKoG1aTXKg=6lIB~#hk{(#l@kR|EK|v6fQ8G
z0C|#5*RomB@~O`~K+EyC21;E)TdXkkT{eZ-eq83I62+CjwOzG&6^S;i>N5IRypc7^
z6y6SA9{9DJTm&ho5~QO02#thS5I>v#40a6<DsicZ8HByg&IYxursuLrlK4jT%7*({
zG`*WtsIRw@b1UQDk%^Gazw8lhQtLV@zFu|1Y#*R0AULc$Zic*lyFxQ@`U(+aUfmO|
z*It)UIyNa`_RrmpC{qPI6&c!(kCfn@P3w3jp(3Y>mY;WCxhcGomK0X!YC3QvB$F|x
zTH%}^_3x4Shzo++ks(+pwcGlj5%0jfNYdPCS^S&j?$;G=d74d4GmXa#nHvd0|MmyS
z{tcVW3wG~A_4X2aeCsJD=2}^P<%u_+?iRdG0~z0WIePSzmRoe)N7Loa>hlVP%T7!;
zNJNZiIH>-8P5cbh=>N4|cSv-zsb=<BNfeC;*?(U}fv@Z*{8R<j+gR?^Ya9Q&to;~t
zYuo3(dr#Y~@qe<uC-2|3k|Z<Yj0DOt$^Tg>gp~2T`?FZJYInZycmAs*3lA3l?Y;MB
zw5N$W)$f>Z$-_;MP7FVrt<=)6+~=Z?^Fnon$l`5xY)jw~r;j;HiO-uNiCFoHBi#3&
zKm!Fz#MkrM-huVPr#{02H(YD(H8)k?Kcd$NjOx*&Qvf2N^D+mO0kSOO0?zuC+tA|T
z>tyZeYjdq3$$VzxJg4*1bERK`4*o?jZ!ey=#|)qn!q5aDX#4m4kup%ZC8`!`!o@^a
zZvC9BbsLYz0<jt)kT#~BYb?k0v}Et*hX<~WRHYT_meeVoYa;je_nm;xfg=KB!dNzD
zxFr%~_eXl)&7V6MpwwvD@d>cAo4@*3)^}!TWCZj1D*&Ym(79&spN$Ha0aaPZFjX#D
zl0SQr%pMlbEirgXQuTO^5ss#`|EqP4TRZ<QKY&6x!|T9&ksJi2#?jxc3vg+Cyu9&)
z*u?;4-d}`0DM9<o=^cg9m+f}PN-%O(B0Em!@v(&<(8xxp01>SMkanILqCL)2kpJ7;
zCUwYIT3X73VEevB`G4SfO<i5f(cx@Diy*i<`N5*;I14T(2*qG96a&M<dDp$zSt<=b
zJpNtK?2-mt_1NFY8;wRInv0q1+{SCCmkJWX64;Mqh}OgNf?k4fqhy<d0ke(bO~NA{
z35NWDQ^06t#UM{qu3qx^J0;&|0ZwXqqnPRN!Jh&8v^y&@_p9n{*InqtIFW%MiG2rU
z8RZ(giw(+a_mhQ?!cK0V&zTVWCoauUKT~Uy>ZWD~Bu@Zm`P*{Y-<$fUO|Xj!4O}Y$
z_ka6#=laK+G5hY(1RwnR)_5?rGd`>y8h)PfMl$`=2eS@mru#?pIVOhGA}CQGq&}Qs
z?Y`(poIIc&7gwgH;4c03RZA^+qq*59XS|q%=?X*$@lU#Vtt4JjPfOycGPExtUkzS5
zo;Pe&XU>)vjE-=N@SX<rfQC3f(B^5HkF=iLgb2m{_t?#}ZcG>pkNC(F6q_#K$eWI3
zzyBmsw6i&h*n(TvyWZ$Kd~xAB0VEkvFX#Pa?ay?)H&&QbJ=yB313;m~tT&Tmw|Tnl
zvSMWn-rj>mTn7S8aNm7f_5D&a6E&#{ZR3b{tzeYZoO{Da!t=%JSL&Y;0f;H^pX^q#
zJ@jMZO@FsL>!Fi8Z|{7@9R$-}@^EpVs{3Ga0n$nRPp6wGOv2Pt*wn?$qfhhX{rhuZ
zD5C-n4d$IZj=_t^)$tfn198NYSMUDnsaZtsc2Fm%>}%Jq6FMo}-hn)P3bmNk6G%H6
zHdo!S3U#c=8**n%`e-Y1L+KIFSW*JnEGQ8(UF;r7;i~9$fzinymP@4>?|HOfXC776
z1Uf5x6BfU7)A!Y29JXPN8P9|U3Yd@#E3_X%WB{%c@$Ttg^>&2HAZHf-mmF&5p~$H^
zp^V*#Y^@<aWE|x?T~E8l?JdKi3*-S>YR7*Cv=3<?ik;W{>Kt&e{QOmq<229WSmpRf
zZPFx_oE*;EFX1M$4rp4$I{;~(H}Lj5f%=unG25Nk+!dYK2g{F>7)^Z4cIFlt>|bPb
zuDqWGPL34%7n%STopCO0V3*~Ds;ct!z7_`vSgZD<r?B))z%qK`PY?2J>^IVUUn9KW
zXFN_N8K<sWf4<{iiLsow2~?&atk@<bcE0z5Tr;@awF=?Gp#j(DUw#bD^F81?mnHzP
z;@{1X=lQMDq~B+^0^t^dPDlh2z3l6!r>FNuvzRM=bQqU;x|{9OE6G|4FtESnQu{Ts
zw*Dv(JO~&!25d2&PMJ0MI;o<m!NyyxCMA32Sq8i!miu2O2HFA$kuY^M9TaeKb_rzN
zG$`D%y`kQKEkygr*bS~YgL~V*K7Bwo^FMY0TAqU5zuuZVlqw*KInB2W;lrN8vOJo0
z*E8=n)QV0(&{+h+{qTHW@}y{4LHiO}T}v9U$Xj+)Q$KZ6k~*ysmcWeD_&oy4`mZ|h
z5Bdlj_Td5r;qX*{X`|D*&htH3w3h(DPA7uYs>hLv;?}z}L>Y?JGraL9M?1mSqrZ&+
zl~@-?4#)BtCEKG9Bf~a8p3lU+oDxV9_iP++JJF|I_J*TlW5)~ULhSv7@RNf4@S_Mq
z@_Y3)HGZyND2@y8Qp3!r<H`fL-EW!#aT23k4a&DXtycL-<jNwE7XC=GpWlDPF;J{p
z#(%fA38YZ251dO>_HRz`4oj;A>bZv?)ORW96;Aer!I($Sb>ChTa);9nul-zCv9AZ&
z7h4ALM?b}05_hxLcW`{a<1^nN7=Dt#xM<m6rZ5qc;sgWv1cQsSb8hs?A%L~_fcl<7
zIft-y6Ic<Q379q;S~Jft3bL}YfagJhEvE9vk4Jufeo!<gG%sBya2wy9CohkzZ1+5<
zgJ$Mcc5^oZL?)+8;o3k0j+r%<aL)W%<Go5puq$azkc8w^+t)l(#&b$^_)F`Dr+f|a
zS07+@|12eQB*tU0Rc#I#1xe0WJy4}3%*fuF|FQNRXPZ4$SdK*Mfo)Hxovzv<VCH2K
zdyamrO@>*Nfa&{{Vy4mAAR{{F`l9*r9EFqezHD(MUE==xiV964{6sA?$ZO*eY6btL
zb?Sa#pls~9uvUZjpp}+&?wEVf64v$$5!=%Os4Zupt;RNY@AP#VP735PVP4*VK$Z;l
z9%TJE+0ij@$7(;;Dd{~XtU3BsK%F`hH?oybF*L%3mzxgqf!ddz;$+7{mW0Pbe(HQD
zxeWPv6{P5nLm!NNFsHQevw5tZXht_7U}@InHX3(+0J)O@4DThw)ba$}qby+2cz6^l
z`G_vkrF+dGFh2eQK&YZhVXpQstv+KgmGR<~Z!EEAVNe4&RpbmSZ0hlQn^F0p*t*3t
zKb@g<;DL@eH5MoI-^*}^1AHnDZZTZf_!~nDRQ2HW2>#76Q-mi>&L;JvuT*H{aili`
z5I*QBTmxElI6KeCZ|VX3l%L)s+5I{%R3(XWtZH3uAi(<)g4-4z+>)k=8T@Wt1%)Sf
z*yi@wiYP7rk#X94T`+Jk>Ucj?)}i>rS(msx*3Tv8ni@P=vsfAC@6W`0zJrJR94uHf
zN*gI$o45gvfPDOt^)@>@bbqt}OmnR&9L4;}wga9PhXY!^Lnr7|imwJpVmQ9{zH$W0
z(uHgw^th9V&?$ISE<c~FSEQJ8^=Gj`nPClpG>25s;8cR997w)4_y(#>fNV?RC{EH9
zZJ6a+!KcrZ+%i~Z6pkz?nB9Sk*>|T0TjGyFFyY6^l6M|V^!NAQkPyxB`f<~&E%~D~
z_VcU}d~*1}dklaAHv?zf{|Zg1CwP{rhAjl(m?<OFQibIr?yy9^6o0G+7bvT&Y~h`1
z4;U2|yyxb9jU=zjDoH&~Jnqw68-|FaCg%K&Aiq4eIGgSk?K{yIjwgW>^3pY4?SvW_
z43^<&n8)?J>iWJmhYMbpAb_wq2=qCT`lNk*U<dqru!QaxBla9tvXHG=xb6OqMsz=g
zpkc*>*X<%Vd?wgyg~M78pS>V}tevNQT6ng5gz>~TFDl?09?t+Z^-;+cw`qd)eU>*s
zFbaWe+q7{wB*1rne;M1O4&xI|JqmcrMy-_I@oAx<<AuiEEaK~6Vx|DCE2?l!X0R?`
z4GjH$z_2)J+h}MEg_GdRCy>_MIF!}A9vs_S+mvOl*Ln|}9)N5yj%%fev96Mtc?g(g
z{wb;7%5~5<RwmDyuse^pAG1JO!>r&nRtu@7?#9C-BcDysE|^d|ko8G}RA+m7d%mkb
zTJcW=aH24k<EP-blSIA-354U$FD`25mzVIS!QnkW4n1NhuR`85>F3X+r>|X4YcbsB
zg=(pO(Eyul9YYRF=cBit@I`T64B~+u4`)&%W*y&gQat(zHcCO9a~1M&V+kak8Kb07
z29Bl&x?M}Z3xdI})&$7*7{nRDY2~@j>nj250%W9YjHn0J<y$p}K}V!Sv_uN$obNL+
zfRGdt(qgwRM;4#KSjxSQXgQiRs*&tjd`IZoZv)a2oJDC$i4^b$E?r(RX5$pbvu@jh
z?EESFCz5uHaiqf@&PjY80@h-X`{3*Y`NMOGkX()?v$cI1@Q)mBx*HeeT`;HN&shR&
zfa!nF2e9W_+Stn;l82)%6Mk{fEST|LdVRP*>TAi~u*9Os^Sj_6)!5o7U(Xs+yfws>
zGgwbu7FNrpNpHI5`cq!b=mR{tEzS_{^W`g521ioHxuE?`$XRYsAu8GHz+-qM`Jms0
z5fUjL=}1?Ux*wmno7F4`j@%4zWicoDM)&D`oUdBGJ3gC|^Dl?22lPr>rmR~-efVQ<
z*yv>LxcpG)HzSVF{Kuo(bh927%_VO8pDephs+(#=JIvu7;kpP%pbs1DLa~Q_++*}O
zdH1sFg<AVh>Pz+%mx*a&O?T63bQ~@MLGO9Mqx58lO0~-;r-~k;t@!k)JveROWMbXr
zPg=QJ_Hq65q%tXR_=Je2Y`{TOR?fV-VE<Ed<!WjDatvU9Wc9x}DgkN=tIQsH7f>z{
zp?`TG2puc7ai`ZAW=ZK6|Bf?9vmG%RRs);xY7Zrn?yYS0vTj82jW4i?cZX~c!a|l9
zOi5C>OqYF+Cy;$|S+~RYe}8r{@^@8NmzS-`bupK+IoRX$gYMUlz2vKmWn~8R@4pa~
zl+v`$o$OH@{oe6=AJV&6!JB8uC-_k7Wo-7CDW{L?)BHI^ety2l3TdG`h$k+k^*ZPD
z3(3B*&sg<@O$qd$T5<ZDpT_YGDd$i^BA$XI@jPsT%Cae!d9RX;*2zCnC+L4WqVCP_
z+N_7p(QyE7g7$Aj9r14Y3)p3ybfV~p<7Kvg-c`7Ei@jtunkzS5`~9rGuN%G!abVj&
z)U7*6&6d!sZ)i`P;@|<)NFH!sL#lRwn>-}>8+1Rp;(gHl$*I3qF~^2mrBK%j?|mDe
zg}6>OC9-Ut_UCy|^#aq`flHr!xZb;J(uH_wtctRqD~OpsUDWvtDMY+`D>r(Q-XSNV
z?KRAv#ttM_zn{wU0&|B%5d6zOBILkQ-NEenL;JAJy@gB?IL-A#M%u?BH#>fjf)1l>
zUV#PyZc7Q1zrZkTA$3|?W!mrIzcZXs)=3yFGP^a-dB?2z_Dab4&HtUIL`j~k%$tLw
zXQAieu<^fe`1~#ZUuP`;A7q9#{4Uiu5L9VNAYmZX8C_jllldGDc&RD?8khc!)CTml
z;s8G;;XaY3TcXbkPPWMbiC1HDLBT_~zK)L9y^WbdK44hPfpe5lUEt7fp9sK8f8!lC
zdS@ppR!$RI<fgiZGO+y_=<4LJZbT+Dqo@csomvt7rv$b)s3puQUdvNAIXAg;8`WbE
zSf3HU61s!tdHXB!g85@XUR`G%{$ILd9d~kSkz7+dIJOLUt!Tfx1KdL-P@Wm#1K;S7
z@PMm}9L<IF5p;jhK7dX%_^lR!(2YRIt}~f|HSuR}1Y4nPTO=70hK*Iu0f#P)*b;vZ
zMX=dDFGw0=eyzw_d?}S(4Jcn38=d+&A&=s31=@?_C#4sGXvc7p-j|%R!-IEI)#TnH
zZ}-}_w>~N>-LAIi(FgYv@<{KZCU?5mRG#QY7%N^%Sq!|<Fm7)cY0>7Bnh^+H*WG=5
z;fcLcuVl6lQ-c;n2&n^Iw~6S5%pyL#efz`=CQU`mNN{UH`JnNMJ#}wAcAtH5;#CA2
zkpLC(`mw97wzg32owFthWRHTwHOi{Dgs)Wm&-HXC;1t81OWb^X!MJ2S%?vtU&crXD
zoCixIYinz(ASo#+Tv|Y{TK`rwnlzQxGGq&Pu8JXyy(PIldP7TQEBUQx$5}~vIX^!i
zp9_I!Z&SdJn+~^_eT!VLMJrB1$O*;(xDiQ|!^QBTZKeg{K#)E<SMRgf%;cIs0Ip13
zV1{Pr$a+jRQY$QUTuM}#Q)`WjMlzSZfB$prcWK6T@~B0^m>Uk^BH)(QXWeY9@O81v
zm9PG%sKo0&HB$Va!meD>%E$UY1zw;(O--42yN~}HpYsGgZ!nJRb={y%m46dH`tS2U
z-)Mi_5CFH=cZ8n2`1-?*13!5vVtH?Oo&i=aj(tE>LpCVtLC3&zoWXBOLncxHp0YKZ
zd)CkSg4o*O-Wnui_!{xn+!D&DnZ{Ruk}~nM@B(NW*b>VeNL^gFX$UqjK3U)Msh`NZ
ze^`=oX&dyRQ9r{}??YH$uO*dLkl#h_y?H-u{AcVz7B5Zt2&FG4Wun>O?wMr4>DP*^
zR{R?udq_Iv&bxO(`wuS-$Cn$Q*|!QSDJmXoM7@9g>gOB%*9Sg9+*s^PcOrj|e2*ei
z;+bO*KKnP>ATn&==K{a5^=c1A`TlDs=l8jL!;xzCIrP4Apo)GoS(^`*bmmvh)?4?n
zGYijyn=68TwAjgHzk2l(@^=g$YMycwGhGcCSTC>)L51GYm0dM=qTqn%yV3T_-7y^5
zli9>acwy3{uMjx=LFQKlEfilM;bw(D&wp`7JCXRy#*Y?qnVM-**8QCvSIjFF_umX@
zU}$8UVFbHp1Qn*%sFxop5R!>1suPlnRP8cGE}n<E<A5e7mx(&9eit-^6S<|O$j)_F
zKhK})bByk?%O>_X%^@cZs}3gYIh4qPlJcgv46@{Oc_TJ#j~1F1Nhw`H5WiBEuzqYM
zk3AByp`;v*J6hZ?Mmg-3wY++z@(Y;5FSpOH|3?b|Oru@0;TT5L5`L<L{+5km-<!lP
zMXG7avW7Y-Dn?#eBww)%C)d=w4yN{#_PL!!-g3Y)pRR7~m9?nQGmchHhCL;2WM#$u
zaSd)cHkHT4yn1y}{a@3v9S$-J+HQqbAX7(1u&8G^kwi6iqOg{rtaTTD1<J%m2PUn9
zRpN@gQ!t?yivAjoBz3q+)o7H5blXzX&s9>P#byIzq^la|PW0}N>G`wW1Dm1BGgj6k
z$L^*5_FIjq)(1H+iHKlDci(_{W6}HDLd15k(L}^7=Z9gX_oP1ZOg<Z(45IYv2#VgV
z_AuY@itT1DOAS*x(D|?KVeh&mQI}rr54Eo!FUv`9SJ>@kKsi}iMec#2f(U$`es&c=
z|BT=#2mF{m?`2~Z(Y{o1GD&Ue`&l$1>MHd;h>^C+2j6eRQ==2x!G>WSOTqM)6V>UZ
zYfld@2_k;owI8AFXHrqLef9n`>_|kEW_n~uEsfkK#M`W=bIMKx*u>8#+*hO`Y>8qX
z6~G9UqIX^@M1rYRpLI5je7aeh!p9)xkEL6UB7T2M{*0u*SSws$$(1#@ItyS-mAO@|
zn@pGDmrD|!3s9hfTzj_XZ^m=6!ux+rBLB>|rS-^lZ|ur1bWdKs1rf8xLU)MWnQ+Bi
z5|?sx6cug5-?#GV8yJ|cWJ{AB3VWLW3RF}?e<f|XKV=EsXD~vpl=m2Lb#v1jZU^3C
z7QQD_RYLUGRc`8L@RWB?6CZ0e^CP@h;Em0KZ&2_u744@IeLZ>d;iQ|=ixcI&^;1E)
z#pkoC(#kp1{EIk*`7o>>w-_5ARAn{$;PQa6hQ=Mxkqm7kS%-jEy+~mXsl(fKf2u@a
z6D8W>tV@pPZp>JjXi{b|aB|~mJP=j(QPa{g-|}Um1eT3Smlqtr)YN;EM`TXSOl*~Y
z>G<}@<7`4w)K1#t9)3?vzo(|J97wjF_ws=E!1+J@G+~m)pX~HqD(zl-o*;MZsCM36
zMMd_OyJliLrgSX?Kd?1bNJ6H@g126KAMk!pGPJnPAI%YaCZUxRxR7DjTOS;uOv`Vw
z`|HVu{PpO@E=dWWj|Y6q{X;{wF2c;B3;-8Dm|@v@MOq8aLeRqMlw;|<EX>U@yXV)6
z0K{xwIrC2Y{e#MxdOrF%8US8jig@<y*>*_Kou<N=B2T>A@CLGAHuE?yl*<sd%$3`5
z|HFqG|Mf9KX=!N!@Fxy*{mjkHLt7JdtHDtmp7r&0NG!8_SQw~c;Cu^CYGCNCHwGX|
zpgf}%xF`%hLXuNbG`hBjzxsi_OpA!|f#P-+absX90x@UQTVL%c7NnHv{>N7oC8>c+
z6}$5C0ctFFF-{z5N*K!~vLv&S^W2b_grvo(=!>O@FBR>^@fm3GvyLFLKJ$)^_LnbT
zM$HY{994!RCI&5n=WnEkvc@@8Tj=OeVu&W;a5(dPCvNPcRojxlCONa*$+1~S4v)S6
zT~EJO;R{CK>`Ixe>&QoB_u2=@>~4qTLGXRxJ+3AfIsSY|LUSj5IUCj4Ew}fREkW@a
zSbU%a_8wH-4cDqixrGtd0(c470_CqOlZ^l%T%^J^;~|^qt#;<{u6(5!w6}EghCBoD
z%jV3~95s_~)>UjOu{5u~TT6GZ1%KhZe$%XZBb<cp(zTfHVLJNypLnKKSf6v!$K?`r
zYkZQ(7!|w_&yMSmG`bRUnT0;iT^yN5bc=`W&bI%}swIhIAI7@<Joy7nsr7PSyH0;F
zFU$l*oMZBLczK<P@9C;aNl6J@(7d>`wszuYY&^<9piG?YgY}&@gNULi0t7}BzpO4T
zjqHsgnwm~|L>YFSh%Wz@e@@QAnt#C~mxr7CP+Z%|9!Vo$mR>s;)t<nNhq3Hg(Ov%x
zz=yuYoVG9i@j!nNMeu|2rZ3=(k41n^-^XW%nH+$0jrdLWhwU+f&G%j_K9!(}aT{_8
zz?S}tQE=q<!P&8)kia__Fklkoi}FkDf1H%+CVlmAJrcE4G-Cz}_j=mnMB4464$`($
zDPiMm3BCwn@9D2In`YOECV^$+`@s2KOA2uvXK+m)Tmw>wNl6eOQiD#Kv0sl)<ldB8
zJktg7tXga01zPebSg80Yh<bdEwoSNMt6TJ^dqnO1Q+>!6CsqJnBP#*ZIu<VipG1G-
zzbVDg6w2MsJdaRR<VQi}V)}hSH1{~Zmo-vz9Or($fFp_*xlH%yME>B*kC}1WN2Lx%
zMgw=kZ^u=EokX$v*}`;><xAP@My6V@9ld&m$oZYH%-wBB@1OO&+9F7O-&0G>Ld55I
z8gT>IPKO2kmn;Vwu&A+<wfv7iPIFHkSYE0f92$!6cyj05x;lJpiumz*vt3Am3v$<(
z6URjzWE`61nUSGAi`okzsPA7$2_=4d{jEFl(PpPW^9HQ1qf9>H{0e)4>+Q3V95(s?
zi?6SaimH3V9a?D#X;4B$q(OQ}2?0?-T0lX%2k9Dv4k>96K``iUMk$dFK^m!{haO;v
zy9fP#cdh%!-RoQ5(&L=7^X&b+&-=XZ2`vX|@|sEu?P!JrN~@1Nn+FNOs|?&^dMcob
zSclvz;1~rZj+Tlf$G>6aJ45q9Pd%xH6s22vuS}xOQsmb8#wY%Ik8fcZ=K8b-ZHGHs
z4t2uV4&rNJ4_s_RZ&_>_@3U_`Lm&MpK06^4#6^^xX>lz%?7a`Z=&JB7&^iF-?U()Q
zM6h?n&N8H7M6#1pe5L)LVmHH=`I27PW-D`g9?FboZDx)(gYy;uM3w*QU?^Ogx$K}>
zikZqmnI0KQ?P7=c|2BbvXtcntW&YXwLL_VblCyy^{l>9-@G~sl#Q)zjcosd2y~bO5
zL?_QT8h;@^@gTB<cHq`b!ddsRRhL)%xS0LWg>)e?-t(oaZja{Wtkdn5dcM=oTR0yH
zp$2|pGX2Ds{jg0-`G7rSKw0244yE|2hs*n_7kizGMr+_2_74~3FBZo38CyCZ-r5Sk
zxas1?o3wvBjqSLyTLKo*{8uFJ{p&dPwpiv1eEaW4wdh;m+y;M93%qSuw!kTpM(ZB!
z0NjUek38$aOC9u4BPjs-LuRbK`%ZV7IO^Mb*u|Z>uLxM=gY!Vx2@J=-ru#aGX__A?
zdV+BZ0$jhlW~%5)C)U`=0~BSqt*tE&ezNfMpXF_BZNPAeJ1qx+s!G7*N&u$QCa_u}
zG&5z6oE4C2cW+-MY25kAy9)^q5BJc_klJ5)FY-KWsJj4vD+8wSiaeeSknDke_V)IY
zh~dvVNp!Tdl>`|)v`peoGGA(IkK}_uY@@9Bv(`{Y1#9Be6fNAm8@)vyy^XY%nGXGv
zlQK>qhZ2gA9%wt6h*;eF$KQ+Ww|BDwOQl9=nVCIqM)alr;CD&7U72FpW#&PF40T6q
zE1wg~A3HL|N`2dti|G-B847L@u-?AJ8E+91u1eSq&pJt@iKj7I9XNQj?U3D=R2cs|
z_bL&6w|v7Lj8W1iwM*{{u1X8TfGqcodH|z+Z1IajAtgTht4BZ(X-N7&Pw(J>CqG?S
zXjzl}mF=%DkI5Vz9W8KxPDeWvWIKUfjG6n&U~%dg1tcBj3E{<yrl|qMFN=uZ;g&Y8
zyN3t#YM3(XB`KY&^v+f@xGueX(D_J_FE7ax!pFzQ2g}OJu0oKL5!UIh?b{00Ke09%
z*(~MzMi543(C=NQAi;b_;&L@8MJp2o;;ntVaZd|e;oFLB9UB7*IO8i|^nfYYgzp{$
zIW+*{D?Y5D_s7KxS|Z_kzzu#<&;LV)oqg;+2X+;5CcWvCn2(%zNq(gQ00x5ITJ0OX
zXB$A)bM)dpybJ+ECwU%pf7ffUu)s3pzm>|)UKWV8_x>yY|22NqkQyLh$F4-{ceEbs
zGWE7cEiH8)9!jYs>P)&IlHbGrT8N>fgxGh(^2e8En}f$ARlD5Di80=HPI?I~TwCsz
zJJH)yH#>@nmCjQAd!0xl^K2-=lRs~NNHMzuYwjrUD~^g*jC2+qpYW6@fz18)T;jja
z4KtBgQ?~Ya1z#)AfhFBBmDx=tZ3FfOwlKJ-eo#=hyhI$*QjA0Ws`pAlM&PN=pC9`I
z#xuXBYI^@zoQXHMuk@8Pu>J4_0erjf2m!jeD95VcwT;dG%<1}5uJ>V;D@*k7lIhLy
z(w^`t@ptag<GwdS{N^xC(W`%J>Qun)2c-Vage=$MK^`&1CoDv0q*R-9TD3>?EtHym
z-t9YA=~>+`{r&k^;Kk1!r9phFIL|U=$^wG%i*+RfO}<Y?3*F4|bg5_YiM15$c@97#
zC&=E~KKu-&zYROcBJZF9<I{#<aPpQb8We1o+e)k;_}61~j;PrlC1h6MnHO@B0U?RF
zZI1iE+Nvuk4WP~n_8(xw!&|hF+ntR}kdh8~jvi?pWCqw9TR8auOzi_`J_^?Bh_i`b
z1i3+N7P>--oP69{y9lh82q3J>nQIPRC~y_ywd*;=3tG#``bJW?bbd{#^zSp22{XdI
zyz}0_M*yFUFMPs(Ch+N8YuF3%89!LXmoIk7DLtOOQ<rt4=!NFe5aM{zzy>fQO=p@4
z(6Z2RjenXIVOF7E9HQ7-BvoVe*ti`pob<~6FA?(AH&BNqtCvyVYqfeOIToc=r=G{i
z#Kn-%T`U^QbO1+}K;V}R<O{l?0njeApFRU$-DMe4KmXoz$@5x=u`-)H5imi7^+#{`
zXLrfydXJ;t)J(E?40qh*XH{KPR&;j$!g@MoHpS$w5^FNzHLz~)y2;&Z@=QUAh%orc
zq=X+vJe5D)bs!m|*GTVh4?B;s+1doAW7rH-?;1`ICLq&4VmZ3tg*d8IR~X*zVDuVD
z{rISkc01ksy4Jk4cc;t<QZ<5$fKA0pz445^U74bJEW(>tb*9?SzrU7H$A7<|Phzr<
zMV35<NTORD6?RNwoi4N4^w?gQ5qqpDbM^sqtA4D;HAnwJ$(ro|qvci52$Ql>pZ1bR
ziZf9!YWwJE{ve*V%BU*s7lbIihKkfJb&@F{nLo9!j)fc^%6S+#w@pb{yiP$g5#<}t
z3_ZQTH$NaYv0`IdPoMtjR)+Ne-0OkQr~;}}btsa`uAp(|sS#kLIp|-Uq%TD>@u!RN
z&Ho`xfo(FjgdgADY_jkedXvwRF|B;Dj~r=ncs)-78uzEbw(!K2KthGz`rl=br4HI9
z<Vv<w2Kg|CH>*;}VwBluh=pTFyH|gGRc(DHB-6(`Xi?j>uo;ZPn1w2{PrM?HCo>H^
z-H-QN4-HHbcb#kH1&3C*zX=yF2o`WX0HP!aJzzm!s&TrhA!19i{9#c0Ggm72KIMs=
z(n#hbWy{?>vxf%EOx+Wny)3cr9#%$yC^}PLf){Bzo14?H5j4j+>>Os>H536920Qy}
z^CA`r2??4#iKE|v-{-rS{am>b&DE={Z+UA)%xQvBXqTa3zWcxQDV&|1iDpY$$T4U&
z&N9APH6B9{e^(E}@gNe*?_WWw0By(Ql8XR(a-;{#@&np1WpYws?K_`}i?dRbrQZH&
z3F-T>G~+9R!PL0UpE3VqG>IX=-<=&D9ZeK;8PSIo+Svuw%di?q%$mmusq4UIy{9kj
z9WILY?E(V?a(B6Oy_4=0`CGE^h=>*bt#=6tCJZj=3BbTb*-<dvalk3LRwF78<ET2w
zYF(?tB!`(J58QlRe~C|a!Uj`p^yU&++qh0wH{0L!5iR9&Yog+2M!~tC7spZjCi{Vn
zem$T>aV6uVb<$x3doQvi*dmT<US^*QaUV&V8W<cNL2~vq{@Q4Bb8|Rkw-|24fdmss
z?)%<+M~zLnxt4^295tf;yZ-0sAb^LcK4qUL#jFo~e?!tp2OQ-w2tz9m6fm^T-+(jn
zy#?6S7x0Zw*$)F&^}SZS)tuqtg_Fx}4i>BS?X?drhLTnAU7#`;&Z?8*-)qH+x5K08
z-=qbwZx%$ncT?=YjD?(o_kgLNbaQi>++;o7SohH*w-rzuW}Bzdy0^BYXMIN@^d<vU
zl*$a5-(>{))De;YN{bWk3b5U9B3*kc@z}_uwBH>v3X&nZp=;^(q!QY*M#jc+_~|J&
zGlBrrB;iIU#DS%<3)g^ntp9f11v1vmOTsa-=vD&68y<Av{-XZWY|Kv?xTi{<0O2O%
z!fA6($(BFj7iM({(vqI2A26bzGrqI;p|Q~)cgqX{WCH>o)#Vu|%w2>d1lWJYcxDuw
zKO(f*5D~o5d*enbf_|1F+!<Bla;vY^t{DOl(brut*<F_o)Auyn6dO)-85)&SwOIH)
zRaBD0U|dP#d%XQg`o(n5=Fh%@AMQ);z(71CcG3KBdi4GKcLZzJnT@?YMq-OK2<S7M
zlis~6B9nM@Cz-;8a5lZZt(%ns+hrpo5C1#1j*kBHH6p*28%f?MN;U_530;8FdhHrK
zazzWz{>hN`G_3)tJ`_@@OUlaJSW27(54ciF0j9E|3RIRvj)UBlXY?+d3_&guejMAl
z&bl$rfmL^O2}R>|F5@&=qXo>o$@v&L=qqKoCy4m}voe$~chw~~;r5fdJxY;!eDjGa
zN9`<^@0a1w4krFoSsaZ<MAtzc*{f4uLo>6Px%4&ut){r%84RC(=8<{!{Ow3+$lZ2s
zc8}`lG)Z*Q%Fn?hpsm&1mnp7CVDyR_u7VD~dzY9vMMxQ?k|vx{|L3I+IlJzm+^k2I
z-h9e)FBPrWU8wAbrO%OcRA8T>2y#Q36kwpP=25@OH}Rr3_QQJxN*-%KJHK>n0%A)z
zJ-deC3VBE|`@>oP)BRDl^#2Rpu$@A0RL=yg1J;4rtqpw!oEp^XG+sUHDJI^H_wrpN
z8rI6yL7qg$DZmf;@4h6z_U(604Q4r-Mf&P(;7nk=;KEHTGG(WVc_lK{brK>aW)D=2
zyLZ394>GH`dpM6g!#|{=q(U6(T@IeUR_sGPp4?daw8?qk@MJ$7Pnfw-HHxmsk8KN1
zQ|A+WS`b4(iG4%dQo%ua^WkLUD=$~i6>x6#>bL?<*77`3a`p3Lb4>d5cV_ZsV#DPs
z($;%lOVKiw-*-<_f-%jq)rJl280qj2`AIWJ-3O4l&Hxv4p_qC^J9=t}oNOSGw(lu&
z?D6R`w6MwNVB3}i%d}8k11m-ote97bGLZQr9k$ye?n#iL0ZtbEF-1Qav@20#Qx>Cy
zb#jWog?mrD(zBu_A#|aSYj;H=_a7m=$rs7~OfWX|GbcCio^Xn<HEG6kIZvlLc|7E#
z-To9%4>8^(SUOHl%_hp+F9!5<s9I4!GXMf_=uDIEHPn!}98WkmPhnkP7G{0B(Y{;X
zi1e**1A-pdUCq92L;G}nP&(u$PWyW86dcyuBw1y36oVU@bf=1vO;3hw&pDO26WU&T
zy0ZC3g2)T1W&7yy@2uXXdcc)~!@rXjnEf*{bAj||uN{B0-2||M<=RZjb|t+g-f(Ft
z5;w5gZd1u3LhB$gL+w-bC-qemG?naQt>mlvp2*`-IWp#g4eh|0$*(Ivu_q@LCqQl@
zSlhZEmm2)tR_HjUD=mnLiaU1lX%)+842mtCK#m~9l`T4x;_k#(-q=!60c}rCEDpJ5
z18f@X05t7R{XBE`v^rhQB42#*#uN8^`Kz{P=eT5(m6(8Mo?IgKM@#=)JaMS-95f_~
z=OCcm4;dOV#K27p3?fnG<x|A5DW>)OwDY&G{>dGkS}gVu{>Kr-%IF6mMJB%av%e{V
zx<twDsHxV8e0&C`kBN9oTlkQmzpP^^>D%L^(c8`Ucg+C4`0*g`A0_a5Cp8g$N3XAG
zQu16pIdOBUMJ&x*Kf2NS$Rb61wEm&kS4MbW9jq6q7X}?wuasT!I!m)S+uK`HSQyju
zKen7{0}wDgWrF6zX;`V4fGANcc2V!V`e$zAB7UbX22LJ-(|H&CA={ob4KEUJ!mgJh
zq<czuHFNRv{rR`9VGr8`2wz|(zG!DLa~wp_m%`>@N)j@N{M?+gxiLAu8GuX2gm(<V
zbAzw`GwLmqwmoIbX%9{3IC*RuESL6@*?e*7>ydXT!qZ`9PRvdI?ZID>>5@mDUE_G6
zH^JN~B`^K6qsFw0e2ijuYqMCv;aKEbC@$f+(7Q|_lB9`k`J4ZU0?g-cDuI*^Kbwvj
zB)Xr}@4w2A>{AB*uq&Nx|M3j#+%ON*Tc=bU-Xz2p+31gwabPYN82%$34rZ>OuB=Pn
z_{fW02n*c*yB`0?=y-HyBBkgyXjoj<e+J%|S^LoGDyr6cFafkgdFOwe1%_w!#2B6o
z*T}}qbg<UWjk#N6y*DTSkt9C?z7!Z%lFv_&cc-uA{8*D>IqPt|9<VeB*1sC`!p8%i
zX9>aj59*WsCsOwHW*yy^df4>K@8x)c3A5}s?8mXR!KQq>Q+`e=6_Q(&lau4HF<wc8
z2Lg@QeU`IBib$m5*PHcCy%%Rb4A{Vt-e*7wFhZEYt-(y1z%{-WWD{gq_TAXIQLG7M
zx+6tg4`D8I9scz2)ad5To4~B+0q~L*Ld|{OufCz7L2_ABL8SaoZ|y58YbJQzwOz|s
zz()80cn}ssVE$L_0xUN{<DRFhobwQssBfA={J_Ly68d6wcoZ)>fsdSsI86K1R&h}I
zp40sCGgMfJO8K8(Pm3D)T0aRrVfHqONHeaC`or<;mA4Ma0U}oeGNh22uMU4M24KvS
zZ0T%hu}6F+9ZP@^o9`@iI+L<N8~Bzr4b5*qxICLa@yu!hx0kwRX=&+?GYGm<fGjR9
zMuE0>kE6wqlUckVV2DC0By^$r<$Jsz{R`<6q_1D~UjH1?_AEY>4gu&5DI~y!g2>5l
z@^Bb`UWagRcQ5zD*%!Tq_&Wi#yMUi(6udETmIiaO0K7j{L|Zh`Ey#y~W!~OeF&x#O
zgz@lAToj=FA1efR5;mb?2+1`n)?ka3{oTYtpCKUGj&;#OZ_umm0?MuOW2mH!aA}v^
z&~8<emGf^3AkUWp;<R*_G~qtsU@Q8x7I71+nmUK|HnIQ=?49YE4l5FN^@xHwVc7@y
zx$;g0EYxAjfbrr+5A*iznqHQA6&gz15O;OHBgPEJawDx|-4Vnq?ZS^By|qg~OrlN1
ztWJ21nY@|pCam|xa%p6QGCPWWTj4qlv43MjgABLI?#Cg3CeC29HbEe0_Vgr_BiidW
z7MRz$uju!z>jo7XS5FK1AMZ`P?4|j~iomF>lXUg;0wnw84b4GH2OZhIdeZkGTW?nF
zj*$TfCpH4ikZUNNm>ZX89k5__gZ{`TZN}FCNF(8L2?1mcO_H9QlcTS+5>j0xPK;;O
zF6#CI6YqVVa-Ay7?{e~Gb1P3hUEN74FcfcE{yfRjs&bxbswgYlsSDx=3$+yM`S$;M
z+rdkJQ92)DxeMjs1U|N8LqpvkDNQIAyQONz>2izYk}uF@bMQK2i|zXh$LHnEbXYi3
z2<i}8(<h~xoq|Uik`vqWb>ky)Vse0+kVOakmV=k?fo(V;*jBv3w)Aj*YbgZbKajq*
z($$r-9@re%(75N<UjC0@$&JVEPI0@845u$jL?G!Y`@cKUJx&;*b(>zBJAWG)>wX~~
za2sFxLVb_L1o>!XayppL_vTdW-|CcwFDTni+cqw)a?+nZtX1{m#;4DWE;wJXPg%xY
zHH>fA+iIbpqhN#jS@uhosf#kQ{F?IU_11<SS>Ao?U8V*2#&wXf$_Z>0>pg$SMIZ5@
za3>fC>41rfEu1a7vxyA=?hDH?MorTJ2KAAdS@YMa@o_2EPbB<W2RyfK9n-le_e<t-
zn~LO$Dkx2sl<nR9+W`g{6k(=%1vj2ELGs<Q<O}L4eMzSzJV4iL>M81;1ta!%vFU=i
z%bMv<#1b0S3!1)}9=OFg6u9C*i;?3qt!326%cFAxYC!)WSz>>_q|Ma-jpq)V#3fAp
z?QB5GZ#yGQDWsJ96|?(GfT=W(<2U1M<Lqok?VfjAh<6z>Sp75gxiO%BW{5UF=pNS(
z^)VzgM_zQz&E)Bv_04{odfxT<offM6-7vaaICk(pytN@VGOzsE@u^-h*4hj+pKRRX
zOD9hOgJnhnYFb@ei*fHQyr1s>1w^pp!3lp-{c)*b>p0UefA+)-!7IU%Q<mRD*bA=}
z5eQNu?{k;B7!388FRd$U2#V#=Nk?|2kiTQ0A8;;`cJelZuwin@g9w_RRsM~yhP|cp
zqJ8c%?SlVl0W9TC!VfPEZN?MP;LHc<z}aw0bI)e*in?4>Vhg}H9Ch}}q9?Zr2ne|G
zkdjD#hzf2D0Ogr<E_Z<J1E9p6N~hlR`P5q<=%jOqk>mtMEHZ1eo+ig(W>jX$@)z}$
z89T)6k}u;jQsr9*p&-pT{91dpS86DJzs-(hH&g=VAE2+q*gm~*LUYh{YxWUc>9hrq
zhUU;>t|IM6v2d0u_A#6xjcH!jY3J~`rrDbWM=RyUTm(z5)({<XmGHG3dEh4DSinW7
zlTL}Y23U_x$tU00P*|XF-O6YY>IfEuDpgVx(Va4kUz-5BB#S9bn{FWfDZ(2;fFL!;
z-F~(|0}lxz3Z3b|#A(IAEicpV?Yu={o@63_C6BQqCy9}hu2vp_{TG-D+ZOxv+V{Wx
z%(q$nqt(-!2-7b$#Fe_clGhObj*|!!+n%%TU_Fhaqnk*2sXQZVZ{=$`V}bgt?Q2}J
z={L*&-aV#^oAbUko*Ypy9-qHuS9^zi8L+m!!4b-O%)2$ZxeI)t5|ijt2=%|WTJ0)^
za%&1&m4w?B8p9uD#r`0!5D|}9X(cHN!TXA7P_go7JZ!tR4kVX&my@w1@81&J<VW|L
z+ypmrDdS3M2Tt*gV_n=`Ty-Lf8<ppLWSE(Iel)ukxm|gs7)2LCzZWrUT2|<}O*}+t
zYrg$fC`H3<?ttb~TQ-KWQ=zvIECv6s?0uvLV7-5g@1Eb;?=kfx0SIf4gD7N#vywt<
ze&evQ!WX9W2or(O@^y&8i9*bW>6j7D3#nz+DdzZvP)7Zat_JN(oFa+nTn&00cHo*4
zR-QDHdr%T2y#ATKMq=QMX=?^=a>iVSP)mKI3I}AG`)~yuIKa7`pEmYb0lBQDq|}m~
zChkmV4x*}&D(ta8vEJ|j^$kx|_E=+Fjkvp}J}X=hVq#+QA0IIb03<N5M5%1c6iXzz
z<&VWqk<Fl;BbC|x`Apx^P5bTkF!c2FK$!iuzL|lmN%+h>TxoY4rK&#joK>$JZqLgx
zhbdbDU$9GWt-AU~IWRbXV^IlsA08Gr9xy@xg6LTIY5!D|k<V)}@o@YmaNdM0+}k9j
zCo#8rQ>Q&QQfMp=+_7h4em@g<#Fg1&GjLSzb%L)<SfM5vVx(*bbBjSIro(BE7M61l
zLV)b<Gn}iF3MkKroX0CJ2Z7{wnJGQZ=-yi3(@(4>%;3N~5W9uGYfiXL(DL$agO_0y
z4L|pdN@w<_Ym@S%?G|rMW*yngSv%RW(z|3J8H^#!$Pp(XK*{V!goj6SL5c30!kntn
zsp^*7^K`Et+!PIGuG8)=s5ew2U+jGkyhm@zfC?af!kUQwqZuHY;wJKO@(Ge|V3NlM
z)=h8!(8%tWjxeC3p<z~xZn)}`nKa4It?FhS4&#5lNo|fs)*!tlw+UF*&!%-AJa91$
zdRX>ry_KhUW82$NN>H{&6r&4S0Lf`F&?-+l5+(nGds!TD<4m|eOY*nXH^dqb?oEN&
zpNJcOf)~LtrmQ3S-vU%KBa}K8or-G~^l=QrFcui4a8(k`pD{)HudAKxsXLNC?YzSH
zjo0>1`d6+9Rb4SLvBr=?;sl48rqcs}P11lfNJ1nHe-dXl-Iw^#s3fecT?Y$roTZTb
z6jGXBKrRJ5gapzg4Yt|+e%CVW{7yMFq<bBA2F;;BL_7{mekg^>rGAh-xD614a&&+M
z5fv9V!|}t#av*Y8XITr@^N;i{&w$B6w*1feMKF*1Q(j`TUcQ3M;ex=PBKKZ@E&z8x
zfVKb*D;`h``2CVZ$ATps;oIZTkn-ZVDNv>VX_<nt@k9(f9tUtdUsR(QD=+S`wLQ8e
z$cC+Kl$`0#`bX@-r$`;E1mKGeHv;c3GG#!-TyS1!93tkeh#-#rpWUl3d#~x@!l)gy
zGD#Xb+aAUo&RI1|y1aGc(pK2zzs7Vn5&)kf=@V84q_yosL&I8vALDYuEq}I(Bt0i4
zCd41>>G4zirA3{WGxNT6GXmCV=M0K-DWnB**fjG_qc7e?ViF+ah{Z>n?tnw7Cs!3N
zM*_#3I6(?GlMD#2s3%`>w6i-#Ugl1JOWl_U{H48{nZcGI<!bS&B>DdS{^FEl-|n3i
zF=mh$LAcVl?v;(`crtHa{9nJEFa*p_FA$VC#j54vAtaG&&@hsamincKlfd5C)ydQJ
zmPJVLU@sA~l97>}OC0>R(8mXrbN1ERTdnOek^?uKu4g?6<={J=)?g#8;afRD1x2$S
zbM?tezniAx<>lV@evK9zL>J~BG~;>3opNMzsa{~Pzx!&Z*vD3+98`Rz47EQf?rl0v
zu{YQe<R5j_S<4P(UwT0P{>A__w)Kh)`^p8-`@3u{r77qMn2G#IY#>GV#{*<bnv&Lv
z7Bg%a0@VEm7UNovaOWZ_WWP7jSvaZL1-?17nQb!38a@--e%PQ&Wk$+0cXu%saeKB}
zehhjb{=f`JiaP&WYPez<LO$O{?fGvXQX7u5p1TQ++^*9wHfIXK`2XIPg`z&(EKMo1
z-^LxU-ykS3zgf9VY0mh!6}O)>?$y&J6EhEW9bBE|nzfmAnC3b3;0l<&<R|<2+#c})
z5krj>&`9^$t%X*R5ynhKbO|eyRv<8hm@}>k9l8g@9nd{|jdGDbneu{~8n4k4ktS3K
zaJX6n*V79Rv(5B+HxMD{NfZl&XBTD1=+=H-GiBC{ha<6nAb$Gg^y#hxRX7DMCY9;L
z0x{%kGLj)StaP3QNrTJvpEIVa-Fs<ZsinQN-aj#4vgRf-M8gUIbS>x3o$0xHD?m*6
zA)elqFh-_K$<-}==GKbv0o_*HT*Y4v3WV4P36{i7(P>X4@tgCP1RXBV8nzNKKvn!)
ztM0D*4|{88;*b+T@zM!2pUzYk$>C7o(BYtntZ*%;j?Pv5d_pl{F_c$Vs^M3?nzj07
z?sKkPgIfACZF6rPoc<0W{h-R`UqKL+0313DQc}{gz6-4D0~iF_x^}bL8+KUJC#KhT
z+;w0xO3T6KRh5lm!CKpyzrQuD;hrZcW;ub+3E;)v%Y*O0iKz0Y)8oDejNqBABOOr?
zCyn$+V!va)QBZL+JR>xk-Ee&IT@!_MCUQD^iw7JrR@*9>J?Y)Nb(3`TabF5nE@_?T
zc7Ybuj#nuCz%q?xzFO-wVp46}Qu`Z9<jFj5xW|<8gY&}tNXfYnPpFZTNS?7kM$t6&
z<bp`(;+K5d{jt9j<#vxUq5wS4QZv_Qq+VvyF91EGGnMRGqm)m{C}v8)&mhd4zWE!y
z;1IUK!PxMy1Gn#^<hzRnX9>(gprzy#GYR;qHfp$K|EdcyJNBG@n_YjnVG%H#$IF1C
zlA&_H9}9Oej}okt`x=`JvlzOncCkQ&?_;<5A8&PlKu{usK)W@f-d6;v?b5D9M%Py-
zF*K%37~f`X`g=TU&FkuTT%N>5kEnv)6F0;M^+9F`_;G>>62MRezsDVYf#)&_tz&4a
z=X3jJOD!C$&iRBCZ6H_Y%;3MUlQeksuQejf&!QumfEz)*5r~^LVS%AVZ1|c~WgHT1
zliB_Nq^P&$#Pjlw*1E|KzNQ7;8P%e+L?41ud$_aZjSMx$!|e0#Fn#mhW#0@j#FP!m
zw#XI0mVzS8!NTE%#DP)g%hGLqvP8}CBh<{fQmGA;l?iW-uVj(aY?jwd^o(H!)6_T{
z+o#dPa>*Vig^jeH)I$=GgMI_@6WgV@zms|+HwR&yI2_;><2cG0V5;4aEKTeMbsfV&
zNd9eNYYX`@^6~M68RuF}pI`aNlT(7ptyo{@SM<NN+Xvm@zlK|JL$Sg}V0<P-Un%G2
zu0)&c&6CKJY;hpCp<*p!h#9I00YbZ>x4^Z$^SXs`4DGsFPhWIB%2ipoJ>!H`;(*4>
zZ<`(UP5Nv_M&IGL)XY&OP^&i50&O+6sK>a|cu{EBd2&{E!Ceso=GC<`XNkj^DZxPI
z5U1)l^Un7=lG)%EOjM#=Hs;Qi&L?eLHRS@wqrZ9O_V?doRp#LJu*$Uv4~{eCmy_Av
z74S8g&K!oi_YZ-8F#TL}{<&JH-q);cs+E4m=JwU&$Ky&DqXRfxbt&S8&Jf5uu1amK
z6O%^O^M$_IiJ`Z^blA}p9p3kz?U}BUrgDb1C-I_ucYbLu{wFqYSh=|cjUUcT9d44@
zhRw5{cHF4UQd0tA(@v3jl#kWdr};Qvea4;ot!nJMP=>BL!8!lxB>keB$Gu)m0{5lf
zXvRTtH*Jl}BG`f`HU<KL#$%v~w1$@~g|h{CRiO@6&6lK9om11&lmJ;|B_x0M?%h#v
zDEHF>b8>dOmxfo!BTGrzxp8k$<0FKNi|Z_bM`s-UOU@XC&RH4g>nC|<c7Vgp=v6N@
zJ#u#OoMcR+FWBY_#Y35H33uYn1$2eTvx;UE3#lW3!~AwOHa8&>d;Ln%Dv|V8NThUb
zfkqI6&{U&KE+Eya2Y@ek7(vaHxZxn&qf0Ugjj{)Btty~6eF|v_p@;#wU?sVyNBw4y
zi%Dj^o#;$argrJ9Cp&0=J}g`f2Mh`WuEbm@$2dTTI%ZEr<pEH>WoKu13nFo{)iR->
zp|N{*N`F%`9U=v92sE20wNPE&A2kyoyeS|Mb3a|8RAgDxzor}*y7LMll(8r8qVJB_
z!KL0)NsxQV+D=gTE;kEqm$P_XBlG5@7f2oRd`Ui#2E44c2->g%ViAlz&=(@pzv4H&
zmB0vy)XV;mF<c;m4A@kB>kgoDPu4wZ3X~NbM4$FLoITN$7Lhns2LJiZG`Dw?$Cx>s
zlZg_g^Sgj9Bk@EaAaxW9q+7DQuc<#UJq;z2AbFpYV-cIuNa^sQz$F2;c@MboZi)DE
zF3T%`&>m(Y6QcfKzxKbRoI+>SKHtJWqsGca!H3I;(z$shkWILegde<5o8IMq{kqn|
z8J!Gr!5(`*eE3iY@OXn`W8T+Y*yUAKJ5T_JPG6FEG~2x{I)m;b00N`~859UOAt(eY
zvP)F2TBM|<v+k6KU<o(mqy{#8MO_tMO)bId#he^h(@j)lg*x)o=Gb1ke?s9B^?;ro
zfGluo;d|-<=?vZ5n_EsZP5!@tx7rC?&0l`u1)bOE>IGf`I^A}YI&C23w1&sZTk|j8
z&sXEzF{L62NF5XC>gp25S~`H9JA`u*H-J@jHuO-HW#0fRLUvzy)V+Q?2>g6YsDl}1
z3~;R;U~eEb!mbAlr_(K1v64@wCyTtVFQ(3<R@=_rUZd;h&ok=xAWYJ}XyJY}xRnQF
zW1ix61yGDU3>S60zZNKqCFbzR5<a*{$h>K_MgUR}NSRlM|C5*EB5{+)IDi_h9zb9?
zz<0aAJ?FXpd%GY2byxZL9f<#Hr8DxSbGi25o?muaW_a}m0z<XYF_eJyIQ%V^YWMWv
zbaN32D0_F^X*QpW4e{cCWz7vBIOu=n#~DN9m4=Uq{jdY0_i`j6Plz=&DgrO9GrtDA
z*?-0RsUUHK8Q~en0JJJ=^pwd(|IxcO%rY<gtL{b<*@nb9@1^xEgZY?tce3IedR~_p
zCk6i98pi}qb-u-nxlIv-+_CqmSSy`=nnuHe6Is!4ea?+^<K@aiQ*_&%4)Gp(xTNBf
zw98_rG;+G?)+=MQ3f43i8!kBG{h5{p8_~Gg${21;ZOr-opKb?`-OwLUiIB>shRr*(
zD(0=36O9HeT+(NP<uCRVldPZ_|8;S%L!P8zH-5Zt9SP0Uw#yO;y3HQ<azwxA@Dqdn
zAIMr|+Mh}A-3}8;N!VZF2#jEEGojSomdolJg`hgMN0?R|OV9`xa(m+*@t~&g`l;}+
z+H)e4v?T+u<#SKY6yLTdsV!&Olz7FC9Q%q;dagRP?a@DWF&A#XK>ezmmC^?=wkM72
z+RB7uz4}GzSHH8w*#(9SeewaQipEjE%+;G|^Tzw{R5QY<c<(_cXvN9w`A&l~ykPol
z^_19=h-D>p4-fW_!6EesWRGiwoqap44`x|U4-fzvHgh9=vomiZPl&I;oNW<hhyHkQ
z<@EgIclep(E$n2(c@iG0?zQJ0Y@JSal$aiHgUMr_z<4KaSk+X#1`ls2+0D{tjwok)
zwRUvcYKW!_Lpd9tuIB%B2Ps?H(A&U0pFi>qq_ap(7&a9G!&pYq@@#r0=s=gZicmXW
zBnCtP`DczL;ZWRu!=bp&%h-0o9}mjmo#OczKEiwfHGjMqffN>B--vH)k1}B2JP+0?
z2&t~b%FIrW#`F?xnl%3TQE@?lhK+tV#mQWL?u)a9GaFzo(oficGfCtZIYo5KLu8g=
z_Aa5R{hA)J3F+kJ0__Da>DaX##3zNi43mBR_teS#eU3_l@^psA3Byg~heF%j#$SHI
z&();PkvK~XZoP8nC)>TX=c*@dix3gtZ|GBFkIl)W4v?koF4{T^nc+FjW#xf^)@7I)
z8-E>0Q+lAUu)ap{WZE!!e=kq@#J>|hul;ez$t-=q1%SX@34Inj<EfU0djFfE;y7NC
zS{EEMKS4BvAN6!XV%G2gdhhgY>&na4BM!Kj`UK0C_naySheE6wC*f4%-1RjHfNUT4
z69pLuIpfGdc3iW^zqfvFo(Rr`2YfAwI;1X|u*ikdIrD6X``mOrBY*E2<Q<%_+biuH
z4BO<JYujxJdW4*`wK~86@&x(>SS59G7%`C;&&+k%%ygKMfqf|BQBnPBAtH_B1c;ZT
zZMG}KgK1-GjdoEW)J#74S1a>r+PQ3SaWfIvMcjV)0`J>!0nw&}^}%Vinb;a!y4M>e
zi?-bTF-Yk`y>)AQpsJJf5;e&BfZvV#Gjz0<I85NGFQE(XEV=}h57$CyGZ-{(gtV$|
zT*V25L~H!K!fyIHlSBsJkICyC;QruHc=Rm6;cb^Ig1E^le%TdY;UOzWOs^GZ{adqg
z7sI-Y;Pyn<yWYBv7+Ob(m2-!l?@T{<^k}CMWVL>Gw2RaRhX@XG@-kur@do1y97NiD
zAQ40$A#eFY^i$Qzrf;%PWbVOFJOPj21c*Pq;fkek)G?d4Bri8__0y^9>AK@*0I!2v
z7Lm4Oa@IYwS4&R|KMH)5#@dpklROkNLS-c_r|+j;_;s83Rpd&l`YNc7t`j+(Yw4Fh
zIq$C3-}B5yxl7Z1!iSRilG%XVv|1<qXAJThgH<;lXg5HZH=oatA2M$HjLMjI1}sRk
z9J+ltH8r=Ts2-5_3m<xpC`BZn)?{QmnfBudI|?ol*y7wL9H&OAQnQ6mtDLBxTg);c
z0<W<BnNU124&Iz|qI0=bVnJC($^al$7B@8RKpS}DM@NT;*YPKnC}b;usA<7{4n08t
z5@k9Ggp8rmklwQX^PJw_m-r*d(>#!D8u?SM4_7Lrr>FFeax|n=bm!?<8Qn#}?xWc4
zDTk-F_=z&)bo_pUBpZbg*myg6xM$8WJKsI3aXTt!4<u-&Y3FLHN~Q`Wj?QM1&<!C7
zKY{JtKde<+>L-5TQxWy2+vamfPxO49ir+Ohg!ou}%|v|j5ZXs?2yV9B%iMI%2`y~-
zEqk1JMZm#1ALzwVT6-8riV0E)3IolPOQ%+Is+(rYcKaQ=<#`{9CA}ho`nV2eaAhd`
zaBdP7shgnUPHTGP$wpfH>_fG$m$0vtQEp$KH^V=RKNQ77Jx5ClW~!TLbBK9Zx1t2n
zVKWmj3&1FfY{@~gA1n{bR%6wKmJx9|D8J$EfO!)%C#vxnQ?eZ3PmgsWlk#%z8P5?&
z&T&Or*7s^_?{!_J4vp!5qH?0PBfFt_B9nSroaD%DDk$^DlkiU7>P%?gF3DE;X~U_L
ziXUQ81Cbx9E&WjO<Ok+W>u8^p(C0vs?YX^6=2mQ&Q<?3$=^kGbQ)R!m)_mI>)$I@t
zxCoqaZohdtk72`kM0f8F5z1wjtu=oj`P~n9+%4)^f;kv3ov_#bU?f8%4`W>;Svp}i
zAymolWApSR#Yy^M_?y&u7zH-P3=W|4K0?69J?wo}=hJ=rvGOqKtiuzmN8P(a((^)6
z4JcF_Fc{P!WK4wez=sv@i0&+pJxGgYX#5dQVkGx{Itw?f_ItkR1awVfzc7Tg#<DYn
zaPdyhdcVS0EEu?Ts{=gG%%EMm+`&;75-T|u<mE0omm`vu-o%$f)N=N|W7CcIwYauO
zW~E#0pPcSPwqGtdOfw9mh0jyV=~J_vN>OHUL1*2>wjR!%JF~Qs)Fe>~^h0qXks03)
z1VUq~PDQsYDKO0H4PNzlUQRKN&G<_Y<Iuj#C%D@_Gw##1a%XCNy96lhk<%BvTl{l)
zj|K40?w&Xg#l!E0F-zJCk~B~TZe`v}=jS+cY{Q>RM3c`o)9=eWymj_e1Bp_Tm%Ul7
zs%*!~o)2Um)6mc~T#aZGET8us&-@Uh|M;<g+@Uy)F`PRCRt@sW7`*WW>;G~l(jNr%
z^-I32a-DBqC?EC8^wfG_=tC8exbv*#BPp*<Uc<t9xhHb_;bYJF4mo76<r@eb(oRcP
z5Uz)reRSov-B()rk(YW`D@Q6TV+vnv?E2oLOENNMv;Sr1SQw<Ca5cVWSWzCwRp~3G
z{1pX!IH&wI%dA=?x}z3PDOugBCqumnHELhuxdd|Gii@{cE=eN;`Uva1J)iNU5e8Ct
zh>@E+cRg*9Z?t$mX>j@aYswF615$!UEXV$iX&XY9<*GC0Z}YM{6o2DnV*3`!*?8r5
zI1jP}ANK2*?jbX`SK_gF7~ePPABIhAEzm#i9h#ig)bR$6m&U5-4tbZpU^;|q>t}5y
z7n66s#s{&v*rbF%?A>`vo@rSmmsf#<FiZ|xdI@!V_6fh;T!km_-D`t+R|&ybypPY6
z;+LJD-uOi#{X>|VQ~Ddzt)L?Uy}T{@jJS^Zm9Ld5_NRuqckX>Q@0ptftg<f$*UO1H
z&Ec!=q=Jum#C{=n`9Q#VqdV|=N5z6~e}W81ACt2&wzi{1PfnYUS9u{m=uLlbaCz3l
zV`Z;mWJ+-)1o+C&x_Z1Iw|_nJP93>w1Yb2VeonMrKz)EO6Czg8Y`p{W<1q#m6clg_
z{Yh802Z9d&bOHE|FwZ6pM#Y7T@{)|Pp&<j>-o&~TV1RgF<3S{rk}WpB78IP+c3r>w
z98l05o0GLe8-mKrj3=^VWdgCNy!oL<Uo=d3u~Lxok{5__A(6j-zxQ$qmFE}RLJZGG
zNZDdKJ&?Q7PxhNnU(EoDK@niPNbku$YPjP(En|Z+6+sRL;ozMa7#bSRyQ!_1%l|m$
z13nMxDPqni!+>_5Dr;K^1=2;aY~YsHO;QS1<Jm`n48hQ44`vp1o;3saJ&W%&cJg1I
zg*`0UlQ&i$m$UeJd6LKkpBcwwZ4wbFoI&i@ufoQZu_$g4S&T5;w9C<gC3kp0RNqWC
zLv)_adNQXAl8Mp9i?Z~<b3e1`;GwM%o*8$tonY_!Dtu1Zt&ebtxIJHaJiT}Sga#xk
zu{jF!vWV&+=-qkvJ?8Bhz5zVNTwp#|_#5wBZe_2lUdyl7IPbxiAKTW%nwx<BfnrJS
zS;Z;p;k_qc6MVS1xve|WC1L9z!|p+aqi@2<aWM$-mGzM!ii|;9Ke)#d_PH|d-AiPz
zp`0Gii-L$hxhx-b<xEab$1HydT(5V!zWmB;9&e0$aME9<syZ-4#nMnu{G;wTG~c-V
zHVOB)wa(lc>F`IZCkMta0<w$Wu*O)I#LKArX1kJHA?qqTw)$zGO)DHZR#xy*GpruZ
z(Ct2HOb2xx^tO&^i*cL3M_%g`oRH@uVdYE#%zoc{gZ`uLWPC(^>WEaP#-T{+;BcE;
zXR+elZb$c$p^WFg90b!X-PeENiF`S@jX=tH5L=1vyg|%e!sttPi9CxGFbH_B&*b?1
zSEA$T(&24uZ#kkHiS~Sl!Se}vG3R9QvyD-&s2f5o0y3kRWpL?c*Lrt;UFyI!2vy&z
zVl_!pD=3xfF>D%8NgI8OU##|FR$Mri_<_a5!P(3cB_$<IE&*U$lcxaA?<Ij4;&xUa
z8d{=fFYC1MBetXQWTz{uW3ZYsth{q<mvlNWrp0MrH1OW$W{A!D3p|gJplC9!OV?j7
zYq~?anhCm!6uz3^PIG))9XSh7ymY-z)OZZV%c2w(th82H$2DqLr92iOV5N*C?*23c
zMaT(p=poJR_Q#(dkQ|s%S{WOUNOV;d36jfh=TkOKwm$70{BYX&U8ZoexGh0nmS}gE
zp!jEVw_$J^-Ytm~lNZU33f~@f-c^qj(`(FF+`Kbx=|GFCzLj@<qKY?TfT`oI-Tb!K
zO97gm?F8%4+=>eKj6S3<0sWtCuVBdb4KK4TpT=Tt-XhBh5OwJZZ-%T-H)2XTq9x{V
zl9b$@T%(jwuy3T4w?FmK(;Io=)EYZcY)<)~7C=Yxl<-6v<?uCXP-|f`Aor*5Z>^TK
z!s5yzk_3k|2*8(6v`-^l-l2i|JQ_2Db39S?1i9~1e{z!&(mZp1JFkf9-SZ54L%k#>
z#{nZPt)Ai_UTi>C{?)5j`Cmxpep(-|TvM4Pub!t&$!Tf|h#g&PY~PUh@K8r5KIF*d
zE=c|NW^|1uW!>WDFH7&LDbK`?>3xg-Mx{Gr;~>7xlPw@-%D*+*l1=;yIMtnbK9Uoq
zHkhJDC!@JpKwi&2M2esd*b0?m05Ro~)Rvb4wjgMb$aPP!Bckt1{CTZxuVvLBf6XOh
z<k$)FaX5oD21!tD;bsi}{Y`S`+fScAd#`@Vi9Rj_+uCq{|J&6*4qNUBgh}(k8P454
z4<`EqufliUER=epj6&+uE;;5orPMl^e|X8eB`d`=>&?$ApH@|J865l()pV<AJZ<4h
z`Xi4Eqq=d0g@o*c^^n8rdu4~9`l7ja@W^wfS^1qR?Z4AqsH)9PWnb{J!_NM-`wUIT
zTj8)9UToL4Nqy|AZ`3wN#>dC+KGx76K@7yMuK6@h&sTi*-hEmYQ=D)tZ!i13YGcj|
zvkbC8^OZIyzewKM+4(Z-10psufqLAA_@k#Ch)}?&M)H1Kos!`8nSWdx#dnq_`Z~qp
z)SS59*&LIlN<*HSIOwAf6l~vqz0>r`Oh`&1QaxDH441~H2v@Y<bmF;vyE~NP`{xRv
zA3rS~!W?W&#Pakdf|%W3_yQ}B?==z@c!;y3!4oq2foiK10zB!cjj8(TLXEz<aK*xX
zF-Smj^L3rf8Q>&8nyO2ASW)FD&Moa!$nC>8vLF)@Z*JfA$fISsK(;v3@mHgi3_8~^
zi*xsP82IcZaTD9SX8UN+vC|g?pg~ocKC7P@Tupy=b)ox$fE`-@Bp~a-<HwbOqA4Rj
z`RUYvWSo%;??s_eiLf}+caimVR?7`3a)0`b3PK*8Sa#NY9r`$Opqd=}*+1xu$G8er
zomuYJqE|V(=hPN6S<k52RUzW$RE67@?%RlmAI9Ygm{T=WTTnG{Tf~iNT$Z~hF7bJ?
z8lH0Hn3VjTT}I12(Gdgud6}3W@iYG1Z9=)4A1+TDhn}a!sMYzs^aT2z34f>=$J>c=
z>pp5pLCrw9!b%Ex9@?(R$cF<)L~Db&zkZMn9EpkjR;v7<N?IR@TQ~fgP%-q?Sd}d1
zLj=Fi_=d!!-jgS?D)zD?we`6`i&kCF_<+%%#^q1gyXVhaKTSms*W5sof9Lk3c))!1
zlAu|0;MB@$H+N2B<l{@%mFdbHPKFtjkUcjc`+=|f_>L52z?BLH78YYBJ<zod$lBcc
zJk#~H*rH3!v$to$N}Z!>>zuAz$uA0Pc9#;HYRQrvij`b%KdpXBRs0GLM?CJGkA?aQ
z5KB&PFsEv)hIO*xHl!~YSd_LJo>jvN{0pnth)6a5Z0me!JO+FYqZd7*YNOuBnOWAy
z9cPOpNdDSZpCzjoc0Ya^9-`zYz^T~E7bJVF{mC;&f!CCZ?uI^76vP!vn>OROBx;3<
zpSpdqpi;DqPK_zP>0lsx^z&Fu=gh!NXfgIUyi;6US5ta?qvo~myQr)K8$?T|(Tb}4
zV~Di;W1Ej`m%Svs#|~{{>R2pzQBqw)><)5tB5p4vmL8q<x4v*TXC{hEDVG#gpr=`v
z8^IU2e;y;b=Im;Y?|5G=jTu?0#?@sPWIomsBWSiBJc8Hz;-HzDo+>+LcKJ`%{6bbx
zpsGzhS@L2*b3}@J&v%m~jM%JW_iTUQQxzU==@A>pZ;_Kdgww5@?upcEflsMc8F&Y-
zG%y&au}>YHp2>MNPEQ2!<fud)-xd%+!NIoxL?anQmTpc|A)jWxEq<Mb?+BDF`gVoq
z#JBnR_!>FgCq5>bj*qVGPiH5jW>=mSR8&;3&dx;$Y`)9L@GpII%PXwS{`vFgYdR+u
z#IBTY#;u!XLZgHE^-=f0nHH+=j|j|uq%0ee+RjRH!tvtD<?F)mkx$3rpT&9h{EZ&E
z%x=^vTAY^#^dzY6xDRF6^`|dY9bd^93K!Cm!wZ=6V1DQ`WEjnLVl&J<o8Mc=#|!^D
z@f$jA*q$bMB9xc#%`Av1!<ds-Lh$>YVZ7w@slw#2KkQ+;an~dU??{OEX}-U%XdGlO
zl;hv`jnmHfD`I$$jC;VqLWovJYMzN%%IJ`K_U^0|`>6PR<FM%2srZ*^5L5sCU$+YP
z??Uc<UIB?-_n}$Z6IG2264VBXX=%QOucu^_omI(svSJexpW{^970ThR`a!b}j{Kwm
zrpkI`$GVgjM~ir^dt#zk(MD__)LZmr#<4ky5*+4`gpn)UTVxZgv^u`mm65v5R;%tG
z_p&6tbTgEE`%fz1>2#Hu$?HenuhpswlLtwW3cm$pW)P7QjX23femE77p5(!lz;BV8
zvP_?1Tq0SM(|hjC74Lq9=Lad86X)I8(9;}r704S#I0p?0<ro>qgp(a?!)BGcZ6+Vf
zLqzlkF$h$8R*p<}ler3aGrik+>7|XAj2Q;)7C-riC1dz-Zb+A^Fm9*Y6qJ4<xbGT`
z4uhpXMob|Xm~Yk(_z2fjBrRVuja3BmtpT#ey~usE>hW>dtMG-;Qs=3ztLwH1_<V)0
zRs~4M!2)C_vezjuMe5v(@>t#Qz=IM^Ke*R(n}^5n-tCr~Z`9%@hcd?=m<fGM$0PXN
zGQ^mmXqd?@eM1-+Pwcz4ca@}J$*hyCzedzpz<%<e+3qK0|9ONWwFjdmtzY+O$ql5+
zY#$4KCgpYwNis!bYl7jWk3yY;+Hmf<bvc8boxr(CIbFW30$9YTkLHs}uO;hbdJA7B
zm=m(Y^>(Q2NYxIZqoY0U+3chdNi@|o^K(*LC8`G7qJjbk-aN=&plZFgx_ajfyjgX;
z$^6?(vgEESmTtFVJ!GnmT@iz*D`!lPce(X>Xk=nNcSt`|^MO4?S^di4wRa?kG(OKU
zg|8_yQrjFM%PyT$N|SkFF}Jv0y-_yeCHMmO738ppKk|EwXiOfcDy|;V%gur#2$HW%
zd`@5I<FbqsyRK~1De1Um<iM}8HX`E8NDL~OcpC^K3b`@<ex{kfxoYmGg4Wl(%mQ)K
z#5E5Yu&&4xJ%5Iwcwr3BCemZ-stVc3>A{{BFwXg+xiCjNkY%dEX&f<pms9x>zMz<g
zHe&dxR$e(g2X6cW!C3S}*=%++`SQriSEevH%FX;yL%m>Md}N%>%W7+4C&f^H{__e$
z=S6QA;fER)17e4f5|x~Dc#pUDE32vZy^)Q|VQNizG0lOG{M!=fn08h@eJ>S|8~A7`
zH3sT~ODSab#f-Bae%NL{u&XPqY`UVOT@n=>v{dSp-{aS;giEcMo1Zwkh%@z=wDL)X
zBU;-~Mv=2`H(59_hO<<gKqmV0FKL0W)9{6|pz{!h?1`60M)f#ev^u+ow6jk~KWNB*
z;9~r8NhDVP<ue|#8?Tphht&g%+h`2V8z)Lbw^m0q85rdI6ylKycM0Wlc4_R0>)J+3
zmcj?T^JUOh6S5!QjePRAFOq31GBYViq?@?aYCka5ll+?UJKs@0w7#L}*-tc$jA{w<
zqpv?x8PDF$BnL+D;t*dS?3m?dG7$Nwpk!LXHpplEIA8R}^*9a)bJhY>Y|Bi2Q-^bg
zy!*AYTx;aGEbDKwhOp1|CsGVj+=;`pw2GIml()aoX|kH|Avao7__4<r<W`1n^fmLn
z>6dfM@y|zQX{EoSB+!MKZaytWth<!-b7T5t=dNUIcXTyt<*MDfWLIBu1&BYRR=ia0
z&LsLZ*QqA~es5Rj4eL|h7YV}UZq0=cVvrQ6DUpU>-cT~!=W$nh6F@!8wQ=W;8O>v>
zET0qZ&oP>pu2by@ZBbXhQmeA$@>&aOhD^i{XHhDKidqpJk|i%VN&E0LpOVbx>2?eo
zwWRX)jxLYBR9}8T>?Wq^yc{^3;<LTw8n-~J=q@&OKczD2(E8U^MfWQ>j<S|K{+F|p
z?B7@{b$S_mUU}rHd+EB#+X8%N@mZBwZG?{HtOE=zi4Qr=>B*xdNdtvN%TfLG70=>E
zh5{#Joew=7lYi@of2rNLkzL99>-YI5Y50tXrX{%LH9+&2&m!l)9ta%?7=31XVB_9e
z9oPjcCpquW{s!2W-vR^bBTd0a4@`&_@H}y|aMprs<rQ9@l3ahSCel;i*a4Az`0JTS
z)bToIJp%eU@^fWF`#O^X?cE{0i3677uCRefm8;l_mfhU6ZO=+ord~x0er;HKBcpxr
z(QcMQi-kEA?B=QfbwwqQA$@RL;O20UI-t+o&btRE7&j9?h70UsWXlBw$*Oa9f$x<B
zMu$GdJ`~sc;;W-|P(l;T0k(@vv84&A)XR5lmd<a+NIlHnRN2K9)XzZ;z4Ny^`&{0H
z(tLXxtU<2MTLYAkX%{D1_+^AZ(0mX|vf5~t0wEV95a(_J27~2x$HZ<#@Y2)^qoKmW
zM-JDkXwyY4^ut^1Ha;0J63MJC>k_bI>l<KL$Pk5bxMWSZg1$-Ue&FzG8N)kakzf^x
zLsQ6$<e9X%n^Rl0*L4Y1`o5u5w4@Z<ioFq)N&Vc7Zr1Q)KM&~&C2^xb`+@MPGIkcg
z2q~-FyZcgnI3Ljm-$b+_6p`CQK}dKqGzdG?6hFyN1WM|sf{g+fs|zqXYy;mMh9^oZ
z^lz<Y=(b%n3Hy}XM_;HL6!;L$a1_KESB&MNQ#@}2$4Shdi4gxDQz<X3IxrcZUtn;o
zR3|IUf`nQvlqjC_q80)j?s;ooEC_5<h&jI!I*WEN5TDS{PbqZJC|<%DnQM#a)4bb(
z(j78ME_dAtdWkOXEwdBQTULZjRDDU^*erJX!X#{=J#=L(g3c@>ZoYbP>eilkxK;C8
z$6uz?9vYR-yG`Yp1`UTHEUXq%5AN*`)hT_o{@450YGu0IUNJ-gkV)Ky5ScR}7l8-w
z=H^MRE9)l}R|W_?&Oq5QBZ}H_g4%OZ6?kwx^voU%1=+egtW3S~GZ2I4kWA-DG_$WC
z<jIRer(h4!b?-iSJGHNXkG~NA8{Z|3yDzJW1h3XGiF6u0;M=;kMY?sDIGT6x2T1=3
z^F?oF%miq7>gwvcJ3BbwBj&dkHpa^HtGoUWTW=j!)faV*9=Zgi8>B&6kS+n~l9X<c
z?ruR^kdQ7>x|D7XDkX3z0coVWOY*LxzxVszd!Oqc_;A=~ueI0Sd#*Ld8gmTVW{`|r
zJYe##o1Au2jeF0gi<qBsY&j<MXDJM~B3@;I6;YVE^7s>ZESKHzhn6GctF<7hiwZmT
z38~uooOu;F%?TGeggsZ0!Men(EQ5Ef!$-IhH{6R2ju1CDw-zhKxl%%O<Q6nUV7j2u
z?54VutR}wI?3|ae-&Iplp)AA<vcuuo7EiW!zQO()rOnMkFZEUXg?PjPS)*df-mKcV
z-OuWW@1Nn|yJUiZc92-@B(k~|YXD`byI~c5`#-Iy048k7FD=D@B=jav22TMrAoKcc
zhxr5xgVvDfmU7b(I3{p2lJ?Zk@T2*V`cR-VZ28a)f&DfkEYU45%EaC2+DIUqfizNW
zB5`$baNs6GX9Ns8FhjDlDOE*}WYRuHFM#TtSGKmUA`cRup*|p$eD*AK)=Z|brX~rv
zh(la}PQAv<mz&^=RDjQWv{jsksIIOK1u)?pHj83|qvi{<D$%~7A*&^5D5o65*aWyI
z7(vmlV+zb*ZMfzpGYU{{tXA(VUwuGp@Hsvi_fn2>dEw~D0mv(e`6VS$%gg58NOob9
zR~f(z+-S{<QYxy71FYA3%@w(C2seq3FPWiYP{hGLqtWhHl_YG1Os!OM41kYrW2PAS
z8RTV2cemH~<7@B6y6~V#u&ZmRG-pqeYA(v*_3dm(UMr>2PsC7Y9zwrgbjifj)S`*H
z`sW{T1VmR?fV}BJenZ1~KS<gI3~^Yg9-s!TSEL~fw~{L9zjAP3?=cjU1$xerU1uIP
zvtIm%e3Br=XIdhuLQP?+n+(=aJbR&7|EjXuO(fR?@RB{y4n1vdQvt5~7=WXSfGF<?
zZG-?0Nbvi-y!+At+QP6zC=o<|r#M9;)+oLgT!iN$)c3&1#N-|m=6~$5dgR03LKXH<
ziXn6978yNA+YtAL=1^ei1o@hXtO|G+=ON@ZgR6&^y$ND00L_{hIQ+WdCSO%vRVA)x
zPo;e5l;~~_umlOLhv>-X)?-4fTwM2JNlQdLzd8c_{{YX(E9b_Qs2-qQ5wD|YgsQHo
zd2+tg6~?=MZ(XJZ<jQO-!!Q==Gfw^9KgJ^A6dkRpFn#T9UzD8;)Ru055}p|L(#5s6
z4S4!>0Na#iM9!Y--QO0~z9}3!&D3obVsSQ^Cb!}}FK%nQq02@-0QwVmfye#KT;4(q
zQ{dsGicekw=mq_-D;zWU7K(fzEW`|AbA#o?QY8{Kk{8kn!7dxUtS*RXOe6XNNS&@C
z@gC~n1+Sml5yLUvfaP>L+uBa=cqu!9r8OeVRRrf>tO$R2q@?q@k;26w=_(1i8jNGK
zM%kaQSX?)iv#y`Q0F3b)>I#XZ^7@r`b-GpCJs&mK2*8y)eS3vN>wM?o-1*wk@f&IO
zj<CsY-O87iMuq`|Bs5-xfo^W|IBHR<Rz6$q&+mxut;o+*ti&%IbQIrnQi(KE7LL90
zdjao)hjzc>;ora&g~PquO_s)2*0Z5rCZK9pT2z=Zoh85>JRmDeExxp~FUumc!Q-(|
zL35dfg!@tJVr06se)3+-<F3MLp}DrB)ADNy=lz=x@F`}_im@r!Bwshl`ER7Bi#s76
zq%7jlj?ifT=IH#zk=WtSO_JG(jp0Ps2aM6J2RQ;Oa{2cW1^%Cyf(p~HGXy@GM4~eN
zo&OZUjCaAnh)FBcBY07m!R&*yjfq_xmQ(p+hof?xGl#oyK<yp4l2X3QXy%%fp~a}>
z+pLBO`fo~kYzE7%1uaNY1O?+{!WYvk&>)tWqTZJ!_un!eDZ!_Muh<}c1j`^i0-NFU
z2U#LJ#%dv+<hN`iE6?6)9BNjc)|j1Gckl(=jX%LOL8sMV=u}&H1BI55v$Ztx&1z~;
zBtFbZCu+jzvun&@O?Lb9xtTXslQQw|BrBs4ri+gEes5u2!u}LH-{WvxnFagft<Qbn
z{_gZfbvj>Z*PDlL<R!+74Wx$!of!DV)saSivihj`w(%}fUf*(o`)5f!YaV6SIUS_9
z<sN#kq%A7t;k0cq`jq$Ht}qhOU4!)9zowqN)<qw!0%B1&@e2Ehn|*&!hB6i^iduzP
z*lEXR-Zc+X^PcYXRzh_c^Q%xg!ptjl<_OR)B`AcwCo4Dk?<6orV+uF%Rya{uF=V2u
zpharzL+>@7mtsDL4^jdAU6iUD(3rzFQhGai<8!&(?dBP>IR^9HYQ7^5waSO#G#atK
ze=~-QlXJg2Peqm0XQ*GEE*oh7E3@on_~t30h`FtkJr&5F!5sy4f#5kWaMWk36q0D^
z!OZ{Myk4%UNtX0*i*w-o9<MMyZcd7=c&^_$Vla#L-b6+#rO-IGv|XqA(_>;<VTpE@
z7HbHy{)qFZC*<IK3RVdAp)S?)!DyMG{gu?ZP!VV(L#k5j&)|Yok=t`Mx6Yh-T%L`+
z8t&P<H3y*ad#iPdArBv~_%1N}p|teYB<P~~6|+#%jxD_N<-PhaaWU&$N|euD>7?kZ
zUZD9>*y{)RQa4Lhupwn1r-Bu@2iA(^a-($;#UZS9CqG+Ps^>DcLE<6BFti4$$ke_$
zoR~9BJLGVULph*i&`socs?vKIlxw0dU!rcV`{@2`@DT+f{vA_`wX><lkOiUv@tK;2
z;>XmqV>xoAw8#?2a?(8Po}*DPrlsL#>QiC&gAoia1UJzI)IGDYY=vY*D$>k{IX}iQ
zG#gn9gC#%Rze1~U>ep`cmH{iP#H!QlDXWRf+yEV;?p3879W`;K&R`>9W^ys8G}I5E
z3NPER9T2Exk8SCO`Uo`8M>z{Gy&nuUa^4Qx>}U|jJ+xJ|C)z>xB0J(NXfPl9WMoVv
zNRn>$BY<)@?kio8`^-|tK(F%c_cw>(n_8*ydFP0!)Y7YdlUzxTIrp$AR)3KvoM0tM
z0tb;ji8PA@R)OhT;evLFZi>bMKlo)*TREd@YB77oE6ywqa)sbMewXG`X=~H@F%&DA
zGEH8B#2j_sLFq&sK>>uOg8ZW!>XDN3d&uL;px)ChFSKw*8U3-x>k^X+_bZ&jjf^W*
zReJnE+wT!N0}1ujG!++Jo$~Gy#QoN}gM7C$9<Z~_ftI8^hzEL9*|0C&Q6G0-@ZC@k
ztw3}5uj;H;id7By?T(!qEx_tiU^6Cqo`U=)mhc9DbHdJ@3#HTH)wEF+8&i+gel^nE
z3j0b<`4XeepIL_QV>9~7fs?t=-_X1et`K2Up2K<W>+H2%g6uipa(4+I3Nq_F&CuSE
z;ZRi5r|TD_wo3}H40|C7g4g^=ByVhaAwGz_GrnBh6qkPT!R?`okUU|DS*fKt^j=kl
z{e1Q%IcP8JXaVhx;?)6Gl>C`nZODOXP|wovAxcrl@fP4v4hcCW4|*&j$R<$Fo7e|Z
zaN`Inv(}F_>=gsx{bCOAD{nAPeeqmaU8vXlUzDckBZ}&2Xw-;~bpR}}MVa+!ZH{!d
ziN5}om*|&hn9X!kKU*X$%a<hc`kUhvc?u#5V<buq&Ccd!xICNA_!GqGw~|hLL+^E}
ziv{UM6)Mv`)^tpnpf}swJo-B{&6G(GbH-v6lF37wVPI<vc|;Pn@+zWkTAue)Xm)RC
zDHVZsYfR{P2%}_RcHnL|1Ntz+6{NqSi@M^Y2!ap3rYha5Cihl~rs|2H4~#sbmFJq3
zlQhUdlXmydV0%u=L0h-zt}3Y1j`c(1WkXib6-v>3j{<iQtP;e>Ax^ZjTy9SEA-Y0Z
zZLUbG7K--S#K*r(H@rBQev>PF@@o!;I60mD8ih{4_Bhe=L~K6}zk<1DY4E(huuw+z
z55&BJ34$>Z*0paZY9bQE>6_BW)l?K4|0tKx2vGT;iomOY4!X)SuO&TGOc!>hM3=fr
zzsyO$jNuJ7GVuVPU&tLAk6+3NFRR#71>VY!GO|-hoOwfWC25{$JZIG>6n$x;eJ@t}
zhoz-u&7!JQ`*5R#+@9Q_nz^OnUn=w|bmpn1TK6hl0Z+GW9uCYqwGiW+cTo>V<#V3=
zU4+8ndGG4paR^(E<AvK{)Yb(z#)zrgiOWXNp$_rbxwUeV2e$QhhuraCwI$(46h2{n
z$rTcqH%Q4WjoEhwxH`Uw$#Hutbz7lio>=R#V~wHs%&J()A?*Xx;>m};KV+8g*qEBH
z$*hq-PuT_|xDZA%z5VzFFS$=V4vC=d47tZ#_GAGqn^dmz;k*zI8xh7rC~PmB@HFQI
zXY*>(f?}xv_w@O-ySqD<qLLEFY(oMX@L9#}A@e1`s-7x;ufg@Wu;*1}0y`%s4B*rC
zAx?^=3X)h$Y(U3$vEgb|fAJ+KTKhT=wqtd^AJNCEI!#}es@baz@L=8sE_p*~o?i8Y
zEQ)(=U6wXxU`&TU3<`$bn=iE0X-^NxZn32${JGPmB20zPh2ak|PG<ZJmG@MZ*{HGs
ze*=Hl=6&sr_3K-iV^J3Z_2@r*qv(T>T=#}N5^I{)iJJQT^!_z3L>qJQaz5ePgKh>X
z=kE-L)lnn8QcGx1FS_6#f6!deJ`dPxV$Cgo_vB*|5+>6uQngUzm`h-F9YNy_;n8y?
z5lKMd?!3mdX-DCW5n1T3qWPMV^9$ip+T0hz{k>TFcX!bOp<c7X0-0|cP*zwA-Fd-8
zj;xp0IY;{FXD%SYjMAC71UU3}0YFf1Lx7)O2Vn)S4n4?L-FE!E?iL1wvze`ZuAhP7
z-3b*E&vT|J${pm2z@@p{k+*upK9D3_N@qRLYp99UwlpU^_pZ4XR>K}h#2@xz_0<q6
z;YlrK!p%k!7H3rXYV^Z?cu<dsy??qg>*=|DpR$ZrcO|9$kzcPN@Y~pdiv3Q~Bltv(
zPCFLw=z8-(v%O)II)@Vp62{0_0PB?bJgQgpxXa%;yw~I_1N(|l9*L=V@(oV17Pbv%
zQQOy7gXqhXMdLy72N!14_FZwG+aa1rLWVT%JRfZn5cwHc_p!3qR?@@W?P)Az#q-EM
zEc&$Vuy<nJh{oR*6WCBc1y2_WrTpPJq-9%%22(Q?q4G?uQUI30Ym+d%agF6t+q=?C
zd19#&0}Afn09pysXFss}fp${)`XMV%!1?YzM{a7ZdZ^Ufg-NZ6%b>@y$2gf{BCR9i
zr|60%k=DVCw1t95=}U|4Yv;5l_dIu%uDXt|%4ukYf9#|wmm^hCiQjMe-*`=;7oL96
z*DS6NYV3jOz2vdgsGa5cm*L~FDUA~<q5c@-D`Zdqh0F|9taEazuN9^53Zjelcw_R}
z(0s5^oafBBD=t-xL2eBlM^T9J>-j6iFS41EEkE>};y#OAir9)JDA2j0F?+Eke5Ogq
zCq82EQR|AUyTx%4E!vfS#m?M4%FLj?&kx3<(+ra`ETi(EnSaJhA1G-^R1_96(&SfF
zIT<za>+htcXb9L>$|ZkCK1eeOvh{7h4z52U#vv(C`L-c3Kj7>S6#0v2So#p&NF`1Q
z1Ydw<DosUo)qUU?e!G78SCnD#H%_&J_fs5>9Mn&odorBg=ILaltY9W0EA<$)M<aO_
z17OTfvGg_ATP-mwk<0!cE`YV_{w`qcyT8|fq-H1IUNuok;y>U7wHtOCW3UR~CvECf
zC4>!1HJJz8DPILM`TeFdILiAadsv36Nur|sx>_JQNELhE<ejNRqD<x#QzF|cu6ru+
zUYh+|zEWL2BrQ54)&*mXXn6?(I@b%Y)ZelPsM_^kIFd}vWoLTumFkwMkyFa>v8PP&
z+k;V35wbL0vg?5LyGN6i9!p_^%;UtRGd}@jM1&NL`}czuHda>Hqxwh{Y^<!Mjvu!?
z#q$nOmn;Wa76s}jgSmjmM;ND}zI2`Queprwpdoxh=7{(rPeQwHG2g(GNRz7{c>&e&
zmkM7OCvkElcJK=_eYL5qJ{){&L&X5-KiL!SNj>5#Kf(=D@rMBz+iQlw=POvuD%)kc
zqR|aGA*J1uYmPH@tWBA3vy)+~W{pVlXt0U2XA<-s3c=x<++I49vAOH1Xtp*CrKl-I
z3$P>DUV7imM3lm@HRCTem-~iejx8p^D{p*mW|;U~9|0r2x69&J{u#pVUmMW?^rb%x
zG;MazhOFOMCQ2ryBEwuC<F|D~?b9=9y)EprL_<oWzhG;IG;t(GxZ4mW^p5M=zn#3Q
zm3K_h=pqUGA1avz4^U%l1182c2z8SZugWWV`wd2Tpjw4Lwep4=KLPAv42ZAs^-#zG
zK7>r+&+3(Vi3zWR1*6Ya1H^D~;i~eovY1l&02@GO!fEyg?#L4}pZ7vU?rN7e&+7dL
zY21mbh#-r9tcS-NhDyp5n`HV<wGU|iyioDH#zsh3YxWy&sE+VDv1TmnIW5yJ3JL;o
z@=GF2L0G@l%Tc0pF~LgjPX?iP3>A^#7e{|rke;ViPqzWVpSxDC46Zt~2c(CMAa521
zfW{B74!YLT&fuDbvpxL5r%v8NpZIN;xbn15{R3yy+6^3qQBOxI46hsS?>zP<c`;-0
zo41)fCYS8N!2uYj6RdoEe3rrI8NL*gfGF}i<5++5;<~clo|o=t-Oo=?pEl5N#YiOx
zF!CxEe+)%J;{?ouA5r~I0?rGDZ8ix3#f$}0#cFwAOv+@N01l$RgK~Ry)>?j>nj_Z*
zEiT_?D_TIgG9x6XTG#v_#r9UJv=tLooNAb*U+WT`cUp&LnWCADP#N=%`FE5sOwmRt
z=8~YzNuFVU$<k->Yf3UU^W;?OZ`g4$-j%~`Q0KQ#bY0gdkwFx0t?se1s~#^<NA))Z
z%kd;Cp7QfIAS<!1uC4jLlK3-U6ivYF2f(>?i=^FPz`ilSqiqmVEVX~HQN6mlx|_~x
z6E6mI5OW(G6UOxKiAWc~N@yeG=%}Y4W&8QfQ#lP-O~uHWxJSIwTvBNWdS_>Xmu_tb
ze~Z9c?BhmaW3NGsH-T>Wt;bFl^9v=js162>73E?_KXdy7jkFc5l6!`?CbI`#b{Y00
z@wj^Ddt_{OX;&I!CD?tm)wd<`=2<^5p1JDX*zp1P2(NMK%0Tp@`p@C!_3$MVQcJgY
zVV?V%s=BjbfO`g~zL;7|GA+=xbdB8%Bm<<<FR8-@<QKSs{o2HfeXtFKW|Y-(SfDY-
zfbhZ|X#wM~D;`^+R?EGS(#{2KX_EJkK{@g%W!9fF&01nURvV#(8r)(RDH#b(Bx|()
zXb)yewz9S%wE@hdAxR}YdS1*oqRQ1G1GZ4WvZkx`NR^+9%~?y}wj$DJCRXhcr7_2&
z)WX|RnlP}~<+5F^w%<mMw3KVJ>}uG%#W;_Mh>(A(Q}874%A@?tmm{IdGe*Ty5;R=u
zn{jzZQ2k`g%E1u`w_hj++x*CocCd|^33g`npaPJ($rZR^k|7;WG;Av;XEd#`zTSFm
z|9fXe<wfuLyqt9Y0PY~6)TijT*XJ5Fo|>(DnY)l+Jrmlq#PFzZ{zIOk+6LVb?^Zb_
zKQ`!%)+O}Xvz0ciwTFVtHCbb`?OAP5VZAs-_X&#^gg=IgC3i|cFGt|jWOa&Sd7}WF
z6oJ3X+!QR#=o9uax|$&XBZhoQ)#jF7xj$?q5v4gjd?N??j<B0FL1uXYvn@)9#u)*L
z$3)^~`uF`Y!DS5b$e@G_ORjX?y|4`Q{RY7+<ruBFB{43PKx;}kZOyruo=HKBTJ2C_
zI$Pd10*FQ$uz@-s4COKXwAJ->V<6Xk+Qg~G#lc~;aq`C)LDok!mo`1~2M=&i^279b
zoEs?ewM@Ib$Kvjzrk<YfV*azRpx*#yoUON?>Tht=74#+{BqDl4)pa{FHgx}WkdW+s
z@09M%evD>k;s~P>!t5O((QVWKnr<++&*&v<|Kn9>Bw$n(r~yzsx1=lOW}ptr0sY<J
zcmNvQ{_Q2YmEY2GwFrF<NwuPyZ}mRl;EY^t0sag37?(Uy04j4eAAZHU-*c0Hp;vD=
zk&~{W4wQ=5GC`0K2Or|SH8`Ki3(YF9y;g9<PEPIcn^|_%WSemg{0tO0<45&}z4z}r
zN=i24UtVuI0L9n<I@o452tlbtL_}}B08g&%?rP@`AiiGD#nWpB>^m1!bQBHo`|m<(
zx4n)(?%(!l<Sd(aNmrP3-Zh;7E4AcVR~ZW+fv#p|O-(}g7|^|fMfd8)gZ8~B>Dru%
z#jmMdP(mh4_>?vq&Wma(u|(4)CAx-x4B%rim*#dE3f<pl8$eF<yeJQA#@7+@xvW#R
zBxX0+wui5&7!sSRdz5Dc(J$*J8qT1#P&M-?E1`8?5Albb8;A5uqoh|#nOY1!i%QS(
z$e5X#H_B5}i9<fpPfVTpwf738Y4zkpVJhq!Bec~3cpL*T6)@hS>L>>0*HfF#dd;O}
zvJeeqXHr!IcsBoXOl%=u-N=Y)LWqpt*;-neS!q0d^ADA<^CVmT$H39svVe(*`x1=)
zKRKysX$L05pEeYVdU6!IGAnU7Hc;T&yTL~GIXoMdqXo}}Uq&#@uTTpIuTOueyZ8vx
zP^^qU14X$J-i!cB{`n#8wapJ9@J3F6VMbeU=!$!|b&Mmj`*|zXiU^Uz1aMC7WQJqU
zTAim{Wc;rWFfUncOEM&AJ6#kWuk4x%03F^hjv|#NSkiAg<VN)+DpF$XB^7MI{)CU2
zvGD?F2$1<FIp3WoX!vPN!i(O->B9`TZ$?I$%-&sbn}(5O>A$d}Z0@3|p&98jpSLs&
zNKLV=k#};S+=JiEsc|^N%0lyy)F4xcizupHDa5zPFT*n=>>0Nh6Pqr~WDOUQ4HH-=
z8gYC8tMG-&dXPLjiJ5Vp)7Loa1=O3_EobZ;mM*ps-;;I6i$4J%w;#?GQ~Q3Cgj-O5
zj=)lA9J4`MnbcZ61<WoKdSo2+cgAxj!kK&Va}sm{LYTH+sFRD!pQbnH1lN78mM%Sy
z{bT2e>l$xLA{F&%yv3h=9E+v0rEyWwgGhg;ES9rnQ_ZP25eGg8cL%)0&h{8n2u2ii
z-X9!Qtk0XnzA7obJCu`C+2q@(ZEFiaCettw#xMMtjm6dG;wwjWFgIO$!mc*Xm(rs3
zAp!B+&PU01$9aHB*^6V;P+R*wb{xXyc=2!n*jy;ud;gBA5S)X77Q>h}X_M1boEtx#
z%KDmbgIK|ZZuUO<<kIm>;&V<Q$=Er?0Fhud<K9a(2Y0J_5?wyWj)OV6=Hcdg#>u9z
z%#g@z1A8^$Mi<Vw!zLP^!R9yPlLsssd864Y{0i6$KS!b8APtT*ZoXH*;1i2mx)B2z
z(Gi;mcR*nu_Lae#1ZV^%D9XA`m9|m@bR{{*chfo`)<Qa&%A7|Vt@l9vN>FV>`%|-H
zk{JzMb535NpP~J4a!nexYdabIL;-|GZWV~9<W2!cPFo6|)!NwLW(n-L7DhR~3;(fe
zSt36OPxOv0pH^zdbh4b1E_H_d#@v^k-ia8>@}<!ChLKZCMX4784J_O*0a(r~${C+4
zDbsJ1qUs)DED#=F-v&c}ie@~q`&nWFimiEYO$XC?<ZLSkLLHXggGA!bJLg!r1nR)T
z;SGC0l)ATT=KyA@fu;(8GgJsyaQj4dgX6YUn09{lCjd8|TI^~EWTyAXF2D%SwGo&o
zysc$tdzpS+4@?ryc<x6E#{KQ*v?)gBK90nAalY4xq2X%2+U~KW^#0RST3g@HfmDHA
z+l=$mXh3{{`?>4^AZx@lD1CHRLRwhw-V-1!*cp@j(6~kRwu3evD|AgvisY$7SA{0~
zZuhHroLNcM>Ik%mp*%)U<2Rc(h)^}2wC!N}p1@($cvPoncSTWr^o-;hn-jVwmE^#K
z__<<f_dlxGc%V2<LpyME(spv%x5D$fFtxrCu{{3<wqJAmLO<+nIL|gz!ZuFoolrxp
z^laKd@%{CHw564mp8k$O{bVDe%r9pYZrT`{7alJLbQ~|wM#d9%;@2n9RrKq|1UM-K
z9G^W&Xbt$f+O3=ywEeD{#y|ee6_>h@;!*RStzKQg;M4kkQ2<^rNK5N{8~44tnWbkU
z!Z{CVOT`#%AexvJgRbS%o@0!gS2=SCBUHZ6mDs^Z*`Cy)NEy@L@sAw5?`>=BF{=N7
zL{BK?(Y6t4f{<XkeIzT5kYMq{8K(5Y5RfK84#~{Z-|u6TTp>C3Oiyn*3GQ)Gn$lv#
z`qbFo=s_E<sJ5ZujdC5m^Q`G;kG#Wd1FqkO0RRgvc7J_gMn(*ERV*D<ZJ&37Za^s%
zOEZA@fD-Lq>_;pCl{_b|usqz1i$O!I9cyL0iv=h@Uo~3Wjca$hZC>H_FXs#{+E3w8
z59#(QPmN3@Jnm{&o^c`0pNtTbZ)`);+S?Ou9MK|~WqN$Ac1U0walfiRO=a37Q^xl}
zzOBOqRSYIayeKk{9~dhcNPPdf2`<tYVd;|eVl;h>Q|eP8m{zLk0IQzVqYRl#5?q?j
zQfH2VmZGBK9nOu|Tm#qy^!$EOHTeZj2!n>fRHB;NpWZ1d)PWr(k<_<YZO1nxd13?S
zhv@nyZihPcYWCE9+kvZrN}NrPfV?U|`^hgqHS`D{w#zGvzpALhy<X7waB3)w_<rm$
zT0+Z;B@0j|`9#nS3A<twEI~>ZOkW4=k?hwm|6(YH7wZ>H)o2|eU=s-mappwjGv?>L
zp)(<DCEcs4YHty7+9%Iysi4)KqW$yQ#+5x4U|N@{&L$x1(>NXBG0ZB-m)kDD17M5>
zJ-CUr1rAfVK7^OlhdkUw?-JSIh}UKvpf39neC}W44)2qnPK^{zh>VbYbd{6hm)~V@
z!S;u)aHpnzhJ4l=(N8Hx>hHOFh7=|rhblmKplj*QigL8|CV)BuN6#u*mg2s4AlMYt
zzp60d_d3pDVa(%d;qL6M<z{<$_%Wrow01UkN`pED9G+;2$GoxqC=<!kRt_RgVgWxh
zxaLn^FyyEyARc_uQYm8>if+1XyzB7gWZaYHYi;-=&Wa0XnaV$DBxr-=MO2uq>HKS3
zD(ciL;F+4i2j{m1I0tcs7RBP@2(-};MwTi&u}@nizwbP+pFGnq$IyUD!#7$LBaxk7
z&Ckag#r|s;J@FjA8}TrPErcr4;U{0nRh|`%bCbU8K}`Ppj-8#8R#)20-WQbm-}O%7
z3(uMdn3a8J!6pXSlL~Lv3lm@wuoOENa{X^4jDFO(YUW9JpYNjN?)Era7?{ZIu&d{x
zB2HIV0<X2S*`w&wlPkvY?U1a9;h?Ljpo9&PnPvIH>GC@LFCI53xNBGlU^p_^sqouA
zczQ3oar_TGqz4WrA*pZAg-88JQqXlqv!=e;)Y3JFk6#iaxXZ78h`*Gi<wo_PTRmd_
zDnRnz<^2WaM)g5f|5j3m@|pu}ol%JR?}z{W*sL2fOL|IiN#qh*%lY@|`|k%GW+`cx
zKgW`7u?plDwgciMA_r>}7{mX2jBq`8fb$fd<^AurX!R8$@v;A3pDe$N!E2NH?0v8x
z4p_Ew@U4Bg_?;J!kbr>1@wB?4Li8J)+3<ySM+{VNmFrfcLTf6rm4%tXx>Zl#9>SKq
ztor5kQq1B3jTTtFu3B?%nSzw~IjFCMoNiC-TmlprPDc0vv>sx^TsJm0XgG#lNAzhZ
zY8KRB7;tU2^zCgsNSm>90-&q_LP9g}C_)B}v?)i6OaSVPBI~<O`|C=h)`mw9(IF$9
z!S|XV!}V{b)X)<DvzolX!BHe-0t_*rpe2A3D56D1R?vXF@E0(NaRuxglt;|W#m>Tc
z1l3cm<={%Q=HsQ&N9#**HajVp5_i$IAvwzyF?^tuz?0+f6IfN(*bGziJM#yJ0ji7K
z{oSomy!@wf;1`eZE;;!da7W&B;~NE@%~YTs<z;F&NKEE%<UW-;&NkpFhhUR(!%hCY
z&gF4I;Tj0GNFnsV{^aXdsWf{0El_L{ttN3x2o6C6T$d#$$rm+_Surt~5I1*sGC`O8
z0la9Bde;@%*GR;w!dnbrrZtRb5e6#BW*7VO5X#`2qt%U#R?0+qIZ*HH09OYrD4+wb
z67$$4*!nv*V2r)r!<l6*4ImlYLkqxd!Y3W<BB+{%My3n5yNB$t3U<ijTIuhc%;V4{
ziTR&r<K7Sh28x#su*DeTC!2r4J|GM#4#t~Xwd`-PbdeNkoDnZ??}qeuSe?)aDkUFr
z<FbtD3%si2as`x={1iX*6K{;dpUATfK#Ioh@!o9X0kNqlbfMip7!I&p7~!x!23<`K
z-9Z3Fg!=~*<x^}Ru@WAriLk3Y@Z1*%9Id&DkTFR|$GsQOS_n{XbGEHK@Bj%BA3y&(
zVl@&_*sGq>Hv#p4w$E6fqHR{E*fls_&pP!c;{XeG8e|x1NqKqJPf%tNcj=&QViJtT
zjh|msv@YlC>sxM!)9;yTWmHw)(cgR>GTAZ+nu6wv`NL3x(9{hKu7|&Y<^PMng@oE$
zUJpPJ97J~JJY-us=bw3*jynZJ`-B=>L#`NlXvZYMOVNlh3(~VB0GxJooq^n+3p^r?
zKZktAo_V^E=6%tGMZYjM#*uvV@KxU@cMXse0F`1KOO3fwAa<%vV5qCuK3i?l$>G$K
zi~}+kQPw747}<Um#|!I=+V}w&F)(+|i6Q{XY5K-XgID3;U8})rHWXN6sMWFa&chv9
zs=!O~ubRda3A_)x11U|Km0%T)A!+Y8AjAVmT;6C?jSkbt3(!y`fo4+GpWvfv%&LRA
z=97=(Y7&9Xpo;Jf_=3CCepT9rAYkfTpjQbzed_q$9nb+C;q(=B9~wcX!3IoI<`Fc(
zj4)=eYYc9@<U^HN%l!XB`@flizWkPH?+2=mn~U28t!e)FeJQjtyO0D()jay-$a9DQ
z_MiUu@lh;{^j+7FT<U5&C@#qld6#%`a@19j@{qg%-rJ@BG}7$H)BhdB!lN_*yVDpt
zM0?SR5kS7yk%#o5KBB5S#r1!m6&>%53_;r>mfk)6D5%WjN4N-|>tyfp`M<#sa2^Ka
zJ(B9Nq$lsmZD@kgf__(+Ktxm3W&Xv&C>(*h-D#-_OWe^I@gI{SyA`R8sR_+Rlc2f;
zOeQ+)`G}hQrv{Y`w^Zq4+dS9~NU0*T144BSw(k%ZdIoi$l73!^;u<9t{?W<p+h)Q2
zh0$Edh1iAgGksS_606aQunf2iGuj=BmEDzrcV{0kzW>L&+%%OQ$mRGXbSRR!S6=@l
zkE^awJ#@F{!-=){<U8&Ad!4^a3RMu-zq$&!;ByNN@^if-T0gSLo+ScvJ^G`EMLDQW
z2n&d^NLb5%d!R`CdnHcpQ{J@W7Wot{KF6p!Va)=i4GjnFP9Q<lU;YQ+IfeeEdhpF2
zgtQRP`yEDsw;^>Z>3&_PdT{%<Vh9)&4b`xjz&`?LS|<Ff74vDq?jigiUyi>=^Fe=!
z8}LiEwVYcb^6M(h4>Bb<%ZVB^($nFLWDpXmz<!V_|DYgNfv>BT%TjD%6#9XukJ0eo
zFk2#4YU?~YWCz8;C$2&FNMW@U^(kfqQc<e1YT28mxjqXC-YTNbVN{0>S`B_vI6nOQ
znLY~<+tC}hFTnw>B0b!)klT<}+t%;TUfH6&?bp4fx_o=`YtM1FJm)T9eQB)diy}ir
zybJAmLSVrfnG8N4u^Hjj1o>nKp5p2SGl%|L!U49f<*%EJ4I~o4HU74GvN0qm)J+R0
z%bSm`#+2N>j<lm+%Y|4NIEw3Z(YIcf$(Z=r##0>C|E9M2qnLf&xyr7Vzn885dER+N
zRJDL7J0@dg+dUuQB{rZK%*%FCFFjd$YD%Jx7Vtn65`Il_$2B-{(~WWJnL6G(4ZE#`
z-7Yy=;s-U<tD1=MG!IB1m<aN4pmtDL=lLwVX@e(4Kx&e?XS(+i{T5!t4#sXQMfPoc
zY5UhSz`g+tuZo;`dz!Ct5sm}s#YQbGsH#Y*kkTsbP7?6eqHQKGTfET#f@Ipzx9O|?
zoZRT?@nnwv*6FvR>DhBwP)14GcL_o!jR^LFe4*r}sLmCER`oLiY^%h6f$D7%2A|{i
zkxhABc&$A=#knbch?>|2gGPJa6XUBK!f^ENu^od69;&J>dDC8)$a7Qw>z_>TD>3tO
z6>?AIGDNWG8Fek=r0^?{S{%^{MB;vmLU?uc17ibsH>#<?rK6TzB|m`&=7xcmrjBmY
zQA?Q#Ju-j$+j_|+ongCKlA7sail7OnuZO{;wQ=S^1nLV5Di+*hS4KP|R5q_~H}Mq>
zY>!k(1!=uRHj-x~k$P%)9>=CLG&x-nF}3#oJ*>e$)R@-MBASf@PGQ97WuO@3vZGB@
zv$X#M>zL2JRtkBG(es!34Q@DO#&Cx8X`!aLK3KxapdJ2-L(}q?Ce7JPovLB}BwWM-
zmke0`B|wvocq6fK7x-eg_J>{&Z~RI(fwcD^vYQAGz1|`!IZyx3cBlyty~#-^oBp1E
zYHb^W=2(9Bbaj98y}PM*ifhk!+PvMgKAnWY$O%+Mvar^Ko*T&rHXUS~9yk-r<_Nlq
zaw)$42@-&me3}E1DXUGwj_Y|4i*vVOhrq;0``@xq22I`Wz9hBemwxw8W6)v0vCZnj
znm|l!*HjakQ7=@q?MhT`6FZp4z8bP|LP5gPqaGiEwl6&{2u#BlBFc_4F6+VrlUAv(
z2?_(nU@tW%bk*TqYR99-t+dVl(x$L!A+X+mMO}pRoNMX1ve&iB5>h@YuO+B>ywP~=
zU`nd~KK!MIHQ1iL1&K0{4*!;-!R9y{WCv!(VGg~T?=&`zcHg7$ZrhCyrlBmAXFmC8
zhkmxUTO};Z&FD>pCmhl|>i@sIm0vIb1P6+w(!jGX@QhP&q;8;8F9uh?zB%u8{qoVk
zTV#*Z1`vAi(B4RF$DW1ljepyMx!I)beVa(jL12$^ySKCPl}*_rcuOnjb{>SI%rvFX
zP2;Ut-n(JO#Cgrj^z2G^iE41w{7byCinYnJ=o^P;CeM9D1v3RD;+w`~pU55}1|U<S
zRlJ(Ga8VA7GeK%Jx1PopTaSburJH!nz%hnq{hiA6yx)S4Ozg#!S<jRcsKpilkm4$a
z5g@)OpU_*YVt<=7d8M!Xdyl<PD=XSNZ&>$J{Q{0?ELkoMJA2^&hhq|yqCt#C5-Gk~
zRIl3ldh~CQqzn9;T`9VXD28e{2L064zr8$DD(co6d-m^sYwtD2bOvIs2@j;Q%oulv
z`js0GZo&jYY0Ocd$|mz-P!~TcVp&P}&22NS1M6L>JF1)UCShviBWNN`4{JW|z|q7k
zFnQq0Tw9$l6V#9|8QyF0r-jauk*y2sGYW5#hv0hJRO%mSV`vw&h@t15N1CCVf^*?d
zigf0_+vetG6tMJ2b)4-*qOq?;v&gOEsda2cN&lDv6n{42mK@Wjb|%2~$9oNnx=A@=
z5)vEWTcs~4Q23K!FgB_>KYuBBc3aLq(8AWX6w@1+Wf+W3cKn+h*eL|gcnVLcUQum*
zd%;^`pQ@UEZ2p%sO6OUZTRuAUIOco0tbn~z;lN^usqd)EM{@%#DqC_0KV-2xg@zO<
zE)G2Ju67@>xw}YT)4yPLJq$Pf?ev>Td>&Kzo3i*hHsD_gFR2c0POX?niDd*nTb3{=
zPK(3G{=f&SElj2AaEm-7Ig8Sd`3Kc6J;-twt9EyH4Y=v`O@QxeE8FGe09WVDpYPEG
z_Y(dWG>ck+2e>z9;tI2Mg&gXL#Lp3!AFJN@Zlus4ZM?zaL)o6$TJ1e`YRU87Z!vLd
zH8|Km^_y>7R5|d9bQebP{CMMlsa*5{v+p!U5D)o<WDpTWe#)I<)@oa4FhdujN@hwC
zv#rT+@5h8({yg{N9}1bD4gz4;%Ex~Kj3I@*)`Ly2hBBWXow^EH0rw5ArIqBLW|I|q
zBw5x|wG+A(FI^ve2x2RCZJp<UJT~IgB?tk9*nGant9j4uY0XGwe+Wba-+Q_OLZ$+W
zkVs~Z;2y`a0D{-UqHiS%|Az|zAw^uGM=>6J2mSceeSglzRgsQv`<e7(0UBW)@w<((
zOR{|OwE%leNtx9R>AO1CnGZ+pdk5$_SY;3A?FnXLMtn=`AUAj)^a{Uh8%Td$T-@&Q
z^kn{)dU#>(A$IhHm_{mNQL1F(s{lk4b*UG!U-Qa9%yagtBA4^fi%@2B(8E)qE#8tK
zeCCOy_VD%*5T%}8a%&DkTgu|k+T#z;6f>v-yD8l>Gx`wj<C7CCgaAv|zcRPm-;7yv
z?m4AwikRpMf=my8so?(X<PA!X2@w175U<-A)#wYWey0|jgKP=ZA5W@Qp>>5QJm$SF
zz1FL@o-!AGhZng{O5_~UtDo#o@a*1SJ;OCxygIz#QJ9;IKD<EKY>RkbF+2Ib%B?WY
zlGtb57=$<9piT)i2+;Wyxn0qn(55?9(dKv-17uMt1x<hL74Odwv(<izcI5&>pE){_
zpJ&i$0@V3ljXZ*YM4{>3b@)GRK4E$y?ogFQL7kwF6m|h<P@9K#3QK5*PaY*aFm5^G
zl7GMcfVMia(<B;4PLo+4?@IDS8<K!{Nl<Wc+WzFGy%?))^hvg43CdS#^dX%NEU13K
z9NpJYb>}%fbfRdJz0X55v_GZJtRQj^Gl?7^O%UkHZ=T?{ye8faXT(ysek71f-0XI>
z;kAaGSa5L<h<z-81097EDXPq>X&pbBSL<z<DK|P^b3^0F^W?-tSIs8u85l?;UbY+W
zdo>%HAh-xg<;k=al&R_H>UPG7k68+<@;+zCy%P%5cX#K_%u>s_Ue(9*9s42rPn9uc
z1~bhe2my+vuVZcBS2DzPNRpHC<HsxJ+L{@*-v_k_H^-C>wPYVs5vD-p?W37{SmFu0
zqiE!gBHy=fkZg3Sw|5()JNGMO5+;~S8$3Oo36~do;A;3n4iGLI4|N}Yp#JU-slZ9t
z+IJKoIa*c3U_b0Jt)0#;!_*Fxm?6azI<xIs?N8nt#q4S?tf*k0-8d1$f|QZh<|$kv
zyW)``q%+f<7;@^WSXiu_W@eI`)OogQ=<0eGM~P0(pCwm*_|Vl6d>>S9p-$fa<44Q)
zoil#D9C~A5FE@B}efxRZp?NDsq@T`a=#6b=D*!lxuO5|aM&i=IM7?%NA#nQwZR2VI
z?ZvpfqN*y+m5&Kha`HwlT`2cEXMURJXGM)AcD=PIi{Q8G+YiU$dBeZj{RJS8xVW7C
zT0}`Qi=sigjRdKwt*ur}o9izs6T-JU+FTOHm~Qp1{#eRd$HobsxV^387B5g7Q$fHi
zCD#Ur^}9WM_Y|1$`Zry2zuq7DwRhG$QF{!kP3pkua3S25eeaxs-h_XJ3gLLIYBBHo
zld~9?7mXbQp_9Vl>h=*w5H|$AUTW&`sfU>v*KCkjgt|OeU-6I~yvJX?iWi-|d0Bwr
z$4m630rZ>{M|LSW?*qg`{1B_lreeqM^XNn>0A449zCj;~4@83U{vLp}y&d(iF%W^w
zg^#e$M?+9)J)ElJZ*yAxXJOApn>7%z+IULrbbNU1vXJtIHxlZk`(Ut!uCD}r0AZO1
z>hL;`Jj`3=%FD|Iy!lnNDP&d$=xUhQ*mget_@U?yw&PHWY>CUa4-k}L!^6QiG{G||
z+<0H;Ps0^{>XOQT>#Zxp9OL%p>iZ%r`S{)VaOCeOe$MyN^28w(oM`9t9HEjut;{>T
zs+Y5gHjOLCue<tcd);d`I|^4;3}hVShc~{~1?(wA@a9%GslihDtL<%W9E;ZM)zo&>
z=TM}oZ3B!%jASp_7(GiU)cu-hpqLO|1oo}f<3GQvtJ>m!Jr~!#yiOocoN_1gqUSp~
z^>;Nc#{EE68a=4k-i+3W5^l{n9AjZv(72bx$r+6I<zT?aT`vY@;TK&Vi%j4*%gpuC
zz+W>ncGJbf0FmoS<uJ%a;#QGKHTK`DU-W97&xx<Bsxo$8KhUW%^hJc4rwyE;hIngx
z!Ui*V<=_=yY6A36=})&_Dn3!+rGT22yQ^QRPe3%<O8w-@{HUe!*dK>z3E5W`YI({x
zk)ntDJv+N5lP<Ewk!eydD3e2|)5~HJFUyP^sX$LJ(!)K^ii{Er=#u=s+<x$ztNt2q
zQ=Oe#hC8V5b1GNha!f>ke}t{s;DR+8_XCL51lm%1z3Va)Bmya<>?(1AnSEbj1L1RY
zRM}&bDM*POt@xb_r=#(dmgmE-_a+%dThDKJzh2cKKrp`!FuOR_VqPPU^HD)C5hS&<
zAbc;gScLHu&h*MXgatxALZ~6?(W`aEgb-XhEdsr%?w1ABU5L&PifkvzA`l=SKfe7k
z?cn>n+pBzrH$1jA7LrnK_x{`OrpkhX0)u;K^Zb4eXPP~B2pO1em>~0y4~27J8xYZ{
z98dqS-WAs8ueQZs7y=Pt2^Zm<H27N&bY&if{4TQP7s-KPRltBFlkq_;L>vneVRj=8
z;K^sP^m<MJ8BIXAM1}0mw_ad1&bcYIF{^(5sRU*QB)mUr5J8X4k-#(~uW3yko0^%d
z>x+fE@*HQ0#^No@&BYWP_lL@wU%LDmY{D-3@am_ToiaWBJ6Yn}FPCF29(?<S0s$B<
zmvD}%uI|n!#Kzg2<gBdO2M|DuqbeTHc(UR9IcaL=_9665*-Xz={lkRb6B5Wc#1yFK
z2&iT}vC6BDUIsctcOfJFu5d(H2slBXFo9mfcFuy(pj=-if~K}S#LcX7wU&;2tPx94
zL*V)XK@-}#^ER#w+aj+a9oVLLA@-2AjPRXHLMca#9}l;u6jMAv*nRi@z2~ok3nb{9
z(W{-Az4OuE(sm8!UgAr2%ljmd3R^q7mU_Uy$QExWNfglbjIhmoE|&bHKDzHo_|?Rz
zJre4JCey>Et^@=yW%}$L{{Dum2i)vSwlT|jAL4ngeV}p<T60nIp*zVyOAov`d4++R
z8@5dt_33%FDtx8G;6n}VMDhX4qBb!F$PJfNyFuE;VMBC*d8k0d+CUnOYqOy^Bpx8Q
zJm|Nfxw6_ukVkaoej<xx-SeJcRBQdriNBN?6B#RKeH$HqQ`X@P3^J?hgd-<FK=t#D
zyfdj!GxoFXTx5oTv8jlclp!2{tCZKe>!LKk@XWB6G{|}^=DDUZ!sV(gnEY<FA40no
zW*QKeMg@NEJSa!{B#3fo58+lhx#Q&m@(le*hn^60)Ue?FA=BT>`{v;>tz6*kCc2Pu
z^@clpAGz`#s;Z5_PgVMj`*pYE4fd+CyaN8pUt!!ATq_S2IKd{L+^*NPH^LwXCufj^
zh{*MC0FX}tW%0M%D(*rd$`E^D5i&g>uz;w2D;9F*)ab9TX0@$8*G?s=^5VtlDX=r2
zqZw`)FIMvgH|!v`kGJfqV|LF8A*58Zc-`9CT9qiV_?z^-27oYIi=P1Zp%-rGmX>rI
zXfOKuJ|WYBai^nG`V5_Y11LLwv9YldY;-%%KYiUe(QY#2OkXb*pPr?}>D~BP;e_-K
z?@oB&XQ`%pZB318j_Wv|3I(KOn=fs^bK9Y&Ot>068b2TW<}hf8dsfnf1xW@h)hweb
z*f51TY0pHtPNndKJ{>*%H}#2Ytut{*RLz1lq*1n61&mLM6yc1U_3ZphnZcl7FE4nX
z3Od3$$d>`<#OPxg3B*TB28Zrv_1aU23FqU-EmZ(>-=gHt8~r`7xqCr}zZ3KX_N&L%
zq{ue+T*3sfALT$mAb@nR(NRFMA1vn1LvkM`4mZv39|7%fczZJsy6mGXX{(u;G5+c`
z>xJ_AtY{e-1)@V6znZuLvvWUuX!0nhNu!eyFHC&OGi}c40=8d4^2KfK2QK91$qih$
z+LnE3SbwGObn05gU41aLOg;a(v`svG)lOF+{+ibZu0}uhYyJ3mGAok!7ANX#tC~f|
zl=L`9S7~GFO#h!ff}$|-zXRuVd*l4~7L<CsoMim%i<z`KYR%5)7SHUs;U}b16mEF+
z2>ic|B+lz0Sj9pbw;23+<!mZpVj3YE#Ap^{{=LgQmD=@v#C3hDe+DCdq*!gBpE7~p
z-?5|Z)A;WVs1~32>)8Cg?F9>6#Mtn^{`+SegJS%*A^491zHUUW|CFr3Ba>+!n)Yw~
z3x2Ywe1ZORQGkCjrRe1n<Nx`dq9*&{fAam{+l8$EF74k}#|Zy*=D+5CIC(<0vq0G^
z8Y*I_na)O(HK4+L%gV|^fRK@qy)mx7(13?EzM&U%_<vj>7YcoFhjI|8EsMs*r#k{$
z=lN71++(cXPfG1nS9zQ=*So<wZE9IQ)&FAN8w~{!0U{X`bceEa{BpY4y?7kNHF!mK
z1e{t^FPOBZJWe_KyH{$T0J?J)FY0&70)blB77pzIi|VYE_KO81$S;S63XlFAhq6q(
zARLnY<ac1{wGm;m!viLOQP9@Fvb?J=XbX$3S_}ydr)^~Cj2eDsneU}#5Qw%|LV~L)
zxF!;h%V0<Kn1keG)#BR8?T0cyI~As+qws-1Tjzr%*K&86@sW{`+iTFj2pp;yvu_#=
z#45zGgg)Rk<F+0ojUE5)42oGnp)&UaBwa(-HxsTb&nG4)=`<h|LT(ITn}c1IFt4RW
zIN)?tPNz4YFDWf85=biRQ_*u4*_s1EFbE{_Tc{Fe^Y>+$l(O^v`PiZ&CJ59j%_t%+
z?m;X;_Xeo!1!n4K5W|3@?!Ny1FM@c51wB2oyZ<dU9|97e9t8`kcA($p$On@;VKlrm
z=j<*e1vAnfeTZfy3~#7EnVA`Ki2(uoL3gDf=TnHKFMN%qG9G{k(K^%48qzUy&DHvX
z1Gx#hfA)=gzeZTO?RD4E2?SHSeu6trPiI2`ECBCFz`O+RLf|v)>)W^wMMXu`Na;kz
zch8<Z>&haQi%|kgd`HOG`axpF)XdCPrWxtIwP(CfU46Y3f(sH{(*8QaEMRC^p0$f_
zs2dUyBi|ZDXpqB91db3xU-E+-{s}k%p<`@0xf(DFuOZ4vA%J!5j7!RG+S;ASZj)(E
z7Qz)7d=vcpzVNC(nOSWCvLpe?Z*KO@FeAm{h$nm><#Nu6rx%Z6tQZ5z!NVXp`Z@IE
z3BB%infkzgdI5aash`121k|)nkNaw(e!Av``-oCQLTodKw(8VGsntNpwByckV@?LM
ze2}G{wm^7Z@m2lV&QzLA>dyA|0;PbHp}wfjWVJ~!UWl4O4*h%52%evOL$nZ(6zPX`
zEg6^d7A8qP`pjsQi#6}$t?J}d5~j>e>Yxm=i}f?VKuCVs`a4?bCN&5kYhKe0&;(un
z?qSs|+O6eQofD-Vc1&CY47mI2Rc49i{vV>H;PE$}_}xo&(~1VVXKmRG96P1=3Fr!b
z{e7_u>YiBrNMVMEnt8@+)d0pf+Ro0-0b0o1+??Z-N$@SUj?Le<zhJ@aB~{4(2@+8)
zk^k>-0zyrSo0DG6r~IJ><Ejw~?L=o%Kd6>otu1VhgtkCu1MYCWqa3u&8Y!_TKEdrG
zMmSwoGZkJDbT%m>9%pmAL(Dk{b+L4CogYuyt$+uS_E6A-BI`;T$6Xt4$0EX$hY%Tv
zb|~aR)$nL{eW}LAIpKl^9%Q!-j74;8kTG`;uv|K1#gB>JAOt!rA9p99A!gjSG`F>#
z4W2PcoTbpKo8FU$a3NkY+`PjWAfXC1e`8jAzrj}PEHRSR7zQq`g!#*JB)q~n@Yf-d
zQHf_#-SfPxr?N<0Lo2#s5^asOaLId6*x-SVyXV780R3ZQ#6pL88)N2tJ#hLo!P>L!
z)lJz(BpT*!7J?|^k|G4ZV^b8p!wwZmgo!iITEFH5?hnX_1tMh%J`<;b7IODN3nhyN
zLnzeQj}{(T3!`Qxt*@QEoh7t0#A3@JA&}(E8Pr>r;~lnC_Wl%8DE88J2?fam^+X~8
zj^OoVv5X9OzZ*Zp`;Gk6MSyVH6`Oc4SL<xBL|1YBK*bHNx~w?zsc_iXd}h%9LlX*=
z8`2ADbBH<GEO+1cDHoYF^mH*XmK<v8DO;XR&hy*u$`m;`#jd;FM_irZ)Wx@?Fu%cv
zN_aBUz_0$jU|3p|Mwq#q)VHup<)kFA+Ib=H)}&#RdYrpa`sgXd0zwEO#a>dC9ui>Y
zG`UIcbC{KJuoqmHH@ysjN?sQuNNTNjx2JCZq|gx)U>|jW-)bETFac?Fzmq}DdW`WF
zc;ytVl6N{BZ9G2|DY5qOYL?0$3C?RWA8n6zy8hHTpStw;`R@bWXBP20@=sKf9m&BE
zu<U&X(^l#7M;X*uJE;{gH?#OMap^sgl%YX_$ZrsMpnS+^E%V_XB`C6<eX_hW{IIzy
zpBmSsl*4ajK^0l0;WKzb3+ldsd9e_gD3CSr$cuC*`TZMkYJ7sa>lg&V6>|3Fccx1X
zG7Z7D3($rp)ClVx)Eed9#jl;dT!}khreC+>lc|BLNx=lKRzb=;v&MVsvbU}{!5!<9
z&1lK+UPj?RnqqAFdKUeFQ~bW*a)BNl(ml?VnkE1Vd7R4R^3=d?eXf}+(GWf;r6G}z
zx=TNOvP>@vB7ssuHlq88$NBPJ+1dG3fe3H9)?fpspC=E|1GHsg5aN(Oo5-Ysxt1A#
zkSAuMWT37s7KCH*@CD+j)_JE~)7O65Z+EcdnAr{5AR~oQx4*~H_P9Hz4Vpo>O6)QF
zWf6y*KJ=+tUZ<_&I&e|A@;t^3fJD8DBd8C2_`o`l>sb1cL6D>K(*wN3laJ0ugz8a!
z7X*L(DN-H_h6&&R&qaZ<ZCc0$tq1WErFISrof=@8+z+j}t-Y*sWMyS-mLPqwPn39o
zK#KJU4Qf?(Wsx=10N5?;yz!*?=*V{u9uY{&UdEP8IktK3!&7vzk?A$Mmlje03i0Ur
zVjK@&Yx8X7hDTuGQ{XP0C#EgMCt4t?pUIKc@Q`v#)Hb)>_^gst52gE$6>YyR6CkXu
zogfSea{rdyj8Pv8k7J5N?1x@vGj+CM&sVJgZjCEM886y6>zH2T%M;S5eI^Wvt@zb2
zuy*yKOdLI*3Q8iwu(;<%5h@7<bD<#KpFmL|L(VD3GGRO@{d_Z(jF_#(R$J#Uiq{r0
zg%1i6B#)XN2O;uQEppo|48m7|dO3+;bYYe{NFzSXN91-Dil^kr&<)u*oOHW5gw<a4
zits)x#6MIaTNE3EId#neT_1R-{MX#E8c@UhutSO;y%Jkl#<pAHrBK?qXBzzr+LgZM
zAa8k>k>Se~@<97m)6?_pHgWMUiU9Cg27u0>s0&;4#KkT<Mvf57=03HbL?7o1#)0r>
zIqv>=&n6LN2)7&Jhs;9e=Cvb04ImEUp`3Q0hW5QFX-{DE9BP5(|3lkb$3?Y9@!ore
zQfWaz+Cc#k0R>43brcmT5hMg81nKS?3_wCe6qE*)7*ddu7AYx_?g2yuq`TiWo^$TK
zZ+-4xcmLrz2Rt)-Kl_QbzU#NvN-tee;lh0wBlb+W>A<GMTrT0;Kc04VHAFw`*!|%%
z9}N4Dp^ofXPvYiHx>(&G-4gOyrDV~oH%uwYTmSrSH0#lpm+RFD)RS%=Wyc=D81O%}
zC%3bHq&7>MKBHrYfxrVBvP*+li|buBN6Yov#HM=qxeo`*Fu5^2V5#p!2EeaBs1pe#
zWle)^ho0U?&<(dm?QcE*=e+EikcJz1@(Q<c(@tCK<YabkTj@l0Siw|P{W!4+{2ILE
zdNo~LASbdm9YaIq1qJ8up9OX+-WM0?tE(F!B@D<dsPKX59aIO`OCDwatZ{C(C);X%
z-WBdND5ys{cGTKxaa2&?;BcMA!r80eX=>~^Pz`DQwnUph-TdR#D^xZ%Hnr5XxFGf{
zsLkUnUT<6MM@L0|0@J(RQv^=CG9N;LK#j;hr-g*}Uj^;bRi8yvf*O_Z+g8PcduE?z
z6Qx}&bMQi_yc|5<(go8`{Nx~M2HS$wTDEAxE!wi-jnf)oE2K3hmbB$X7CrJ3y7AD?
zBq&`N{AKdk)?art(`R-%>VvNT48xm1pY-*IbpxCu>v@Zl6f2;OIg~Wf)a9t@B8S_8
zl}UT|?p?pC0vj3j=c9N_TW-2HY^ozz4yXzmWf3EPy}ronvX9Rm;Y#4;DFVNsV4NWo
z#;F5iTci6Ajl2pppPyZ&Yu^6_O-N9PshOF<8<ibXV#(iKUwNw9MwPUo8?GLvV3uiu
zVrEV&;075WR!)yy_ur-~%=*q9xnTRS3O>gULv>$~l*&p@1SYUll;6g5vzRgv3=V`?
z2>qPTOl6i#+Ox8VjjFdkmWgqXLB3?rGsGUl@6&WDS}+%y7!^p~zDL6@De^h_#Hynx
z=&YZ@H||v~3S+La&YPs*H+s9f_X@z;>xFMx1pN(yyY!fR*Z`1mWL-qc(TEG+BgtpA
zUBd3M{R=feo*H0*Wm7zOrR<!H%pf^;4tk-~T2IsI%V-F)Yk=J^%g&zjXEtz8_JwBU
zQZ#`RG@5Ykk2;m!69z`*!ci<^)sq!_t20g*vyr7=CVsl2;f_mskhk>l!TH1DvDp*L
zTNua6rw+F}2zM$EvbZt*7N1qEEn;N)bB@txifXrs$jx_|7W4~E$=%?(q#5_n?(vC7
zk3w5I=auM>S6A(<b}ub_7YV&L5}l@@`FQ8#&(MZ%7p9llN0zp}iG-f{amwxdbb~Vf
z1m+zov`9$VQgY&;5JP`DUqs~={&VvT6&4E%g6Zw>YX%#zRMXatphb9|M3GHQs{C$!
zBl&&MMRof`G=HgZ<`NbX5~7O;M8Sz)(l`m+q5ZtF6B*8?7g#0({Jq0iqW^n_W2unP
zz9U~f`kd2Wn8njd?M1{hHVT@WZhRwsf`LDADyLT&!9490oBi^%D?Kr3z`Ce>(b%he
z?s%KaAl~y+?Q*WPo`k5V)5p;TKbGp(#IGV|BeEfVbL*fVyMwL2d*{xZrC-W-p!iZs
z&0`8-rJnV6O&|v6aA_z)-$#=w)^o<#D*5-kP+jw!vXdy#iWrzh9$Mx%=^KzF=;gK7
z($d0~MZ=H_1~o^u2Y^9yn7a>M67GEO=_$I5<wfCI&B}h56>t<uM|MBwlHWH(DoGKT
zyOnF-&g5o&6sNCS{-7+pZ<3umc_VIV%>Ho|hu;!seP#gliX)}^_mO>IFu&w=86(>|
zz$eQ`v6CWD^xPGGT4l#WhRn^)`BB)Ta7Az<V?-|q^op+$P`!uCZU2txc&E5f4D(qf
z;o=rQP6i<WQxmO!&ao3ioWCO@`_#w*4TATygPP$O@RoBxBq)HJLM2sJ?iUl6Rh16=
zX?#&pk?-hwNKjS-pjD862JZRFBllRct8#{Ixj5wBn_1b}44M_hTV~#mF-s#0hGgg7
zsmaN!IcL}u*pQ&cPxcF=9755vE!~)g8Y;)qdvjw;z0iT8BEHdv1P7w_EoAeker_@N
z8>x+s;EJ5(=a;kv0%;Gy6O9UqtMB7?PJW$OLXbNbU|WEH!^%~6IDd=3Wsuv;rK8o&
zFeF?Jsz_B?)gAB3Yx9v^7aS(XKeBeEaW_5}|MmX(*YRH_oBP%NyJe~-w#71h5)#Y`
z+ZF34!H1fh9XJk7O-64a1E0$K*4dbb390TgCE&xDe*US#o7?M!w@C~Vb4o-2{ob+r
zbzkK7gxm>Ca9G1qU7q*WGS4CA)09V!)H{p+hZC&^ZHvpus)dle!MQ1yB3oZy-%F-a
zJUs6T<f@LxdmoR#rvg?6lc@4Vmdp~i4ruTk%`oz6PUD{znX=)c?j{;Y75So1r`Q68
z{lJM?knMZmxBo?jlQswFY6ko^HInP>xMc$XQx9~qsvPHUgR)imlR|-6E2~w^FIR*Q
zfKw8w9W5MJr)thBMKNxE7Z+XaNuG_Qx3RKejoZCLGy$<h%c_{tUx7f6@Zw5gAD2IS
z=1em=jz|#IG--(dQ45a#sTLQdn>U$C?&Y{bY~ix1T^Zy->L6>#H?UUP;aAJJ?>D;O
z-^;Jt8E954w79%Gr@YdYY`2s{>}=Z9ZMi9gx*whfSZ@tgWwITkSnx77n7E3AZj(`#
z>uZ;}X5ZD6r$jKf$5ys<?+Hc6q?U43uOUTXKea3B3<t1e)3iMpgz9i|niZ((8;Jbv
z?nt3%r8(C!T8J+h$wL%G*li`12q=%Jog2^m1I`XYsMidUJLfwsDq1gWxEdZAeX-j@
zNe0aD$|qzalEgY>=`&;q=dp;#$RosnWN{MW0<$olvBq{Mqy}?xaz@T#n1dQ=QPz~X
zUWcj~0*vmBxt4MLwiSYpFx|z=|D3z1BY)pdiD956soNpw=u3-DH=b2iu4r^@h0R>3
zpT(@_NqY5B$Ijq74L(7`Pu)ykZO@q&BC}{$ueq<2AqH<gIWE2a>%87%S|@}Njja>N
z=n=oV4|x2{4MdgKWRN509euT5Kvmnb!MTxV?%@X~F*ivV?h0{cUh+XMwUi$$<UT-f
z`c$SY^>((pOAUTLE^W;6dD1-Ucfr{AF#fV^-h;l)-oEJ|<<X$UC&Ngu9{e`XxTBcP
z_9$(<?|)hV4L4r0RLk$)KI1>k&FnUZc2x$aVDfwrZ38}+B4_dCoB2Ifv0?=gHV)lM
z(k`r(h??=6`ktlcgm_~kY)Z-7>OGvR)ESA<JR{wj7d#aa&w}_hLk-veF!JwnVN(*7
zuwRoOeQRoFGW;IW0q@-t-a8iCH8oVSTG~wa^=Ob}3eC|bhK^&AT5|F!%<t%3nlnGF
zjH<tOOy6#+6uZ;Ta=%YRqL_|^$}1^)f}URI8=Vh=Q@F6+5rpwT)zEtK9fNnjU_>O1
z+w8GM`+{Z;8;oZ~9{Y3g$<ud~4m!`43luhYBm1ODg|$~3gxfjuCTBgaI3#3uu6#a9
zzosLR<!SrovR$Yjb$n#G7);xq`+q;|6_`s+<eyGfb{q-vr8#!YTlW2@Jd2)keXyB_
z6Q<I+NG}LoQxFz7A{pR2R~|9z_>|TJGvjA{>EJi%-Q%2|xeD-{Zi^?a;2*^-{vn8I
zorE@We^Ghj8m-5%bR}bf6wX4|`l?T;K8CHWK)qt!HvvJ7(&85eK0j=+=M`<9UgbMY
z1FrEY$!>rM+pnk|igdW3hqVJ?7_)j`_UAL;gnxuzq_OM0_4bRwBz#R8(O1R#d^;yP
zaRpI6oOOF=J8r8X@-^}QcxfIoo4|zHXO|P;quIW5n*`pn4vt7$W;BYTqru0Xa&Mqq
z6vU1vtP>Y+#wj`xKT=IQ_AA9drBnoOc>+aFF{mf<-8HG2q%93i*`97mP1!E;mE@5|
z_#d!MAHmN!(V3_Gbf)I_yHyu2ZZ-CKBiQ!-yRWui%VdF6wc<SM)||5U3yyah3hRG=
zaN-SdI~XKFgXTn=d#J{6R5@7^nF0xp8eLk}Jt&==y4%LnlL`uun>1zJDB`mi)yByf
zR9S3^CNdHz(B7QY_K3Z7{uph{fz#z(d5;g9&pks-U@K6k^{}ZQibOaM`4J6xHX>$L
zPh7K{-jzlUjt~qtdN>X2m`(C)%^m|4)YR5){A=kG7l14gLZ=(D+;2am8~@kx0NW$&
zBgOy9x5udrZXU_)jWsVGqyXj?voYe;-aj-%g)}xc@&g*JY6Mf4lKgQkPcuHI*5@b@
zgf$vplb(w#9`S5W#UtDxB6y4T$Gxi!XNHa!gMyTi7~s`W%!f!31Oj6$^$j<-rJaAP
z0rSJ=yBntF!g*!7FkHrr+-JGQyU<KmHEPF9kIS`6+ud96YZq5$KnTv#?YX(RcSCsU
zCg+MrIq9eZXz)$V%`)K|eBen?JRL2*ggsKKiqkK1rM9uKKoKpY-8&<E+1c5x&?{En
z(uD>yGpbGaU|M<>x&_&~1U-bNu=J-FX;{U6Tfw1W#iV%iH0JyTkf;|(D++)<f;Aov
zXxrl2g^gN~&>`n<!+N(mhuF8$O61uj?YW<GQt!#12mF0C!dUo2B8aavn%};Cdv|F;
z&S&!H&p6#SP*^<LcyPT04ho$iz7tYXQh+1S2o>i=cazT-1Hs>aS1H#;?gt!Msq>VZ
zoqBg5+Q0gW)>PiW_v9-d?MCI0YOu+xpk`{NsZGjn2#$3kl@I@WkoRyX&!YVmo^?y5
zSXf(^(^u~yP*U@gb6=iOyEc21Bj)^NO+OL@zb!%U@qOmj>69Zx<a`1#ER<9l0^-bt
zq7ql`q(~YXm&_OkD~JrNr|6*7owrPo3$uZ|Ic~6j12O{im=sfNE$!?QDUzrVxIr(Z
z(!>TcBUCk-NS1WIyqu!CqKEe;HGZ3d-T^2x-nVV$Vnz~-JG#0MA*^gwzKV!YH@8XI
zgv5vgN}S%8=F#@1CJ$nCf+0c*@5u+I=mG2znW8$`85z@_h&&1hMEW(7@JU?#7DYCV
z0`@Cf_!Kd31>EOkc-kfRi^n4-;qMHqq;=Z-Yg{$+G-Di}{A*m@>a@O@<}<$J?mbqg
zAU#Pmes;1O^IOe!tc;9|s{BMNp>%4-U;0}W#wIbJtMw~+9xBhyhDA)?7jb_?^tI+(
zry%bi4Ueoq0&_CAckl>;w{#>z?1Ws!D0wJK8=ZRVZGQU~Nrx5piv>K*iqF~m#h_}1
zvevH~WDvCZ0n~dpKHLE+?8ZrV;XizMK)Z36Ufud)6Dusnf3E){@udIb;j34}Gstwx
z|1WO)J~&vpGhZeOQBS+05l0b;%$ZZj0IK{j!6UGC=^o_QTZPW<2;!!ZdJ=xL8Xx8Y
z!kLT54>$`I6L<B0E>8;c#rouulFe(HvZMq#<Q?i~`V5xdLt0+`z)mRoOj>cVK4ipD
zAQ$l4xp>bDxO>QB#G9#_0rUflP{&s>p){d+k`*)`$VJNgLO6AV9TB?Iy@mhWG3lrY
zy`$gO`4KS}@k(I?e}Fj^TKQ^Y4ao14EWW56YjlDt)D|(!B<^~exg>dwk8Bnw7S~{Q
zSeTfY6b;Uav(qR>1>GYfTtG;aJBQTN)g1<WtcwRVu~ET|f|?ajs+YeLco^b%^*uX|
z3wH*3XYdXGjqwr>7C1q?B$Wy%Vb9n=H>!2ZYu>#(%C)sO)1A2i6}goE@6t&w042k=
z8{zFk1~7UuxMF>w-d02fKSKt&R>w*1o13--A3O6L#V_NeUpzRBPWmzFcmbtNLQsUr
z&3=DrQ6e#3ikVTzpSP$>X})8cm4Q#S82z}_Gw5~+oBfv-nv%5<!j$PO;}K0CLJQ^p
zq)vl28ToVeAH{EXalC2#3>_?m1DL~hcXz*Fd;W$yZ>AQjvWxO&X_t3D`J+~M+h4HY
zc#Rt${I|<)zImGVUlh}L<Ui1aV@Dp^+K%-1vuNrc+c$VAFqk0c<H>|OhY;)~N0d#6
zA%N)Yee2W7d?KKtSI;ls?6`EtdZ+3Za@WOiNfYxcgds|}GPBr;_-maovJn;#RRxd$
zUJC8OR}<M1r&D(KAm}Yc%-Z6l25gIkNnW4`RbNE?C!&Z#Kz9doaL-LJv&1Muow8<*
zUWrKU49hD{*Z7MOTUBMnS5DP)Ym$6_$`F0_=5HS<;$3_;p&end7LV1?OvW!LvhiRd
zFzJ6ulDk*WdRGzn4fPo_Y)B}?{QIbCk;`+(Q-OC>)kmI6`JYNXPn7~Lkl8RJ7rliB
zuctye;y;fmXwgWV%9I5B+#CE>70o%0sDmbmpME;8itj4|C!bqXbY<LmfNud7Rh|Mj
zQY+%vNsRT*&_$>{s1|EISn;eNs&(e!*dPh#lBYpfK>h7xkXsD#zb<7+QQLwVSO^aM
z>P(Hb1t@djxf*0LH?~LOn(HZ7F#MKB6u)o!obBS*yPo^$sp=%lds^n7Vk~0veRf<C
zT<zGVDSjlnV$Ofgjz+XFgsCif%6+H%aiF#C{2J$M*gdU*0&`U(kW3<kA4<Z*aYDIP
z6g2EC&f{MXBh~5Z@3f$Sj4Lq+isFd!8(Z@{lXc{!04pmi`_9G!1;WV0G%~m5DR1ZS
zURxkuWvV!v=6~T$=PWbP{}0||M{Bp!(l5+&^=B%CTB~krXiS9gsp{ywg)_&742MC&
zLC<OT+`NLt5f$h6eDHdV{RwvPE<KRtETu?K>{oxrzhKGreTUpL#`E+u8}Z_$kr}me
zDBMccw}_i<?n3OwMvT&m99*NAm(Lh?%V^V|DZ1IzlH1K$RLLJ-WMiAPbTU}ugcQB;
zk8iZZzN9?KP{f{p0jIs)N-XaPexv?^7Gf>q&@v{`x*{4{>62&txiZJ4#fBxI^9z55
zCHK*b;WrJP>%Cr~Br6Uj+`i-TqWgGIkTxOg?6QTU$|n?vHwFUXah`cwj_?cu^z>v)
z-pDT@?x{2!(yNDSMSL^x@~zFyXUgDbxxTwg4JK^?88G3CAd`+AXMxFI9k9yXgKbKM
zkQ%3~NrL&|t-^LgNCqc}sA<Z}RmyLRk|?;kOC93Qp{3aNA6j%{P{@3J>F6sGg_F+L
z9M(H&9FV-H(Ry?-xB7B3EP3VjUwjYzJ}UPijS6`?Hu=iCQn$oQv4`BwrfVaCm$Ayr
zPlnvgY}=Nfy=RmwWO{oblCfc6@MIF}nyi!J5$>emiT$D|Bb|PqJrB9FhP_`H^#6=K
zs9fh>HB#BUa2y$+n|hxt*B0>IgdMrPeQ<$(j9v|EDg~Ne5#<7*-KMeM4^04noBlE<
zQn8nXbM_}Th9M8N7xaVTB%5~tFdxTFrDQV*%?qK`a`2h}9*f}jq9cR!t*|T-(TiQB
zwtPAe);*1TfIwL`Kaa`RclW=86(R%m`zG<9d3N6oPCK`K6jKXcbia56fsZVM>G(iv
z>RB|96w>?dfOA}><KvQPAWmF)h)@9LYW;_&PoM7X=S@M>TArN1hU7t<>7|}+>8a!%
zxmVIfw!_=pxE&OjLvz$=ogR<tB2+|;>I!-<wMrP%9Encl99T*%mLqI_U0+^)wP!B(
zcIa;y+g23$R8+J-Jng)`<C_n=kia8?J!aN>qB~kB=XBM|_5AZ6(Ep1b?(VJsMGuD%
zgpfW+FFpJJ*2C5QKlE@=RZC9XcBo_ra5MJRkL`5_r71`sIJ10h`3tNqH(jdY!MO<Z
zT+f|;*Q-~bX5hmE{Nj$A+<RZLLxxyE{Vv<=tgJs9iAiZOO|!j&pv?cZ92<3ZKd?zD
z;bv^@0-WS3>dHyI6`?E-?I0cCi`wUYHR+AUq?Rn~$UDxg(>vpeHz<jR;VWa9{YG2M
zlZte5w<?m7?%y&uxzkOCVDbeti#W_eT*m%WwjF)zf2+x@`Q<6HiD+R@Nj1N`wkLl?
zu!gjKbA@m&c$yHs`@a|>IMLvB<3o}XX`LJV)X#pK2hi9ByOn_EJN<8ia`mcew4L=4
zLN9VZ=GUINf%fg&FDVGK2>25RaRPsj5A06^gf1dlrp&uJ>TC$Tz<eciO}NuCZfp7W
z)AAO*oY&wi+M8K2)8fEI<b&mSejTmu(dQ7qxrA=|<9Ua|v*)mBM&iy8**jL@#-=$>
z?nNQ%eM4RuBFR!0F!KM0Ry|R%KUWNPP8yu-{fG$QaeDmBrQc6@|37h+Z5aeg8Ydn1
zydidJ!ZV9huTFOsK)H$filursVWfmyK@>M7lQWoH;w6KN0SYv&AO9rG^KSpkQ{oyV
z&8R2luFX0ocaj7=4>l)5m{)D3!T$Mr+S%>$#Ho)<T^o~TrZMuR<DQ}4T0>lNf{LkX
z<!m~${A~Xe-Y34O(>Okz*)IcyRuw?l5|2f!{l#1Y_k@^lt#o-c+rYwGQP(Oy;jA^Q
zVxH{mHVE68c-ini55(1eu51w;eY`(a;xA(n!V@EwiM!dlE(2^s|97FBOwMS85zF0N
z-9k&nH0Iks*o%sXd4p$7!sw|reaN=IyHGOA6;CF3X}{;gq`dkN`~Va5jy4w^Y^M1A
zqtI3c|F9%Szx!rP>_D?b&JOca4;gL~l!E^p`A^lePz`-q7Dk_bO%cqIWKe&k!pNiP
zw-SP0;VtQLklmqv=g#VCX$x+ytq6iPUU?-Y?o_z`0HcySYXS%f7?F$8xDPaeCh)c%
z+%-=Wr!!$HwumuT<@jL0{UA-{)JWk9`xWwANS@p?qKDnql=`D|{_q(lJ5rGo>(^0O
z-+f~bk1M&P?>?C@*9811t)f`;$}AuXAwL{p`u|YgHTAgyNwY<&4*Au!&bcmN%^P+#
zzR|sj=5=L$Yf?>Q7az{kzkU$%O%qh>%_gCGpOzLD<cGX!{@T$$xV#pReqg`x`-vMl
z-7T%7@Uj!~$9g;peFME&Q^FC+3PrrJF!w>Z!#W40nmT=Eg8*@S=;J{K`0RGOIFcCZ
zmN($6ca(UGYj1kZT4WCsWbO*;EZ5c)6{*IE9KoUk#iQCrsp-i;Y!`9#g8=MTF?jKo
zK-Rmcn8O5Fq*=G(Y)C%?Q8vfF1>ydi=PYN8&;4A_QA8sngPT1G2TGPi5W`tiQC`kb
z9m{oK=Knw^eWyNm`$~eK;ch7Fc^T(zoW>Mg32AkUwRu}HpP-lPdcEF4pNY<OqWSI%
z%qNAhV7kY{3e3(s)Yl!lF5z)OF__Vx$yu+y@e2Bxp}&@};S<SUP~Mm4?rwZFYAaaz
zZe+$A%y8KUN>P25k8PB=H`7>ZbMACR%N6cU&jQD>&+7d`9E&y4KQ&6PEU{hm_AO*$
z+s`_&Xq@+Sx0F-wn(_IR68&O(qxvTgLYUC`DEtatOzQ>q>NM0|X%U&5s`kf+`Guze
zx|CVkR0<>KLdp=t|9K1Jgn0$s-}vY#=n47BDzpto?sJJ5<M}J8YS=b6-!w$h`LsPT
zDER0+AE{+_GWe9&^|ss-B{Bn(nA?XKtUw{-dC%?_anLYF*QEBHkG#<qfnF8tO@C#K
zU4Ed=Q4hwyJ%0)z97a^{5$_NjXh#FVGcw1e4^ykp5H%xdC$_uo&qVwtA?p{Yw@a3e
z2)Dwd6g<1$0}%*C-S~-Ha{(vh<%<piHWNz7=zbbVpIqSb`|RsetV%W5(_<z<MS>T|
zC{fKy!uM;L6W?GLOh<LrLy7*MYi8K7OLV`ZFH@ciF5_a8Cwd(Qs1apLeoV@$n3m>-
zF30$-Ya%$T{Oht4XNqqY?j%lv($FiSP7pa=tV*1^Szn2o%&r2`gBOkV3@^WL8r)Sk
zy7$raMr!vtfNM{wAE93MRne1B>%IK(S4V741YxSp%BFdjon9V$UzVetf70pxwZY+(
zO$;v6UwrV%v5<92XQ^8r?zPz~mf3CBu|A#9nB|tH58ST)aijLn<NAKHWJ{qO5Bp-t
zS=W`P9DOk7H8ptx1K#iMPutYo&+8^e2(hLde?gl0vp-K~>KteT|0o&9JW)1S?(*;t
z6_;S_O?34izx;HFGD3zgN85WgK+4H5pC3cNW~7mNYYDZ<m+9pE;KvVhVf7jrOpwEq
z$7nh+(#LKi-Ycz2?UUQ$)_u@1WY3kp=GAG9+r9_Hhg;(%N?ubEKBUU*e;nwG=~%vb
zwDaW8&rc~u(XNB|g5a{W4s@qjCNJ>_9#Oi6eWh!#HN;Uy-9d&t>QFq)z1<j~H<h5_
z@K?NQOw<qsdET+QHp_tH43gq^OyD@jIW+z`?bKGrc%-b@l!}5}cef;clX`^V`ZZom
z%ALYw=lw)(P(;bZ-`kA{_O+P!<1son_g8(hNBx(7zbh)4xA5=#g**(B=I^@dV4=>v
z!d$_4Mvj;1tIVBaop!qWE#FP+Phf7kfPy2lU($SDQvDtULFmxD`UTq@T2uWJi?&^P
z1#H>SA>y)&XILkEcn{5juj024gErI&vVxk~ST7f*0d1p)P&UKa=dX?k&E(x{^f|iR
zCX;;QhIs5*nfUdl=NvNUl{=pV^<{nAi~juBP$%DQqZl$`<cF6w#3Slee^TGO<51_K
zly6-W>#K3MS1*G68q=LM2wAwiYhzz5Ri4c`ryRzZRE7}^G<4qi_W84fmxxKO=fU1a
z;Ep$NIZzeK>D;|LRUmoP1wwBe^g7~HJi_q_x#_fvp8n4|5yBo#RRv0RbGOo}H%Ftr
z#oIfr7^*bVLNu9#=<*nVyWhLO$MLL@zgr{m>Af!vlJ)=s&wvORg+O3(Ds5SgK0-RJ
zv$PpjQdbq|jGJx?V#4(}FfmY!1$(3%%7r*xHO(PLKgX0$A-5PKUQLI@905xB6y%lG
z;%r~YMyxX(Rk_5`oq=K19pYRgc9sfC+_>cD*7u_1McRexdlB!};CrBwY{s_bjI*NO
zEFE)0A&IOPSMSE#@aa>rh#(PYJ5Ub$rLEXZFM#+xV=^=}w7GM7E0~{?BQbGvQ^gLr
zW)P!0IDDIImFsb9a=TubW!cR9_rd?;HWQg6+V$8^3uyYVRx@Y+t0s8G2v6Cpr>AFF
zEUhR8ss`XeiU<J`R{bDDC0qvpCWvre;d4Mfte=Y?Et1v<cBdo~Ou0@mb3G1T1Aw-<
zNL;E#5eSkZU+a*Yn~rNdj^jGGjM8W|Xa^0_Wdx*R7BQWOw{s6m<8AK!ceLwq7j-V?
z++VCoQ0gxO7t|Lah$E7~?CwHAn*%e*3gLy2v@&u)MGQwTyh-g}kS}>(c~!F*$wR?s
z-PL0cY`X+KpviFki2d&7lCU)U^WrS!S4Wvx0+wuu84>14lwMqeBRo;;A;LN2;lqd8
zV%B|mih|D>HLqQ}R`HLg_u+kGdpmdxZ~oz#%!*NigM+gH)NJc+XI9=n0>ivWG3x4j
zBbXCubCFH+Qvq-Gt4oqtbW&<)#oOLimC3Jm19e1@PO-rs(++?-cgL=Z{v)Qf2^Jij
zTjv1J#Qc~UNZ3{dN#(=Ly#)F^+t(zrD%1F`tmlKj!WAYcAtv@`%;c1h&G6^7u8)iO
zfiF8Rw_~W;#N#~MYZ}}<;TsxwngJdx7&^6H*xTP<^{fcwX$~-Ahj+B+n+PbB64UHk
zHQHR|1uEHXpDG?q@Z3L<%B}I&D8EG>hUJ&(IjB&51}#Bf^YP_0bDj9`biV$u>f9t<
zHYc^8kE!c6ym*LGPq@?tA()|lCsIr3bJL+2CjYiC{q4*q$Lha5Yi2(98f=YiQWKqO
zW5DyPcmG!8PQR>KS&`v9V(4}5U!ba<MpP53440TT-4{(iL-vx(=T-XUgLqMo4B`IO
z(k;8|biGFw1#v~9Li@3(iU*z9PJR!riWAWooL-{DK#0q?gbvqSyrfC^Z4~l!DN!wZ
zhE~=i!>><U&QBg!8VPcvn`PO1j#pX))wk}$IbvLF_m`~!U3);h`?Ndf`25GB!*g{n
znla>^vG5NPRMRjRAtf9Tg*1OJMru7VI(0E-KCyw^XG96--z&w57>K&+WVX=phAKAV
zyZgUC$4I*oKbP_I1^*h}-8nnHvq%0JEqZzvBIymjyw9?6J<0RRmj1}efPd0h;AvEd
zakF@VBc`&HNvX~L_^=}@WjAR5xdHgooETk{*3>cJq$1C^C0le1HBuz2Uk&>}IWn95
zcULF=)zx!w(DCCzIMxt?)=4DGLz4T;`RJv$251zx9tL{!!=YlEf1c;x8&dZ(E0#v+
zKNNjiUgpM`e=+>;>t0<ZN@4%I_>~>V?c%^^3CNg%d>@p3LXb7Y*q?#QeF~~Bp_#1p
z%%B!z3!uCs3Lo)N=thdXwmDbk2Crf8YdMcYKtupjhg?JV31gH<{HqzjXWCl4XUqKG
zxX_<s{@=pmImc2n5ZdT!>VLo+lB*r!zJeV;JD>z2x4oMC_PIgcwAI>(6e6V2%{6FU
z<OA0J+u;iR{~ieY%3E-8-*0{Xq8xsCh-7m?%KZFW(#yCPhYn$qdZ_Q^-Bu=$4+*_E
zOm|YVAFDUrZD3`^JQp*cT``hbv{9sc>_p9}2-27L1wPz(!d>|KcA>Te<1xBOg}#?9
zre<wU2aQjzwVaPM{dMc*=*G%3sipEM*{b8bk#EgMRw{=}=0DF~3vQ%DZXs3nrT1>Y
zy#+kbiA0s-@tw#|#WK6*d~YNeYkC}aqlUix1c2;Ot?oqBLzI1F5EZ!hqPxCT%_5;0
znB|>>p}#pgf2Ue7r%K0nBBo%*Nox9hhHAGXMF&+DovCijQ2VKKVyO$&_!g3dpV+5~
z{Z<}@!fT$K)hjmg?a&?~|AEelc(?uCt%<YE+lLX@lJ|+5*{8Y8^z=ovJ9V|q*bviy
z6B-jN_na^nljJf30!LNVVI8P(%vimE0Lx@EPG1nxiE^=9EZ-;;o+kp@znA*K7a;VI
zv~um8i-9hrR#bMVvuO^q>?UxW@o_sPEZmOd9YqA2J+pGRPhF^&J@TIx;J@!o`8~l%
ztq;b82~mjn2fDwlyLHu%+5dr>sdN!aCLb1ZU~Ty+krB{1n1^nVaU#SVVMkKJT1`*C
z6^WPA*3w!p9*I-bB+GvH$^<kj-A&dQ3|{)T{2xj0&Aep1jr_9QVaP9+E*(yskKB)u
zc?pO4;K2x6xgicJ#>p|$(B)Qc&otTW>-|swp_Y?lZ1}={_TH?0t;psEOQLrex{`+C
z`!?$b`3ed8i0{9YPGW-TnB2=iiUj@bHAknj!h+aoZ9ktE73GLOX7#S$O+tloy+Zi(
z>1i`$j#+~(1>cMkY=^ibu~M;~tn3YtDFSQ$k@=q9x&F90ba9kJkoT#nBN|Y4EW9<s
zxOEp0G;a8W@eOb9UGwcupQFso6p8H=wITuQX+s?#G#-JdYq3ng-0j=9zq`ucij!RO
z7~$LN1*GQi`}+w5*Ezi_=(k{zM89=+F9aUP2Fj_xk%u7gR~^nAtb2dsf#c)cj+}rl
z@2xjD&?ur2=mNdg4PTr)5S#9V|4d%p>N>@pQgb#{o~$4iUr!+sD{Tddo!&NshYue%
z6I>l`^<a2B+JJ0U1ns&hDJk6q)(aRpFt||w=#1Ux;d!f_@wor7G8vDv*9;a$5jMWI
zv1&)TEX!Hj+R~VdT{^il|LTE`T&)OFO`7-d)2G$>BA-|ZJ6qDl>sne``9VVzgev6L
zT$9J0wV)tp$zprKJi#-EtEX$`t{^jilo=V~5MHw4aFRYdlPGORAxK`K#JSnD9X+op
zOPSri#Dpuhe$H^Fo`mJ-kS~RT;k{!WxiFDgar8BFv?#HG*dWiFni{D0q{Xtx`T%QC
z`hpMnLCpRcGeeWx8-mTOB-~OZdvZ84FNV4&5y%fjL6bJPkv?{%4@%u9BhU{N&q=Ok
zWtSr4W9o)rFvNMGqh*uFBArnno@&8MCZYG`?d20MecY~U&JNnhe^+Nq>(qhx>pY3b
z1Bikw9Xobx#J)W;DoTqqDShe&PiB7pd5UUJ1VZ|Lrl8dvEs}aU2Q#)^qr3O9{Ql_+
z7hICF{f)5?9(+I(yEuAfWMoYB$MfBREcP8#XrW+chWzC7_kd8&A~|RXN|=3k|Nc(_
zd<d)oUur@#LQKllBTjkw0rqE}4}qr0lh_Lrg}&v04yA{Gf&XO*9aQ8<N6Qu$bPu$2
zbhcP<dI$j=LA~={>wo6N74t#EINXccJ%V6=67L{<fW65T?qCIZMN*x7B`~4ncS^|o
z{4Y98%B3(3G3OI<a{CJQE1?kUpa~hH34E8WP<BRZdp_7AhEEZ%8|9MpP8=kQwhyC4
zlXKgM377v|J|ScyLX6DJT%-gP4k}4HNn5_d+rsn;`}o7{E+nGHAmxV<z&_53Hn6cp
z4Ha=PGxtuj*ZlVZSW{tmC8-D}Dbk!jFY99N>gswFbc#^(?jpk#AGfKyU}qc7n;uX@
zC_*MAI9ONOf+$J(=Mp#ZFJ|+tI`I7@bo-MS)B>OgVHQ?Fa9b-=;di@leiFL1HU!~G
zxx&rc(2n8=g!sl?%Mur6E4?N*mnLQrF+}^1Ynk6sL0%K$yN29Yox!1U{5KDvk^f%C
zEX+Ey*C2YQ4+=?5%<WSc56*DD_Bq$u{H!A0+|qkF?QV`Ghq;9|yVHCOAshe7_{4;2
z+iNSGsKcNPh##DjlbX@ejD8&*?PpAC@8VL(cjk=M=b&;bf((9}qj=C~=5(bVY?7Dx
zibMYU(ja3*aWNpNyu4gjvjOFP9lMS?gt#BBW^M>QE}9L=aZi+e8VYyLOK%@wevDN)
zczA3LV$Q2zE2#;E1dk08_b!o8N*I0LfzHAI#tMt=Af9Rh&<c~5aHb~r#jB_&IU2&!
z_0<ZfC4LP#YPs_qe#A+!xI*}!7Qvb*d(A(on&*^J?%kIay65?cpLi~i=Wotw5I^q&
z##|}I?(&U>KskFcJtk6xYYzJ=PxnVOdUyNUG&8+j&4~qD@<9p>;-}^1(A5yWw2)Jx
zSF_adj&#P>89MfzqugJNB)aU*3JHp#lK3Sg?z=HzQflsBTN+>!e0>2k1NMgDWm3cI
z_XJ*y^1kO4sz}Kb@lv#%H)10yd+7Tin+`K0I^8bk;Q!_qL!SNdBsFRkhHR$uH;?R_
zTDx0L_K->VKD4`;@?BqF`^}$o6^2iLPsu;pU`Jxq;|^36dQQ=N^sU(XJQ+s3S=OK*
z5<lD3D9JD)WtL{r<xd-%%a}Yow7ULMtvZ%A7!jZGZ!cY#LyY#12RBPwIBHfeA6$56
zKX+W6`dZf&(FA?b?uo^6@)}x1iaFa)R2gALZ?B|Y>%Xwl?_xg49vSqIpni;>lk~IT
zd{L=o{i^~;3<r^G@kbFeV=Khx#sVj(3KPghK4rS1<H!TEwtBxCr1XjTOhl$-G_K%6
z7gGBBql`4~!Q-$K3G?IX*Qd6agT`N~4eCAm5Ya+hBaE{|v)(|7{er0+uKh{hMNt!z
z{Ym&;m!5+bVTbNG$K}o@&cG=)da!TBf><ybUo3?28a#h9lG5b~?C`>@1bPY;SROkb
z>qXp88~ZnM;?FxxbWR91%l~-fyK3CFtXsyj_VrUB*Yn55H&_A%17|i$g1ABONEeiG
zXySDDAqxQtc>5`+5aKJ6ecdtE<%T-WkipOB!_3GetteS;c<n|_Jd+3tp*nOvc*XG8
z@Al%e_ttOu3Ha8h)g_{?vUCbHZk`DazV-(x36`N(=TiGPa!tEHs!5NYlK7NewD!T|
zn3)R^B9wYjd353m@9n$=ElO&46p;-c?I9a;{N;pi{6d?u%gfuXqilh!ZT8_G$M+l}
z8ZTqTSi&qEDzaWwBy%NS{qN&eV<m>Lj5>8(<v!bqawFm9@<Wfm&~C)Ntvn~+CFNrD
z%$9QMp1|h|yH6NHP7=KD;NI#>e94|>R^HZng;~=zxwo=|cpI@hu71$B>%J;LOW$!{
z<5esho%X2pUdG@V1p~XqI(ho(Tf}AbF%$^wCEIIb6B)<rw%W6XciOo8R(h<oi#Kd8
z7H-UioWe=Ks3gXgj%6$juhCCA$&z%_P3gTJHrZQxUD2w%u4HJqD20?Je>$dA8LhMa
zGyRa_41>&YW(;Zip~4FMD{kTTU{fGbHL1Mj9uWeS;X=PG)fx?y(EVjjPI@VC!-%u4
z2b><UAm!1`pB6JgJEGFUKUzPcE2DD-J&Yai?%kRCm7rF46B%YK?V^l0cD1zBpOY-L
z55s!fp}6quQExdb1Jce_Fkm_FlyTH$;+4@`4q_O3dTQ}O=CeEJa8kJa$yfRfcUF7l
zNQJx{##+7pG!pZy7?ogtvVI*$`*$O}F8$X}w&J#&-JH^Gt9-YdQX0wCMsGxJXoj`K
zJXLaC7d%_XifW_ItvIITC1X?OTbGH^ex8$RskX-QVkqXS(ny%Wd3eeXyypE}mX1-X
z2P)s(4{UB^Y%b?@ekX*h4p-W5#rrksI5f@!y@rvmHgZE)-+<Pp{b)q<TwBLm&`)@;
ze^2h&Abs}i8FN>6_pqtc`}9o7#hR~*>*aEM@9Tq`=M&vp+oMMv8qN$~$hNWA?-_6#
zx#d^2^|f>aCx5zw-tFg~<l6m@#y_Nlsa?LRUe?#|C7WbffuJ%9@`2cU5;^&x=ZH$4
zASGrz=+3;RpO2AqcAC`Bb!IA6sobw4JNC|l0eB%Obvx_&u}XPo^gw;7a>G_~Xk3a5
zhv0%9>Xy&GW^21RmC@35uIaa8c!&lxA)ku-n~n|L49p=l<pa@d0044j3OEH=l%2Px
zRom}klii_~MVNc%VfssJSvtM%o@*p-+_kIPTQ?1FAVFB5+Smo)cxggEq*?TT6eZ(3
zA)hAZ4;2F?4Jhw!=r3+uJa=wsNPEA`jHTWSQtf2^fM5O?c6CO2%z4Uc9@nYgEx`A_
zYY{l4oga$hti2enjy_Y6w7QNO!5zdVe^PSdCCpA4de0mrpa}9`q>VxyB-u@ntO6Ou
z^jn?Syz1G@PaMtPKfkmTm`rO>VY|Ayh5S;)C_x0NH~CYn+cs&7j1T(xwyu3rDLuNT
z?+Nxe(^NT2bsFdIJ!NBK%)pUycySwj?17#qgoNgQ7S)cu3H4p|zS>3N?Y&_FM)AWl
zi@dzN`_;cB&naRn5dx*_RNf%*o^z^Yn5*x3F}MiP9m+xxgG2p*2!ew^VfizhUCOay
zSV~$$PcN42J~hG)8chGCi3ZxwQesN$6>jpl*ef;%4Uf5OKTnj0L@NrZw@8yBkc6=y
z8Q(VT8hH5i=)x%V{Z9{c!XPau^m^PxtBIlz=^~o3=3}<Lv}+ezN#nFLX9zY*4Fn%Y
z>E9kkQY}=#uUANKNR^Ji&T(GMam?Crb8*ZNv&p**(#8<b&)v4M!^~v-`t@r8y(vJn
z4R>3ooefN!PNv9x$POuuAxb|rte;=0u!rK*fruaq2h+Pn_F#9#FduA9U0vNuDncOG
zR&I3-4O4?)9dh}sb`&BBXdKi#K8O`D`m$mDyz6vsZ}U^mXN|6ci?^}hn{<#=o47r1
zyQ8I*@NkfZ6y4g?)HG~>m;G*<C7e#3%<uMc^v8DIJfCJ0zd<pQQj9L4Qb6N2uxc}!
zvo-tY?LdZg3QTm8y=jNrkV;rYh`%PX(G$A16|P{fH>He49ityASbnLro)fC}LJihn
zJbO_NUU9p#%KSalR*v^hSO>P|k+^DD<~k?6WtV~XCAx(LHiP0egTE%q{4%q$R*P&T
zBqX>ds`$S9-Xy2D`00a&`{lGF-Y@3}9~P@Wfw96qes!HtXFrh1{d4{8)2CO=Uol$a
z9-hN_ASxsxRMl7Un+9>;Vnus%V~-mc2VcE$nA#1^FST}XW?&>RlYg<mhB9g&XxqOA
z8#%>u!l~ewQth&genX<hOt=tk*0dMJ<Al+B750<#GRssh9DB9r+8Zm-vIRRm1BB($
zgu@vonoF;<&b$!%b%7~@QHi9zwNK4{tB6d=koJOTT|#iFn`cC{o6mMvlf;o#l_VF3
ztNxp$>G@rob?aA^@VE!AivL#4V0{7mXz-)sr?x_#do3b32Lw5G2&sw*J6=a7pW(3<
zzw;v6+@=2pSLPe9G&i%Q0ll-0W~5V}X20OimX<1BnU<(1sXqH?Yq+gh;5_!M4^mxt
z;##Wq>u7TuJN-p7{a-WPLR2-26Vzt|U2}AgjokDlZlG{6imP-eFd7}NvszHAD|mQj
zzKHKU>G#C;H_}@U=>g|>Y)AWI3d(PvZ6*Jd{B93F@V@#-Ox)zv6K(>I(8lY9uA!(?
zt;WQa`B-CvwQsKzW`<r<=nVb)P{A}Xr05?Y%sGGID2JV`oM)S>VdoY7Ug={;bR0CM
zJH2fC^B<U;baXZuHOQ>nyPY&isbW6V+ORA4rh3eTvprHY?n;m9GbPDTk@-X}uAPUT
z+@X=GN1L|xqt0r|HYz8d<5}CBw}ay1fiY&Y_t!vCmQehlIVu9Yr}zgt@31NR{r_^5
zwG)8zaC+3WC2%W=bks)NR#qVl!CcRbb{n7HZa!K2^ajn{Sz3p~h{#cgT3A@kW9!F+
z$z#M;`+0?)ac$#_7L6Yrx+y|6ajfjh*GF!nKE-?Cv0Ej?f(2k3`QsLBpJ9I6B=Y?%
z3Op;%D(FYriETLau58*{-s1gn=H!!qM;rbi;Zj`ualg<IdpYrL5k+~_1M$pXPn=pV
zH5<ko*MqI0aTgAmiH#l6M}wTr`QJQ$v_qKQX@kOB<FZBU{Wq;99Bs<zhpKJIl%e}o
zp3Idu->#ZI0gmy^B4!2p(217p@iphu491_X_qHZ_%f#yl7iGB^e`|v5&MBmnUiKN@
z^cABRNf7PYi!@dC&zZdDZ_h?|RzX**6*wrgX%M04Oe1J|FZ<aqRsODspvszZQ0V-p
z2mx+%__ke@Wlt8LXVja|L_by2)ja5ejO!oCpBp=hXoZ_Mzg0j6<e1@H8}WHbREhI$
z^m+p+5+w7-ugY`s+jrqv7_u1@an_y2;H%d~o=g6?*SqTZbs5{F<P^=xP#Z@}c$gN3
zMH;URG70K2{3E`=(mOwBVu%#I*^{%vCqtuz*j-2LHX}I#%^9bdO|uLonus=mm_69u
zjOyD}H47)iUMs>>pA4~1_}y1z-()Cw_!dJFW7>#&4F#grC3w$~8EfooNlqM}lzI3B
zr^4i4qsfj8jECo3Px0rG9@>@l+54<hM0OHP-!QQ84h>Y=>pTmRBf0LTe2`4DpqdhK
zTG!y7`$~e1RP(Q$TTWCssU!NP+C3jG<o<c99a&&~Hl&O=3}SRGg8DM_#;*{TBS{m{
z`<NRqssqlzzC_n;55MKNr!h#mezYUGUG~#>iTUO^h04Agnatj)TU*h;Mm;a$`%`{?
z*pl#}BG3XYBHnK>X|u2gtX0=b+_hNrD22V%*DLh&9Z6px2K_{{rJ|FPP!#&og_aVn
zr0GD<?SPDEDJ2e9KGY#oEcm+~*5o@meI;oWRDLTquD#XN%*30wE-MN$g25*(xER0>
zC8)$S_sDTik!vB1D$V(vve26{5*n9PY)=fuR(g3_8yF-kEjJ6dGeqd2Zc&J^R~sP5
zFw1^Wbar9&mv~27x8S{ehUzxe1s%>cLE<2Q==X+2Tj$=u-tS@8fU(8>4rKkR07O5i
z<>DSokU~i8Zu*Qb;zsg@lv`cXJg70os5W4)f6kR}pu<fyJYLjXWw$L}#_(vwZDa)%
zN=`uA3*mC{Gmxmavz{>W56exdsa_Dxwhvz-N0)nsLzWFx8~Wce<Ao9SS$oN4<Q_SJ
z1(ykc7!IHoD70gN`BC(}_n-|(*=^N-346zV!gUTmq&nie?iob(BF$k6p10uik#hk7
z8FQ9iK+~n+Rij_O9@Yb}i^kawe45seZoa^Pqg7}dDs`$gpe~f~**M=j2Pg!34t^7W
zmUiXHC(tV80G9JYiqJSC7M-#G@i-XC=y}oPLMEhMgJ`o~H-pJq6G%3ab67V`9*5sw
z?TQDkp@@PU`CFakCxU%8*87YQz+)6EHRzfQHf{{hmC6f$#~=InMa*+9RQf_?a*xz*
zWuo&SQLl9iDjl5Q<1-h-VG$@S>HY4VIR5jwO<5UTxXoyEALCGEdZft|0u7kDhets}
z16N9o)y^&(ZS{9SXU(6Q*_w4ju<gesfaYgAu67>z*aEebE>MwR@}U^mz^FP~VC_df
zx<d+Bp)jq{iq=yYx6B!KdY#zUuO~TS>u3(YxEy4+G+x&P$(c$3)D|5tU;a{LZ>!-2
z#_SE6z{fneod|$!pNecWFR-z)eo1hB!XH(!)}zNK-xV4j&Q;qWNqrI&wOl<#XE96O
z-kl07;eNYU28y;e=SK~oRa~;a!YmcpjT<-KQGBnOsxg=M-WXMqbKhwWldF<Ria{f(
zjmhszHp8SQ`1TjRi1Kn>rYM$xbYV3ceK%$7ZSwCvKTPZ@x6E`kG#HTTF?*_z$({F5
zW7z==v*B<8FD97#{})!@+sR)T`6U1~vqfQ;<DikUFuljpyKuQjgL4556DVEgLz~~;
zoHOy=Y2*oCHM31;gFr7&g=9jCAQ<aJ3;6oV&Ystt(v_|atV$0#vFK-*uitgmUhLjw
z7^o&22wTu6)><T49I&@FxE+9oLIhf;A_(+fO5nKe7tS0iMM7C0ha$5d>O~{hgPl?j
z*+VwU<<e~YHf7c5K}aKSn2&{pMPHelFmj)0q{8Gk%#%=`g0MF$A~6{MIEY*V0wiRb
zVYIg5E<N=VOPd2l3`dW)JX{@SE(JqWj)X8yL&HHs$fuRxwRcY&BZ3e}u)2c2uN}uw
zJ0ADKRu7i?8mHo-Yxb#4*^dR}Q4iwFX`#?p?eV1n*il}efjviD(}$&c2k-fSn{HDN
z=5_NUGRl`{n|1N{DpyH%ss8@{NPyt-G!x?S@9{f`^?cJu(^z#xhz!wO!_QEPPo0?2
zz)Mpqup`=0YcKJfg;Gb=6CuiBPV(juv00>MceM3$Y*%m2C97d7{I}jWd)iRe7a@Y~
zs`MH9Se||rY#XV)%3{MX{_|+1*vFcMjWl%0$&$TFe5HlcdzEL6@_*WhSvhvhg3WlC
z<KXo6)Qh#gCzvUvcZ#kS!CnIVu$SgAlUp)=dUA4-&+sCIz{d{ZMoWfmnQ?|0Gkz#-
zrWqScsG^>YR=BGA{FE25pCpm$!Ws+g{P82FfOy*jA59ugAUr5SV+7BE$JkTm{ca*P
zHv5kyZ+cfsY_Y45v_gWJ_n}I+v41RQZ>nq!ruz#c6^60UGsYDj2@+uEXA$3HyG#Dy
zV<^tGFi;YcjVmRj9Dg=$6AN0}#zGXcP7L1g)6LCIC}SiFaTXcD0H^{{gqh8GxHjBu
zl13^qYHc_qd(~)hCk)9I_`4Ah{|MHE9fX*y!i~@(C;9jiJ0Uvd0x1dBp|ILpy2br+
zU87n>LW_J(w4eS9O#bdh9gud?%UJyw_NtzWwE32{HWjoz1p=iI%~Sf^!ox36O$-x$
zmBK4eb75@B&zgN4o0}YRS>)fK7yv`S*0#{^W(TE>M43H!uy-G}5U}3-_IH=?x`(R|
z6G|berNV3GBYegg5Eq9G&iPj6?#xergsw%qbTt-icM9w0Tnz`rTWiPbLINR}qKDdb
zvp-O+t*H9fDS!>qQ=2eHWXJJ%1JLBYoLX{^8L}<bjaozDnvwEfj-_&=O=G;ecdf%`
zfk``M1+}2mC<%gyOytX#-nPYhvYhP>hz-NGypDl^L0`wxKYKC!$;hTnpuPlJmO}h<
zNV^I8Lqdmuj8XXcaZxi}sOorFW*hDm4pgq4y?s&aPczM7G~tl0B1u&mKAbIJsT69M
zog%CfMOpt`=UR>2aQi9He^DNnOImgLwyy5BB(8y{!_6)nfb$|+n_>5VytWkIpnvJ_
z-$&g(4W>Hrmnk%@Luw**w_DZgbkH{R7xrz*y{~qco}nFxCt{``T_W5hUWqSsv?aTy
z7WaGdolhf&bPqC97Gr<;OxJCV4p!Pv@6nCLP-9)yULU`jsYu37<FVB3HX9Lb-cMwt
ze~9&yh*!MWRB7ONxL3I!9d|`7IYtQO`&#{F;%$;4wn=yF<AK#M)dVKyyvnebkJ!5C
zqs@@UH`1k?b!6>9GpfaBb=M<?k3)j;w&cHQobk?C9$zIobR0HLsw>RD{aBe4>x%8V
z`y)}`ODAshKp@Etech_qiZRJe`+IMWpn=m<{jH#&Mmb_0De*$M3-I5J4psi}Q96d?
zB^ib}^I|MHO3yq}rn@l<A;|YE0+VlNDeJF_vcUul2OLVg(8wotDf?vf>Et5j4jm%5
z&5qlzb}JK@PdIu02jBK%ek_bD%NHs4XMa)zj-Q(Sb$WFL9Y?Ex`opA@I8^ko-%LpN
z^6Cxn!P-zQb#$}d>gr0)#a{%xiJWaa<sg%iGE!Eh{Z@J?y%G({=lGRA;E_C*9|AnS
zHr73TYNl>$%YFV?&>;_xfFrht3QLhvu<kb3<1@9kFvlSvVN3zsUc98-T{+mP_H!Dl
zLRYTTDuy-RyFTF2RLje5B|LM^MPt>=DSZx?jFO$+PgfZF527w4p~<)f=-$Y;@$2m1
zOg1;Yh(_YZ`vZ#yw0ykV48Ak!3v+pp=)$_|ed=9F()Bx4UoH#NQ!}Bv0{Zvq!1U#4
zzE?3Y$#MRqL&JC-b;a@W2#@_IV)7yap;rP<eUMV6$SY2p0_%GJsfHNcRUPmZa@<*t
zsa1R~5XZlny{p(y_Vj7nLo=RWDG!hOL!CRJM=C>~@0*>zl|D3Q*H_GhU!JD<90k)*
ztRq@GTfhzVo3NqGoJp8``@xEDom-Y(y-Kv6>wp$nOD?*5{Q>C&-I@%MSc?Ytj=Y7S
z0bahp!&<sL5JC*kTlm|A-{_RDmnW=Fhqyi6s~onI+A>HCdPp}GDX0EwD{^A~Hlg5z
zSS!oeqW9@>{~%-S`mO@ofcuxysP{F0RGc%pa;Y&|sAx&1oc8JBj%3e~>nE0%Hzgv2
z$jeiPhH(7*X@ddT%o+;E4mg(7uzOgu{>XRN-i&19E*`pnvrNC1QQP>m!CUp)d16Dp
z4W6yxs3*P;5>q~&peq(0TA`y?Kd6()#CE(Fzy)Z0+6xbQa3VN=H7xv8E5G-}3oVzY
zdoSDY2j^)ywrf|}9tE6bnN>o&DR}fhEr6iUx1U+|+kBXrTpEh;14@MpxSnHZ=@h!B
z`ww?gKKMP?5=wa!6>Yxh<8OZA&4>kC1n3})A$a3QnKs*Fq4tUYCkfT*k@(obqis@f
zu=|0Gl0W!|QX(hG{mbA)l)gQc8K@{or~X7=UHgs5JJUG{(cqQQSGbio(&Lt*sVZ{k
z-$_;xg3S{z7H67>HYc!?{U%FJQ1Wj#V`8GAF0C+Xy}EO*dYDcf-1LWPeegv;*n7oj
zzc&nCoAG^Q$wTU9V^|_6ygj`getdSN;XU(|VI;i5g&Nvp7Kl>S!{s&P^68<NBOW3{
z86T*4<{NMU=Yuy$1)C_-?7-hVAmSBFb6V{e+4=?1TDbszCWtXuW=${d>uCBOCd9dx
zy@WS#o$wk~p-jA1%`COmIeJ7PWLoKPVRrJEcOM#YG;^PO7POrkKd-5LL8{q~TKm@)
zZiw>t1XSiC&16?;czb~|TDC9t>O1oM(DB_6V&p3Q5eeFwTn+Vc-IQ^v>SBTWoORiX
zx@pi%zDSIr^Mf0{ja<e&!tH8R3*P>_my#=mbQ5s8lvYe<JtRWGZ4yHy|IjFqXh6%g
zIw($EHY{b_dLmI`zlZsN^!h`tJ~oX1qqld@s%GVGC`z&d2N?`cW!D9XgX;VC8zgr_
z9vm+|SbKVqPP^Fvb_$&&N@t^GzUp95GcC;PplgpJCg!M}-`3Obs`%jDGop7-_V&8D
z-x!x!D3APJ6dxB?KJJw9PQKvX{QbsZRhE}xt|WfK{MR_-yf=db0?3K^aV*VK_QAN~
z$D@kqP4vtagTQRna!AJDqe%U_JmIxR;nk+UQ6r-y*gW&wr$<a_DUGW_yKrgUDV1s<
z<17?>7cKn1dVB9^xW2z__#jA>AkiZsOo&8CqSuiiG7{bB4AFay-X%yRe2o@8h!VX7
zVK4+iw1^<;=%V*J%CqPDThDXf?|T1v*Lv^uoVBbO%<Oaa-skN7Dc5y<%FF4X`+$g(
z1Hcl3AY93lQ2+H-Lq86=wioN#u6m|o<9~V=cuRtOUBrMdoeIzsvFxrJQd%kdDbg4?
zZffftDYP^-!=>3O>SFzCo#86H(&ERadSYhdS%vW=+ZYf@P50e2Sy}n=>C>k!v!erJ
zB<h*6@`SnPWlq*{z>9xcE04t5c}D08{+e#}Zc!vV^ZwFIm2(<5&HMEJ?7I5<6Nn|e
z#pPkFO|sn4hGN6k<2541hT*F?WXe?b=wZtljblohhtJMud^-#_)!mvoG?{z#rjdiI
z$}eZT4=qPMy#Xky7UwF0cv%-1`B-O)dX~>-y$c(pZFdHQW=9>O=gs2DSdqe`4()TZ
zn4vMg*X;GNf3)*gA|gz>+VP9$2BS;EX&F+Sv~6AwxCnvq6%t_eR~B53z&BeYznP}@
zQU!q=HRMDyyQ)lhWQP>8;WD%@EXsO-Jt`>d+*omUkKhg2Fus4!c`HKQibPD3b>muA
z^So=CpVJj~cDNBzpj=Q@bcl$cm0`~?R1XL4I?KwiA@zf^ixI?WJq?3(s#*`L@9Eb`
zv(_31p6+>*)7;=A54tHA<|hQ|!<YAq2~a-=n}6UHj=3(idzVu8X1=8yxc}_CLPSAt
zbkeNhaGjwPtyyVpgRtrWs8g4Gpr;ChKIP;XogD!QQHxhg{v;u9CU~@*no46m=Su#t
z|Dxu6TLDxJPKz=#r;Q<I;Mkz}F{>55Oh-4D!-3hmuyf>hnXb7=ukyvV^mDLSDMCjt
zB6<q#ToZcUT7P6l{s&bN0=Mi=mzKO!`jYEj_NS?6wFZgb`?gC~H7M+f1qAXmc!?ks
zsi*$~Vc|XXe5EHY%~AYJ4tO0Wc8Msm-p|IhzXoDg%kwKLTyF$zPeq>tyLuI~2DW+n
z6<~O%#|`r7my^C__b?tOg&62gwZFW@$one?QMTAP6}bNiz<<k8r6^OOH7q+F;L(VX
z4ORr#XMb@qwYcz^K@HJg+x&PkY+z)Y5yc?GwVl3Xb;HduuzW~^WTG<oAt&SxK`1`O
z)Kdh2qL4)9HzjmqORk=KD40%nJbjY&_lklka}RFfEZXyY2pPKjo|GjMfSF~#<HM68
zTg2cX!g3!NI3ph|`zxfNyO6uP|6xGz<X4C3ZG2YB+k90P)B4WW1tVKHE;?bwcNo3k
zj7z(kZt8`k>=(SpP!p8}JdvlrK_3+|BQr<;AnsE(NefdZ$k03Dgv4?Ziy8*0re3}d
zxYPuLE<xX+4pMUifF2;}=$c&-$Re`zX{&=l+26j(YIKXCF6#WF7U=S4!;oZ(-Kfe2
zU#6dllQlM8WWP^xpYM|0tD0|9#2_*q%8WCi9r*<EfTJ*-mLv*8%4VaW00sFH$IW)t
z(S*tq8_W*f-mh!Z-Nfb*drz$J1k1#!PjCNvi}Q1`-Me5xrpF?i1HVmw(5-YHp~3jZ
z{*GT0?~3%l+-^>H!rYWWR8KRtap%}h*yWLOZw1OV$BUxv9r}bcYx}m=tc2rDimAIS
zn!U%STA_66t+)JESDb_UN_1H_pKwx+n0Y!nMo#~<hnjiwB<eSxLbaa0z1uU6l3L2t
zizl4ZRe+ln{5rCTZY2N3dYUm5NEJXw&77r?pOtAGbv8Qmx@!_j@EoyT3AqNpKpPYD
zMW6?}WCr1A0_a}fbKA2`L~YqY#cq#1pOO-j?oKB4l^9_5j&C_B3f!bq&>%$F!R-Mm
zxC^w(slYH%o8l%xBgqXX4ez#JW++7N8OdrBGxs|lc+ed+SHFICPOo9b`EuHb`qs<o
z9H0Rd{EA?<F1sx}z-^p%F02Kmj(HajwOzOuQ|k7LZ_o3LZb7*0e#uxE1(6BayCw|?
z9^qBS?AlhTnr_E>#t`JE&GF{WkT1oE*bo!8@n4`vWAj|dx5tZ{VjIMA7701Ywh_O1
zi++${?aWqA=6+u#F4E@s!dp$RN(TR%_f#|f-TR|NE)F*i3FDiJq;)MWLTZd}6gq7y
z5`>_yl9dI^290KT?i883{bp04*tsGxdKGv=Ci2D2xw|qKv@QOazKO7I`j4iMAz?7U
z_;g)nuG9NyRUIWO{!%NufMOk3&cAv@<YCL2FIW<-nYoHDd=(0DdGX9RJx1o`b~I_Q
zFZ2MC%?$d%Ks>%1(}MpIe@m6m=ijt;lmxS(Ui|`vT*J!85BbF_TGeO1vW-IDrYB;{
zP1n%Lq+O^ANvpuQgZA_9?p}$+A4V<Y21BD{P1O9G243$11Q^|f=t`1jeGgZpsMOKm
z>>Z|>Myi;pLb{_n=<DH)>c%HO$8U1g8Lk%Tgr%K~Z@6nAUyc?5CtWJBi<F=bkgktJ
z##_AU5MbTM4i*iGPo#ansc}B|ssBB^q~OPL<ezd`B;3xE3>eP;&gI*P;$Z=AJdmQc
z32VU{*%)b+IaE1Q%Ep^c;SHGhe1Z6*dVJ$n`KJp`=yHp#88fzWDTvq5^tmK;#PnO;
z1m@TWFSkn}zk6b~*lcbl7F)+2K11nZ2~P{f$Y>3A1Oc6puPryV`l-`g*UR)tXI;`5
zBmdDetrluX($y`c^tPekusd_k-xImi1L^^19RdB`cAr92!IK5m7uouM{)&gpj&2E^
z25`*x-FpvDWoC+Yo4vw`zNtk{6EHKqnE{<?YHhWB(votmznre(NCJI;w5eEx?{o?c
zD=s$z2jNdU@E3_7G#V|#CgUdtfs(`6F}85Bi0ier;Y5D-z)h1fX@>88r-^W`7K8sm
zr}zqu>H*Yc*|jDri1tmd)Jq+D#weldlTO4wSBW1^&Tevooq&s15C2?tYlq1#VfYCb
zxk6~;HEp-M-TM<ym6cENHhF17$@0RMPf)-C@~kb`6OOp8g}D<Pz1N;}K%#r4*gmv}
z0Gbi}cr@zmBKmX&S$twc1NWOI0H*p8!wra7Zht<`yh7XOs&-{uo^5B86qKdR>zm`8
zce*$}5I1Qxrvx!JZ=PDNoESuprMca|?IaVzr9E7Uy&bftOno?dpXrK38hk<~n~J<q
z%$@#iGOP3b%_UU(bQsAnPgE02lw&|K;b9<9nji=y#~VE|j@x%rc$NbR3nGbEdoPuX
ziwn<Uybx6j`_fL2mq(A*GpRlz*JIuR&3EeY7%CH+?_pJX_4Ga_Wo%K4u&IMbQx|(H
z$AljtQ0cgNSMM?D@PngNv~-T`1l{=G1sQeO`LM3Y;c=FAsd#!;9orCt;L6AL0a12Z
zGi(sREYV-i8yM049C>u}ZL{~gK&F!{&4><5r1X_ex78irSkD?E9IKK+>78^Z$3N8^
zE^{cZq|bNd!Y?;;&%5?qAi$5|BErHAL)8L<?^+5nKUY-dxO|fg=zC>3*y_ccE;u`L
zV6@J*U7~4zzuWXIWP%8Ix7P_2EP4Oc?HOTSitNWh<wio+UPI}aEa#rzz<J+SyvL@{
zx_=~kqvLk(xiCFyrKFhHphm#(SgMSVwSP6?hQ1&X#AW=xkLVA5vs7q8(_u}SZFb1>
zR<+<nvM8XTCrvVy^fa9-`|A`Vbd$l0Xv#@ObwIAt3U|HvsSyx81@=QyGdm^V#;{l2
zr37J4RRU{aqw0P4OnsI-6y(KxU_IP6q0Pxf9X>~N_@Y%#y;b;dcH{c=W;)Fj4r9Ha
zkonb=RvN|aS>N%Zov(Sb<2y@hE9>iZGL0H1{)YuMdG&Q=_iIO{d#;$gg}6DKk^)dH
zi~6~@pHdc9E!)ujphM~~F>?EMCya0oB_$}u5}mv>=>26B`q$|)NDC3FIP7&)^_5=#
zHU3IcYyL?beJVMmk3O%<^eW&jSqk*5(-9s?i|Jr2$ht-POwNJ7X;I)dUp<!>LSWB~
zfH8VlSTKGT=I=?lNlmH&N{l?-{6KQHp|x1_?r%=#14}rqPv*~{^+@v%&Ci2Io$p9x
z9lJ!U=H6MlbyO&|eEJ&dlpXtH`nu2k3&0qBb$Z^Eq!X7iwv;muQcstiCdl(xwp1n>
zSCkoF@$zNqvwvS?33}?^`c3|kB{IHccysyyG(vXa5=HRSl?@|+HB+(kZw{>gYW>+k
zQw_Mrb1WX@99d^P{$^)Z$*LVV<a{E;Pex0_VN9QR+Zh!i;4<88HJs+)_MFqeJOLh0
zY~s!w4x^4dNT>TOMQ=1%X?X-pveRE|4;;TlJTSUINt&D_Ca-DB$dAePnf`cZo%l{)
z_>c-v6IC0Ifh!92apc;b@TTr3Jhf2mi#JAR@ce^&tRIP4dy{`P3wPQveebWe<HEad
zRaWLjeSyDmWu=qu%y`%o4sNzQH;QX)@NJD@<foI(+>%tguHsw%<<4VuPJR%2nsz!Y
zljp#Nb7Y2c4_wfXHBJpw{@nt|LB9!%7Sr0SGG`d^HhDZhldX9B!y>qOLeWA_wa{N7
zgZnMg-(?Tm)$9w?g}bm0(`c-{!B_j1>eP7w`$k9a?1`jRj-TXORrtFS@=!kt<OpLK
z>Myd)dTK)zlC2A+@At$-59XwR^D7wgI#-2^Dx+v=@*njIM+?04(#mj;2Dx$Ql=c*=
z&6HF_g#G0HQVG_fU9vrT`;_vfUHgf}hO9cHhS}ivAZu*ej)=1tPL!x~x#0c9)GU23
z@2%vxl2Oo$^ERfZ^7M@1Exx6)eN>rCFl;jJt^18E-JM@0HOr9DtLhIi#6j)b-ven!
zwB+~yfXJ94Eg#9x&YZ*IoyQbqwYAkpiR$k}{?gAckhM~bUzn0lq((*;GJGvBT)JA<
ze_Vr-OY+j&u({SIS?Rte&m2rm%rd06lHJI7Vt&y3^={;Fm?-uRwNh}<9=qCUKE{Ld
z@LQyKmtA(u=TeIj<28-unh>2w?#kv~Qr(iA89>@{wMpDdE-o$2Ke-@)r0Fq~tze#K
zp%<Ml<}%X>m<?DX{{v)6YHI2OnG7cZ{cxlZ5HO;ScukfcJwMMv!AEln=@AV&xtWr)
zYJmzJMSTJx9S#$sjL_|gn3OTLC0%)sk@U)XF_37slP|&%i7m$Z4)Qd~g`aNWMjC9B
z`VxXdJE~W7>LuLdn&T64jv#)Cmskpq=86Od?Ad~v;&Fh6aXL{4(zK}1)8mQy*&*vx
zXb>s@CY#qbA)@Gi?}aSD6PMhu`1Ht3asBNnh|-i&S&zq3G-QQx?@<^bW5LOnr;j7z
z?Ts&)FvGr}#9j;2G(~f=g1*~>B3u_ikhw%{(81dCQgJ~qZ466G7&QiHJN5Aaq;`LZ
zM#B<@#_bKz<|PXfJajWub)=;BOVaFRT?r2pw6%P0*?%<;Nn*qwh1RP$evW0Rx%hKO
z>!aF;L-N9LF{R^4ndh4~tEL~KGA{7psEyvslR&qV2ZJ{hc^P9Iss;2ls=k(Oo`haj
z4I@3`qrVh(bEA29`=ol@V0pDy^n<-o*{<;1yixK4iy8Z1af3G2oPQ+bK2<Eldlmq+
z%-)xc?nrK`ox4{a1B9;Yi{~J|SQTKRe!GV>h*<RTUf~IpRHq5BJ;wRDR`jI}6fe5z
z`UmdC&l1<?TXt2BE#iE}*WWeJ1wCjS0MZ?>Sx&|<Feq)rPNF54^|~0&ZDRe%d;$vz
ze(%m5+f-4fN%@xhN?L&T;P>h<>r3hJdY2(cPLm<ih^9bdJKR^&dUI1QHPgsr@{D<4
zWTX*T3v7uWk`D#3GyD+}K~q8O>beQ%iJ3(e%Jh-d{9m_PcK22oq2Twlh{u>|dy3bx
zkb<DRYfT^OIJd#(O%~Uo;o0NO$kn(+t3k-%=kcJGhQC}?;N0dh-FIKA7N;Bg@8d3A
z`RnYGuX?GG0)eSFU+cg^;Xd?nWK4ALTW6(3blc^w^Rxdhr5pwMdGX%nvH#*g0d<o}
zke0p*xew$jZ`2u<Btd?V>jFr@-%KXl^Z;=F9@>0WAmgYvWHvoX#%hSZa(RuMwwG8;
zncDfKl#dSbipEt+`a=nM30)JVnUF&J^Tups^_v$H1nd*E^8!*Onz6y36K@gSejuMW
zILCLd26#B?I10wCnsAp6MT7IL{Tp3YVW*xl5BPhBhRFm)>7OG6da+KxzuW;!)m_1%
zMR+P$B&kuAYC?Je{$NceJ`vZ@12KK$(~2fZtKL*oFauA3lscSFCZ0nq3yBQ$S?&e>
z%7y~S0Z4C>wyhD*muSjt2YDIj>#qw@g(lXjC+h2nCMuh<X*5cT*5qVgXSqe%1<eu5
zwmXHXoMGbyMd;^VCC1$r0bcR8Bk&(nl4p}O`2Nzj{PX#8;7Lbvs)e|LTf+GO0g~dc
zGT;33T9;f$S0zYv91vSyp6AS*iVUq=F1qHx(jZBIiCY%+omW{b6%zZ&1=|rNkv2OJ
z+xPw{bKH6NtzC1i<4I^=)Hnm4^!8e{sq+nT;oTZK8U=az>c66-lj3d*t(b{*aUfUx
zZQ7ZQbQ30_%a$bobQs@}{d18}X3oj|_Y<Tj8P-a%aHH`@d|KM+1~ks;I=j%5a;>FP
zH8hcQg<NgT<0c#O3-ix<{$xMtx!3~dWl#^!1DFL3A1Y7SffwOlEFoI6!mtVgpykP_
zx31COhJqXyIwsQRBu?OvT7zp2c~h8LiI5I-=lILLosELGcZIiH82oAncnh^kwH3z`
zy1#wP?59=?jpscp&9_}-_lmeppOu9z(JPZGqQipwFR*o&tsam+M8HJk8YJx9f}YXx
zJuO^j1e#C91=oAC=R7c2XzJX@`9}d*mR3ds{-2yScMy;KbM@i{4$j0dx2#D*YPzIh
zfAnU7{zVfvklz;LdMF2Kf{H!?RccoPc(X=T7h^OG3?#EF!wyc}hWm^qwD3j_HBBgA
z;tRQXpU!p-7Euxu*LN%NhcR;-4cous)Ixc8-7u4Z2gw6h!+h$Tlz)GocDPpH;|=1o
z@{%pId*wT04$&*i;;FZ@mY1DZxAmd}oeoelaC_Uj#%ylhtxHH8DCmJwVMF_g+ZqUn
z;9C?o5@qTZdq^3-bk%x=w~4zfjxe1qX<WN8dtTY_(EnQ5m%Tjx!hX$mZt(yeAo}Ml
z<f?wo!+8~{8er}0@wIRI79NMeY@SNPcl3kHGe$q(nY|J&vr`x;QzMfec#ZS1ye+Ux
zYjcU|M(!|D)<1NbtfzcFV7Rp9<w!n7adw_abW+nb@53)f`e^Zb1akrFaHw104D*1H
zCX%7xScPfPHYI;DJ}GJGZ_$)b!M}qF1ftJ@0vi^G6^g5u{^u5Fw0yqP+aM2tJ47a7
z5%z0EYc;)CdM9@HTb@Z@l2Ueu!`4R1Sj87@t$iP?ZrwVkZK9Es2>8miic)<e%9f%a
zR(ySr8fnzmew7-QCl!vn%_%w>|N6GL%nceN2@-rVYkBcHH0dNo^o+sm>Q$Hd{C+VW
zn?pXRa+F=rYBh;D_C7uKWDz%g>Cj*$FOLeX?Akgl$z=G}4ISnbF<E?+Xx4byD(zT^
zXxabXGvb10Y)`PYCTM&2A8v=D3UevbY#D-%S1MHDSCQ=OO(>ed88g+ZhPfnGJhx^G
z`e*OEJpHvu<kzSwH-PwWj6F+=dN)vy3+VPaY7CYVF*BTS9gqd#PXo@szkU0*P7O~$
zd|Z$%jf^Z{WgoX+E_&ZSyCu$6>}mG+amXc(PsC16$LkIyfaQd;(HnB5(_`p8Rr)<S
zg6KS4`n?3e;%$w&Lw9@IZ(VMCYUL-YNZfhzi}>LEJi%2nkcI7#0kb^y{gPGq_}L6y
z_GVMglGj40XIb_chg2}g`SjC;zf;eS@%?W8sTEVVbn78~6FmCH<yx0mT)2PFdoj1C
z8uD&*RW1?1c2A;PgtOa6u7eUb_Ow??CkeH*we}t$Uq`HJ4wfFBo@Igl_=_mo;Xlay
zH`oS=`ZA`yy*EjaXR@N`3RG(3vy>B^kl+F52~^ErPT}0vm6ynippBU>sJneQi-#mq
ziRZPr(zb3gyc~Px+IsWWWV~C)3JaYQ{y|A=omb1d?$CI$SaE2S-^h;@TD*}LY)%|h
z2+D<`uEt>hq=x4!s<_>n>UGPVdS->PUbT}FkvI8!cvCKgjWd#niXkSbK!$G3@LmGW
zzPu=<MSLMN_iA67IBY5VqC!@%1w~L-=$J6%RU3G|o%OHNnAKER@C36)Mkl`NTARGq
zeF5M@23d;}k;-7IY?2+=yJ}uZNXt3l*L%u=uv0knOCBZ-xV99+PG14ArkeQg?v=o|
z(jaNENCYtH>q;PG@EF#YsGf?w-ZX1eqXC6POTG65lv759hcwiF0Lo-Q%eToIw)H+P
zjvfOCu_;0D<sIanEuh-L!RNwWdJCu*f!eqjiY7JWp2D%PlGVev!<HluAO7B_X*?S>
ziY&=E_J9fYH>v699K1X!Y{GC8Le`T1rU$QrxdW?WJNPlCjj@0lL3QWD(*RKJ@Or_q
z2?uRHIK&uj1O2Ny^a16S2yk5?D|ZoW34qwsr>eZr5#w1_#x;m*%a>60;_!8V^tM-4
zO<MT`w2)we?*fH_P}Ganbg#1(2BqzO<vma52~nVUom!{+MmCd^wzE{PY*){uR{9YE
z`U64RQbGSL)M$517Dqbz<h@L@CSVIX-p1>lR{@6Z8o0_0RY0S@5M%=dP)<Fh^Bx7e
zSI>_h$7X;%xL-AFEcpr3E8i-+5@I8LB=Pyb<4a7Pk)E5W*IH_uf%O^f`qF*u{pJ0u
z3QvsF=RZ5usmrs|cH}9>-X#t?`}W{u*MKAI`YzD!mZw+7UN9jtYT&n-uz=vRD@@>u
z-X?rtZXPps9u2}0eFd6<j`$zJB;>yo<~->1iVAev+uOM$B{3J8ocv{QLqJq-<Fo(n
zAg7(R|Lc8@ASb|a6$r9e^R%?G;`zLM9iqK>A^pt@oU}|e!1AGciF-2O=YCxk9wcHK
z+S-i2#`EPH%H0O<um_bI>ut9>Q3;naO-VgdxLGHua?+<QU|2i6{*C~xfYf3uSK~oA
zy7bXraIe87NU92Wbc(?_{cj1Jn^41U2Cp&Sxfn%1sDMZd?Edo5@Y@bhd-rD0Yry-Y
zXj*-02_Ret0Kek|=_bd=@2LQwm_Goq^z1D5u{-H$KY)0G7DXIJch^6+gR3u98^q1V
zeP%z-0{EpdmadA?C+e<Wh!h4RDV-)mhtwYsuIfRd<#Yx4z%UaVeLp$$?yOY>86B~;
z(;u*?Dh||>BO<bRwR5fi0xGMXCg25xfu|%1P$+mgUGD}(2<QWqSoORD9ipI!Id7?S
z?wYaU1Pgy?>LK*M`Im3_b0q+`zM&hKHsFvh^Y8y<MSOOMI6U89m+2^}Y@|6nsKwZI
zI=QZPT>h3zPyB8T#&6<z+eMO&+BD0j^dIVH7avg#h}k~YkWfw%inCEve)dr%Htx$$
z{Y#%bjFsw&E!KW8rhhFdYl$Rz*p>^U;%#ldl<A2yEBnWMZG8Z~@90-$UFA7HA>BCI
zXSL8RvzWe^zB4hHlwFC%ue#~K^i?`t6y9S};+4#}v-g(V(Pu~C9d?>pP}%ANGz2`3
zi<52RZhg|{0-fP?T?J#<!1Lz*{{Fj4yl@?*+RsCEw>tA;PY8vd9ao}+jCcG?E%_m6
z>bJ~Rm?)d_!x1A0f|5l6`xu!w@Oi?*X~5eg=cPoN?_%E;5b{R#yWh}7fX%#gtu_qz
z3;bdOL>l|okNm$Z@s<4oMUN$YN~T3c)5wFG1@CJD03bM=ACejqUWdLP;vtC66CwXd
zD~KqkiIVWg`yc7OhzM`U;)n$ZuT4Q3Q>279%xV<%gja|}DVT}yY9h<KNlbV{b^AXJ
z{(ovDX9-G8Yuv9YH9{$QYYvG2Me)whz~ITSsv!X$@t^jbx_UWrYlxDkX#JHnM7q5Q
zT~VsW4@$KCN=ZQO(D57You)fp&v@rUV1iXc1}T?G?tUJ$ZTK>De2UPplm0Nb)5HBj
zT)%iI6hL$v8<yCbjzz}Zp<b1oxWxpGGsgjO-2}IVYdLW}h!ea^b*)#Uy3)}ZY=GJ_
zm;tF&`*BDDdwintqgHt#7+IEyubRIRTd7>UhfW{7`;52GhL5UC1b`CX=L$j)3;{xu
zCCly-RhU%=37VoWDtT@jC*lDBEqOP&m*g&}io9GUJ-lf^_@#JYNorIY8!eJXs;v+5
zw2FPgCkwed(0#A&=Knhb|35N&|F1^kb4i>%z!;P&ZO<w%n&|N>Jsfoiyw!OM26S*U
z6p<baCbl6b&Ph0xT>BRTBH-iLnIF1O=mV*COB8G|QNz-<nB7r=s;I0GU0**IaKAG{
z=)>!IUxIdSL)8-Pe@&lq)ew_#TIkuJe&4<MQkk7vjyHku-aA!8I}3eV8b45~+gh>S
zCa;a8aQRinhDt8fi}r%QPW5MQysqq9IC@gNSCQ`PO4h=Zd>MlNfh8p;*A)yb{a&xA
zS(5tqNom+wJe+O`FSQ0!`3Z3AhxZLzoNWFxuRGQUCo-e%jQ>7f*TI~or1X+BO+6t0
z|M7@~rCf%j?0&-(F>86rRK$gIUUiFJ#Z@Ayq}p1=FS#dnFmbL~(_&?;Q5NfL@F90v
zUD5&X;~N&a(|m-7?tkED^3ec;Atm`PD(cLhcfJAf%eJayqQ(iwrEC>ql{MiQc)K%2
zwS4Kb#t-&keI6y=8kOhI{Tzxbo0R~p1VbpKB$}X6J*<yPnv;z6%Zb~D+ZJO5#l?@x
zVO7%~e%2_I;CRp{3cT|%0B@-pr?H-%ntI}wI!2X0FrJtCY3TSfC~k|wlEpyGCdIiV
zP|sL{uV-Q0b}x(z@dWHafdxy?%ZS4Ur1O~g7KimqmA3`#iUY+ZB<iYaBR0FgjIE^+
zq_TX_V|K-*s;=JX_XuM{yVFh>4fvO8R8*Alq_vuEx~37f@7Z>i{>uBDI0LjFC~g$!
z2#Vhd*xl!mx*9o=m-;16BkbC<i%m_N8KnOZG7mzX3$VZsb9FEbfE4UU)3Av-*N>QQ
zcE@W>N<4VrGInzO=g*&?plnl#b(Q|cy`1{;IM=?RVgmsQug$4)Zntl)@4=X+z~G6n
zoH$fQQqs{N7y*InV$M^a2{jqu`+v;(9IEuQbK(xZ=oiy)A5B|cZQ17g8jwi%x2m$T
zLwb7pS+M?!Z>*t_oqzs>V+{`v&w*$XL+|fzczQ_eDn-4lZ86+PsS_CISa21f|2v>2
zMPcmtpu31yu5EE;!^-UJtRUjX1W@wtr4*=l_C@dTf1L`kdw4)ZPW8$Ib6PWSZ=L(7
z8PF^7J~bPFa4L^hjIGB`IG$S;zg*O}dSzwXyod4VTPfu}+5|lY;tK8OX2p9v=y8|X
z_o9v?fb=?TiT;*k;FUoPS$(6YF6#x+g){>tZ#y*T@JJ|PABHI2OH*D58T0@J;Fn+-
z%IM6@Ol9i!;Q8B#07-;xF$Q#OL(*swP!64xRa7SHxI0bW0|<W~0VACUoZ`-o-Q8Wu
z<lYO=PkK^g-9=Ed{kSCw!#M5%6!2z;L3y!E81A|2>HgXnws6F~%m-BN4MQ6a7@a<j
z)iV(Xz>=+Rus&7v^z`(BFdI4Rz-i<=&zCyqrc$!3kZvBmmucGFoRdnJO_hEPcjx*9
z#wzwi-*e|$$`8`Q{BidIWdm@=I-bs>oH%lDjsx9+%#Wynfq~(dDS>{AX`TTM{r7*C
z>93qlEF1yu!q@!%`w&D8)`lXpiM#?-;#`OLV(QM)(J-hW=E%;<>Z4P<w<xHyTC8mG
ziow|zsKJ`lOej4B0^x7sG#ZoUtVHeqY@Ug=vvB0ZZ7hkv2d!zn7(W1IeAXN3`1l2&
z31@&l{2BUaz&B&t!M0fPU;}sVYsYdisM+iSIA>0rC=VF((+RlSbM(0X!32sgM=D`X
zJUcskn6Hk6nT>10B7d=pGWH9gz*giWJ$3%hzZ$S7%RAJdQ$NGNvoKaQM6GNYY-M6z
zo)b6Y=>a8}gLXB+Tsr~^@ZSvX&YwSjI#7xX>#qph@&?<t1LCYXnfw)lhLZP6YpeW*
zxG|Ukk?Zh)dLz3A2W#uU=i|;T)B?3DC78zFx=(e}owyOa^XS}A*?}8=o6RvtboDuL
zv*4DnkZkj(fcR2WNleqgnSNFw_*J$O!J~@Lem$o6<jHqTPfCsBM4gLrOEX15RWHAv
z4RBk)?fVATJ1*xP2DW`CAzJF*r@Zq!z#3?9{M_hV4cH;0Qn(u|EJwDndQ;C+XJkmB
zgr1AWx_Q6`)3j4YFM`Fw>Drbn82=-hAbo<6lQQ~ERX4evDR4WuinZ?|jH`fdVGIEI
zr0On9|NOGFv=k2LwCK5I2O!uY;XKtO3T{tja03PA`NG}idL=jfvX+1xo$~z<rlCv^
zjK6X5-aTo7G~3=%Y;xfje3`IVq4e3|?1G~?$8L9QdQ41AhNA?HkcGXyeg4q#`uZ|9
zn%t5YT8X2pFS{JC;rKVo<LDa@VnOQ7F=dsPmq!9k$NfE6Y5h`eO7*x)`}MQkm`bI>
z!Q)KFoj(AuAM>O|_S_a-1FJ0^I*uHxal!_^crjxZ`En902Hyb){xm}ThIop5#;wgl
zgz*P+OE0fOoJiUw=p!!$IOCu`E(xq$foTKpJXqM-0|~Xl&x?nSk3r8xZyn87{kT}T
zBwXs<bF`T1)>{qRP(^F}5cu99K-SUP+IkM^KCnOCuGU{+2SXbO=<2)d1dtYd4Cs};
z0lJSh0E#@lAp!VfxM$nX5NG$_5DSnvfWH6j<jJTv2iI~8R^bWsT-$_l=@?;=;jSct
zggy&hhjb0Vh+|%WkK9F6xXga11pZitl+ih`sdW-GWYW!TU0t^;ZihVJ04d2egolTY
z@8Zz)r8OP9h6B*}_`j==FeUljY;b8t2$_Edo@0<oM}+5Davf9@P#orGO|1nRm)z0*
z&Iu*LLq$a;z_z%eSxLOd|3}TpSC`pZ!m?VJK@s%G0z4OjTwGj|m;u3);heV^gzLo1
zVQDYG(l6y10h9og2-pmN7KFcw7p*Wk?E(u3uy#g)4{*BGOd9$bv^^;yBC<aX3hQA@
zhU%m{_YnMu?ReNhyheZ>m@t#nS*ntwm4i3kX~Q&$8Zt3+CE#XPoEcwOE!zb5@&AqE
zSPgBjyzPs~MN7q)b9P3<SeEx|c{(*GWGEYoFl(P-iJmCmM`%x;BOD@%`zTu>>NEs)
zey1}O5e4(E_bsZqRr^;cDX?hqgIwNHB8+`yAODkcOf~-e2MmjzQ^nNEB+acMdQmgF
z|MtetQVxBJSmx}KLx^zKmSWCus^!s%2fKmB=$h0QaFqULx;ENC#ce#fHe_Xmsk5On
z52ryoNsc*nM498ioxoj^pR2--(r-nP6WS|VRzgS5?J5i<&mF}2Nu4ry;Tk$*&fJ@Q
z11h6udj-t$#1?G99|siE^U(HsGavrFh{9#0c&+u(fuQKU$B2jMIp5fmM=Jb|{K%=|
zq?BJ%J&49^OzG>hp^&o;C0Cn;CwrH9yj{?ezSZoiuG1m@McTnN+)2Hi@JKhbVanWe
zx(xgOY1uYF%Td|2htpplA(K4M;Dst{b=g>18Z-W+ir8L}KAo&K!P_!=VVz<ExDYa9
z0~_AY*#6^gAyYH*DC4mhd0gVBbK(sFf%4KFpW{iK;`by+Tp$0RYJJklwi~S?8+GZ8
z@J4wQ@X(U<mBA=5)Ap^|oDW^`$9|YI@LxaBBAsfB!e=D>^@Vks*Z3<BEf_nMb8gm$
z`-}>pdN{zK-5_{&yH;+h`2JhRL$Hh$H*8>I>;5y}QPiqgV8)V7#+X%@)l6T)ADNUn
z)w;C}<9T3|maX3=d~H#S1}>R4YK2XjC)4}3O0jAteh_i$mvMRqJ`-d?a3}xvX{q8%
z_xjI$b*8@xtY)8smKA+R8x)3e)`<i*X2o_Aj`hBv>%Wej{d%|${G;7Rz~}r359}WZ
z7y$i{bfb|nxYVF2HeN4}14bWE_(I1j?#N6IG|ma(4|`z-$0~^@f3A#`F4+WH_W$ed
zLWQfEN!KnJ#2Pj{&Co{UHkLdK!t)#It?&%~YzNuWr#0!hIK0pCX?KW&&VfC?Q(+Q?
zK1oauh?JHs77bK(o##9ai}kC^!|kZXUMJipCdgB@bbjp{edA9jk7DYIVFCzOCl&sH
z68YeRV^X}IFM8vz<K3hb?rBe-YIwZ)K!`sTPtv+$24eer_xr;%oQDNEnompxOFB7=
z2oO>|=L<vop;Y)IaAWNUmk4JlvsFUlU0q#vw7L{y{VK+8z}o^645pOPxV71@J6$gZ
zxa$j}52ARE?2iRD&y?8lsVdv`GFVl@cY~0X4?3mw>OUOS!WOx`%P?onnya(VDeC1A
z0c%3-X#$_p9RsFfQl%zZzRjn1zh6k}n^X-x@J5{V+w32`hXrQ$ZO{=KG8v4Am8!TZ
zqIuA4J>$D%l7c!M_a=sI(HZbIeBY-g$qfa>Wb9mKH-me$xue$BSghfAJKg}{*lqaW
zpmQC+@w|dd6&~a%PC9vXBA~n-+%5giQNqn_<17}&w11lRZ@58<(l9BH@}sCws)evr
zEU_@?dx-uspucOv5*B;1n7*ef5=R0-y5xlLM8e`AVlruN8MbKQC;_(N5ST%%6nc(f
zq7jc~Gsc5ZBREojM5UfRvTC4_!qjKJE9{}ChM-M8(+dJn(+@lrQG(cBR3`uT+*3Bi
zut3313|JpJXAzcYKjBU<^zf~bpSpEbgwOxaAF}6XaGp+QlHy;L_Xw>c6jbC(9-D;x
EFJB?s?EnA(

diff --git a/reference/plot-19.png b/reference/plot-19.png
deleted file mode 100644
index 9a56809598eb815bf8d8ca8a53263dd75c85d3ad..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 38148
zcmeEv2UJwqwyn0^+UCZD0z!*f6dDx}5ip>%BBBBUk`z>wC?J^vx@{FfKm|pcq@p5O
zK*_;C5D+mDi=ZGuQh|g5ih@5^Vc+ie?tkgO`~DdBkM9`W1y!fc*=L`<*P3gtxfU1q
z?^B*RO<>xD2@_`S{#ohZgbCj{PM9#s==;fdhk5;WH(n+isVFPq75|2IWVqvH>gk_#
zY$r@ue1ZO(=-1S9e8PkuC+t?*rr{XeQR@_-cl@SecSO`Xm#eStFZ<!hw;BOEf0(iA
z+o`z^1q+WZtlHA?BI;M&tsfj_EUZ!9wQJ|M{0sQE&Y84#>w>cjuc!VvY0|{wDtj+g
zZQd>M`Cc1a`Q>A|WuNyQ)GteE@0DHIQq$(l(k*p~ZtE;cvC5M1FKy|aNMqaDc1%T|
zewZ*pSNX|wdcAz`8v}av2$<9T4gGL8z=NM&f4ui?CcRErz5K&>^!oGai4yd>b<b2s
z-q{ZB`A?i;YlC{&u!e@F=1xsb%`JVecia;<*=lnMC-B%Vpwae?$7py`(uQg#bE&9k
zSZQGD@4pmE(!@?HT=e&sznnTc+`crdR9PX+@8aB**VBb)K+Zq@+XIxF@X034u_K|N
zA+cE8d(x~;Kc1g0B3xEhCTaTqL8U`IqcK2z-$fRK<<isOEwM`>jHMZ7RuSj3Xv2xz
zn<k-)tlT-w@Z@CaU6&T}DGNIta~&!+^p&Y|4A<fcS?BT{(A1od`|4>?9Fr9l-Mo3T
zQU0Yx*5x`Td9UPHLXRImPH2zI$Unz#u&`8dY-AVbrlPx>o>SAcgt1Xh!H1`3jo&{w
z&~wGlb=irM-{+_4#W_qdJv(||HE4mdkhRjKg_5GCW@e3xS&quVwy$J+UhkN_Q(3uS
z0XuEoaIy99`v`?=@9E9{^ieGXr{&oLkK^+`75W=<I#bjP+@}bg-m2&L@tmY_{w3{f
zSEqOmb52CC&7On3v)#JSuB&E0d_GHMAOs)pVq%hR$XFTB(5zo9xuCRT$By?E6$PO#
zU5iXYold;`T}te*uI_Zn040SmwYI3JD6i|+uM4wJnVPmGt69zelJ%P3mJg5O_1P_v
z17C_IS{$lRMYt;UwbhrpkM;_4f+dZ2W#7MlzaV>`eE)k_b@g{AIKTgNbo<*;OmbmP
zPrl$#f^mUmGVkyqJ@G)S*rDX)<dM%2VcAs!bSrNCZDm=T4W{#pzswoqvRKyUavcwM
z(Dy%UhMjwD==&~@-!S-*d-u~RE-@}ofBkN#r?G%N++R^wXK=Z(tgO*bF|=<iIVCH$
zg2Uyq+|>q(Pb3ZDHZ0i=XU=@G%Ha--%#!PP-I&_e-rl?6#LL#Q#)0OAEjDSETp1Y|
zb|vd~DYNv>cmCUR_=k<eX|Fgp&#0THn{2Mrk;Sg-8E%i~k13UK{Wx9H>B}dBkSmVu
zu_^f(b^UQC&CNSB>t`zr9?9$tVO{!e`U3OhTDuB;e#OD3v+rQ}`^pbg4t|{No?11a
zrZ99&$o8eAyuv8Ujh#_xU9i}=t0ql8SwHpk(aaWk*2N1VZm9*y=MJ?szSs3r7(RJv
zv7EH9OGm=5{b^FOHhu6|xcSUUr5HJX$I+3YtZG(-BKK4xWB7A~vZ|_T3Wt@=>Dn~(
z$=~?nvvcwTW$)k5^Ixgvt81BdeEJ%RkiXoU@{=H=cM)oS&CSi8o(tC>`?<!RX_0W`
z>8~fv%)Ir)`P1~o4<A0<fDNKc+q$Pv&24>KTbp!MeMePNW3^elswE36%$k+O&X3ZJ
z77Z-ksjI7-+5K#~_1oPmms=RPbNe?9w}!j)+qJ8?55Id;QzJBG@tOTw`roV5&vbrW
z-Nuw0t_9uUDVL|;@#m(sgbL2y(!G@yr=MG2xOV!PQ)le$cbS-&WMUCnW7RG1Y<YQe
zE~fF*3K3ccl5Qh|$8P@m@%e><5}A4-?*;3;e0_aiG}*ktz1udwyohHfS7r6~?rui%
zqbK(RzHI6*7mLrFGk3+_)`6r_w}GUdXVWE=R8&;3Nm@HPE=ie{=c-6#%}c3sHWP9i
z{Ky<h6;wZPfG<$rUdM)WKghDpI?Po;1aqfmX?W(%&LDG6mqU)6WmbQ=V}F@e_~XY?
zHfefi&Ya2p(o}3b&{368s=!%hf&X=AC!1Ft^1ZuhxFKtQYufi%VwK<AQ#f}cmW_40
zZYaGsY;2^(qAfZ$cAX;Z*Ilexrd5*R^f6q3+1Xysu6G~X#kA9R{!#P`zy6s6i7`*k
zUtg>9<$k=i-=YmGa$JYL_#~G$*$f%`%C0!&@=CVFJh67$l*Iv$I8UZ1JgiEvv$t<n
z+w}3byIkz6w8G?bZ$*6773w)nlLo3x!tT~~e|+BS4ilK{`ue**Jmcm4u<T9D>S}2w
z?Zn;HIsLI#wXF$mt+y|H!QyOP5q6<}vz%OSM5wdr+j{<-!Mu`WiKaO-8T|&UKIwgn
zJ2YQp(tW?Ott{HV6VJMn`Mv@>RahlZT-CA^W(NjPrfM{^_my|JpTg!eZk~i&LP5bo
zdFL~(y+QSt_O3{J!>RvrJtY^GqOQ}#q0Y%<hKPb|c;|+jd-ZHfqzXGTSlKS&od&Kw
zd5$AJ1xrQmBqpx!${8Ci@m&{GlJDm}()dY?BUA5c`&8o2W7&I~_ht3J+)$buHrf~Q
zY2f*6MWKw2#A50Gv{-wru21IgV3vdfeFib*LFO=SVfC#E3LOuhH=Gq2`9QauSY__o
zB3HlLa*XXh#=ah4pz+t6ooMB?*uL^;;ebPSZ-_<u8g6GweW-aD&MoQO*DB0T>q>9G
z+L>nCLYoG&`Yvqr%nX5zCzCAF-F)0DA0F))W#_mjb!MF3U3`6=H<q%@<Fw9X=GD$p
z*X9j{&apx2y1Ifbp)Qs-9Ne3+A7|d4NMCg;?qGr9zGS~m4W5=N^8(&B1saB$yK~zO
z>yKqTHmk_@k*T{;gFWM)GL$f7ws7!oHQxLZzWBt_HjHsQzHa)tU~fDuk9A8ws3u^0
ztgp)Md#hqxq3_`H`CUXorc<x;3|*N=qcPc@o@0Z4W36QhFiFZOOnLoz<~6C-@&iNV
zDcMq0!isLLekN*`jsr(sTsT41WI77%E0TO&Y(D%lWd&0rvwM3yn|X-$7+x&(TgGe&
zai}}QRzF#=*g-FmQO{uKdCeD-uI{SH8MULWuat`gJdbISQhS1<pJ8wDGE`npJge{R
zrcal57-e)Od&4L_h>B_(V8y$0%@WlTDlEEUvz;a8+3;+5@+A+Myn!0q(i`$F_Ao2T
zx#`ml_1%V=da~JB5&{AOMuXiSt?Qgx^05BWt2RjY4!Ld_8|f|VtaGbXzyj*ZV1?GW
zbv~|;W@gKnOh1t{=DgciKq^YhuJ2Q1MChL6+3-^j;^X~>hld};#`;uOS3ii0+c#=A
zvx9u$v7FHSLqF5)UMX$KSXbKp${{=Hq+hLndV0Dge?SKC18%n1aMUJI-mdMv>Hu70
zqf0%<V8-Ij8-2HYc`?{5le5TAen5MzPU140{7zO@7BeRfKP6jlWQ5bB`+fY;t^A;x
zw%dP2UgGiM2JFP0<YsDr&ABF~MUQ_W44&terONs8?3$!E3eqE+`vFi~xPw+s#kyBJ
z;RZ^QJ3p^|XJL^e3EM2IyTM4myBvnPq@?6cb+tfjZ0wuz^6Mk*0+mfRX~%Mm;-2by
z!9@vXA38r{(ZNRrT|V9t*_*0H-v(y)-RA_Z)wvo|stdb0cjdmCV}<}Q$6ZGnMOZ0E
zJFmC3v$rp47R7=)QasghiU%GcjM7}A@)(^rb#=ni7p(PkVP(m@+SGXI!Gi~$o}QY~
zhw^#~L`25=4afTLb7f|H?=sCa<1k!vqP~_^%C)7fa(Ocpi=Un^NNUe}JbV^*QzEUh
zhU|QAODW5Kj6BdW4w)HT%V(hXt%^k6;|hbI8U-w)2KY?x3PX2!Tz_szNXV7qCScW-
zV*{{&3JVXg`X6;!NY+hfZssRkth?`t+>>KTRh36CEN*BG_wx<1%9bTpX=cbdY%{hz
zP;ZWpk5BDD;j#6#*AF{3UDI;z*9P`uzMUo4`RJYWAw$QTd+#$3Kb6QHG)mbYmok=a
zus<&N`1+fc{F=<4wyt|@E-e_wscE^&l&L+Vo)}WvVm#dUxv;ahRMFj^JJPDGxF?zX
zpJU$}<;Dl$zFvzRY9s+_mIjq%ul|Y4tRFq(xx`$<O8jv7JDuh{Hwp8SaZ4;aku^a}
zM`savvW}E`x5Cm;x8ct+kL3nGob~Tijnz-TJyRliqkTtZY)qonxETwUwcVp#_~fUf
z&we|`X%4ayKe}A7!ftV$6{B6Za5xoS&3vsv<|a{57`+QJ`pX_h$Hn_csz(es4zw3t
z6UzuG$&OurbMO22@0X>x*xv8M8mTl6I+)GxEH^%zyPoK-7^rh$9djwXy4)&#?5_X5
zJ;_$-pYI!-)1*a{C~sMldh+DS%KSB){6!|gSIpHyHcOSq>FYi-eakWy;b<pvqYb%H
zOr09;Q@Bf!#(W=dPrLS)afY_+jC!u++xro(liP3J1CEzhXzs5t%z4~od9Z?QUNN0%
zv8N)*U*?`-CcpF7O$7TFNd}WYyZfBMcM0p(t?R+&T%VSeaBp6|pNPm(W#7G)YN1=4
zlK2DD`^$A3VITHcI?RNzV+>TS)H!tI^V@qpg!L-;RdpX8+2%Ts6xLb%<shJ)lUmB<
zuI}fv{mxD2zy9!!OweXK(?o86JQiM@z96Cw<1a~sdV*?m+6U9cdml;?rqpBfm+9(!
zzP}jTxp2Tlp%?4qXXa}HJM20aD?i3Sp3CjmJb3=___vv>?P06(9~=m8$epWpX^ODp
zkE8G?%J`=gqnl(~7Nyw(yPJz8FDRxJcgfT9_uqf7bRXl^zt=i`{CJ9;vWWXgk7Gqf
z-Tc{#Bd2ZNJO&nw&mJ}ncIis1&7Q1?r(t<i!`;8|m{Ie~n^#KWT+$S~pG=ZG@$yFM
z$Mb>>Wm<-v3y!5cew_F0+;n3cf-p)s*H^A<q2f@X+8#~koPfe(>FXL39D5onPep_n
zRX;k_(-s}SLPb<5X8M?cOXuU%-dA#s&qYR0e;MvDqp0kQYdu0aNyc#BvL8b>SsSG&
zo-luTab@TYbNBveLm!~EjdhNVJ^+1ud}Kh#%LLn%>qwq@dsku6^TL&3!RCsbjwDIz
zhFr&<Jbr)cM=2=<*iRxWL(UweBeU9){^A4!R$c=(?d40z8$;msxXMie7-3zT)yo@_
zL&uA+TO{U8ox5V0MVM3bP2qz64`*ku=q%mR`%;+cU~g|99g&vi<Zlt?`lZnI`0=I%
zB1vz;mD$<_$!!t^pAm%atKNeB7S!46KHB1*JgA+VlET(DyB&<>BN%%Wcu~HK`R2`)
zdb9W}XZbh`QZc7M*kxIYA=ZKFK-Tht!AI}*wcE{v!D6H^gVbNJw;Pu1gkjps+H_Xt
zUQ~2+BXEhkaZ$kPEK^1E%7>2dyv|$R4Ul~rfoEtfOAlWe>S**bOi>YW@0)0WJr;1`
z38z0kyRpT|CY~*escM98YkMB`w%v27uV>C^e|!QWi|gqDKIuUC^?gZ*5iR*IG6TH4
zyar3G+~(r3Y57^?e7P2%J5#3Shilia>1wt#k9?S-*xDm+pryr6fln|dEUPQ!GV5IS
z_;#yX1d5>2nV>`s0;YsBpWmk9It`;7mLFjxL90wYin2Lf83mmBu~8bIkaO#8$&>q8
zp<Q$LZCXO1)eR}LLob76H(;^NCEo$px=_YaZ|{wtmXHgsbZQCVIR18yeLW>}c^5(q
z?MEjMjC={qxl}*WEYpK+l8dDp4%moT&LI!i7V%%9>P4>f_3a%h`ZhTO1kc^ZIQ0eC
zCu4aO^WNBCvhQV(j0AR%QS0m5JRX4dl*sVw3(2ig+TXLBN?nZlTHnw*|71dFJ^pMZ
zMJWV42eIiFr_|V}Y#FFDPB{7I&O!xON1-#Xw^;*N=PN4?zgf89#9qUhOV)^qHO6Ih
zc}0Y|ZMK9_bR7J6q4AoQ!NuD<F693<b#6nYv7gUewJl`-!y_Y4c@K+!a=9JP^R2f`
zZn`RjN!U@~4jeA4vXS@0ze@m_{{m<!zW(M*=`r#OD)IqWt!I$eoA!5nQqd<srAUGv
z<i=&s{v%l>-bw#$V#;)X<n$xOQqt19TwGiti+@ItLXQgA|K7djcp`iB4)nl7FO)P=
zQc}WJS2i{_z9_aP4)AB!3jd3+XvKjN5_YmX%U-{xrOL;bfY1r%V|m;I$H}vWg`}jU
z_}+{)hi4yBSD$-!%4}axPp}&&Pn}wAy)#c8E_CM1nJ_hgMP}tO^d6bRPcN?Ey*F?8
zU{}}TJt?DH&le|>fcuQA5|7^#k9cr?)?xv;AzaQV&kM089i?t)D1s%9_e{&85POT!
zd;0o|cbuPb(bE&K?4+foIK2lb_58wYKTl6`Pl%^f^bT122uQ5~FC`~u4Ai_l?tbd|
znSu+>PLB1Sa$Q{iaD%Cr*^A3d4&ZEjG^k66Cgx~qY{BIY;HrFl82rWZ@^X65W`^u8
z_m`68W5U$}ul{Ym?f(j=_#Zs||CpAXd@bE0_8R8Bspwh{LA`fsii>kNwX8d7X|e;g
z4)v*RQPIw21pp}IzkOj1MGwVj^C6tjb#2`t;zbPFfddB;${zP?=J1B2e^2^D2q?6@
zLDHS7iiokZs>i_ld6F%ttyW~$*!CdE$cxcQG6Gn`lxOAH{qbGFJ)LuA&YwTu`t`Oc
zdG`Vh1YpV7#%=E-v+5kcdPxHJW9?Z#o4PWWXIem})xe$gSd9%>^?+?Fkjd`xSr-=>
zKg=U)D{s7?oJo{sqiq;RkIZ30tvz$6zP^4uJ4Dv@1fr(=!JfuO!bZSdN?7smKm|ld
zVGxoIT{YT?#|jsez1QOM7-93hrjzPux*EX;V$H#v^r*5c%F6N%C@-GR$49|OA!q#3
z0g2Ncwc_*^5r^vONu1uJl8IA<e&FLH`~KIX0@vv-;NuJ5<pU4l<&zA;b+3}r2<+ZE
zr|}sLTL0kvck`IU9W@vpZO?ySUcO672^<o}Z@K<qXSAE3pdfj%ppF*7`oUd9!pWCw
zOrzDSuCCs;eh)?ZyL^Ro+SYrm@h`+ymkwITOi)$=L6G^|wRwgW39}b(e)QY5fJ6HF
zwF_d$6fV<r*xA`d7O%Rv?d;@hovwj}$BOJb`1qcmpDA{9d0vHG2ln0Sqs#AU>goo$
zo|ljTl?)PCldgt;F)bN3!Y*J>r;4qPrGPI!UczqC@+AeUO4%b@izSQ1KUi2<`FU1$
zxbMxF91qm5HPLDbr=+F#O=V?jp&N*y6vORYRrk2q_(d*G>j%DQyfXG%#p)6K`(HF)
zHOdV9a=YW3x@~96XF746(*)mDRoyf<Ny^gAyK|X)V*SmM(Wak*N_ecce%{-D)i?E~
z?PmGez6pVT7-nuxz9F4Gr$>;pnR~Q@$l*wpR@;<XdtJObl)5C`Y_C|ILLcWF7`$|7
z4`2@ZFhWM92P~!r7mI(;7YP<O$aM6#Ky)K&q<@kbl;kkI)(W>zVSP|$PC|mDPH&1L
z!tXuV_iNw2odXJzOeiQ4BDY1$L-MlHHpEZ4^Y0j5Si(OIZ?NmRQ$VPNG7F2g;aoF>
zw;;0kJt)ZEDN9J>(4iGM?Z9QmCBIA2Rno;(U;j&M`^m4E`?<GZYq^N8ErY>G-1roB
zG8~a>?)>>`c<za7cY^2JC?gY?dZ!`s!GoW#haLu{5Mz*)jw31nnnjIjFUpD?idkhX
zl_v%4^y%)s(VwQwe|LGwj;mG;((pI{(r_iAh`*2NoF}JXVX+S8d1=9A<GunRV=n7<
z>;Zdz9o;dJN1&cWYF|C*Y-72PxhN@tW)fe{;<?Eq>R2g@rOmZ%5<EbPSqWP7zE@Y<
zsg(Q66k+WzZoKr1^gTTKy({JtF@l`~;O6*_i+F?>SFj>ar-<nN^+-=oKi5x7uM{3@
zc%v-hBiEI>R7c$$v50lvd@Usmx&xFLQBX3S9E2bR6isUxGVHc}-Pi{svtArFGcyZ5
z^LBS(XFva@VF2(S3Pjvpzw{EmM(k5>QGfpY`PAm$%^JMK;ymdRU8#+}mTVT{>Yl1n
zH~VnEpyDPF-nQY><E>Q2x4G_`#}loT%)c0&(030lZIP)y5uY{S1wwjx7x2;zIS0$N
zhS^)dJ*V8*c<LY#u4H&MrT#RY7Eg5Hg*TYIs=2R#?Y^ib*mfV~OS0-r8gefZrvH`A
z^Ggr!71_)j&vyN?^*@oj`fvN0e@F)H|BvDynvnd5PiN$|d*UAJv;d;u3+{<8y;3sT
z&iWRpQPQ#d!<<9vdN=p3sET{=U{!R2!!@`Q%MN(iLVMUB0A%_chZfpffLHt7Z}@OU
zOBTWtX6G+)!7amaFqBn33@$6(Cb(VYo}y68wzH%C^}i8^7!|eVs8n8=on<`8P3f_j
zXY>`BvjjFiFw9(_a$5-Bjpx`1x>O9s#l?xktxSzx9T#UD9~akU?{Z75r8K!I)Of~w
zy!3W0-ipiQ4Bx7<vjH4E-mh{ydhZH#dt=sS$?NTvOYg*N&-Rq@%Tg>+X2iB@7h7bv
z4`&88y0~T82yxE;GMf>Pam{SWh>JgqK`Sz^_g2ar>u3sZch!0oW+Ap_&CCAd4}=t%
zaWb#-6;AIuxC`EbNXqxemmx@`;0V~(zO(wIsp%RRO8=4*#<`w+<474q87z4!?P|;-
zZX-r7E%ANRJWco*n-%TA21RWWTms-g^`4&ENhef!vk2etMXL_Uy}yIx<>d?h4w6?@
z>Dc$ydh#t5<vB}<jD)~F?welq_xB$Fm(ItBvw_+?)Zpb6k)FO8Y~~z3KB7BoYV_#6
zJ9k7Wsm#~jJX{v7MXp0Y%2Y#1NyjajO!%BL0nEw^GZtNSU1|X|_vY87SDOFO0=#*1
zfX;vfGx2kx1V7=xfZTgtNPb$#wT?3!b2(`iP#Sihw`?ab{69&-|HD`O|B(>j|B+E3
zbeJ&w3MB5qwmHK!U<(b~=CmqjHj|*irg6sqsFvb?ehMjYUX?sBFt7=<hfA3~&^6^$
z`_oyVnARf7urg0Bp+JF!6qifwN*WPzrut6=`QB7jRYlvg91G-<AJn@l)En0ixw?4|
zQ;KgwC4~iIVq!%11!cN}$hAg3Q5k^;gI((oE!;TLTk03?yV=xCg?D`Ge6w!L4}?=Z
zJ(v1zSr0pr9OY+Tev}tv42LdD^hSmil%IprZyQb%8uD;cd&0~EdAtzm2uYhyFNoAp
z(Vs<Ee{$kltw3hb2IwF>Jt>bM>R)JXr<i0?>dRr}C@xs5{R7~=H;AI8EA|C11drJw
z-OmFIudjKq!vHTEfrb!R@0cZfENIwys439DbGgd(+^fq~8m=e{U(6b)7K_UsZWFE;
z(1!)mL-r^rbL`jY;-32URmX!&g;pW3Km4|UB~YoQp%IW7YNV?;$}_PN*G5kVPnkJf
z5$}s=fpBij{nYfh1OosXxCP#CuN)9xa2p<LPir3jBH$9)ccD@5z=12~2*7s2GVu(%
zEd%H^rQT(dennb#VqzjWfh9pD*J~>?vRjHRIt1O45o7R9r@#Vvq03l&XJgyu3N1}d
zJGljuLWGQuxFxR(QxlVSY`C~mRMZH0HlHOs&YNqB;KbH%cJlLRFPbJGPE2u6-#fJe
z%GWyLK{mD<jydlQ*=(1@rL;Eb7F6rpLB<D^<z~ooKW_rp;-;x4i4;cSVwZ&elt#UM
zW1nmH@7%dlhw(#4k*ijyv0tmZ?dz>&^}b1Ib$W)DFZ-M125g2Kx#wT@HxJuvc#V5!
zOtJc4qR7!vQrdfS{qblHkCpCA!o+PtxM7U2#e2GV>W|H@LIf#!iI~=^zmXQe1f}mC
zmx>xe@jz4>Jf|1sxYUG=J1P@+N*tsw8xX8lnm>^{)B5^`tZnlO7F?z8e6d4)J8wp+
zw>4faIT3#Q_7X%aPFc?VZCA_H&D%>gpi)V&X(~DxsJ>ti_jf;Sw#7N5kKl*0ob50Z
z0Te~xG;rvi+9{7ps~?+@X-7JD)zJB3{1%>;mPVlmHp0>8=XWY8LB|kt_;D28(}IWl
zB9)Ke;@VlP5g>2><BKT&)Y;id!40;=o$PD{QBm@y2zlu=74dD!k00+u@JnPCxI|?o
zr&s+uF3eJWb);jq?S9F!ii(vcJ+X%$uK^`QL!o#V$U{kKqm&fMGbnN?Dk@sI#ql(V
zvE<}k2#4ssI9X!SKwjLZIA~aCDbyU`%p!|FpP+zd$p;@h{+-^Qn}GL+`yP$xJJd=w
zmjn@uR&SnxbQKfwniVW7bMv*)v#WyC+YGbyD@=;8BI6@8ikv#~C?Kn-V2B5{^GnTF
zspvmn+-X;D7m`Q_#;&NyqC@{$Z>4vvqL7I#cXQK7x<pHYa-SdjsH7dEG_bd_#Gz@!
zKwOntnh)c--TmTI9fATs8<!5$96Qnu!G=!K2}DkJcl!_bB^W^A<mvfo?T@r9JS+|j
zOk`~+-7<Ia;>8KiXUXN>kTf3duwEcxAfzVW4;S8n5QH(Xi4NX6w2I@wCa#sofFx*j
ziXax*G&<PWi0nkyKyQA2e(F$jP$MNcl2;hxiM<SwlTqlosHUWpi;QhuiTHTL^IG?K
z)?_P8-xl!be0<iR-v>;?`pv$WJ07?Vy9TaE4P23YP>lnuqaTuGmW-z~gbDfqc2DF?
zUIoL9QA$os*)kUECXceWwk@xGAhK9PH;9nahZr4_Z&6m)0n}El_<J6i^Kf0!<`43C
zP7Mgnn0FXt@hAh^$_Q8X^M&+6!B=r~aC^WMy52*b<2jSZMSq9?E22O8?tfD^_zydU
z6j-DQPm<F=T~*nmzdbA+3PKhA?KvA?hq1pem9KRwnKWV2mC{YjSN|KmAr0ccCU^XY
z4g0?(y8P!J#rs<a`JY^mj)}PhX7FI(8wpscfW96GdJ;oQPN*2DMIQU=NG>$i*o{U;
zMXeW^zs=pH4zb4YPSvJch)4Q!+F$nPslO^nQ@q4@eQR57GG~Ws-c^KVBVjh7?Zwsz
zVrl}Jl`<pdOuJbNTaibRN!OoM<P;ZZ9I(q!q~*)aw8@v-+Dv~bQk*HgCH4LCWovMD
zP5C@zC&38jmDl^)!g5BcteVd4bc&COS(`ETZNMIPmoAR3{3K0TkrvQoFwzvKfdP~}
z`TCX{+jXAGI0f041KWjg@ml(CSII-bYa9$bFsEsxDmmEpl2Lnm<Ju3Zx1lss;w3Oz
z`H8N7TTpr;NfJo|PtTJe5n*r04I$wS71UfvGOkM)8noFIZI3kU<!g77j2&x-jR_xm
z@cr=fUW32SNi^M7+`?YE>yjtz^N_2;^>w<3=-cj&b%-=CAqxE(>5#tTUqw26MDPBK
zScf=e=yLYxsanqzlwK<vo}idH+xrx<3%eht!9}*VwyvTi1>z}$ReTBJCSQ3mbtq`z
zp#bzKS{n}m)w^VOMfRPr)TiJ_`#+sHiuy3o3O?+11`k6XSvpO)8$z5laj6Yn+7M6P
z&LCa!gj?RfPRT^{)^ksSu{z&(vEYTp_S!FRZdi%v^0vtMAszF#e8>VFcLxtoR8-jM
z)0IIaDi?x-(yuxFd?saLwkz_q$6Kq`I6jd>;+b6jNSjuUmvv)4yv^HtDvfu%#l1j?
zL8P*wVW=fcqVGA+0tbX(0DOD^b=J@&=liQEEQ-$=b%t)#kTR2=o}>g_NGVTzEKKTx
zw>ztNJ39Gb*tePboiN;z7S)fSD!o*dWKsYsf2k;4z;CtseNyh~*_CM$(8(6(!Q~@c
zKlge#WP#7lgAX)%6)a1cT%zfW@OUUk_3pl^H)LoEzEFn^c3Qa$LgSLRhKGQEeC)NL
z+x{{1G6nOs^!Ol_@gr|iSSo_7@j{5Xz@=}wuNKyI$DB9BpaXp#GGV182y6J{ujUJN
z@tKSd)ig$ErNFi}B46Q0X--v{T8)l5JSOwt(ftX+Cy|S>hH|;!(W!SvJq>w<rOJr<
ziGmC!aU8f-#PS#8-TI@gp`uh45A+7bw=%ijjYBF@Z_PE6p8O>$%Xt$9TrhFMWYI8I
zk$=l#kGi&Z8ys!Z^tK?dkvv`$KnYb+NAFH`LCUni%B?pDS&!X%kO|VYJUvOP6&;tp
z2D*hSeZ3?*lqlvcte@{G$ewjT#;&JPWNcF$z&~W0I$wIhKn&KjDDXqDW^f7Vj3kDT
z-TMQD1IR7a@*n3t`n!jpPO2;PUs-_lmcca#QR{o^-F;pf1!zC3Ey65APh9J@0B=?%
z|B7`2MW=frV|1vcU?t=c2(Ic9G#L!=#u6Gjya}JQ52qx7Ph5Ii6a_axg>FH^!;q1w
z--_ZLGC33{lE@jcf!W;dz<aDuCMTZj+Mmk%dB4ziif3hbsnM!9C`+|<dEXkn_rauF
zTrc$}*rq$+koGBZ`-GW>r=Nben^R%n68pM|ci0T=?`AuiV2;egpuLn&R&0++x=NbM
zjK3a$JFPqB62=(#oWO;aC&Zle8t24$AmhjrY#~m8;;3Ym<;jz+30&gRI~BQ~)&6=w
z2J!&zcA=eV9Flx8cr#R)i5nxKk$)$0o&%NHSy}Q`!k}Dk7xO+}_z)&Q7z?#<U}UJ*
z#e6I<XJmQRD5S3YTW!XtCI)9vzrS~Pa)fG7xiLqA@mZyC02=z~UqeR^&2A4iHk>z)
zU*ZW*pr5h_bLW%Hg>FWKL@S~^P3b67|5<;%>Grm^1_+Lz0;WX;T}&!;nhornF~$gX
zIFBEnwWa&XCg44PdrJ00()Z|ZCpbHN=N-YZ(O$)|4GOplIDOhx$OF_K<HWl*iCGt|
zj_BLDns+0U4m~aXpwF5cHmH;dVP<Xlr`p#6tJNEziN8d`P=+SYvOV!ATWQiAm%_Bn
z$OA~94?9T4fCLY?B~sOP;ARUCXYo#Io~AI<u_8J$@*)9xdq~!ewv1anoz|q~0=#2d
zb^mq!h(=iKQ)X4wBGRT);lOD31;u6j+U<o%6TH`2&U-R5Rx##uzk<GGmVDo9y1L<+
z*>t+CecPTeD=QaL_7ZZ1)k9TooKAjN1w#!;M20`|E?IO4OJ2}2b;(o*hpOX&r0O+>
zkm8)%Li*SvQLEcKhTCDHsFIAt!@iH=^zK%X)v(j<qa(*$kX~4GhD~oy7*WTWkfdC}
z1SNZ}OPJpCaL00VL9TTfGg>QN{I5q+bXgS&o#^ch-pc(DhV?}$4(n2RHb6e9)5oB*
zoyPrv&NM+O_ttiYR8nt1ZK4MS4CPvK_sMn6Gvu~uHImW`{^7;3^+C1nkPe*_pGx04
zvK0f!fW%1|zTYMTa&htO!3G0XSZT|2{W)~R6Az8<uhYNE_t&teoWVvv{&W@kXoB>e
z4-;9OI(GJU2j1D9?9#r&oAMw3R?nk}FH7Ht5(1z?0O2I)*JAOEd5uST$4!2uFx?T#
z4bt26jE&Zhc|-BhrOJ+F$67JO7OXowkB_fH&sKGYfJVT}MRds@`$rXB*uu-Sbed<6
zd`@fGHTl`{pHtW7p5DFv+k#tyGryUbSNr2Ouj5a5+VjnmSfz1MX@oDiyy?&n=kDlD
z+%~V`md?S6eBUmgc<0-zKZ^e7$+zmaAD%6jx%x6T!rlEy_0*TkHu!fbFq&NzG&L5d
ztA)9R+V0D#?d|kWDZY9iV(fg$0-?E(^;b8i3(8ntDwfFEA}YEMcAmIj<=r&G31|1|
zG=GqXyxIF>7oV~aAFqbsecy<N@?tv;>mf*bB$Q>xk33j7sZLphvC$IZdQZ<Akl>9B
zSLFC23fl#|=hyBhC=-{uc3c6`nsIeXCW$buouJFgGDyp)f<oTAm~!2bUv1E%J;L4M
z+7EA~Zb2bJgAGDlYi^ff6lq7Vjn^-EY>WOhku_Z+bHAP?Dn%-cp{L)YH<O0*<b&oH
zj})*3sllZ!W1~u0d&^#<o<$EfH0uv`>lO}S?+_W@Z=EKgE`BDb!2zt@y|1?Q^%^JM
zVxAYsf=#xFbH^s0ABgdNIv6;KuJF^zYg;~`2xu2E07|*CEx~gE?iiel&NuFSynR8~
z3lKwDiKow_x%%r}%>vM)s)ljCgNiiV2A_*i0@OO>g0S<d?2#r3E6L^0V^iv?St7@I
zmwS<zZe!l(A2xDK&!N_?C#{qr&MO4r_Pt9<9SgDTtOLHXekoZH5FQ}B?(m}rj!mbn
z#G$SuvS-R8E)`#p75sQXMEraWp3@m#Mo?E*?v@45#b&A(#(${oh)H_o#c${;sQ@>^
zPJcCZW$+JFx6qIj#x+kXYCUj!zo)B0PTAPBbI&}Ixuo|;{eUpR8_1z|0Yy?7N>2#O
zbtP3*@u-mu-u6h0TC3G<nT$P=zb>U_4v8{jv=c-P!1=uZ8$^Z`62<wX5@>m6^$t}N
zi$R^57bd5;afiVyoj7A>mj`L$LQ)oPl$Oqqj?Wed)U(-5S<Vxsp^K5nTw7(!yO%1>
zau0_jn21V~FoqGlXJqcXejPQ%kz@+seQ`Zr?HCNoqlQij-?5F?lA=vJ)6p6{(i$?^
zR~2OFIbW7P9&~4k*|c-pN!EuAvib8}hgwC@T@7W=uzR9TTF_xnJ}mpN?h0mkUQ5Md
zNHYA|g&^GJ<0DO|MhwrC9_d~)G;gL!h}|K$jERb)5aF#oDy8cH8<-2}%wmbGzJo}i
z6D~D|c2g3{U=W+!>T(k6@$t#AGHoH{7)X5uWsNW&AHG&IFrP=%{&4(zN2~0iW&{fh
zyBoz8b&DHjXQpNXZEHU^)m#fKLTp9Si4ua<3RY&d>2`cQ>>pBA2Dxj*!S{a|a*f~C
zjq<q$2$8$&<V0YMi56_PwK!V^{g=P^5=rOM`=+f~aa}B{#UM{1a|$~mRkeR+K^7;o
zA5^F20rdqM&)w=q?m37DMteCY{#eXO6L4mYF3QTz&R%0$^s8xgDF{?e&C(->W^&Rv
zX4!C4fbDZ)?kA?Z`&ykmxoU_tPcvz)Uw^H{$&GcI`_kC9Fey+A?gYCR&vwA`N2Jt%
z)PY2uJrqCvVtZ>1C<G$#@dhFZ1YA&<8$qYc(>uaaG-APc^1$e;;Hnl2%UO>heR0u`
z;7oXnAkyp!v%EaH>=p+VC`U$~D~{|4GZTE7xaEh>@<CBB+dfVK`yMZ5W0329bt{Uf
z_GLUS+n{~iNl{Ceg<vfEMa*T>%>&u5qk>JDrI?i6ftPP!g1ws8mn|y}vjCA5-DS5)
z`yK=zD7OkL*Tutd?eR9eG{(sp^FBLUVUuNA({-<`Tc9HA?3fRaK5qyY;qvOzAd;5I
zZOHQ`p%sY>6fA6QJA+z2cJ5|JiY4}yz+*qz8f;UvdUxP{dxBC|#V2|MsC0;V5#m7Y
zRahQbcPh1t(M=KWf{?(il%Z=1D*qzlMFQT?)ox(Sa-?8^Ct~Zc3sDV{-kHlhdfG~h
zwJS5E%@8DfR$UHl&Efubjz-0iz$Wf3wt3X~x(v1AhX~j@e|HGUjy&I3$7OP%fkSAj
zntI3l!Fgkn$2Rc4E8DNO;27p>2hCSr^x$}zh5Sda#M=VCjDvllF>BqEE%4|^y)&Vg
z>-^nvg9Tu7@{L`#X8i3lrLkamL4_k{d#;b5%nH2&(tXD>1rf`8q=dCfwIX<@`~k&N
zJhzKFuPsnSr!G?g*+{ht2+kK^XPgKV*?_t)Pftf^xSc}`+29%xIrEYzOw*GmJrPax
zJCA``-XJw-u6mt)wk|1>R&S}M+~&>e`?vkHy3D3Vrnsf|Z|i)@*4z6Zd|l^<o$H0V
z1z2lV7ISo1g>4G!wxLGnP(-khxv8nC*F-C<UriyuJ*($U%kz*iGiI<oQlg{9)U#Gb
zN^qQ9zNyP?nMz2<+R03D?V;VMcy$G_jMB*=Ni69+ne#-ar@lhgM6{t(_6*OQCTlfJ
zv|>9sP=N5Qm=hC<JMV2BP>j89DL@&?4F!g?k`j1Swm#dc`q64PEIR<L3yLEj=V(wx
z_Jm{lbOPBC7`Cb~tIq1j*6&nh^59G#<~6+`cnf*$V3AffQnv6(C#izcsiKh|IUkgG
zdV(6xYR@K^jp)AyE=CgUYJ;mj1>_VRBB`|qk{U*6#kefTY)%M5r3RGyI#~~ijP-~V
zkGh^isY+>>2-T6;g;0g@32f3v2IO{7cDe)ynn3s2L)b|i#g9B$@XZY--W+K2S({Ku
zloHRI2#=iAPEVNF%@KBo;tdv~`ohJv%Qj4`u8xtWM=2<jm-^*-^A}LkiGqEsGs5d6
zVy3u#Q|qm>sIi=Rk^?sP5=zmgNLtlpk|W3jDJ_YbjZMurj2lo6($-T@K{GdD@|{l;
ztx$^V;dX#Qi9TT7fb?KtA7S~;2zvRb1eR@#xM%_R0cP7~lInpg-|z}M?w9LSoH*g3
z>c!HhB)%b=jG~2Is;Uj+Vd33faKl;9`=e0C$}R*OhF8f7A9iul>@7bb+pVHM9+BMA
z0({vJHncZVh6h-;&VG53&i(IIrtcYDY6k7&*a8q0_v*|MO-FoUjtA#SKx7m<hU4G+
zp!wr9?U_4Zfr%XfHhF}~15`kHrBtAXlz552;yiUV`S}3Cd7L^3FO`u$iB#b|l!fmJ
zyut5a{0#uv$N^IH4!&J`hQm@(thp@!F4oOAP4ailmY;`JZjI=FniKEK!?!WLC{&;A
zKKMMryeiQV`fw2vbA`uQ7Dp69(D{1Xz}shg|Ih+V$%83fs2;J;2WobS6>6bE6hWaR
zEiio<&4<Uf$ouaFGZ902sc&$_$=tP4nfkz_HGoX;(9o&hFySrJTOt_tC~Iug%GzmU
zu!pIKh0T@_AN#aYakbr7n0p8_g6wBbYUbl;2N|j;6{j+dzv4zFV^*zQd%1-z7xp(g
zwdVEf{cT^v=2*-!4>~)cHW}wfsj<Am0^TTN#eg`G6yy(792+VvF!qx#Z%-#cJW}93
zqTDlv5>Sb+h5O^c7f?=kT*C%DL76;WNj|c$D??EZFo*)kc{#&n@eQbI@J7);;Y%eY
zDi!lXY8YY=y6S`(hnb`EAGsAZ*^qt}X`EeTNcs3^KhvIz_q!==ahRYFHu|3+*6lm1
zT&Ci6dpxy#X#5*du-tB!1<x{A7VkEn%4Y!4u7}U#-5Fpypx^c4CZ2oSGSdCLr(D;H
z=gi0LF!`*y-cUB=D@%*|`*;I8Yf$w~T^H#?sO|{9Mzi5F?}PQX*Q$>T?)SURPy5vY
zn~X}xad)`A4uJJ*j)%uD8*4t%3V0R%YaK;`j%|?<b$0b9h#JfMQY3~)Pu8WaZOv6#
z9>%_Fpz{_8tDH8x(VIIvnR*Vmwl&h$0P+MqVbPja@iyC$*Q<+#GFOJV+DO*OH+U{>
zfW8m~KcQi5FcfRVQBKTlS~bqn>_;TQK{;DKm_oodTBn9%!xbb>236K+&BeXV5BzSt
zX#;~cHA<n^h?j=SXH<we$OwXOvmQdF&Dy`NK$ODI3oP&xG1Pn2WFsF}&s8p?0-wLe
zCXA2g_`QlD1mZfIQ-(5nd}n52iA&4K@CWt|jns!d2#gd!4+-yL#6=E&K3?H&H>kDO
zh_jCuo$JyX*>`jOy}zxO<+B~B(z)Rayk|;De|$z)dczltPw%g5oO}c{oXEF9RA>b6
zVx@Rsw)#erw@SJUba;A_RihfaHGjP;%E=+<5#ZwkgWwB(A)jmjQQ&dCXxESj8mejC
z{bQ*p594sd$B9#&Lt4H*ZeRSvE}`O6T!Ecdn%5qmBaFI^;RWm*+qhaHJbY!FZ`cIz
zVgtl&Bh%_iUTtoVp^BwOsuPZH4?g{DY7eno3Y@Ca>(Mj5uG{}Y$iiDeGfXi%s4e?I
znZ)h4f1omFsFRQYr!sJgM)q*e@K*6n3O*r9w?(q}x^oimJy_e84~PU9-o@v(-+mu7
z&Xri<feE8@EUyVR<(hLZ@6Zudn4t}?Hif5-cpHw~`P;EY>vuk<tR^2HwV%iwZ1BG4
z#&h=*?#xh394Hct&jn%FGng-v%}2o#X}_(BC5=$%kF3NKg^g`jOM@Z((|97rhRa2!
zcd>4bJrOY=wBJM&aE?mzW+{6$q8i?yG{|?sPP^;2sq*If^;wuZk-;ZZ_GInm9pQY$
zi|4#N0;=Q0952iiJT$WHF;p`Swe?k1L+q7NVncb)ct_q$){iv9mX`d`S%?V4o6SG%
zT&L3--e+}Bzh3aqLYBFWbU}@aL~VbbQ`GjmS={dQZx^1?d&;9~|D9$ndj^)8At<$c
z|K9v2KN?z?_m75BBH-Sd@r}M;Oy>y)@vKQkSN+~`?dxlq`zLt=`8l#P{Z}^1UBxsP
z^WNxtXa8L$5=(3U_}zc?INSH1X!+81Ror8gIK%YK7sN%^XNmNMs+PTW5f2m-uYiL8
z_WqwA{tjr}Xn@VEH=v*)TlOdnyu4=Fh8q-;>|o_*90S&Fm)Bv9n5ztFKDRFYqHU<&
z%Qll|WJd?vuuPOC&p}Vq=F?(D%I~i+T{{DH+K61Hmf6<`3JB0z{k(qt?SB#Mkd6<j
z#50R-?zbq&zS?IXkMg!FXQtAxS<tEr2nza(FXf{!&|%BuXk2Qd)ih*URwJ-{dtBoj
zNahv#+l_cP!~sj~^|221G}OT3qu1md(@kB_VdQPiK_n_|Q5|r4C&iU+a0Yo|@=>=#
z*S`4T7I(g`y}jL-r+K8z3%vJKt$>+jr#t-bu#e3|NjFai*Z}MCCB^<bjpx?!(b3-H
z+7QeaHW^}vCgels<-DX`6v(wLqhO*bQ$*J0lShgKvW8F)V2$q%JTN~o^XufZty1LV
z<f8A~e;QXkV6CO4RaUl9Tl1Zp_`GQp$X;*a>dnk@7>kI`&nPPV;#TgGh+xsH80bpe
z;DHBjtTv+gMYVQ4m+`74qi8kT4U3JpQ2t!G|6iE+e|IjTRMPPTJS;%i*<KaD-4U18
z{QF%MkEjp*`xeNus)XbI*=w}^(9UZHo|(9tguPoJwdo-MLd9C-SgBb-?!~#hIwm1x
zc_vkN-P!4F1SVEjzlGX8agjaS%H+A9R&m>jtaQ6}LXj>NoyY+kUaS4^N7DXL{~JVP
z#O~%K>#v$e5sNL)4tgAXq4xw83q}+jQI#?!BLMPIUPLl10`;i?V8*0pq+UNLiQRmw
zWfN7$qU>^=^*A)uEvga9q&z$NPqb5OpoLHeQ!x`#n8@MZSbuyEua$|XXBe^;%uIn4
z87Zlo%vjoBTfbR$h1E@+C`9W*_Z;O8_fOU0dFa=26IpVg66Xhk37Q+ji+zr?ua}v^
zN+v=Cz!xDLy7s*ANT%e(zeCJ1U0so9AT*m*N5$J{AJpn$Drk<@7gztrmvJ_31oOcV
zsHjvZJ6)48)(v1_q4_@GyMJN4P14w&=InU3`CAuh%^fT|J9oe5uF0{%YgsKeZ1E4B
z_hA|109nW3)a1;Nrs6>UK&{xr^P>;NQv67d(-<N*=9ULU<s%0YU%|0L**G=xn5<WS
zBoQL5WnWEjd4@#Ln(%607bxiwC9TxowL*iL5TEQkIT{yS%ND-UIocnpN`zl~7Tm$R
zr^d@Qall;~3pWV5_*7_ujHtj4`Fk`@nlzl{LuIq*PjQcO&8V#eahU*IOKpZIf5WQ<
z*3{CUPbBRQ-on~i{=H*kr-()d7D6LZ<NDAh^3d0;lXAe>taJ<N+49EEZ6j$73<sa_
zQ*xVBgA~2TGfyFl|8Acgno<#i6(nsTu;1tG-<X?t)Q^q&Q<h_0vv|aosE?@?u>CFf
zAv{DKd+4}}0RtJiX}*fod;jlr>PmIG<VBg>cKM)jzso+g2wFb`!%P2QtA^ilwnljQ
z{Ll?w>`4Lw%6(Jev}>oCp;)o&)qh2<>CSY7dOfjhwJ3EV5R^7w<>YfJ;134v!fX+P
z8IJHxjD=C;DTpBRNWt}ZTxn$~fIG;{JNs8@?3CBJ+{6uTs#}^^n^Tvxn^mO&rxSMl
zsR@&uPL~RcIk&sq+~wL+Guu?olS#9w=Mu`cZM{WiuOS73gs5npYi~)SJa6JJFTOd`
zG2=qU1tFWCiK@{Z+3qMyomo&h0J%ze3olMcT;i|3ao)6ZRM6RglBi2mwY`&HUt{33
z131ZqV}(~eJtuNK1ZlTDkvrZst*+~17g-fl;&*AKW}x8}MUzNmnt3xTs0{<sJ=DdD
zDgrXIkews7py5!FmZ)e0z`YL}6w|#^Jb3ux(M$cO94gSH$bP&8G8h8etT>kvsft{r
z)zb1T#(5$%!L#*dP9JCIK_}I>4^S^JlA_vc@KU4Q4mYPeW{Az3239G>+1*eI6gxO%
z5t6!+7@4{ifbSxf^#(etQFl73C*29)wR1e;teWmvaB0z8hZ^ew4-rM+ni~j-{n&ru
zdlq&Nj8hkDY;DJxlNk&Cb$a(7+a8dn{eMO&Shidrr{)4lx4>dl>7y6a?m%Gc8>O3m
z`sx1}@AywI6Xb;pL=M0suF&75NxJ<%F+Hf$ub4jQ_uuaz)AAM5FGp%4vBpIynN#5+
z{g<DIzIRjy8?g?|907KM$A18Js>-GQPD)BhXn<+@O-x{f>?sO@cGSLl^$Ml+D&wSo
z*t{$g_3JKvFYL#hOIQ5)bA391_)JI6H+A#cT3bDlD7ahvU>Z#Jp7h^xTajzI%?3|s
zMA?`VTmCSHy5DA&?N_Jl0Zj7$X2_g!3$-U^v;N8JM*icH`#<2^KmCIJC%KU*@z38}
zeT42b)Yp^HT>{D$7RuS13o-iMSVQur3<OtZG0qofyquhg1~q&2pzeB8R;HQR2`KAD
za&pSJr9w~?o{*4Wn#?01y7%1-zC%z{65SJ9+uFRSy$CN=N6a8~o#{v_RiK6$)Kaa?
zeEbqCZ_aRJ{i~!k<epyWwed&h8m<0}2^%5b8&x^3+EH82-@Rbkxf^Jks-#4v$E<XB
z1kWU#gd7TJe@^}QfmS#$PA6#vHXy23Af}Ozvqi%xC>lT#UmdjPxix41G!Y$_Mjsg*
z0>;-D2MbqJfh~f5651V=V;OM9d7b@t@84fxYPvqKguSYmvyF#S&<3<2vFE}1sG}kd
zie0GNCwQ}#W%8$W_^YLm2z&WmW_x+8w`b0USouY0o;luT1np&|9<HmfuZY#v3VwBa
zrA*D~9YQuQmM~*hPp9sJcU6AAgp6Iavj{qqyj^~5JJPlY5uo(kn|^8UgJfgF-GH-y
zg|=URHc&|0etZ9__6-_0(GO|aC=#=B$i_B6g4$*sf|O%Em`x!*J}T{p5!5=9L9IO~
zTR5yY?%;nEJ*(ASUY1q+rFqE4<!Wh*cQGUDxAXvJEC>h?;sz`c+y;tT`Xb}@02b$s
zW~8Slx7t2zuX6#8gaQ3cEGnkU@Sw&!S)Hk%Q1aXpLm$_{)_NOjBXM}_v&%DR=o)ge
zAzQ#2yLNi#HUFjrg$7om{(^$xwSEq@E~V9u6^t1C-y}w+T(CV!s+?nPmRVfwxU%3u
zolv+0|BZna2;*K{S&C{Nptr2JB;10hBS$q(;@yS-=M3t9Leep0dr?rTjne(Fwy57o
zK!ILtNIH7Bz<^KH(vU?$Hm|sYIFYtgjb~KXkAfEPBF)a=QI1)6>*u&J?kmig32`e1
zjpc&xX|nHY?*oM-={ndA1vbk1ecN^R8yXtA#+yNojU0*CN3S(2kxruXHFPSvB}&(!
zq1(Q3{Cg~pjgDzc5LV%{1O*lM5qPJVO<9#hAt5lrcO$ULHhqVhQ(d%7l%zLM7-|o=
zhFkQj>u;}EsdwObCw3$s-%aS&fpTZKCYE!MZSYc8gAhme=Cll|PNgy}&y-#-<N)wc
zgk;YPg$oo6m#}tBx{_AqBQy33iI4h+mvH}<K9=d}<T0Inm|rVfTic#Jm;OxFic)LK
zBMr>&2&=WKHh)uxO-rp|Jw4HbDZ70X`F7}Id%PiChSo3m&)w#MTq&i^a6AcFDNna4
zqLO%Regns&(2P@k(IsK|VSPV+v_iwST?&h(`S$Bzw3^+Ht!wBMQ65I`$ocdY(rKfH
zFe#5qM1QdRDAG>$z42lL8+(QazFL}`^^3Z9DeVp3B#9<1BBU0gh^q&Zko?L7<Hj4t
zz86tyxCYiH+BAc1X#y*t3L`Z_FJ@jU4~<N#S_;s}WgVm>hrawVm=!JKl*+;x|0FKp
zZ%)~6H~*6-=w5qQWffEjUL=oDqEx76*&$0X^p!3OVWk=@K0ZJhvd2hh+h(PpD-yPC
z&sfrXcN~fk%G2}ZVh;9sgIP8hPxUMlrx-s^BbJU_%TiGkEqlwG#InJGV|(*m$uh1-
zX|rBeTU(I^3&jOV5DSzh>$D>@HD>7QG8e>i*7nz<6B`UJ?`Q_Ld3}=3sJU3d!eH6+
z9x~@p^L_KoR9&7P{cCtdt-8SpNML7}P@g(#oGmI!nbLZ0m&4q+2Z@vk*T&xMz_z14
z7%TMm?F>N{u?Kt%d;ziqf=FZUNxf5)Hu9yI8Y?4vR?Bje!AXK)k!&IH=XK!W<0Bb2
zDO1Q?_$iJ$23zMtU}BBkNbw`s-D^pJgALm52(t31D<9cEYT4#X3NnCAN)qi3R6oGe
zddlPssY1wrhG4Y!aA+{vw@F(<?BwZ*`UJ5C=&SZ7-<p;V4!;5YhEb`S^lGyhrFZCH
zB#b0T6$kAD3QiYk=~?f5Hed4)b`NreAnqyG|66oq42lV4Q3iti9EF9bjHEA9e9voA
zC@M-qolQvdVy<lH;;9%7Y$+%g;y~zNUduBiKcyfMgOZHI$#n3cIK7*ng?N%xn5`M<
zLeKv$ckv%Gz(w;3mhZf5Nxp}2^kAtFtyW?y(<P9<Yec~VtO(gjQbuDuF=(f^6G`;1
zMV|iR5qCzRu+4{%2Tq&f?XNf466UUO2uu<mA2kcVBqX~YtE7bt9PGvtQBj(9N(}k0
zd+=*{ydjzxHj3cR(dzOd-j%_@D^a!pVE~Lh7#mWp(dDV^83QAc094_;%lAbn=79~s
zap)OM`2;M5Ua$rE)J>F+kNl5uM2I~tGACdSqoFTNP}t%`R9!d+H!5vevk)zm(2t13
z)+7w4D=8;0TDbHNEdYEt`r4p%94;|cOI-))LlGlnh_PCKneF<cscE9^x=OT<#`)_|
z4uo!lxsYAa$*D~h_4UQQQ4KE&keAbT0n60`Shh=Pz~&Ozq(L-57vZzH^fbWu`>)OX
z*T@-;D^_Q$|2IQ^3f9_E{sT~H#O&%!paywV^r@f1>&8hFjvS&f^}eJGXugkN<K-Q*
zE_#cnuex$LWuTIDoz(Z4BoCyi;GtExVlv|j0||cpgDQbdn-5*<Vj!VSsn=Lb_E(BG
zM^qlJA4UtSc*6(Vgo9bjtLo7iqJe5P@2Lb*BW2!mh5#fNHRX$u=`|}GP`F&td18_*
zEBbThD+np6o|&XI$kgs2K_?8Q5fs4Q(Ea9<LIDlkvJ~p&*i3j^UPC|7g4E!Bj$(3c
zfB#6$VRweCM6&%paI!G2Fl(2<U}g-I3}l`VW4y9bi8P#BR$jmLnsLm7`=fhqzh{-4
zN|bLII`RmxoGOICkmW)<+<0Bra(DXOL};u?{Ku0aP!Cn;(3X&zlP4kx8_Zp4gwjy8
z^xTPJWoGF)tf8`7@52h;ZWaz`sMKUSUVd+qTv8hH$??F<u)-+DmKWnYZ+{jkDpEiE
zI^E0tXL}iFg8qQ^4a~Y~W0xdqqli`fC5|~8y8Kk6UFSwVPAQ0k^p1qIR*&z0Ene|<
zzj3WyTQi;Q={ZHj^(P{AW6SL*sS4@V{sG~lgj$+;cW#AyR{KD9u=l;F+riAb&{0?b
z>LcF+f#8D*JE{?;Y+D6qt3gU_KRYu%xp$cAE1>m!SkXxWW75SrFa|#tyS1CmWF*LQ
zVij9jPB3i^`kI1U)wo%UTFsK*wu~eP3`-rb;Lg-Y%l9(<9Ivrws4jQ~(<OAaG!_Q+
zEyyK}5BhikA%=~7S~*oqQy5vY9%`-%gIU+UnOdtiQvJkt(Q86F8LG2C$>$sn=-r?r
zkbm~z;tW%v*;j{OUhi|f(r1$R=+cxj8G!>*WJo{-Ujmm*4HFvCo7M|>o!Y!NUR4pl
z2!6W&-SDaH!Tef#v|>eTfd-WJLbFHR2C|W4wn``=1GURXV1C%pOOpoho|oVQ#Hod~
zl4)(ZF><3sDU)Zb#D=04AyDHHVkA)@b-<;@&!ofN1j%x-eazL?tPBKTa@6_KtxV5%
z^Pa3)L^_mbAT2L<Grw4w?p1CHij&=71eLK?U8zgzY}XHys3$(Me$;-DYF(m}*`wK(
zF7E<+vjs}q$(AUPriaSN$Z*2+k_P!}uVg`5rxZb2!zv-U(QcPhR$iW;1C=QKy$))b
zM@yQ%p*Gg1)`LQp<@N^3ecX|bxOpvk&9+8}DXH58?7hTiYY{xytMp(Sr4Fd#!O9jC
zm0azJ57^UR7c6~mbLkZCo!$$0$}MU_MarJfHbtgX-G<1M{sw|;qhP%!Pi%&*45=_l
zv+}9W^B}4t_(=>1<hl8?4adU8@ZBV*()b1UhWB1C@42N>5GPP@_O3$Ga(RPg+?#8q
z8n;b!red$upWq~)K{n76$B{O#+)u^d(*cAfGKbruc@MbOl><QtJFZ~hQ3*BiJBQ}z
z`~+!x4s*U##_B+?not1@l_;-vFpKn7yaU{zOdp7@JILN2**Va!x7K84Q;oeL2MLFY
zX>|I%VG3E@&^&0wWbd)ygv2u)+De_m&iAHg0O3EQ!mww3pWjU<v(M4o!R97g)~4tB
zTDZR|9j&Z(>Um{*C%A>-Fln)WU1R`eRKv9Xj8@_3<fvfIJ%M4dh=W<P-dgIy79~QW
z{tTsS(@0wM5v4<Gc~_1f8;;){s3|@6MPlqADKb9-gM988MLEnZ-i)C9?7+vOqHx9d
z&o)^tT4zUHSGYuMyfo$BX8Ti%X0PeL8CzZ8ZR55=<(=aP<?@3kphoIJQ`~%%LU9un
z)DXD3so3F<LS(g%BLA`=XNW3-@{((r#u@br&=)z#+u!QH9$wxVyIQmX?$xkIwj2@k
z^_Ix;I!?HroyHFLvSEQ*VXA>gasIwdN2BXU?muRC#WN8!vD*`_JxB=7l7ZMB9(X*M
z<3`=)xIqVdaWOMHG>ltT(R)WAKV)=GnQYFw-JL;u`epW?+G6rRq0B8b@GZ6Krw;Bw
zWM<yc0#m{viJ|-yPn#QGV23{&f`UQv25Sy!JziDe;wIJFZju;!_vn^YWeLS^9e+sZ
zHx4@AgU4{<_xwo^6;PWD^ph7riGp=78Ub6Q3wS;le8ch<J)}CJQQWCCuYN+^sb23v
zOm<{tvWts3o9Tys@4_N-mz|3bv=5Ze<I{Z^UQu0r_+L%L;s+QX+yj)dMD|=P(~7L<
z+}h&j^K*++EgIrcDIl;17!PjL#}LlCi^j~b-lUviX9JDrsqYN$ZEi_f9?4LT9G$$z
zr)5T2M0IKJ!6ATTewFKMFUB)E&3GaT+2eu8>!6BA^iD;Nm<TR}Y8fOR^4g*%H@TL^
zPiAoUmiwEE%-&<ykti*!WX`R;e>8JdMcU2eo+m<?n&?9;7K)z6sc<@t)X(=>z1wv9
zGaQ6OsnwgGUj_ttjRdKOxvp?Cr$<`cM-h?KqxU70S0IXALbft`<CiG_q;j3(I;&t}
zJKZ?=V<>MLji(&6k~OWZ`(3nd^l5-+%GeR+NQqfwScHIc4q<F+>ltX^yc!Y`V`wC)
zpq54-mc4sjh2*N1gxK=}S6fL>O#PzKrxMi_kUdFLa}kOLW>D-t0K&^HvNMa*ZjVaZ
zthkieUIo<_JJ2E6KvHozqdfB%sm53Xy-IbwKwVn>)lJD<Sw&8?L`xp(9@I1EBq7A7
z+`VwYGaElSu4a_2o9<Nsf3dRqt(vAe%gRKqy{YKkQi1@BS2>FoG<O#o9eavK=X`{f
zTsh2o-!b;ss9`2Mi$#B2pfc6kxCW562Z4710$#ei-GPS>;F+i$bXWB6zgt~!_MlKk
zjPDfY=HxO3gMju8I@Po)=L2-mUqQ;pxA{?>qt&A%K!hPgTPRCNEZOGRt9Z0C9_DM^
zsQY;B#q3e}s*z`>K!2d)PmAH(notY&KxJ17{^w8^su9a@>L^ElU4Hk8U(k_nHM3V^
z2dmco9(5Tfnvcq|3E{y)&IiCc-oE5w=YaC#>|Ik#(Zi9i$NmRvfLLn8ubUZ9sWznR
zsev;!cAf=!S|V?0mz%d|#t-xkQ@d5P&5!}c=wbGu@;H~)&VsQa4MVfmQ#s46%{jJi
z)X|<16_GOP{FWOGUqt1PU+4Znk*z{xf}7czsz2twsUi$>kMuihCy1<I6u9mVJa~ZD
zjT(tap3)pr?3_FX7#Gf1XRU*yhUV~;Zr0uxmydn(+j7xeOMafY{K&Gm-+ce<>dqMx
z`R-1grhPK8P`qi1clz6HUbj7eoEUJw`_??2{mVA`eEZY9YroaG?EdDi>iQP8vVGUS
ztDiSGG|w7x3td^d;U=@rmgQ24{;geUjwpzvzOOrC#dT%4MRq0ZZ%q{ZuYH=?|BINf
zzD~?+bJ^rQaT1?K_BLDePZV^V{{iN*GQ~;+-Av>@KOcmO`)pTjMmQYk1C7Y5CGGf&
zIJ?fAdGt>5^O?;Pu4kkOx=(%G-Y@OG`t*s~=Ic^!^)-ph>)9GH$!pDg4dSUn`}Ei-
zBR4nq$F*zsrpLZTr(FNzv%mlT38*=~MT;)py7l7&jWx$iOd6qi!$V%OFL={0NM&j>
z+EVnK*0u~C6}uN2?)$&0yY6?c|9}0WNW)CXDl|kgB0D8X$V$kbk-bu8nxv3~lt__$
z$`0A1?3IyG$d;^-kyPh?^*P`3!}$x&>FT;ZSD&kRdydEBzVFB5aX-a+mHlS><pK5>
z`id@|$S^HoU}Fm|E|#0>D`Le=QYG}4q&LOMQ_q`?PCGXFz7-KyZJxZR@Dzt{^ZE1V
z?K)mWR8?~ivo%gjYH}L+&i|DIgM%~K!-c#p;<lz}Jg&rNh576iIxd+9uQrdSUvwP(
z8n>)#;u2QWFwS4Fd^@G##C*0yONoi0A-mUXe<&lLa`)onNjJCA`fZW&D?8+Ut^?r>
zU}D|IxvbW$X?fQPg~;s-(W~V>R{AWH>wLq~m?k$DC`5W2`Il3h(tpZzU}a)rBA9;(
zh*;M$C>=k}BPG@S%nhQzefFEz4^^#LTC@mDG{5@Pn_WHo$fz^o(rEC9Xq!`viUVNp
z6@Yj*;T4EcVBP3xwoBPl&z*x>?Q`uum&(Hs_!>S3pS1LLN$1{7srLm;%b_p%7dbE7
zPt-n|(M9ejoO?3Gn_61v;WK^d`xi#=_I;t(Mrds$FZMsQbm%Wb<@!KqYxx<Q45;OJ
zZiu((^7KauFdj<i)o-t?+i>^pUGqG~Iw@+8pG{k9I^);W)z#l!rP{o|+S0+4ZTvbq
z5(1yfIyZi};oc>icb9Km{!G`Sp<DCPsx!ZaCgvmjrs_Ogp0`hkT==RB>Gdg+ay(tl
zwAW1UEquO2+<Mrtv)IEx0zD8sva%c?a}{9kVIw2f`%Ma&j~+c5PuJyS?2o0!sos9+
zwEsY3mf-EhEAlmZw-#^69wwJsD{m%WYv*u$l3^OlHUV9&cY&if%}hb1Exh96;y8)p
zW;>@Q<^~%Jn}i(JN(*ebZs~^^IGxSg#j_qE{PwyHbWgk&#$cj;M_l=iagFLe8{rA)
zeh~t8Zw+L}_}xCg3DIfMK5^!ifwoUynvrDW0w(1sj%c^6D;V<V@^%m&E|$FWa)Q=-
zdU*I6Yz?fCxL)XIaARWa4!*cS=wL_c>2u4E_J*a&|L~q0I-WY7v@dMzc-_jWzYEt{
z8?_Ms&!pPC?hnq<I)^CP`V;f6Ht~XenB9qpjlBg%FPCQ{z84+NxCoR4`XIJiSXg{p
zT)d&0d{yt|akL_k-RC^`3%73F;tuO9G4^iKXYy5ur0Cb*A>}buZ|HB{l6A(?e=xnp
zx?gWc=7x+Q@-2zuKwlB}?_W<d%C|(@J`tFiDT15)1fwusv;i~!1id<wszXJ%{#<g(
zvC8f%_TEZpz0?yb${YQ6F@FP6ojNe7QtbZD-bMXpy~AvON${OJ??~tSZId>9U0FJc
zJx!@5MTh8i9KPebJX5Dft@)~pTUGu9myMrni`nma*2o3B+>e%j1Pi7*n~hGe&G(C(
zE0i3~RZXr0`@9NKc?eXPY->K%Y%8_o?d@&X`6{;V*Vrv92bX}c@o_bI=TPZL*5^aQ
zXIbk_W(|04n|=RDmmI5`m#%N`5-4amk-qmtyx>{bi4f-K3r&8$U1m_uZ-qhL2VMD4
z;pm|^cpCY+7ham<4Seb*L3`o&Ng#j<4CZCJd_MfHsw(Q1vT*r@F2CgL#e=buog11t
zi?Vfhbo)9+>6yJ*I2fD%VMG1lV+}-<>W#a18NSbc-a+0Uh|TYl)mY;W`?JGdPC4PQ
zzsUKvT}eqvJ0&H%zOHVeKby0hF><W)>eZ`BqTcByMd#d&_U+q81cJ+bJaf*r)5-bl
z@N(q6d+Pta!kX&UQTj?s>oNFf@OAS;<C^+*oBR>D8MJxr*B&Z$+eXGw0gR3!ubn6U
za5R=CO-(sY<lbv=Zt$L@Io7ZLQm;vH0kB|PG1#(fm7ZO^b|hfoT~@B7!l}Mn!;Yp8
zLdvX7_%D`Nh4o6$rWW>>w~u<<Psupr9usb?mM*tU{k)W8BaiB-%W-Tzv1jeCi>l`=
z_UsPdtZ%#!)f|nZWw7JTP`CWd;uXFcy_1VKivAM{skBpdU*H&^zHt*3C`*K|;E<3?
zw8M+o)H8pd>Ei&msew;blNUL3SB)i@`b|E0l4@XSX-N=dr%~=>TxLH^uMi;tX<7Mk
zVd;OvCI|ba(3lJ2rqY(V5_@o?8)L7P_%1(1F$lI;^y+%1?p%VJ(S7#-(pAh<(rCYB
zyz?IxzzKXzz9d3P<T-sh=f3*;*G~!p^g`RQ$4TC0+9qz9mR<Z?VKIxMidDIUb8q*^
zcjPSd&!zq^FiT6ALw`FETG2D#6Pa)`OOw_(f_3*W4q)N4e$gr}a_zXal>g^8r}OFK
zHe;ij0wl$H!(M%`PhKw0&5z%AQ*=*;ae-xC>kxt-HGvQr9ot_--ApcLCzjK2hm_A$
zy`ulXm#j0Y{)4IbU%ERoujnWG`S~$$bB959BHIuERw-dKdL2>I4^JMhpfS=qJQ3GT
zD^<5U$Ce%$KWJ^uk0^gEdr`z~NQs%5d8e3I+sL(pWR~NW_Zh&`W}ZI~A3vp=$P&l*
z2B@e6ciiajA=A;v?CMai!NEcDtM31IHSfzGJh&B+=8fy1V%CPYBV%K{$U}g|)CAdd
zMYD0EXpv@I=&@!_ufJ!~2e+3I**xL$@9k*TjIO@fG2yAAs#*z&74!HT==H37rsZ8i
zdmb&T%8t)|E(?dB&@d#;HnAhESw*gcl)!iE;$$z<1q1}>m^e9?0THUX)v#m=D+ook
zbQX(5ZLOZ;aOUwt+kaxKU0lCpn5%8G5nqjTViJ;3dZFUJQND5M$no%l$JSL1f6Uy@
z$+@4lNzA!dDo);4?`zb9M~`kp24;jw-39{3@spLCPsp<_zum5*wqh=?=trcy!jF-;
z=ZvENVBOzscyZnBvWHGF9!ML&+gK`^nl(@Ztv~l;EQws>$ackJ<Xnj3%8wsW5Az(l
z4(!^sYp0xCvF@z7ZI_eNSq}_vUcqXp{+V3#nef`2YbeyGA#$#f?X}aVPNA%aMoXA~
zY>T-VrE24*9((=zdfg0DlRT@@nQU8YYx9BfD|=!@X*O?u3n|*El&Ug9YNT-xRlx7l
zpY&HhOR<Iu>(0&1CA+tt5#`VILxwnpGm~Xxa)*{p<jCs4lh@g@?)~gi_m)*<#%FsA
ztQL=j>b)eNBnXM(5Gnb5o#7mw6Bx;@iC=rId)VX7LS=MZ=H39cpVh4%oqf_Sp)XrD
zCP%esyLW6;6-apbX8z;Uskpj@nplgn_0-hlbWA+m36|~Kx1+(8S?u7!re;zW=RJJ5
zUyvg&PCkXJU`+7JONGeTXCtfSA8IsCE?(JtIENxWLs;%tY@^ZFb+d&Dfm(@*RGwcu
zHo{HPy<w^H?OSM3n;bvB-dI;x_vYd(Hwl88AIm<ostMEh^>n<kP)O(}6g7*5v8Lqy
z@Q@Gzz3uD=omE5@9|aj_eaxn(9TqTpI1@5m%UH(OKutwu0SQ_~Lql0rbrbwY!m)a~
zFZE~<pOk~>xZY=494k7&soE<!7M42qGjHwp^A3-W*7+OPubX4eYXvkl>*B1bajD;S
zvdDv}M+OcKqbbX@j%<$la}C}O11kN3`!cRwyN1R>rZtRnNqK%@Aiod780kFp0!k{L
z33#7vqxLdi=<FvKNM6T)d*!?wr|UFN)Z50fM!P@Ncr3%u&tDGd+1RW1aav?=#6oB7
z*}eRUeqr~l%J@_L(ntPr$jQ%^Y^f`vqN3sv6=nRnJbOY%hY9l6HN<~i6hg@B3n}#`
z^kr`u^%N-&4O!o&W#)_idrrIi52V#J>&^Ubq~?EF`t2Q#d)>aJf!)0S`8BV@c`XKY
zg@1q5s%2gIKlZ2Esskp4>qC{-g(_bgt_fE-^oe&_?bNBLuvtWr$A&Sv#%|+z!78!!
z^#w)P2|ICiyf;wUo_Hx4;Oxk=)|~<Zn-pm@gN`XFnXeOOWQ-D}=a#!$TFm~tIbPnO
zJ-gb>kM{ENOrM}c0t2PYV>+3~%^wDRJ&knVNd*lJ4O#a^&&^#}QhWXP`?R&UC;LB$
zkDuwt<JOOjjfEAM1j_JNscG4dy!^l`BEk?65pfBE%p8WI29faM>2lN;Zp!6JR%g4O
zs;jLfX;WhR#5cIi`pR(jEKc=oWe=rBcTGgU$C6Cv^HUt+P9K@e*5)@OagEix-U_Zg
z>WJ-nDpvDt^uxDhD0SVhdrA8>AvK%G!4*~(79SbrjFI~4k(wri=x=z8cqczUjZ5(y
zHPhCu5z)~%aY*ERzSE%RxB^yX_)PS=|LAT+)ke|FD=+2kfyiQ>@Dd*7i;qODs5foi
z3@dYU-Bg{b_Q6^KVPSd{imz&<yaia(aeSE5m4)_sN&<PFy2NzW^Ocg)NSu2-U0`r<
z?djKsLdZ4O5z5r#<+hL3$CmA9VPR1%G27A*yOo{&Yo2i1OfQSK<XG}NZu}-Z%tZ<N
zTE~ZFnVAgmWx9gz0zuL`EUf-&&9C03`ua-f=@bMo?1HQKyxPf=6ol>zEeed3U6?5R
zl<yKgFko(6;<HWGYpNVhFjH-dTNxw#yUEjB{|fwl1NZ`riltl!0;|070<Hc;v~pO_
z38i$eg)tOpm6#oRLvP=Hiv@{@ilQnpr7tlh@2>G1SBlSm3$%v0_M;WSw~AXph@Ny!
zwibJPbiCAW!Lm5xCKG|M%srCkHLZU+*Dw_2CW3&jzi9AB<cQJ;ZFf0}wKkpTmb$XC
za)~JKE$|R(h9wbl^q1CWj7&_tzK%+|UNk)Ki!QpHe{lo2+_=D<3jFCfkcT$nA~-dE
zBf0R1i!;N9wh~I-Rg8N-fwvp?0g8zEZOqInRx@GS4_3czt<}uvB5fCl!2S~7hU`%I
z7Euz2oI!9*I~DE3z!>55?<e|*R7OD<<vSl&R=!#1jpc_Wf-+GK>YkR~{_%7z=An}(
z^fOGU2y$id{gm*%2WDr(LC;B4ch1WFoDt@1D=?8q-|^3x%~!5mvFl3bLUqK<eoIMX
zqdJMW$*sj{sI06c&-rl0wGHQn!8t--%R3CO!U_23vsRUw^7gEvdxGI^C*4Ow_JQFC
zhlig)-cS5s--?ilY--^DCZ@oNWF@#Ec-`n;zXRGfQiE8bFtm>Nv>?=?!#fd7^SyoU
z3*!Q@r_Y|j0>IMx+&RNLbseJ8mw(FvO9x|LDn5VS0l*i?K#Ah$HRWeVV-A=&e$>Pr
zxQaiRo9~>TpLaB@HgI<rTbll~;mH@hulMgWlf}tHLZ2YA!N#|C%2R*#p#F%Ph6Wu9
z0)eIeKc7O&;s+jAR8%CH0JIIQkI~XNau{Ku^;w#_g|50yzkmNG(>O?u+n4t|Jq;r&
z+@Bs^u13zx^8e}euBPT04m1rN-OaKx1(3@i94m7S(w_4@CF7Icb6S&MG=D$FdCl%7
zB+wIII#MG|vI-_Y@?9^-CVxZL;ggbLixN1anD5;84L#y&T^(P(*dYiO!~B$`>%#oV
zNoX_2()9R%2%k@X62xCex*!5kv#Zah<>k$e*vZQ)a12LbJwb5<oiL|pc$CmMv7SJ2
z#XYhC-ZH8LL$3V?CPf-ioZ$0Z7ias0B>rH8BWa>0j7w1B7eZRwcR;J#PM`8l{)t)J
zOhbb)(jO^3^W8;h2-1)G^q@E)%`qs_xJ`A74vvr8x|a%^`6n3mL$P<pae42L9emXg
zhZ^B#VhD9^5$V3ZtWWh)LD)N+(^FyGR+^*r_{5~wVw3Yrl&pr(WR&k~1aU;9NMpE+
z97r}Yv#>}cX?~TBpTB<<VpL<bZnodeg^w$<7-ik2$lNlFqaSOM!k{<;V)^5c*ObBS
zg@^4KW(wzh!5=*BxONx#BuGEHO#MjGg@uBGLeGO#gZ9LHF1+ya!>E;|r4mA>qmBEn
z@aZ}Qp6j&@4LRpIU|qz3RB+&yHqGSXb%uJ#5Ewp{r#J$<Zy48dx+@;Yqczt-Kj#xH
zJ{Y8>rJuqAhBA;W9SWtpNIrRwj$mfblS)T*F?>WB_ytRrdOpv%#!<q5)sx}~HdEWc
z;7bSWB5U2ID#*yvkLTV90(03C(tRS;aRsI~nn~N7)V)R=4b9EUKbNMpwX_Dl=fU+W
zgkl}_gU6*F;~IV#=zSCLO9p}M20l^bLSKn5tEi}`MvWD{OIY8Tn(oI1*|<MwI+Hz@
zVryfARtn0l46}<qMtK-73jS`s%fnNK3m}7Rwl*;Nj2gQ7l8?mq>dBs*6mip06yNu!
zV@pkRN<%rN+rH+Va`FIDBtoiRzxKQ{J^0msEM(2-wH>3Sl~r;g|5Vh22f+^>u)S<>
z=_}mECF`~SLs9mwi}9zcUyUZn({alM<1Q9mSUbIYELy1a_o@StIcZ@&Q@BLK11NTT
zw(L*;lpN+^f^XdtEegGd$FC^~HYwL9rxs-Y!JQ%k!HE#ek9ImSGDc9vJ6r7G<#peC
zTK>`wxl21sV&daX9e;zCs<n%WiCt$%pRU-<ym*rYa%6CXCaRdyehV2j1fyiC3sb!X
z?bhB?eR{60ncaC#k1nlzOT^#d-MjY$>h(uoe6UYV(Y3ZtB6fnI@;QGl6+$Y#p^>r`
zAMhzoaa*BAyxduPU-!j<3n!-Xvhz1&9IWY%EEpOd=0j2XfRmuGuqNkWKZEN*L9CM{
zW3_oGIMSbA1#7W5fARLn$cUxesaUb&y@eOlWb&YPGNFO6%`;S)4>$K!wD8WINBnS}
zhCp-`GrD3PUPh>-3`A7St};Cj{fC86`jqD+6#X_1nNv*^L0I~A*u9aDw#_4-A;N^D
zy*v@lJ%kzp`4fFv^uhVrIn!Sv2nH#HN6b$Dvtbr6ILYSP)vNEY&+jlFdtD<X^6NE>
z4y%-KPy|CNYB1d6JKx>E*3i(f73XRQO(EN2gw3LtB26<`0lFAKrkD{UqB@sBNTs-*
zF?6<=keiDQNLnP1C(HQayIfm<u{JqaFjal+#}?vEpN)*$ww>93Ku!KqXbbCU`C^lv
zYh6;d`xifAN4@_7)_UgCw8e@vttpy2DDIb5`QQypT%^u4OhyTET#{f9ed9KKv(5mm
zQ~R1E{Z?Ef7rJxO0s{%R@n){b$+M_IW(kME^)j;VM_XH4k~h5XbnOd`X^Urq`}CF$
zeUtIHBOfJ5yqg=YIhDRwLBUM?e4En4S%f1mmq}1GJr21ocQ~dXvRb-xmzWBp2xLFE
z61|^X+x^bV=w)mx-ayAGA|39=R}Vd2D7_^PU{-DN5(Ez}!xJanrw-aKWFWM*$pHoA
z7OBGm=s6L>-V3_y;#FE%sn}ZJVM}qO1Btd%tS7Q&X)J*}V#hH~xBt*m&({B0Ok_uk
zMRp5;Ll;eax74)u2CzQ;aw*@TtNA%~KRT>7KCz>rKW*Go{^;>zp)|Ppug0971R1Dt
z3#P_@h`%%+*;gh13JWvy00uCr2xO^->^}IM4xqJITsjTHuHA=!wh;pWGY5L~kbcxP
zZij|OnIZ|X!xQXY-48OA0i{wB<S|aM%8tsfnF+C)YBu&p=`?A>`2CT|$=m3eQ2X--
z-iR75EpdRUyLQ7e@?)Np!xD=v#I_#`<7#gONYc;J#j#^YIEr$eLvlEIcpGj#VU8Z4
zJ>#NNLEzEP@ESNvmu=YGu^s{a^r`nUcj|Im;_Av!UqJCjk_bc@knS>kRGb7GFQf3`
zhm1-DVjzx>pTBmh*gltihu=2}VyC=(-_SaWK5vcfIlKQjChENuweR33%EP#1<4hAI
zR2zkiP>xXZ;bLO4TyURxvqa)*kX7O@8UPsC)Ada6WtL-)*rTCCjkf1saOux2FUN#Q
z1ljH6o##CC4LxHNcKyZD&;;*3dL$^{6!7QwuiB=j8&G)xw5@xb$D1DuX-JBT6j;7$
z$G3g>_}m&*XIo^!8G&u=lFnP6tgxOd@b;KWm#Lm<+~#0)=r9DCwn^Z;ritlTL|ci8
zMfBL&x|ycw2}($DDFpNmU3xC>yY%|}PYl0O6PRPABFWC7v?ouF*suJ8>N9}n0o!HN
zpez=DmpOQ_J@Eotq(g#q<mGD7s;a7^$vS6^^-m%X#8^V6V*{>R`}Hfi4D!u0bYHx;
z6UXl=_D>KQ_|#(KEF6AQluv)kCrR<U>guQH)BA=cs84OROPaID$;pXGNSJMtQq!??
z?)i#f)#h;d^5sV{9yPf&m&!`#sySGOQ$d_5DJl0iM?xG23?q&JJydy{RJ{-MAFig&
zcSJL?sufA)7(;xjK^dqC5!+@?q8#2Pluuq9t!%C4aM)rN(m1!Uu#RA6VUa0G+q6yK
z!Q*pZ{MQi`6%{O9R(5yYN3I+Rn;+{oxp1d1R#3veBU8#GP45}_I1vavSEfYXgN))O
zXoyPx@Pn52C~-G3(#oq8E*&4APs4cTti3ug?WwHu5aObIf*$&qN2eh&?kVv)uBlzH
z6npT;MevM?3kC)cK0HtrbM9`R@T^60c6Ltx9+Qv|j?nhSQ|60jq2+jfva%+b_Z*wb
z&~)_-*NY7+jbL&WPE&l6lKKL+b=hA%Wn!w&-edAmbH>SI`$NOX&Tj7}UmaRC#ukv2
zuKnz0QvohUe>iP1$ju*{FVZ;xn4Hq?>k;ucXn7VD8c$8vjfFWC2=4J^KO4nTByYSe
zvd#VYt)Cy?R8%n;UCx}tKqqXHC<xH8ebG6<7gWR+HW30BJo{kVAJG}AH<$ke^!Qpt
zTTdSL7`wEYC(bb3nASJG-oj|!1%=)=t*!m_Tm?3Q!-JapB@V@mggJ!I{TK{!N;58C
zBEF#zFT;MX4;hPK{!;-k)O-8i4|@iV+Nm|#NNWTEY@vi_E9ACE>x=##I-f?dj9B6x
zbd0XhbK+jrgii@~ie>^eakhni*c9RWagI!x8@++A^V_@;)aa?TJRKH+*(jL%ZaT8v
z_QbOv8fj+OScF!DQ!#AZM91p=%obHKZZiqo_nA>dP=S*EqKm>xt=FD9t>z@Bq_!ag
zo_91+P*A`lR7|V~jLV^7Y8e>NAhbuhfTB6yR=Y+1n4Ip5cR8BH`ZGGeni4A*LT}P&
zXlaRYe~LrVyutPw_~e9-8IDQ5W8{BMZA1hrJ_LcJ7xGIpPjqJY8TUpVD}25)y)llf
z3HRPI-uGNvahu4UbLtMu{C+HOZn_G#!Z)fGp~PB3{2~IseRF-}OK*5O%~0Yzr`p?0
zp_|<9NWhXmMPz)@53A!NZ})!ZZF=e)<&8fZb<|-jliDZGf;XzSPb>n?7_-C(7PxoE
z$X=WQA>2R&0&!ZwIiH4>_BhB;hy8`D9c^2z(hB*Mk)G4;c+^Jh*gj&t_C6oM*D!7w
zyqsln!`pBz!?e!CK`l-TxpY@7N?-OVP>ImLTdGXDWLo~=P}u$G)xiL<HFQrnxO4L;
zWl;<1#3K#y2njWfyuNyslB|BNe*U=bp1Np&%H8A*{Z#AMtI#*1HpYtSek-qjK>L&V
z^F^z5bD(RK1ke@*f>8~+peWX6c^sSV+n>`K<~!3q`C_33aFM&>Dq@9$B+$v9bl?Bn
zE5s-OYSzG~)Gy#t*PA~2Yi%vbv>E*^6$n2j+|T$No2-1dhbKF*UGBUO7NcfW{b;zJ
zE$V?$-lk2P6t%P%0*4f|_K_@+myeHvz}wr~ey=TPMlMV&`n;siH!2H|4}6-K-2350
z0E&MTi|9)jk9J~W)ILuNH@sp@3=CRROa0~LM>K;*&fqZb4HcHQd`I)jtc)9WuQ-*|
z?QLx&gPc&~-=~j-%yRCNNzFH~u=ra{M+hAR#UeuDMO>y6k344!mzDJxQ@>qO0>idF
zB{T06*P#lX#J!|QV_fLD@z>Ag6N-wID0O)NaoAl{kA7|cLA9`|2-mr`zMdq!NYlpI
z0m!hpYf#7tLg)19Ias)pEyz;d^VV6S$1cclrJI$7VL=n|?_QXd1*B37?w-oIw<UCl
z5OM4lA<JNUHRJEIAH2bUFI$Ply#wxg|7)_8L+Ad93q@Yj46Lj{-@ki~jE<h1g-(Ul
zTL<B_5@0CpP=HKByj)%l6cU~iZgD$?rP<HIqu}Sb231Hcj4UkY7Ax%2!?ek3_BMPA
zM4oT1yOOuhb_EC1yfP`$liBdyM;_7o26PT9aE^0biCvQs4<67EU)oc`izzrxYG@!B
zyb8TMf=yVq{F^Td)bT*UH$1_~aile-2^78(!hk^3CV)*m+CV`TuHko1d+MS1oa<|A
z4P#@@`Gtj6@I#2+ipe^O5NYh%l5Tl!&I}KyNBi~Q`_){-H1XFJNYz@&s@ovhfUV&-
zI$g9MN<w+`|G5<P8=VE>z$IIvRH+~qy?_7SQr<xSmBAGNBpK!yX%c<}tnd&jXo~yM
za``Ea?14|pDk?PC7RVtK%#RfikdLAg#K^#aD&+3zXQ)G>kUM~6Bvx>M+!hqsO<q-8
z0Nt`f`>;5BD6@XDYSeq^iK!9w26lEY#KoxzbWZLdE6WQDkxQqr5?TfBHsUV*oIs*A
zxY^c1k@DV-^xzbW9mT%O2XF`Xjs_a0DZ9Fg0MLfuwxU{EAn`XSY;j9axNK^FUoQlk
zfL1;_(7e{^&t+x0%d0CZPtabt0-N3UC^)^gYo$Cg5!9WWlA;z`d6T;$&mhzM%6<CI
z-|C`k=CbDH#G@VNCzX`aYp6|uK;2VOHdM`@d)NlR3n$R#p|qNjvu<u~m<%lKnLr0X
zIov?BkTHVt0kyJqW&!AsNhdm;uVs|3s``AnFMas%VN!!xU7nqFRgXprd(!;}?Z6j*
z9AZD@7!0c4-8-zErbj*swbE7g`;7fVrlqqdQBX0+cjkh4z-1VKaM%sbo|RQ9-$b5N
zQo;+Hof;nx0zw|b%gye-cSt~prW0lRByqW2+WT|`w&l5ec6O_M1j1(lZKRE@b2mpb
ztrhq3m)5s4I}*11FrUsyLTWfk-=1^GZ#^Sp-M<NNL<c93te8P&dygO6F4u7V`t>^x
z9by5p6GAurpC>qa@pO^<QSyVCgg8zHTt-Dm=i+a3b4O8~_y%Y`fZ(E`pq>IrOv-m@
zKh$fo%5gh10PCTmqEe0yRndU8OMnOEH8mTcZ(zQ1-Qi!~PxqGkb3oMe5ZI$8p!X^m
zr<sz!P@-?;%IS1w6!y-%)Efk8Q9*Y|`IU9rXc1UhT~&m%E+@z7QnJW`=@JfWRINaL
z_XOS3t~`7&Su^3VP7hMeO^CEBD=Rx8J3;iVgvEpF?=|n#y&fk)_(_A*On{UvoKg%Z
z&PLR4V`mRRrJNj6Q8{_Cf|P@W!^txVkcdf5vbC4|$UuirTJ6ZQ1-npINtb@5Yt*!I
zOwV*Y8rU(UH&Ky_KA!Kc;>kaL{2=#X1uEoU-5UcFC^2H3yE?`!fF$>oEiIbt_ma^>
zTNQ1VZM`w@=?=wS8-!pY5Ea+%4-OB4gN-$JPx;>w0^DTO-X4icA|g3ZCGWm{8v%5e
zJyn#fQy_d!SU3QfDC4J;<BaepS>(-MWcY+2LIh5BW$^9VwVt>e5phK|N}!JWHhQbj
zrhWs06j{$Yz!8a;o;crflh{p3Nsm-fZ``;+Lql@}v|XNyj#2uFPO>V7diMto(a$WP
zx<z#f%qIDdQJ&+zl0$M*n20^+e7=_|5D{k(72y|pX%9Qz<DKpJw|}n{FYiI5Gjg>j
zI%hy*7mg|_nKru82^-~Y0U=oMl)48JWdW|gg67nS$B$`w)?a6U%)x^+d<}quQ|=N2
zK?*Mr?o=8AQ1c}MVf@zRG#9o*ft$#La@2WCNoi9#5BuL*k}64Uv*4+rGg~ydWXpT6
z1+I%EpF%(77K|({c}fw1v!y@j(5R6qlYoGV2d4QyNe$#du_syJ?b=!uv{M{O*74rY
z0$@=IV{dX3dIr$)ZZ&+M)yf>60c_IlNhxnLjetJU>N3(=Z9>=q5p-+b3065PrL>Ox
z7ZKYLVj<I&V_oK;&3Vgk^7^PlBceq^23cJvB{8y*?Snm8A%tTPI|g@x)nX1LqW&IF
zhiAl?gO7>-{qE$~WAy%CCA7bZ{=X|~^E&@u{`7}l<U;PcaI=xWHbW<mYbd2FnqU1d
DQ!af=

diff --git a/reference/plot-2.png b/reference/plot-2.png
deleted file mode 100644
index 434c6194978ca8edd7ed96091d44409045d9c858..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 27051
zcmeFZcU03^*DoI1sN*2x_>4#&hX6VVs0c`Rh7pViD7_bvUP3WIXu$>+${4D&jG~l;
zD!qiDC>UBq2mwL@NFWjj5ip4mlKj4j^E}VI_x;^<?>~2~_m=NkX{YYJ&)%QC&prq4
zU$HWm_*Ujy5C|jzx%j&c2(%{_1lo0N-#0)BIrqsY;NzQX7UsVLpTM(e+JOK*_J?0|
zjsk(?Q$?>GnG0*aAka@B$nU?}#TKnl;GNfF-04Ec7=4KUj-B2QcTfI$E8_Ltn~F!j
zmveaO`*SPE>=4Xsw^Vq~)$N-f&Yj<VVXySn8@EoDet7lvY3Z%x<djEscieS)e*XR#
z_({X(?Fh7@``H!-mCPNhc1=VP`brh})ltLTcw-o=(u(drvsJQFOx^!``(Go0b8$E)
z`TVtfI>G-l{LQOh+^gfFWlazuFLQFs3Ltu~Uo$WnCJ&N$exD5r>U--(8Iz&g+dGow
zLs>5lh)J6LxG#2x=$k`!vXto4>%#tj78F+P0Ch{X3q?(n(|`E16J#(FL3Bq$D^^`7
zUHU#R_*%q1G}n$*MRmdYtZ!=-8WPko(4wMc4dP>B3H}QdEqD?)d42n~Ypjv^lx?vs
z|2z!E$Y!9>Qg(z|jlvOn39%GYTU+CaZP~&wS}-jkPf7~yCD8&tZSA>)$ugo+P!ng#
ze$TLVJ=f>Nv5TqgDvdke1B-yL^N1`v#7rv=Gc;8boHysv`fCs2%K%yvwGH=!jWQY;
zjFe?gwFK*tqeE@2hUafsa^NFYAn6>y;we&(%x|f;&DUB<-w-158198fxjqu!AI!F4
zFt7AJH5H!qV>zv5H%B|qnAiEbXbrdW-aI?iu)1E;;Tme^aO-HMlp~2u)9bx4n$jKM
zkzTtRo{lr5x55RTo46k5BQR!Hf*DGvRZf6NC|;>z>mRhU_MkV(rE4t~v1(-H=F9y)
z3rQ)Uh$(LL4U^^D61Bo-661}NC}D|o?c90xGuSZusWjiC<E4ufzl55BLjg;a{3GXe
zhH>5G#pJr$IO>sJ#D(S5-w?W`EMX-_cW{jIwyi7r3^g5X4Xq<!$P-3KI1U8Ee$-34
zjNiMk7K6*f*hexZW8b^Thjz!Gy~+5wmNGORWwT-0g(``vsg7f&MHwS@C34U1ZVP4$
zHG}1FImuiv_FM{=6qlhgTMUqNi)~Ug!jf42OwQnbekx27A(6v$UQ^R2FD%F#+A}y<
zT{8cQaYv8-_{UFOHQsHh=_Pbe*{&o(m*qHjC@KhZEFo@WpGhhLHw^E+L1E13IM2As
z<c8!oAxeh6iF?=;t!qnw4565O(<sl)aij2Fcx~X9WS^4S{1L8K4|13JR8V((U3)9r
z<R6nuJ6O@3<x+|1yFjjWkE9g<RGf>@YBp(Uw>Pb5Z}C2|x<teU@54|#U#{V5y)FBz
z+oR06Y)*&v9Zy+UTu*yW%M=Y>(l#teb(AcwXF<^h_I<hu@`7fGU|7vU4iR5i#SAdQ
z9-~px8ZaBZn!c7I0)r4etmR=Xj>(HS&Re?SdrzH{C&Hg$iCEGS&q%8o_hjpBQIc@#
zRL5rxV!FZhhH=eYyI%R^jwN}S@hGYW>7>^nGuE1w9V`b(Z{nr#v=rgXR$j_v;Idw6
z+;67$+H;q6cdIq{<zgG#Eloi$TD+TQqv&#ox^{M2)KJ08_>XHbSy9YTCoOKoGEh8c
zc6G*3F9~WIdj@K@Z0R=^O~E?_v*FIch%dO>Nv+=qsECr&l#t#wNh3ksoF6=OCzh;x
zCvBO3nLfFaHgv(GL$=x<(UZQzA4^RQ`7^iPFw{(9WBe4=F)_&u2Q3<*%VK!VYrYT<
zZJ+I!+R>On=}wcD;WizE+c6!~(T82VV&&9T`@Svq*_v=8ZB5ca>a4-DFSSuyMxkaX
z_zqTwtlxY2rO>D?xvJGRGgRu3o}fs&DN$HtyfOYWe@Z%0XT6D6y)`0DH=g*^^tmZ1
zCf?PAR_w2Jllrull0Fu7ZKH?6_8ZP%Hb(_LhMCsZ7DjB}T-3b(LfU_*GI%`oXYH_V
z-fLLUbVzrHJZp*BDyGS&yJ|At#I3mfO}!A_JF!=Gysg+5VcOm3fFrbUR$2kdcKH?g
zQ-``W04~d-whXVCp}5261PfXjCbZVlSWBO_*x@$%-N=^0QRrZD%aM3yjU(Z{#N13(
zU6=vsQVAzRTH7e0cTF=tVtdrQYU#B!{hSlU#|n3P7nL0r-ZGi|PZ1@n*L|Z2s<Z8k
z?&bK~qeAzcNy2%X(Gz96+HsE%14T}hYrS10dsM+Fx1XV`#VLMQH%{~GjO*oa2}>KY
zTFr<^E7a)tMD(?WyqqtidOpPO|B)2l+tmRqh#E(`>QQ^{Lv-fQ5^r@;nAY1^5gcRW
zZ-*e)aaJB3JX`g~fVJrM@UxIGAvhcyVo*7`JS*Tv^j9*_dOV$uarQ`i^y~WO4I^xo
zSS_87%}j1f=Qa3^2prh<QN7qVMFQD2=PAj-C1F#roJsIEtj?b{r}gMx7}ANH;iZOg
ziARfkxthRzyXi)L>_lP_smWfqB#P=E`lrxh@iWT5K}$ypCK6s%pRa^ge%=zzj}+6O
zl#%!;ZJt{HGKyK<^}_o^?g2Cox=&d9@j&gIb1+e%D#>s7i_h?E;w$}HUoEZJ7J$N!
z0EH7fQE0#!y?N*=^fNcte-d|tohjixa3WYvD|m<?6DHSKVzIbV*wWeaXc61ID~gkN
zt(P+-FbJa(0GOKU-cV>5Eco?Y3fz+~c=yc~l$L+`BxB0RCC2DMTyHd-ZJ@<5`h1Tb
zocPn#2LBz3FIwhy<y$*IYf32J|1RLANbcou?~*=6MD^+i7Dq9ECvTQeMV{r<h3*5O
zm47(uCJ+<G9dD90*hExgeMdXAw{x<jP>-|_#TnvFap?XbGbe-@2Jp!}gZ0$Ql=_v%
zU~XQaqF-2^b3+xPv!|Bb$S~H*;AQFgu&O<#>hAQoSRhyImiZowduqwP)!ffk9lbk}
zPZ%s9z5{{K8lkO|BP0hxsMD?(tySCN#MjM)K|R6S+6T1@X?nax^^Wjo%x`7=H(QiZ
z!T>iN|CzjyKZQemzg*vSB!a0>z1LndWK%dUEN3VMhv1xHJP7ZY)|PSYrs9L`D?RTI
zVixbfI5U2SeL}hMajNd~pzds6VW}j@OI>&+5z%w`MWN1}Z!^(JP3^J8T{E$-H!_Fy
zI3BVU3krbXy>{_q9>7gfp`rQCF=sR3JH9v!9eBsW-V|`<Fn?Gc-I#JGkX7lf18u|&
z*VQ!qZKCGTL=2Ti=p#&+K+_kn#M;40p)FBe)~{!k`KhTLYK3A(^7VKbwZl`dod)8p
zp>*wo0F=s=hK+@jI89+I8y~bYgzwUPBz1(#6jhy*kLnz7yQ=x#n^lEXL+!_GC7$$E
zZH1jFCG3yd8gv;2v{O#FqSvHO?kMEzL(i4ZRMS7ZLUq`FP;`@YAvT(Xq0>^XgpMyK
zgbhf*YfVaP24wu!YMNL*L$#ZWT3gjIhZ}4cKNrUIyeVx{@T0t5(OmqzDH+9ZT~-Nh
z(wZ4Hj}rMPFOmH}FdkWvJ;n6&A$LwD5%L>Y{)PM;hQBU?Ib9quPM8}B6?}8C+|GK^
zMNf9UNRKBVr!nKvp04NBeI{0#I!Xp)!+vQVx7lYu<7m{daJLgxBM3=nmfInWO9@$p
zimy<{{xg`xl@4A}qE9@~mmA8|4&&<HSu1QhJTvO!2`rxe?oLp*)!Iagbcie#|1Q)&
z1VNG^dz|twVpR(M^mdqiQhR*n*|DFtC-wMAQM<HP&uRuRuWWt6cco9~FOQ|vb|c!i
z2=h5QqYoEX3~INqDePED{%Y7r_(t_!jm{T6N8rq`*EoAmN}U<QC8>3;H+bNT+C3a4
zjk%FT*G7ac3ZcE6Wjh>hxYZ{PD2<o>?W5^3LeY`ZwuO@*t@?$M;^QX8ky_o(q=p^~
zMZ@yA^X%sm4gCj)jHgFGv1b$mIF655x<1=|gG&vQeVb7peO4Y%pQJRM@n^@Io8F5m
zJ}4Z_gzm$!X1@Q!QW5vKa4_`E27N4{F%z!C%h|r*2{=9a5#o<gy@>bvweeRog7-bq
zU60=^l}L;E4K|b}*egS|R<+D9s^3asAYsD>oF61SaW6?7BYjoMGC}eG@_RnrPe9&k
znHB{`KA}+dgG0k><0<a~a*Gu{tp@HJ>D^b{tTt;@?LEHe7u3<Jd9b!jaYHV&y92;y
zE%tO}d)zQQDYBT<b=*X&ylknjO1jltA8~u$(j+NjxT}5_bsmkFyq<`luHZ@+_jHoi
zN<2OT+8%OKSq7Bs^X6ts_{RciUTx?;8==<$dMvCbt{gS<>gYgp-BvS$l;{3d#m$A%
zG#Ar(ND=A;g^jm28Vtn&67H<tYS-f-+Pv{8OwP81+{O}vrF)KQtvj6-S}ORHS^P4=
zl3iyhC+ugq7T~sDpTvABQP_~#3(ELub`L0Hk7n?fxGf$@ZxzijkbQ+?XB2brj6!t0
zJ<8z0R&+tI%x_%=_E5tI0%+hD8BUe9ozz@4u^JOo6E*EB<3UoD@)$X%6UPkeT`t6Z
zv(_NfHI6Gbet(5zXdODQ3+Rx3$DC;%yU5e%DN|C)b2QwxNe!5+jd)u|kfiTouxW(7
z2e^U!7tqN)AkbyQvN^RePaSv><@=A6r{3;;wlZB*UL}JfadXDg0pXe$l0}cBTwih3
z-QD6zKBdJSOsO?MR2>6vrl!<d;+ojFaohtc^;0`^x=v@2A`><a>q|j#b0nS&cF64f
zlcc+SIrqnsr0X_hBMd8G-)Ta?7X}xD2Aoyc{INNw=8Y*z1ukHpwm-#b1!0%h(j?6+
zWM`+;L34+wyP5A(g6A-pR7Y;*a8SOtFoOX&s8=z^o!ok9xsKXyyCkc-n!JxJR}$kf
zCY2?mR$T?xF#u;h<^fRq?>^mDjs7Cu!w1iu``{!fZjVj!1cCaJu>%_7KFOs@z(N6!
z)770G|J_mk_pNYvTjDF#T~>UCD42ea7V|OTyXwS`X5VLcv8{>Lj^&}((spHa?KQ6z
zc?VFkw$YQGmn)AlcM)5l;c`1PMTOluTN!?-9iy3a`*@PH1>n7br_R!<N@F?vIB$;8
zMO-50*=1G;U_ru%g^KIne%%>2<8K?ty-liWAC~DWvg2=#MQ8>^zg7Dx@Cg;%=-hdZ
z>!MssdEC}my;S4hs-Vp9pW?9+azeGlxq{C4j0w$uc0w%!HP@Xwa*c?!mh8<TrLDiV
z;mvseDYtGit#zi55hJkk{3qpb$JAYuhjm7vP`0XVB1(T!<DWxmiY6(w<WL;>Y&#*U
zw#Mo&#?%ECK%ZLu2Le?GDXa9;d=a~#=z2hDyrg1B?RaXLPla?$Erx1jr6;Zn8IzVT
ziQqj|Pm97bOdr1#3+VESYxIteZgUK)s8KS+y9j|<bu&sD_)(lj=(}Syppl&ODTk;e
z;}6NSMY}p=2IW5j;uKKb#Tm+odSP3?v4c@*wWZGPjANQX%)-vlg*8|5mMGDcO!;V*
zjtZ>mpWmUhz3sFkH^56N>1kXjY<izhVie3O3_u_d=p<EuwzblG&JwJ$JfhXf9~Lwb
za8~>b4{Zhf43XTSDZ6%);jEMnxQrbvqEfo3(<u^O6*b*PFX>o0)BNUwl>3e2v%Y!X
zf2t|zh>m5_NSY1b1Fb+_ZM!=$bRuyrtGSRSVyIiAIYwk?z!QE0@SLJY?b2TYeOmRh
zBi297u;dD``vie1c6XBJcFA|0kP-dr4jBCAFBA0p6ju8X6WnsH*V*SE;*408@ebAp
zcei4rK!iQ)&5=HXKa>?kosv(E)X59{(1{zqU*#zmzq|7xH>HvzpXVSREqw6MCk^Ta
zTV}1Z%a7uDPMddk#0m|G#{SM1L<OKe1zl&?PJ*3eabvB5cPC#dO}sR6y8!y;m6bAJ
z%zHlK61Bk}B*Z=b&C!2i@xOdQ(9DYiUdo0~x9<0Do}VqtGJ&3b<s{;&4zoXY@?N~1
zr}27zd8VOwiB|M?dgy#Gv8OxsrJowN)*0N_6?>;%)Y{AO@TV7tNID)r@$ipC+ec8c
zJHc{XRQ3Ot8~abmvi~9lV_m*d=i||QBAysQ)Fv(vbBHgM$Hkhn626nlY$@L=VC*0w
zs=BTX*Gt(W?D!YhWsIr2jv^xgReiUE#fb}N%Mu^U8VXkM^K^^_y3YZT{}?N(K0iK7
z(TGRSR0<j==T`kOnbrcg0coNIT56FiN(O<z$EP)LcE!5v?NVZ<qo+sl)Tj8?Fjy8I
z6|YKW+3_#-tWZo9T#Q2Je%vK$Z%~pPvliyS9~;|H)gJtiXvS`4L?x~v5A0PX&v+O`
zMQ#%=Gle6JBM(oAlao$A?xennAG71>VmvEqCTT}W4%XGfs431z^{+fuodfL3C>+ht
zR?lFMNzaRot>X`1maYUee*`GJ|M$X!mzy%lADl*UwQ4Qrj)*9#`yjG6MkS}X`P{gC
z`Bd%X2eC4BKR67S9m=W+<yah^Juhdngd1-ab?DXr&$@R?-H6q@Yv2g2=nBd5neK{n
zhvnfnFz6Y@=lY^XUcO8*X66ugp>ZXKKv*SD<Y5k=b#<l_fB<1O+7spi8%;USH9(UU
zPA}fZ3`n_)GL@jZhW)WEeKU5FK~4^H@6yM^?D;DyeQnKgPV5IGzlde%kgF`yOc2(D
z*%`Iz*WtSy?Qw3DXIq|1B=M?O9S8Q!q;d=@cpn_<Nhxi&g%vUuiGmbmd1D)FM|<m5
z2Od=@aaXMIK1InfdNpXO4!r7?kwoy$%+vF{u|vHgN<cq+dU2ZWm?;Y%B&Yw~nB<A6
z`U$sNt6M!_m)MhsdPO2ncDL*vFf;UDKiacN%(PRK4>S}FG0`MK*BXi74-VL*$9F|V
zpaJDscca>x3{)d`&_ImA=bD-B(PDf8$;toa?jDG;&%jf4JQ!F8X{AinY_Y=r0we*y
zoKJ;55UbxQnTB7*9UZczJITu*7*pyQ5#vX{PK2=rU{cD_-+Ic^-o5_K+QI91@en^*
zoDE6EJr0^l_!oQvI)wE`C-yzJ7mp?v1?WV~h#fCD_wN{NV%<b=b~q#=tGm9>a{LHO
zi)gCK#F!t23itVlia~uKz@B39_k7$9{!*iF8-5qBTKB&>`cEwWmoJzC4i}vE>sGGL
zoKxI49k@=9P)ZXkTSq~1UV5w_yY5ln_WGkf5vR%S<|0*SXs?R!qfnpgW8w(=#NMa7
zMrGDl$-@MmHn*LU%#Hf$r$HULqGJc(sXG2&UJn0K%Kycbu7>b1wiqC=+AR{y1RU_=
z%byvpXvJy;2TFLlPX<Q8J+Za@=~f5~#$IO>F;5&8HFVm;%XRP8Soa~0k|qSL4CgG)
zequ1XLoQ;F%1FxUV>#08NgB^CiQ0Pk=D+MP5<~^3cGwfD6wU;ibZ8!o$^3keA!UXZ
zhkZ`o0QMR*Z8cTW6)O~6`Qhvmd^6p#rKStIzO7AU%H#&0t7*4nJKFkehx9~Bibn0$
zPyxUZ%J@Xse9GWR++@`{`rAV*tS}z>utBWstC?WcOs@7x!lz8zd1J7c4!{j-P<1Vv
z6m8}6<C5%qE>qDBstDuexw0Vhc0b01QlD7UqK}DU90Tj7na*Cp!;Dx#nVyd&7q>B3
z+8blV-j%*_v-?TAwY>MqT-H^h?#YgOq0xctXAB>Unq6K>HMU02+t*$km;_QhCG{&m
zCkHu2u1aBCPj>tf%Aqpokx?O_=9_@i7%mF(Z0;4PWyU6L7S)$!VkfUUHo}FKTbxII
zokMTu>*a7)i|bZ?&YcaF7^niR+~5X(iGCNh!6dRAP%e@3=M!P<tO1Q{5Dy#v=!=J-
z*$GuZ9vCz~i&MgwJ;dWyt!=;c8Q13&ROdB&XI2VY=SI2Jalf;ZyJ8p5V3zKnRgKr~
zGbf{H&l6VQ3u>N|vNK<Lr;4BG)D>7&=r30_Y>h+t<G!R9>LqJw{x>!k@*L%HD=x`F
z>)dyb{Yxg{^x_tpO>-Yq$brV-mGiO@T0843MJ}!;#kv|}RcjpaTCgm?mQXFC;)93^
zRF#L}d#9MdeX)u5OGhJtz&hS>cc+wR!rGH$DX-_#pUZTaRBjjd#lXZ7oM0K3bbNMC
z^dF;E2BD;FqtD2^PTCb9JSM*q0V3O$xwoobXl!nXcw1DjaF&>RMr7OF8oCbMyjv-c
zAMc#k<fPg%XR21uvbQ<J4Hc%)cyq(K8r!vMslu1sEUo`mtn&_|PB&NDh_Yx4v|>%z
zsrRjYw;UZ6A2(Xp07s9taWU*+M!Q5q1foj+l-^>ZgznE@rttH97(`g>;b`J>qGy-|
zTda>6+MG19tnT1>{&nC~G3>y(J~ZG=y*yhNSE|uut0|J5B&McfxG==v$<{*3EH|Av
zs-IBWEVKGuBJtHPVkhJ9%4u$yiB@$SjcNQ+#()v~Fn!w7h9NC+7O*~sB55MYp=V_p
zZBn6N;$O5&V);|oHQJVF^5h@S0$kr@(9z4<(`Q5CHj_s&p4Eb;tqq2PQb|lsCk6=3
zwuqr<SqEX*i^Qo>j1~q}<Ke)!?3wD2#8_bjTu!#w6jyHW%oJD*hVZh2E?U8Zf7vwO
zxo2f`uRW?JseTK~IIKjf8+F%*4{kh40Ww9R$8X;czt;ToE`ST|@|3k5-4Q_W_jTLd
z$4F2oN6tKP02A-fr(P$HgOYuQqfh39hXTQ&0cKfe`r=BWEssU2`|f2q8dVk6E}{Ju
zq&t}&2e=Ky4GdXlH+Hm!TzB%NGI7S9ZWrI&AdX;??O#xKu-4>D>UkYe86ENL7UBhB
z^%0M+dzWvgYLdC*5)DzB!A?=!qRPbD>Q&cbCrV`;bqq)$v1*1VO2jGya_IYc7c~o=
zY7ecZlIyxw_h%*MIpVGqJ5x&I>|uY##j`6IM~JQ}z~)EqD+ech_IAdjYb#2~r(C&r
zB^tsd;qEXYBaWDb&Kx{SzfCkPD?tprg?$YyIun)Bqg;AID5%i<90O)QBajHSSDbpC
zuUHf`6K+BL8VTI){WjKw8osM35E|g*GN*~bWLz7(?W!)?y@F<)bR4?l-`-4_Y$q%o
z@Sjysx6l%mg9c9l3tz^_Lpw&_Xn^^{m;Ty+$=L$Kv5en~1qBLuNA(p)xuEV5Y2b(@
zg6N~dw7}4zbu%&cx<!pBcXuNjBk_QJo%K7Zwdf?+d%>jT=MD(_GlZvzCS}-tMcva-
zAW5lUNEAG@{2c$rj(1FnC}y%wE$zoo)L=Bf1Sl8Uk<+Kdqv?28C36qjG6|tiXMVJ0
zHkhHs5bNc56v$nMq7#`aTk-%9LZ7*an@8UTCeO5PvY!4OH(x%5lCJzJZ+Fo!fW?!6
zVXDif5GFS2;`SY{oOZA<WmDD@zvC29LAc1jk7H64Wb!m__M5Hgjx9kY9dYXKoyyUw
z)<y_t_88-{b_VuW0KaR?nvJkqA1^kEM_&1sXC}9dC-Tu9mrI0WG~il*HL;0^!-Pfu
zjzd%>#l6=5C&hlB9Ng^_-PxT=&8VvsocCmkQV(UOa@BZAY*F1RY0{q@<SP{Hi|vgt
zThlUq31Kf-8APrFO0i9B_Q~$vgUT{x*wVOZTErz71<xTgMt}UPv-&Ty#hkV~?c&)=
zW#jLQyMUVS|F^rsv8Us+My2iUFux|mVSY(Ve&oc{J_q6R2;=Y1!Pn_|X{W}YQX01S
zDYAQ3Rd9~{k1rm2kz%9SOu&g0x;rGAv)d));vLwkM2RCVBf}exMAJNX03o}7IchMA
zjbRrtn$0IYyF0W1y9lS9ww8f+UT8(YHhiV(chBS`=P@_jiR@hW*E$`)!f-lAK2aM=
zg;2&Xyyg_id$n%S#{dQlDET{BhtZzW5r)wjsEU%Qb6T+|8@?BtdZ>5mh$P}mj{$c&
zD$l+5m3!RM^WqpPd_d#2JLA>?jgQ&38GQV@X2RTo__+g~txHK<oRH6Zskr@ek6q(N
zQT%F~dm@`SpaH~%UcMq#?9%)$#)SGkjc>}HWXxWqd&7;5-k(#`l?@gO`0m_xl6$oD
zff!~JNuy(Tbp5g$UB5VvQ8}RDGiPPE_PUf2v%S{GdVotsb5<rBR(=4ecjyO{P13Sy
z=1Q@x{L#P=@rsO}>UCTVcCz^N2g|ZmykvFZK6^S_@`G@}&rXXiJKqOTUeB^AQrcM+
zai1uFKEaY5Zvtl_P9Sz41dKTTr5&&6QLoq~d1U}*fT&1{c`r~T;+4|Bs={}`xWM5I
z|0-|?<rQP)P@h;6X7};TWat;N>a89d;!N4Oj8UR2h7cNFBs}+bi-uNac>3_z7+(4f
z@tmNtw6?F?1rRx)O1hII9Vogr00QX*0Mu?M0mn+f)BTO}Q!x+!zPdB!WN}t*sfs~H
z$2<5ts#-RqfOtzCeOQIbwc9IY_m$Msfh)<}&x}39MBVTe3g&<cGt~~E6ZzF>`x;N@
z1++%=T%KME=f2nL!?kYFp7Dc;C5M8=bDdjE{}DLi$iEpsMcY(EYgX2Hx@(sbh{P;a
zrmJ1df53zq`GWy7ZblXD*LU#+*U)_Rzr&GnB$}VkD3IXr-r4aK{$3*jszoJ^e9<WD
zK_$mA<1eynMC&k6$;mO->lGYgh$H}-&<221voJ`p<$jW?Rg=koK#}G0sdB}7)|4JU
zOpBOA`4@HoD^&}iPxxWJf)O!k^6%xE1wkgI`Y^^OAas9wlQXcizYO`X2>-A0KRiox
zfqBFFr2AGeUD&vRv1Ub#vBBS<3rft=@61?bqFWplF;*q|i<%fUjLFAOLlc6sE%u@1
zuT=lo`(12p&8j_rYOGmSF=8-LUv9;LZw8xT6#TsjrUtFxN@?GE&KM1ri-Mch#rH8%
zRT96B@htIx43|$lteVM}SYBTZb@C^<bjSkfS`}gmqm+nKZxO4ESGjcjH!PQ$kbTF6
z%5$r696Mos926D^1Wq3yybp|<18U+_;6z}v(kz8(Rpns}-<~R(&%g*;w>%{*+~k%H
z5(m(9`o9uA2V*d+y6{feaD;Vr*V4UEXe~ARb+fsHiJl8gL|S*hOt1viK=-6(oxN1X
zsa`8BY}Koo;d6_A#KUDJ!d}|uLKa<Bd%AJeT!nzB+^S||&FAbF^{grYEKAp}FZ0y!
z(PL{?QG`m)UVpg_rY+M1ll1W(-_g!w>igXY@^x|>F{<~tIBDQynH%8LNw#jP&5s!5
zmrt1(`Bwp&fZME46T7~Q{XRppiNDJpc*u41ylRAQAHGgTV5o*R@&+TH2N#WqFSRq;
zHtcwmsNT34V5i?~Gp54C42Km-?hZ42brb{&L1|$MMf(WOO@^m#>`K1S!Q)3q?P*Wi
zl&@4|+5|aVBApFn&u0r5Pw^<(h<V21Lp6W%_GeppzEV4?QtAb})}LA`2oDpr(0z6@
zwtP|sg?)X7{C$vPu}9*ybRP~tdx6&YXT)|5?oGSSydII8VYuk5l(`qw{be|4-+IQI
z=VObTZawP(!seSkDA(eacR|r*!tqZ+SO^6eRK4S#@|SAHUY+piB?-m$6;D>T&tMZ#
z9?d=3@x2$_bzJq1qkU$mQ`c-4QC3#)p*;@5KKUM_FZYRByM+pt+t^~rsxS|u3527;
zaXO!rOYK@t*M@xgeY5pYuYSYaC@C}j6yWI}!Z1MuQ8IlfEV@Co^Pma^+!!{|5%vC5
zFkEQ|D-<(qxG`v@zx3?xc=KVm=t*DqNW8K(oRapT+1|e>!mPX`j`1ScxphDj*O{3>
zQ)*fZ{%C>b_bDP0S~?fhr5s3B>vNKbR<)YeYf#w0?#}LSPHw3Ev=^fY{o>MFEgL4o
zd8Itwa2zrHNgp#@>*3-a{rWV472X|R&Uhv5L-9uQ`|H1*PL^ugbe?J`^5kFaTTKfE
z4ort9{ZSNK)Yh19UcI`)=38?PFiq?zzvR;$)f=2if6YIi7B6Vm!YxgN8z#7wGZ~Of
z_@GZanX=lva^}eDs{@)VYnh1cGF~x!@v#wIuVdbL9{4@q@#&RwJ#bahm7{-^RTP9E
zzehV&k?a+V8lg^n`&5#mG3^5J3F1P@%#`O%vaL}lW7EQ{N}M@z`gP!7Y&ofrpYqte
zy7@Noc6rInb8q<FKKpXZW_M!vXk(@W-A;4$0Q7o9DdSd(vO>_l+bvPRT>{{#+of(W
zTkK)W$$oBM9}BHi&|`0BGaND2j;xBL=f^q|FPTonhXJn(JrWAW9*KR=SoTCpRG&DG
zn_%a*eFG)Bmrp@GTz(|ab7lRgU5)LlV;s#w!|_O(`)?+0l=2=6W6?~Yj)^n=pQykc
zom-}Y00=Ck99I+O#&;_0z)$dG{Mb}<9BC&*Y-tYfoU~#Qb^2ku`k0krR}vFpwUx}h
zXi9{^;8sFU`WSFe#FCvul*f=AR4qB4@4shsb$I)2l;_%<5KUB93OE*HjHa7)j8;vc
zQ?ysh-Tb{6?qzZM9#q&21Yy91m$nH@G|u&!yr>6AEl>p0+9z>!hxVea9f1J;jW{0f
zz6u1eO@rb#-8!y|e;sy&-288&vVUEDO^0PQbvVSE$pq(;G-DG)J(F|J0Kq$h@#Rr(
zf$zwUGorHYOZETE;>iGkdW-g89oCh7*p}>R%XenK1Wq8I6L(>1QTmJ0S@WU}pou#G
zq?uHPdGM*U&0tKby`uB{M7h1coz&UgSQ96sj<}hB`ycg@@N4WvhCcCv#7m~j-0HDB
zJK>Y9J_d~;bSO{zn;YcSF@V5-glI4OF=~i5A#-akHS_P))wWDw$NeNJ%O)KB??HD|
z+0<U05f+f;CZ+$ay+L9_^HA0E4g@-J``^?6Ua~M~#{V#XRgai7^!IYj{2*Z%?6$<`
zHV@d#u??xeduZM(#(WkhkQk5N+roKqfQR0DIYTe;1#^@rl>66yVP-2e<n2wyDv>Pv
z8)exM;o|O!eG=ynt7|ufk-APfb(OW7!xsKNrS{-ktULP!!zwE@9@Al|5DbrDcSAfu
z0>$_k1=Rt2Qo2wpW?-n~k3aV8yZMtv^7)CU`+ooa*Bz7J{TjVzcf^iEG}VFIKYe@8
z<PPf3FJJyVxwvC@$)7*H{rScIfa`~^XTH93{>a6*agfL;Rxv$ubfd7qyR4(V1Kok)
z@sqGy+C;5#tKuyOOYRaBr7zkfow`$27T^@&fy_cqtNMCSv9j5dg*17{YTlWrj*(w4
zRPVOTn02o&qWNI^=8<ywv{|qPq~=m2tX6a=3Ih2AD7|!TDnh=)wl6TnB`1r@$^*pT
zyndUKc^OXzm#c18X6wPZEWY5+OONNXRz9i=9WRr`ZBAvE9qu%LHtog~X!lc<@UC$K
zl8|*tiI%UMxEgJ}3q4m)C`DlKlmT8g!W?N4KTvDgwqI7<TU$G4*8P(7&3Gr)=H9I7
z<M=s9&CK;;BZ%P7mV^YPtxV9@VDYzrUtaPKJ6pa>I)=ZEbaV2-avgIOsaIUZqa69o
zc0X0sE*GJNrxq+1dpdXdn}&+xM5le0l<Q1t&VptYj~t0&*HoTFO<!>hpk^xTsShO_
z3vZzk<R!-TG_3v!#%0s~2-Lm)%D99!IvTe7z&s5snIEmBvXTlB#5k~gyJM{`ufIDv
z&i|>#qHS{R&1l#@5NJO5mzVJk{p(Fi<Mj<Uqu*uZslc!Iz^Ki0TV6+uj^R<39fxeg
zAxK%PjzcHYw2;$DQ|bA)l#tuhM%mM<9w?f^XxQt2$f0O0WfWkzUXtA{E~82%prvg3
ze02u@mjSx=*pUJ*66{|tM~<NA`u>AOlBqc0M>z(G{BU1&k(Nal{(Ki3&s~&f#hSl}
zRmy!_V~V65lZ0P9X!xl+u;n=nNZIMGZxRs3Yl!h@K-~_>*z*SsR_cEBHDDD*4db2t
zi*wGUf<TLAfD{py>$BALL(cJSq%PxS%F`Dsk<>|c4p><Fq}`(!Ig71%vgNf{oUo3E
z_uH17Vfr<$ZY|e(LO#8EZL~F#IXq&nXe25e)FQY+KE1C45_9URMd7a+3rE@|6J~d*
z?e_p?`rTI7nRmr^^QLK)<4|m8?UQElI-a~!_CqJM_9X6?h2G!PbYw-kN2tnv^z~}N
zQPC>1%BSzCK@M81Z)O|(0KM*tG*{lVS%@KAinRatk}Jn%(cU;0n{#s^T?J>PSUYmg
zC|X5^B3pdfDej5`*QmZ?IQKefsS?Wtx5Xey4oNX%?Ps_{LDzI49(gnYgrhSd-UB7y
zeRutqE~LhKJq@f1hC3q%bdGQIWWQd$4@oL*`MOOqof@~n3X5rbuzkbXaCpnJ!G_=(
zcM(U)-<-{vHTnr}1BH{?aqfk*rzy?iP1AX6fR}-+5SF?oTMbxu;BZr`px`#i^1bLg
zNTsTBH>)kjSLy4G50yWGpFFjJ)U$S~;`&5j*hF*RO=Ah7IAlGqXsH}4RL7fq&K3S*
zwkCFjjWw0Kp~#|KS`Q@S;yMv*QheR@Hd{RVl@Yt0945A%!!DUQhc{6LFv9}UMR8T2
zHKt67Tb$w+T>4~ei04Qo;k5C^apJ@dNM?-o4{?|dx_re}DONJ9;;UN>3IdG4-Bg_8
z1YH?DzMtCX2MbEok_i!8FUa($@=IQC74=GJ!|qz}DG>|j2MP+FdSmq=USfuN>wLq-
z0SU<O(Qa{Yao!yA^~~)?_W!V8tobxaSP$mueBqDF`07Ls=QF=r|IvAsSVCCE;DhH^
zWzUo+oxtGNbuTv@6HVE@BCY$<qQe7qJs#?O>+A8ZPxfw>l)0Oj+2n81kri0i`RaVA
zd1ieWTP3z{(VTv!D>kiNG<I@^)(<U-bbnaE-3Yh3m#jf&F`OO<8JPW9%+cV_O0lZ+
zEwI$VwqZ-Pff;<J)rh#_<f;8sj(Y(uKRZxHFaxf&;Q!8Ct+HZ;yKPvCW#4}JR!gEP
zcry!(%cV6zN(yM+km1LEr755P0ONA}%07RorxJQ(D@3wALNP*adm!aU!`|*@T1cOd
zS!|Tu*>D~hfCS!TgP|AKxe%*M>($v(GkL0$TG%oo=N{OE5?-tu*lG;)clsSyKof`|
z=pXw%F8=(|0D`iN+yHAr)~BfiRTX#uQdo4YgAp5q#9eyPbhAB42_FdQnm#X5@4Ao8
zDtI47-o*5RM$F@ASC<&UCY?^K1@eNIGj=ZC+ReJ|j-tUK!xyE8;s?N<m)6l>f^{V2
zbHO;N31U2Q#a(iV?-o`eCL~rAe_I)E(gJl$VLTR2*jNbCD!;F1FHLpKD#|~D)P<Df
z(5AsLD2Ll3IVP_@nk_EDLOdRw3GG3E^;GcXS)HfxH*G`2Yx(Zb51peiVaN_J+@;_B
zCxjk;&xym&3`A3by34@d`}cH#5szsq5aS%$A7I6_{XODUH3rA}58Q>U=FsX-=w25u
z$HZ83pl-4v;6bY)#(&VVPki-k`QATtHZ1Qw0r3!1Bk}la66DWP@K>AFzw}k9{%<R2
zw(Qt(qqiAsAu%>@A&vn8?(Noubson*l&#tDlOTyty$_h*%G0k|Hfq+nLdvE3KzG4J
z;_v<~s?89#eHB>@Y1aJ|Ygs$Ip-K{m!THudbc*B*l2`tE3+&z1|5qTEM^osOgjc{`
zhuj#;>aT|yW701ToK;mem%%gEFN$anAtiaVx?_yilXxOlEd07|FaS+@fN!KhAQ>zC
z)q@a^Tv|LtLHqgFiw}U14Wu)z*836#T}F5^vh!Gm*v(wYx_*tvu~^^g-(ab1PG(>2
z)U{f_1J2_JIIqD7*{?^V6~F#mpU5RJ(nq@T&PcLSZIJC!U_B*#1cXXbi;C+H@c5dP
z{1{Abl&u3LJ$bR+2zCs;vRD9?v|8^66E3a41;--|jqzR0vQ9m}0><cv9M)y=-Gctq
z5V7nO@IxoF!_{|FDY0K&6&{Vm0;U>&8M%x!Y~!{BZlNFw<nUruZ8!0iwv%!I+)_GU
zB>1MU;|i|}!MMj~R9P@clF_K;ddVZe_+4cymS1P*7m>xF2C91pjwQFmyj>rF0M+lZ
zPso(y(x$R2Wy^ACd@t6>9Y<BTA49q0D_%^$#Ux7E=eCgC{Ppmfs&H%8rdC#G)k_^w
zH7H7i#GchB74=`o@kGAZ2TLbdu0K8JC#pT4dEJ|5?C5M9_7=tkZ?=IwKN=w-3Lftj
zBUt_}fKA?I-%Z+~ahwJoi!7pC2L84dKtA}fm*g%@hg?MTYyD{DHdiDoZRTs1%_F2R
zde@Qe3lN1-SE`Yjkt4Xx7C9r`aY59t+wG6p(3wbNi?$s}`Gl_PS4V$A{Bhp#f8N{v
zkNxrg#x2BPZ`EpxRDJeXbGH?fKaIO7K3DO20{LXq7nNoBVoo1Q2^Xh7>%zgYmGgQK
zfiiyl8%A6#5_M^P;Tv%tBKfY$OAGlKt95?~BCG+3D@sxrB1tFTT~G-Yt;YZLNG1ZV
zpDZBJDtbTR$HDPg+o$kNklgYKv5kB6LOxkhZIB%*@bkz~<n$BkN#W)78ZgUx{V|y5
z0u^Uo-bw$^nZ=$3La4~Bf(2`akw-S|6_{`_(l2dA?0K#oq(Yo8t@nai_(ePggG{Ie
z?)mho!2OZ9^m*~g$nQ_XZ2;`cQzK|zT3-RTSVvyzk2nX$k6=-*aX6ei(!VT?>b~|C
zTpNOYBU-iY!7RXt(z6U(*0<lBD2?^Zp{YTl7tG7uoL2b)3`EfADA>%w38Rto<Ulbi
z?|uaAI6l&MQxmzJgv6ST*K6VjvO3$>JId20#V!7v2M~wb-M_Aw;Y=FPt^68{p0^kc
z>)zR_kH3jss6J5|(N}+_xQ7jiPIGp2x4yXE5x98ZW+`&|CuL{;_72~2<^rT7{^OI>
z#doHX#Xv0962qb<a64u!k*spLa{b&aTI+nq!29>xxFS(bw2gtrAkU<7FhN!8(_l@L
zFSEC!zRabzOf@^#L1(%W=ms+oBn863v`sBQ4r1|^VMuk?MWI0-kTo@Q^nm*!JHhye
zK<8kHN6hTay^HshG6I@H8$OU{1f2d40!U4DeLox$7FQ}uIZkd44X$=?_|UIbJGhYN
ziNb*Zr~dW-0+eK*C@OLG$>{^b2X^e0`$H+y@CC7X=)nfT#AIM#01|l_`7pR@{W(}w
zK5FpvWWA4#J7%eEi-Fm=t+@RYgqlseuk8T8g5<7Fw`rNn+qcBO3zc#U{}niq%L2Dp
ztq;jywk?qH3b3)OlWkG{n9#yK9nsAG=Xo9lU_>5GDyS-w02aRR>C2D<#;<M;b#RNF
zyW1w%TF#<ngHe_B{qb3bAKxlv4g%D@3nKfyY$ACSvJzSGko8AkF2E!N($yGBZl3;V
zrW+&@z=wyL!!VM29++#fJ|~0eRL1*cocnN3E_U|jt^+X?QPo4uu5%FMJX(#mgPsPy
zr6pt%SvY;Yfk!3eu@FLD+hRd5l6D9(Tof6VpS&pqpH1u-b-UG4z^~YN(sIJ&F|ApG
zwAqug9Na@f(jKVQy#3M9?KHFb@_SOs$@mYObkvN4s~?e+dLo8s?^q2^k$O`{e`o#q
zhn|Pe!qU1PU5Aru+${hx1T`2nEEZ*R3Hn{Rv@(&{9u7T*zz9#EDwEtEMM1cQGQ4Y_
z#-+`jl^M0ldP`3qY6R8mX;rT3efETK+~S!{<RdF`?LP-YZ%h+N$mxc>Iu|g|U1$|a
zS*3JvUF)Jg7G!H4TmIsL+$+k$tLyly5RFp1vS<vWc#d477yTU@(q-SU0`T$PLeqV7
zr{C2vqf`w@B)%(=m*G|y6#zvxlvDPp_y|&`EoWv_;Q`Q1D~P}+tvxJ{Hh^6~T$a9i
zdcP}j+v(D+_Wc8usGka(3;ncz#T^%@J=eP0Hw@N=)~h#s><7=tKi#%O%1zq<qS1rY
zV4S1Pk*<vsxz0CWNbjy6^>pyw>P@~5#d%wvyTdFZ2ab@>JWR^pg4_n{=`S!IUx-ZV
z7rMy1+73Utz>xCHTX%6MJ%WTyh=z-?rEV+-wAW7WQM+Hx)_q7Fl(6x5cm#g|q5({E
zVnNR3*$yY5s-6DM`$Xc?3+q@gtf9#QV)d3*vtGQ1XEXey{r1O$R+}Ns7d0&PGLyeC
zJh5OItcRy-Z5>$Ufc@=(i#cX4;^3{jEV|6Cqyd@F{|!n0&bhDJW--jA(kBMFi7FBY
zX&-|hX<R5Qi&x$J*gnIL>Hqzsu!IQzP-ES8l3tfBd%Flp8#$4;Y{z#)N;Z?+M4$#j
zE976db}*mISf412-21z@Ol9R0unlahe{<?Sg#z&?iM%b|%qN4(0`ha2CpH*VEj-g^
z^(TZ|e*%_NHr>Qs0ef1n<EaFSZJ;=pE*|Fpn78iB&IQrCysYGrG(TWZ9eD{Euc}wS
z&Mj2k3>#`+&868YHP7FLc%1&w9vXbcOsyHLr;YE)npVPBBSWm$i@}~3*Gs@Gt4QL<
zp^|)^4_RQ>tZh4_x#}h!$Sh6fL}p-h@viX%|4AEUY1Z^<Wt|vEB$>);R@T$S`)3tB
zrkO%&<X;GiPpkpE@_0@OLK$zO0yjl&^igi+Hc6wJhX)|R`evc1tkHM9#-Y}5cVlHd
zB_4+QBhtgYWF5h=K2baB85&|eb$ZFC%M%G$>Gg$}KGnAH!8r1mSz6Kw{BmkQPdReh
z|K57m+51reCl(M)Zy=dW%M~TShQNx}>jE$eFz$~XBj_@pv(Q9FGvNvRqhpJ7Dg2|Y
zDifz0gY_(KJX{K2vfVY<l&b8k@TnF5J)2T(mRM+1a*ji7EL}RP4CsyizS|5+A#Ld~
zQ?qsIp0$m>5?!Jq#u{m!M%o@PF4+?9>OdMs-)pzFj2ueV_k1*IjF@_0?Mh!bdse<w
zhky?Z^g#{FJRP<!PXM9pvy2yOC;z1S^i?&4hQ9s?F&%3wqGhEsEjOl&Q^WF*oicV@
z^ImeggSLDreA%`IesIVF*Is>e$O5Gw$i>1Hwuq38(*@<m+TCu2&v&!DeOQ(T1sNmi
zHv0KJuNPxvCo-EG*UAu=()BT!o{BN5E@v<MCoZgQ%dn}4dt1YASE9|x4+w@Zv%-w4
zuBTCDI-J0K$%_LJ4U5R%Z}Y~UK@>K!)L<($CkeK<TEfzfXumXA(fgc#j8T-g8(MsH
z$Q9LI9TiXlT)%q_ypKWRSwkD|Ypn`0M-17OI~J#IXC?fMS1faT5QzcvR5!0pAzljB
z`gMjw0%a`&^v#I6lwcb(XStlnKtOwL+i2`ZkK>AHDf6sLY6TgD9gv!eDqb;I@^WPO
zrG#b<BR+7vm+l~{RVbjzpHTeFDvFegxz3JNM81z`KNZZSM<+52PX-@W^;{uhv5u_z
z&5=VlB!i<{WJb*rkV{?7+SN-(n||B~4DvxyAFu02;lR3<dI$#wS#x21Ew2g#b(k4Y
zfep6!!?Gu!%6c6(q$<@|^1Z9tFXPLbWsHx77MqsWs39Q8J}$$vl9w#<)nj19$gHt;
zXhChY`$yZBA5n0Iqpo<tRGFQ+1Y4Ybpt0G;Sot?j7d<cXG$){#koi^h&{E6N*B}3=
zO4P!dqF#`*5->KLM@lU>^v#S?(TewQS+s6rU(C#L9JJAIIPLZe&U%|J_XO|Nr8dpy
zH^z0=Z(V9TV51-7u6;RS>+x$~Q+oEt(q=~Cp3s%it*a)5e3R&xt!vJk-CuIDU;mo-
zLnln)a4Cc5jN|g)XD%m1KTDHyJ;9sIem&Y5JP)f_PGq3HKQvtAGtlj)UB--IXQtyf
ze&%4Z)_9s5)8*gftA*mc6kRHX$Qv6Hk=aM(J)O&k!F+2>Hez^RXi?c!!V^im@%T@F
zkeA|i&XqT`YvzU8__>Y`OO8XY59DSa<X@UpgjPKty{c+_dobR=@&3^!=b4iMJ}5K<
zEvs!uZ%mUxx!RArO7pcd_fwxGr5!o_vT!VuFsimt16h*Qc%b5$y7<Hz>p}T)(gqiY
z)j#b*Glto)u<%DbB0iR%2WQ!izJSLz?0n!^brNTY?+Vnu{;H<pBx#bq(7(P%qQ%$2
z-oOG?eO16x8!#b;yQ(lZau$Xb2_3%u@`+hA@(HB{od!+3qGO3kz~{1|JNj$RH!S-C
z6nv<?#>~o-o|G%aqh|m$ESpX@fFzmVb!9_GRvz~WAF2@V6Ks$(7(=0=GugZ##<Ik8
zW1_r97Ik6Ok#v9LXKrOy9Bb&|bgk7QM)+Ib#ZRvioMtLbd|V_R-x|nDsA)+U1*>vS
zbg1K@jwM1{*YtwZuBj(IneRx_#is_M9ddE)kjL?Mxic=M!;<#7*0Svv8lGDv9C05#
zUC@`Wi?xoq6nP!GIg{l!dA@k`?4^W~$26T2iW#0$u@C(Sh62yCl~xI1xkO<h7!Hj7
z_^M^G>BiK5AaJ=$S2Ge9N8QrAk+xfNFx5c0EdsiU%1VGf2z{L5dAW@YwF&z!A`aQp
z5T`p{3|TEW^O#>9kK+}ev?}Vl;@stvzAjmLa;9yr8Mq_sm^;%2c}#Z5C7`d;TS(Og
z12KI0XaA<S-rChx!!t1miCE^ZZh<ppO5Gc-O?;U?i&u*ukhHwt(<!?k7w6J{oDTd0
zhfC7RoL;aADzc#fZlh{U<Z^|VA_L=rfX@=!JRFRQ%l~pR2-j|vcBI<cs6T$=DH}s?
z6YlI*nJTiStLL<{$l;W%vsP&t)^J;Q0M>uT_s9mZo~6s6{2j*)RD;crSH<9}(n%%8
zzW$V9@C<>GH0_lY$r=uiLa6(OrX4)m0cfo*ptTzb-j;7Fs<-@JR`5PgWE+GyvR=3+
zCY;%bFqV)-sh`N>Ym`rWCt%@4w`1~;+S6y5+=(1o#FEETFD2OgMrYNk5EF+Zm#pJB
zFS9M~YC@!aHw~OEuMOW;@pPp>w|3XQlyKyUr@Otb*#@;d>;o1~<q<<Uo{MeP5~m!e
z;$GHJnbD5@3pVcBVQC!e^{J3;zm6i>z=@6DV8n_wy`O!-z5VU}s%<e2=I7Lv^?t!e
zHB9ZO?hWOAu|%~#&xR8mICBw+{YlL>tK*6N_WVCz3`>|p7stLkjQVx4No(qQ+EfMf
zNLqt+CSkFbSUa$S=*IYDf7!lNyTWaV_~1qTp|hvABJ5Vo!rGLBi;-T>YdSk8()tn{
zs;|kP-$38!n8`oOo6@&VWc)b)+8>}^E#>}5C%3XN7Fxnhf)xT*$^;y2akEDq%BGU5
zEHQ2%G5`9bR0AiYny&!8_`*}mvZNy7!$#}a!=e7YJ&X;FUdE5kp~9JG!s9iXj{PN9
z8Uze^L`VL0X%%2ag8duuN1LwBoIL7}+Dbg{VkIrT@k~Y5U{6mk{bQBN!PO?nW4Ohq
zgR-TgS=-)7Tbm!$$#Kge7Lk|x-=YsRaNOeFrf9=@POZU%pRbJChQ->`jfB`;jAEE&
zSM#tfD(+3in)96`_5*L9vxG#33bl}q^`!U-eLZ+%ps4b6sg>tdiKd-_rfjQ$iqmms
zzSv+J{g39)-5BlJv~F;@%I3A_)BdB_N7C~prnr1s`)5|Rp~k`@h&TgNw?bqSn{jHl
zq4(A{FKxCfNoZ77X)P;BjLYf(toisiE*;f&2Ou&Mm1|>$p$55Q2Tqj!*&Z<Q<)rdv
zfWzVM3Lls6hPn!+43M2T8}8ku#YFRzbb_qff}@7K;CjDXCGUBWkAcl%YgLAWo7MW;
z#X^%qmbimMwkUmRBjbmFpEk=J`T29Dc_P%~SrTj%Se+GMb)M^Z)UmRg<S{j}%4U5I
z<CDd-R%725mCcLKw@1ywmTsJlI4Ga}el7v(U|>rxzi}ypmDA?qYv7nULWQ_qT4z)>
z^NTA<R}t2}#kN6V<x@8e!n1p=URJD0UoX)wwd|{?&bJ-~bCwP|$L|lxTl`jD+c<M2
z2X|%x7N@oG$87Oyy&LDpRXiCg(txF{!Qtawx>z;-VKTzyM0{Ag_JfvmQOGd{XxP;)
zqGgdO;Gd`$Dcr>iOODX%(}DqU=;0!(IpJ@|SXo!Tr)GBlD%)=_q4ug^F=t&W<6UoH
z!C0ZdR*k$F@Q@t%KDSC9`<XMQSkFfH$4ORSO%hfy??um<uAKuY_7GD%ly5x%<}NAc
zFby%!L(hheeUZy0UZ?H)CYu>|D+Wow0<)T`ncf{#^u?}Rsv;_>3ZZEba))DO>Pmmt
z`MI($o$&K#jOwQT*c)S*35;4O^~94xI0pz^6mMe7JnJND##&SxOv-KLfuO)W2e#@_
znm}3|bJ4EU<y+e|Z`?Z4GMXLg9Ivo#nX+8rXz~wSquHjbV?QFXBG;;~6(CeWOeizo
zy5|{pX*DMlTm6Q4*0<=(_GMqJku%nSYnHiS*)O<c${Q(zpiFVLljB*g(^j^<=>)9W
z@@h?+6`Rd#HVp}OnbJ>Pd~QwoJz){+X|zOoQp2v?Te(3sWbqll_AcQDO;Ivl$IkL^
z@oAS5Ys3ERaz1q_&DhDAt{Rl>IdjP9`WHW|Au#oG=D^gKqKWs>1us0uJrurdeJ30e
zGJjb{xt4B29r}fHe~=3rRX%84BpCoY5%9xlRAb=EsF4~u-h6;8Luy^CREafDG;F%b
z4;Vc=`~+7O!aAf54R*79O|R?Uf*tN8ww>`edP!c?J1d#UDEpvA;4>h&gRp9Ack^QI
zA{_9&r>%s~mlK1M?~c$8aBPqpzpyDmgIj)Yj-QUR^exuUnp8wJ6<2GdqYGab3dm~Y
za|R8?`T9TM{a;K`Nq&u-Fv3)u40N=7CAW${!^0nH);(eQ#{UnE7$968d3MbhGCEc6
z=+L~`xjzK@|8;leUrk;69&2xHpH^<|by=i9Edh+EC?H`*YcF~d0yPL+WKtm@LqGxn
z8AHXEstkIPfD#}gGEXuE!(gxqF=2=r=0M_=gee4S0tq3IyhGo8?;p79y&vBBAuH>w
zm7TrU-uZsNpW&Nx5e)^rue*4*SqT+W-`4DA_nXeh*N{%<bPso0MRH4hOCFaF#$7Yq
z7)JMn1*%7QHw+drw2lUQsA8UMY=|t<pbDU%vb&spU}^Ezmi#)r|IUT0tAB9p$I7)u
zaemDSA-|v7{`Ix0DwyPV_Zt)5dEjp7qwQglF0VDQeG*&kneqm1WxP8=R)l2Swdynq
zDa`e736N&n7?c5hshEm3kCA@;0@4)r6aKfo)*s~c@e(cW;K@tGFBH0#od`sfy|Zrp
z@LKK;!d7hg(nN3+1WIUU16Di{2DGFlACC<}Rb<*?$FE5-CGF{ifEh(l$0StL+M=Id
zW->5i5Q>S+&)!$IRu<pjCR>MPg|*4)R_r`8JM&0k{0i^lOS6!oxcM-!>x4LS;Eb<6
z)gcVGE6tAWr++QRFZcCij9UBXXZx6YmMa2eF%7`0>rDR={h-b0*8-Ab&CgPLY0mm8
zuXpGgRQALVn3XQ1**l5*jFso{7eNExG=ob>Mx?@^MbW$=CU;J7uRcz_#OgvJd?9S9
zi)J`%b$hy2E*Ff%1*15U^_O4~scr5}8EY!qZ_+l4UhcoI8#Cu(Vh3sMlzJfjC^qnQ
zM0V&}#Z}U}iv&;W$XOroV|L_x14Yk{dgyp9`gk;J-}P=IFF($8gsX4TR7>dAuKHR@
z&u4sRIGQ!Dk&M>MABcy%_2=Ha#JJcPombtnt*t5h#O7{qXdZ;F{0vmq^Rc0!Rtx0T
z12lRN-?aK);3Sz=QkbJHP(;hrVn_{xK6kf*FwjYUkF$5p<Nh)CM*q@8OK2-$@AUVP
z7I}uNH&88|BUN-Qw*#sh%$?=KjnT&2h$C9wofP=r;B4StgYM5+zHMol=SD`q)C4Ng
z(p8R(3N9V2EcF}?$#y2c)9bFAr3dB9&cx=oCGEw927zH_NvDyOkIilFKyywP6pPY<
z2BV*r+u+)){xU?;<H9>05p}%2B)c!reN{Odp=bQmt-e94+2kKbbX~)j{x&dkGG-FS
zEe!zD?ZmUb0dFs+KBrhBm0bt<n??_}G(UOq{)~`5)(~u3@@af997*s|51SW46~h>m
zcw>XD8`B~xuH`e)pQI0Czyn?-+J)AGmYEv37#vias(vA(SkV+&==wBFdtc-PFU~<#
z4EKd6dw>q*@zyIeJpAn;xbOtgT8wYjl9q|j^v1G0qpju7$S7|QG()s(4hRZFjhL0(
zo}NoYiYOQ^=M@)<_QH)a_M-JD_yiCl@bln}b3!W8{2?|5kj21_A1qTw;^yy>Q5+^H
ziUTyq*=88mrsFkoQqes;T(tnCAIH8@is1!Fri0jt_?7ys`D(SA*C?2RauYv2rX(x9
zP4izKlm9n7)BnOkdXA1$LJ~m>`bBMS;8rK83(CFOk~6-z87BnX%B?8ZUG_Unn=iV9
zoTg1YS^|K;ViAghTXL6un5?Je(z(=zhz<X%vBykQH#$Bhp9s$~VH}6}sUp0e&gG;0
zifJ94M+6!Y-;UNK;6N0ZYR4WUr9ztt`rC?Z=t@JJ;uu|-J`|zF!fSE<lTX~HZKoaB
z=MXCX1|jI4NEZR`B=jUy;P%a-51dLm{;@0{vo8fu{Vp=}j6(4Z?}>(CV`fQWGT$=t
zxQ9#zgT$J4Ai5w#c+u}u9pgz$DJ169^Px)O%?9fQ5M3gA0hEwrhm<#ALO?cu-Vps#
z0b8l{Rm@H}GImErf3;5pac_he)qbF%<SInqMkc$&kVHT8qCVHjJ`4Fa#$V@<x=yhX
z81Of@Ap_F4Z~Qc4<lv@L*x`R-KnE|5CUslR`L^U86T<$#)Ae@`)Mc62RkUx;@uFVu
zQJ>+Y1b~~qsm{=}F1r>}LKw<tYaPsE<=K}VzBh?oho;p?4svIZRkwgs312%jPnIa%
z7{{n8RUQ3fVDPZ~H>8u&=qCk%(#+a(7?o&EINsD$Ue~9rgSVF-^TRasJrIu#_{OU1
zN_1TiV!BSx`P8#t=-kC_ZQkq$6flN0`vBNw&29qH>oK9{I9**1rf*Mcv<x4TsCfrg
zS$lLPqV(8l#I&)v%@GK+8T>WAr+`#6?Wf5Mx7qG4vhg%{k<x!?v@>QjSG1Mmj7=<`
zy1jXK=VI3tQ0ON6nfw+zRt`JKyeAq23=2k{{5W#L_(mQYAK2*BNh-2=lxpCxE#Rvm
zBpoU5BsrdB4+TE7FFTI7%R`$>>-$n;#7cdq8mzKT?MKi@G;V4ZPAQbFNY!BUoA#^2
z!G*&vF`(t5tYKSEOZVY?`H7I$>~U_mD)^kYQW8KB3hmgno}eRSath^46-mx@Axfoj
z)e$Km$y1Xk%ojxh4fFL-%qKp??`*Kobhf!W?d0hS(y047A(1{1JZn=m;|io(u;)By
zGo4)6TI54*{5R*GEb%%?{!{$ru@rS&p0hgca^qG?vw|2_rmb&pf4a4w+ACv`HUY44
zb#ZDsQhr!K8cEZ)=Kf)wC#@^;@!&41u5bQT0%`v}KV}!cM{}}76oQ32(2NT)nOLVd
z6P(N<Nk_H}vl{%`qi-s`;Xn&TG~cZz*Gq1}4k*GUx|n^R5I{V&cVXRmEpMYLdgc~q
zg;?9?ObdhyEnYYbokTAdBLI8|=rK0l7ZWCrvsS&8Uqf-wRU0F!t{pi2KG;xs4Fx8A
z#45mmAzO8M({xyvqB_oJ)burxa9Y2_=CXPLC+|I_m}>!Y1(&yWM%H!8Y+HVjHjW?8
zCo+M6U2D~0RwG-faXB)cBfaH%ph;wqG0iJ(TXTKzAa-9D3oFfn2%irI<ft0cJrYM(
z;17XQt+BiTz+;3?wR6wNUsY#LrVzjapAw57ZYEAa`-#FBQY<-b%L#$P%7(?H)qj#N
z6cQG@f@t=V!ePcmFT{(^qJ1+txo_{RHKiox*c0zr9#MTf#lOyS<_q&Jz1*g^7~f;#
zlSXX65SY@Lj}P|eJ;|OG)RknaUxipH8|sM3e@>U@+~^&}UepecMQr~?N5<`gW-qAZ
zg}vUR#2!=ZXC_qdXhlmAenMIgF3~I}@o@3W4M@$kWjk1=-c*)Sm_!#WCo}d~xxXe5
zJ?`je^3P_nrB(`;aIEi>Q=-aCXy3f)LwIu(sG+v|?+ug>OfzSi$Zf`nN?1|^s2{TS
z+0X6lJX1?V=Y@B<oAXT_&&K-Bj*j?FFEgmaK4Yyz<t}L15?TMSVTmR7B?f6Jc?!|8
zvU(0(^VlOmzjt{GrUdY(z4KH%G2pN{)*GCM1W03(1qS3XAdi(r6uG<$Qxh@pL|@RX
za_S$~{bKm)%1vf!1!nmrb-1k@Gi#MbpBPmj?ud0UL$}ZJh<{5q7_s;~`a;<zbV^&f
zwKWjIs7Hv{_A6fNefEXG>WD-;WQyG|p(c2bYI{ECdu*89Aa={zW6A~ob%d!}%nBCH
z`!otN+vbJgZeA^VGF4{2wx>`_x9>j9yq}IXyFs<BV86ey8G?04X;!H^$s#XG^MTNr
zq@TvjNJB%jw~M5{4#mIcs8^6dJo|#h@%wGJ#XC31wRDNOZwiT7<2gnNaL0F|Im-i+
zuclpDGwU27lae!cC7{WKtv(`-VV-$d&VeW;Pai*ikWv3`o_2TPm9$cfa%cw@5y6G!
zNv+$v1uI@0ntz$2&Oo+3;v!^@<olTSQ_*kVa0MJ0F>fNjrW|I0md$m~K_15}Sb@z&
zZR5p!8X14KF=q*oRsBkz*pD;dt<|vJKlJV9Z6)^u8X8O}adI(b-#)eI;)HeQ=J?Y@
zN^|Zx6+Wc*C=J!=`5fjOJ60%hI)dfg&Vi)n+9TYc{K2o^j+v@jIPx4<p$nsV^9?zP
zj%cwk=^IJNt(8i(7*5r)=~3({JJn0_`rb&8m5@&lC@w4=lI+qM*aKOlFW-y!BuWbz
zf1KH@<bDqn<L6<;Nq#Fnu5M0QLi|p=yhb&VxL!aHbj6&W*iA}peSzb4*yJv_8V2}k
zs8EQ10!#It3OlVb_jGMg$Qc}2>`sotN;3Cu%)Gx{&^laWPD*8Pg$1O|6tU`2uyTSU
zb48e`M+98duBgP`USITM+NHA1?{Lw?ph73X$o-nKRF&mIu!A|@9lU#_A5nY3P$#qx
zvPMEdA!+2+i&Ms<A8)~gZj2~}5e7WCf|8Yz?aZ}S*j8TymSQyEVb~fPZoH((8K>ub
z`vAXi9Rq&a8Ch3fbMpiA+yJxfX9UOvd)$u<kxcUzT=%|S6pMvu?FEv#^9A4MFEg5H
zDBzp5I080_Y=Hn5XeboC>BdiuL;NRG&l>Exk%TwritYR)uNv$@&CMOy+A^Op5BSp{
zWD3c0i^D0f^&8!C8(rpc4(ugFKOOPl^R-CM0aJAK^T43i@ajRdY<%b)eXPKR?wX|Y
z0P>aS^nBM|f0Sl*)nLwyaREGFsv0DVSeYq#`;Cwcn1;7=u_o|`^l%Qkz0hshlzoOG
zjv4iz>-v>{LQ?2hp<g`&_g6M3gtbR?55}gfWiZV$Noazb+`l6hRGQg^|3<K(X13S3
zFV$#CDG>Q_U~VmrHyR!^QgvGe;Ix~EB>GP_RMF%GJ=3J9uyDRaqOa`Qk#XA+vTL0<
zzh>$&GxcVn8S={MT}fb(UoAPb*gyWLfy}J<#MSgPGB=cvoB=f%;5fxRJ_eCA6%R6Y
z7ur{34(H%bm9{`DUU(>C#GmIJUc3T!wp^QJHA$r_2L2(3M_}0E3$=AMrmS<_m;Rzt
z@dH#1%|j90;dSG>>rMjHuz&)fm-}$z5AN*6sXKXYvdFYi<_V~(;dakK1;KzdRaBn~
z>+~8VM325#@Gn+s+b05+5LU}4r_Zl_GnK%@L>hScMkj}eswxP_x1zz9DgARMNk4LO
zD)u5(K@%X@<&RW74)OMyQCj%o;-*=-*<q7>BX5GCEZDS5+EAN2`uuuCy-vng=OKsq
zmLsz{6>#kg=YE&sIUCjO;XU^_^<xGB5-c;wa2*gIre>nzcplQ4R&hgh;?5?TzJEwW
zO<bzrk*kWtni3pl#dinz8hnvTsw-~3Vx_D*N(_H3ZNpxgd)K&{9!M1_;6XQwgpu6m
zbq3PZ8=+$jFrT&jmHV?Xvxs^5PkesNC4xXg^{C%JeG&Z|&7s~^r{^$a0?rsLDinoN
z=9sk5S=jV|ZcFm6HDjwABK{g2k-KqvMU_eFau_s85>BY0*v{R1I3=D_^}(0mi&a(b
zB2Sly;9uer$AdcrLdfYCdwQZCDDg|uG#zRmfoLa`UyFUAilpb*z7X6u704HiLxW(g
z*Pqoi1`}p4;gZ$3$pVo}a{q|_e#>#(k^UgsC3Bt2pKc%QWT^JVfO$UYN(kyjzS@W`
z5&?RSGU+2&;{36d{`dOxPkd%yKyJ+xO%H4ZZ3Vr9gb1?eDVl>r%Mf2iHKIG4la(r(
z8~!eu17U(0vdT{#E`Q*?<%LzX!Hts;a&>ot3AR;bEB`wJ_ipap^SU8vGIGPH(!nGN
z*vXIL(Y*47MalC2h(W6`?{A?snrQ)-AM!`oV;j1#Lf1P##`+g&;)n;Ne3bmpyz|M7
zT`^SA1YAydUM*8fAobkaqQGdB4kH$Wms115uUi!EtGg<vUQ6rhhRW2+YGPTn6Wr6f
zt5UfcM(T2z2|{jtgat{D8Qm8Zk;4ypyKdr8jHdBQD}Grb$@1z)qIU_=9(zlH%%+<$
z0zjghY~bKpTbi3ss%5Qwec27@a5Syuor(g0N8L{B-A5!&vA1w-;m`u{-pp0VHkReN
zPioFeiR5ffTDZ$ppV7zYEu>B2wt<Tm(`70?Rcjx#ssW-Oiu8cvNWC4;C#pVrcrB{i
zR*l&FJ_UcZlp`yye7be~5)v;I$R_k07q_b})5SXSEiRr@^?aJbl4Y<M6RE9XBgg>#
zoI=OveF-&r({OfWjRA$pG}Gr7iUKHjqoH*`@HJ-N>&(eiH0a}9B8(($Z1jfrLYaFC
zcUrCKz$Co+wk@ktaFPjQ8-020?AI*o>H~SfB!9bC^r7>>Xw;LTq>BV9QdK#6Sjc_C
zEq53mALY#ag(4n{!U&%>rTCR^J_`S9dIct-MU~ZFzUiSGCpZ=+#VWmAn1G5h*4N(t
zZ0ky}U0vppHD;5mD%_ACPV+I3_J*sA4oY=09$tm|I*#Tag9BU&EA%*0woZO~O?WVl
zDs<PQtmhoWS|Ige6EvUmq3)Dbm=@Z@=WyeRK0sDN!e`c?)MUsq!B(*P#)n}3`7ex{
zxG3O!TJ-4vl6?ihpl%%_rP5s|<|^DQnjJsTC#@Ia7EnNmf7#Soa5HHqKayWUj$78N
zM%b(QfMNb1_buckI@vAYM9yr42Nnv_sSWNE%qQ75gp-v=dv}`muS~W6O=sXRl$-w%
z1?6^kP7Vh`$xU#1ks(z?taRg^Xb{7+WW1iX*wZyjlhU|64QQ;e=|Z)Vz9r77t;WW1
zx(Zh2otTLrap*<?Ky<=JkC|0wzwfj$`AP{1S)GvMIw*M91<)406U<e>+>{mBxoJMg
z@8|R3#y5676rDtacRgppt=ll-eY~5(lyOA-z|lcmakcSv6ngNf0NNzPlULeuxN(}#
zRmBS^JK$f!wD%_5ev1hS;`k3_Rcz8|_d#1Bbsx*4a^O&!5{B?#EPM!6(=D*^t7kF$
zlIO60FnA&bUvUTOivyPJ%WdM>yc8F$v6%CxxjdtK0<_#nvom=+FzY_BxVlHT5aIx~
zFA$dZI^u(D9^Q`4@3u2>on!59*uLMBJzh9+q08bS9>UpIe9<>uLOB2G&6_y`XZKG7
zdm$Pj^zClX5AwBdy8k3|1|F1?sZ70Ib=h7N<BR~$8T|fFJN5Q>|Ms=cl)l)p?Uc%R
nq;2GXRrPNV_<uLR^}pUrvY`7oE?wFFPT=qEKQ{bu`R0EDE>2Ny

diff --git a/reference/plot-20.png b/reference/plot-20.png
deleted file mode 100644
index 04f079f59eee5a501b34772f8e29be0d104b1557..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 57251
zcmeFZ1yEIe)INGJ5D^JcQBp(&krL?^36Yj=MY`+IWgsd7f`oKQmvonibaxy;<<Ond
zcWvMJ{`bz@Z|?V>xpVI~^PPDeU*Vj6_WtelTkBcR^Q`?=K~C~KJ~=)Lg*q?wQ0y@Z
zb=nz)I;oF`3qKhROIU@!aP_4n#o%vv?LMgRg1^q%Jk+p9p{RV3KRAKdHA58Y8cIs+
zzOr-lBF4p?wEsx>NR2eoii!Cm!;9;sc`ElA@lJpKdWQHS3yZdhxJcHE$x9b;8vM@T
zF~)m8(zrBSnI@uPsCT+mm)L-$MU^2`)O0R(>fmUWb%x*cqfD^e!`|xHKuPLAshPG2
z`;Ima`ju$jU-YktuU-H56_>Q(i1X&(SCb&Y>IIU2Ub_sq&Ts$yo%_X;m16(Cgx=xk
z{(F?`Gz}=$e?CPL6sYeV{_n?{B=6Qr|NF72q$`s*|NY#$;{X2F1rfEhw!W-;@`081
zRl7sIFHyK@vBmJuH9Pe3n8xzTY<Cj%gr0_dY9QI)C$?LSKhA<8DNBuy7vc18HVB?u
zJpH^#SwUe^xvZ&}-=p&HQ*PR^J@ubnjPh&QI3X%gJUrU2yl0ZuWWN7gE64Z$!YVMI
zY%{v&xHRJ9H12YSS*ut~UH!_56DNqNsRNAKBY!lI>3z%6D6pCDyNyR1-ExnWm1u2k
z?Fx_0HEQQcoKvSyCz$nT%=RRo6FNSy*DiNxPIu>XS|)f|hk9A(og^7}0`*>k?&q&x
zYx9}8iB_0$Jbe7G9y=Ic?g@8Ndb#K?5uJ0rscS!cX|6Le;$OO>D(39WbF`J~^|U>b
zlP#*fgHt)^K<IE&$zgfys!FD+xPk&PwfpM*%Iy*Jp?t%0?<JC2UCgla{h7Zvd*#=L
zo(0n0xN+*oj~}q^XMX<tX|piE9KKF2<X&oV3>U{Ev+#DKi6p7}O|{2<%R0twrK;2p
z`{?b4itAbfnc>mKbi2LjD1Ds!V=D?88f)&DVT<q2Po7zY^GnpJa7=rgj4vuG+VxqA
zlz@_l*nA-CzLk}gROa=wv<77axF_xG?37EITUzo*zH4ae(mU-P{2gM9+^2R6YmSe!
z#_T%<Pk1Zk8n%X|PhX{>VLfs3q{@ctTynUWs%lt$rM|wt!QQuTVzzTVHx*OmsC;(=
zs0B4YwHwjCBHrEIP4qfGBJ3@(7=CPEWW?96+uYoo2&<BT&eW+Wekdd3z$l*$Z(bWV
zj^kwtmrEsudv%?jzHxpfY@5++=*v^~_7vHu#bvF`jkMXd;+gHGQPsvDaCMhUtT1-j
zb<wN)D>bbieVJ;Vb4xio6-NEjyu7>(Jy-lC@(h|UC{{RH_HW~#Jxc<|&DcjzJ4dt|
zWmXmx6sT-edF&VWbD0%Z%L%M`z56>9@oi7#5=l!-*VhvXzxN|!H4B>=%6rzGBuSJi
zLCK<7h<BcX%fVH74o+&d>jBY^b!;nPuWTg8S-24|n+?A9$_x9`c>b<?y7%hUDTjpt
zS@f~n_F^5zX~MtCZ6_l|R!;6aU7)~=Y^{=vl>V0Z`!83I_E-32o^eGXU*f*MR1&!~
zNIDZlm!6*f=KcGS#su@hoS_;oVMeu_tletOVvE}wvX35pg&>R>Sx$}f3~oK#-;sRs
z<VhY!YdEXqM72jybaYmvo~r6)*Uj1ID>cXC&B1Pg1B1;mLN$M1WeWwav&Yc4x1>Qa
zG3QZf%IOy_v+BNwkJAYY(-;+8h5uj`A<*z4GrTC7@T91#t8?5K{qpqWEtyZ(CI)kK
zDB!fjB_wcRFRpKF*sf1WZO-*drj$9aT`zZBn(t%oU#Z?_$uB4%p`!Bl*x#n*=cj<f
zmX(zaRk?A`_odVI5?*CLm6w+{RBB5HG5Y06-3jd)PpVrV&%%G<EIMCXS!<YjSH{Xn
z&z?Q&xZ|<AZdhnJswy`_eEIV0-Hn-Fu!q>l{bA2FwX}Tgh~ZybttV<-;zQ5BqTsel
z8_52wnx+06-q^Y$8^zVIjHxiBz)nqFqou`R)+|IR425ZS?+MsVO*MaXo_v1hLrBP}
z_*I@<-Kvpf)05f>g9~dN{3}g6_A?zS7zw=RZ%OG}b{_Z=qCZtDCP}ClnvitF2?vgi
zX+!uzw2hV8o}}Qi5a~^kH5|GB5|@^lx!pf<>}qVO-15|}_+NQknZ?Bo?`|1<hwFOS
zG$flahTMY#cM@u9mJRdq%CdT`nqNPEW?7}{)_7L6JG5wR<Ge|D;5!l{ca4*?XefK6
z$gKYYx7FB}h;a)pKS)l^g}GJRkr;>W?rwISg@J6Ufp!D-w=;2Gx+yd~Hu0E3&8E&y
zvi=N}kNXSVNxraLj;-50(yW)vQ-fu?<FPJNChuOiM{%Evk)t1!j}<5r+qWC=IY%}n
zbRPF5eAMZa(A~R=DY=SCuQA1#d_%vXYL5!c)@AX)VaPf9+}YfzCaZnuMC?CTDAs^%
zpw_mwkX9mrRgt&rdxtv{wugJ@T~El5kW2HUj<yE%Bq4&Z`|?7&VrUwuGQT0$PR`62
z+k0%y_jlKLd1clduD7z{ojX?$G2Yop^s6y2d?lP!&*~l~T`3jW@+VOThlhEdhpxWp
z-Z_WgYmGEmOSsjYgPXJ(;!7h%(gE`X1O(rLZ%W4xE-ZyvE{~N(${E828+1nV+9TzM
zmz_gMXxz*on;rux&aTkZ(^Ckl3YN`7TRU2tTF|u+v$54oANP$^Uf`)ym5*F%eOtsm
z>^_8b$8h3=Vr;qG^0$93TRHY_?LvL5`?^$_fc<pa1wj|j=HQ!lyV!6&2{Ex(^IRWq
z$p*K&?JUEVd0oGWYz;=`v`IEuUqY%WI};NI%_1`_TNV40kgzb^q;7Sl_MZ)Y*!89$
zx)}Tb+|#FNxwurTh24LDnde%5+_A%FH}$Z8wD;tR6RX{lG^WK~%$-SzuZXX3nuS=j
z*4Bzp3OEyEDx3x_XgU-+ECmc(E>_H~udlPYZCmVCTTfI+F4pKbyi3!o5kg(NdGoZ3
zi%aBG=1qAW`dEQ~Qp6h^-MDSYvn^N>I*G~Z{pET{7q*kNIQ>t|6~ltuP^&9m$D`?%
z<lI&kTe?d>Ov84cQKdtiXj_dIXV1?zCDIX`!^eM@JPc8zts20my7Bv0U@P_`y*yu4
zOlAn(qd1r8tEt6vV`T+K`-AHOj~_`(KY*RF3I!a+AW)ntANM^eP)PG9d&Q&S9Ek0D
zh*Qgy(Mp$lGPg)k*ZBCzIn4)@+xelu!^W$9LqgY;qr=~qr9p~=gTr=(Ks6V(ZnNd`
zx5|#3^*;2^^bNm%l00&=A~{T7Kr+7u)fWYO|JBZf7fqnx_LU6f^cP>guvH%IEg-?(
z9+Z7q6fc^`R<ye0J%@5q?_5{>{gb;#UwzII_VVAmu8_L#m1^MZT&jdsD6pOoDm$3^
zD1RLeN=r)%;Vug0AHd0!^wke;P&20-&CvGY#9tX1Qrx?u>6|d=a3jO6c>%@@>2Q|c
z5G??V$vJ8vx!dkwoXl0N)J~l_qa_%!GN|WO6tTR~DIDJV?E8D$PpQNw3XIxCu`Qw9
z$<k!*Tm5Q7MdoCWm6U?w;sRg1IFT9t*wfSVtl6)t5T^oY&%dFZA}u6-?Nqs#4`E@i
z<GhZpV#?74_#aYIQ=e@w4!?Q(_C92oF90y2TrO4_mP3%pt<<<~&D-(v^N&8&Eb1=3
zDQoRFY!ODs{`_mLrlTVlWcZ+{C_l)9Q!_KI;q2FpBUmqcH%?D7k8B4_3#j1Z;dK>K
zYDP;2w0sH<exX%jWvXeU1`994;<Y_&+%S-ulG4)M=UPQg#qa3wX~LJr^9}d7bCNQy
zxej)2;T!49KQQ!?zIO=$9v)sTtU@5%`uO;G6y${OBSq$SpVXpuzvR4sPq4W-OnT3#
zjiK11TO>m>UbT7Mh7FZr-{|`C@_l`IM$PKzXv);$4Kanh&ohdOnQFNbfz*Pii@0m#
zoMx}?8F!fIS{ZX%k82BkVZD9(EW}{3W^VMkT}W9@SHy(vrwRVlzY_C7LL?#^{K#zg
zw=F-(+=3#2M>}<go&E7Ia$b4M@lk%;7!!~Gb=h05#K`%=xh6s3{3Prp1kj7kW!<3c
zD7$2wkt?{Tv$NAI!{_to8>!odTv6F6j-@lPZjaPmmjFd+v|n-uJewIGRltfpi7jNA
zct9k0V<NNtnT74vyh>n5<JXvtv@{4?CcSFy@vm{Ap+2`{BPRj6P%~Eu^2HD)N%#k^
z%yh)u3HKrX#ABQ2F6_Me#Mgyc{fh|8&eAB^Xo+>BlDwc##Y!~=L~hrdL&J*{wH)o+
zrk5=2rV|X!=(<)m(TNTJBs?*nE0fH>AO|BwGTvoldfj?|XN5H^2THBo@c6-J@k<`@
z!37(Q<ED`1KWmSd*{708>uWhU6lx>IO=;9(q_A13A+LFOBaH!8hQ{lF1i%0aUy_>J
ze4s7eQt&=6(Mq9d@6=dnV*u5IN00DP@sOVzsIRGqtP7CeGkQOdH?+CwklDR_^T*7L
zFQmO)p?mj~&{0aMa_K2+)99be|DG+uY8VBP@WB^tIayie;%fOgAv+f+338{Cgbr38
zl#L~dpHohMoct<z^F7W<+$QYZXWy@}vUd6<-_CJb8MoVokW-_w)KF0Hw*B?prCgmW
zJw~$l_wFt&3k%^&<pv?D)_^@h-0w0hxUXc(Y?F=?#^*A#CW?W8OKW<}-_P%(Y4q74
zM#*q;0^I&Xxi}$A%+HZz0Yb?DN=4aR$`}iI(^9!H-hYyB3qNdAWnEqB?*UW`%GbE)
z>CfCz&HAunDVy9?Y^kN3sVZB&0`<J!MVE|O^A)x=ymhDLiM45Op{Wvw<;X>p4OCo*
zm2u3RZ;DQ}`_+}IomhH5U;0?*6QOt1a20rt*J_`i%%;{(zVWIy%DVb(6t}fmzIApd
zo!h~|0TPz(o86M@)=Q%$POjhb!@j@2N*A6g6Y-+5QlKqXQ0qwcs!fUH)i=s=TTfa6
z3Ud%OIRVyX38duhna42|Z;unEfiPET?<ED;$@lMN-9upKtFxp?uLc<C>96nprH)JF
z&~SNw6?p<^ve<I;H8gRwf`U{i;wx9)q@|_lyX@}oYqj6mgQ}nOwEp!_IXcsQHWK<A
zsP`lk6z1qmH^0lwStsO@{o~?jRI@ct$K<R!OaQ8cM$gB(=J4`NXIu%s!OYAI9WU?I
zOx0{r*a%2|efHz44>faK-IH2q7^n7=rNa}^3j;hhlio~P_wVCeASLat_Hcu$FWGx&
z?n0e@mD?)1UmH3X$A)~v*245RAIlx)u6O#tho1iY_Lf~_1aL(r4d3Eie`dJ$5g?em
z`rq)N$CSdIhFWL?Ah~jHK6AD&{o^!xcC;i{TgY+o3ZLWR8wlBE*NYb~PC}8L?M)@p
z_1IELUEl0O+tGKj{?(Et+DUzoZ)tD8gwpoR)+qS?v>F00KzbLzo%HAA<l%PKTel=q
zWMyQOvj^dp80<kCn32LAA)|_I?&v6v;-aO!|9gMGV$29?K?auLcEK={>(26$L&Wz$
z8tW|pl%s7<|1@0hvIhYgZ##EJa<($ol-bXe587CfbC`T+ceNb)QkTBuFOdY@i9>l6
zE)g{oU;w6xIAO1GqiVB8={YE4j;_bOsq!VRD{sOE-=-eP|6P8%kLyV)g&410LwOzT
z(#LviYozx0zJ9F_agaf0NiQG!9%|BXx{J8m*1QiKAPMfETxi{>W}(URsVRfa+3xx|
zOWgD1ub`oD`J(It!TQB{P5(U=zbfZaPqH-Dz4#_JedD35g#}X>iw@%i5jh7QAo4qk
zu{q8g?c9Ec?N8XQ-MaNXeQC>8&~b6-c~jHF{z5(XwbRvy+gix-Xv;}Xw@0PAj=gyK
zGU&qxZ&*dXm|VDul8%3F#QNtKCFewz&BsdbdLC}6Ax#TX4EDAb@|;$5;dY<BfB(Ko
zS3C}YIv)U%@o=_j3Uy`mhX)&-zvfcozDIJJe=+N4T!h{RDwu?g4I6TWpFUlw-0YD~
z(<;gFHBYO8_U0MX_fpp_Mx>E}@;DhNd?aFL#|~+?`^(b@V&dWn5I9IUMN{m^LC*d|
zDeNueJ{!o=P%kh#A3!4<0C};{3PT>l@AO@owdxfd)qD@dBp)Mb26x~8Q7GdNI3rwC
z<>B^dsmDH-(C^8Uvg=M$L6X$>#l%iMe*Ac9;x!@F{9|G+CME*NS9P#FYXIt>0fa}+
z?d#XC#L(hFW1a-5GEF5jj6MV9<>hq%=YjOC(;YE>(CuD_atgUeLPmxl)^pEvbH4u^
z)EVg9>~32#=aO2-=>1J~VSVC=MV%{%|7E_!rugK#MgINge~}1d;qg4+fjD_|VAh-B
zUy}KlG0)>4YEHZUKcSs=&(gjsIR|U-&nxE_-v776-QSyt<iE%h28RNV@%0H??Gv_1
zAeqKl0vZ9Wd~0k>f_@}TqaY=#Abaq|>2sAj+HnfH9aMTirm$Wiz;#$0>K+xh9fj`b
z3{YdxZMxj>VjgYkHQZ7O<67=X_G{m|X<1TQ$^&n%+?|fPdHWWLVwuD5)u+RO(sFX!
zj<!wWhGjI>ZaY@+4lzl|2Y5f0eiD;}YC<`_3;fO3L~*}Wgl2=Zj(7I#{f2j!NG@IS
z0jx`PM=b~8moQ5{#0(4!BKPmtQ88`B+<YYhQMZ=sb$AYeZf0f~aPHyF2J4PN9-<lo
zPF7>3KiGa-ub_W@KW|NjX(>l9C_v3_r12VqEAb;19rytq-DZ<ep;;2YqN<*uX*fWN
zOIm6@j?(h-ja>;36yfa%U>VwLYm+I-!9o?g)OtrqNH^#BSD<h-;{=5A6c>=DPOeeG
z?y;jAvq^z1`QN^OFRv6t9RN&3wqQ5(A<EgBmmm$mez-hRX!^2b-1Xe?{=~7(kFTe9
z#m5ZT%lePR^*V+2vsw)~BV>5c^A|5Ose2C%sl%?O6%e3=&pNNxomqn-@(fZui-#jH
zS6Nymgpf?cp_L}NaNz}97bS3#4)ez~^eqO`fX2Vp*BkB&9&YyDZ0Xb(%+(9uuk<`D
zKX3!sRZr7VR4G?v2RO1U4$_8;g>drod>S>?7be)P-=Rj!2VN$<Dd(wOrbrft3(f#l
zSoX2u=Joa4mnwBoJrY1Cy(Xdw;I*F~()L=NypOb)1;(APb#og(-F?Pp)OH0rm5Cq0
zt`kthzdWtSwI+M%jx;iuc?Mq}AD4KJWhr_2@cZ#;X$-94-vxo*YHDvkje^{yxp7cr
zK_?&HuXcO16nMslf$%~Mz}HhBn_oftO@uVc)oZeqJh*;yB!7{qmCH<++;x3J05Xa%
z5lz_2Ona2CX>STG14C0fAJB}6YPq_<mr5qyWlKv*-3Q?4d3@;9_UZ0P)EA%xfu=g+
z&9!9BL9R1?i;L?rkOn1X%TVNb0Nn#-d)3vYkfm9KZ#i0w{xktSXLpJ$1@wPk=U0G-
zdid~REdY_gIM2Pa$45IgE9w!8E&wHh!oqG%RP=f-=c6m3HqH~xBs241m$&4SPX_AE
zO?4dmb_yR;Qt>&|PS(Bb&er0db)Za@5K{(ezz|ubKSVGcU48l{kY7M|NCN(VB`_$;
zfqiFKeR6Xlx26RPJ$(|O%+;Npl1Cq*(ew89=GWnfQ@z|710-NYr7YAsz``@ndrCpM
zxMx=e8%vkz?Y`MS)?Bx}bSu}ogN8?_CPnyX3_CmfDirDyC<%XZtS*l9S!Ur*%Ls03
zY82#&Yk$#z5c~~*d~?e!fg3?86w<t_aOr)FjSYZ(^Axy7=f%86ixgmW?;99=o?lu3
zlJj-l6OqnXK|$SHZUI#MuOsdm1AsYk`}XadjAuxKN{Xx}sDbW&2t+7gv)91cx>h_C
z6H6xa+FQ4U-TeUi+N^2<LK-1Ts5bZiq8Nc?gnr59=)i?ZtC%2LvuIuaZPiL{&Cv-3
z=Q|1x4i0&fET|(!8I(6b)>iDoVD1CFguTZB0e8+5Qt{mg2g1dQHSF`IK7~tczCYdd
z>m2QF0BV6dH0e$}jqvmRn-5+S*v@NS7L5l08X4A4BJUMT;V*%-)+hw^Bc&GQeU|hV
zPZ7cp5iIZR@Bf0kg1}J2_#p#!+I#m%y^asL!&vnKAZ)L*u@R$ye4;AefRMv_%Ebzp
za$_=7GNW^H#X&Pbb`t={9|yY|Pz7H!8@w2{kT5d3HISoYj+TRNoYbgb*hm+k(Z@Or
zY=NiH1vP<i1Hc#RD4R*wwTNOUxQ?k^UHLM_(1Hr=jM<~S^*nbQ=9V6P>~W2`5acxL
z!1ta;m>R`qF~oNCNlwh*q~z)z8|^7*>3tR#El?saanFBo*)ZyZ+n_!IP@ykdiww@H
z;T&tFxA5@+=@}yG%3g7nOTh8sc<bRk><4IxLWMG_<hzYJnU*|jBt(e-Wzd2RVd{7S
zi<|5Z9e(y4k%>f-Vn-Q2@KmYA-3wchOL}*ler~t}@Mb^uJly-!D*-|epyYjVj*KN9
zf}F4<jt`xoE#$m<?K&M@_Q3w`ZURD5spnl^u2>NYm&j%8Je%$H1OZV?IhpWd;F;gm
z9POQh^F^5G;Ex||Awc?a^(GiCAs42|M4X6Uy>jE$twe;u@PMdcGwz^NPE)|KwY3Fu
zvk}<A)e-Ytmn@m>wrE~b<BsV278Y5ZM`?QGxk}g9bD#kyw60vk1#a0FnBN3wbH`aO
zkSL~qefmzwVr9H)sLY<;`zxvsv>6M}tA)u@A8$EhL)G3L>MWhjDvtf)28jRtmAfZ)
z#@%w=yP6;lsDut0RE4CZq<&AnB<d=)eFlQmaFq%l<T8*AXo22;`{4ueD<7W(Ag_MI
zg(G@U)F{F~Ew2DYbp{0;e&x_hzA=~hNPaA@(=sK{OHbIYP<Hg5K?xsjeq@TnZjT_s
z)t|QH+Zz%SOnFg%y(ZF9SWHx7lJ&~%?d(=zUnfBuTneW~LQL#kv0Q#mUU2JT_3ku9
zR7q;N{qhZZ`d=@Jglgf>oA~%^0_#n5pV*$B=;`f^VaZJWoSI4p@s8?(rs={=I50a`
z?-|7diB-CgQ$Y*|1c-b6)-7DfleeLdq6+A`AUX*>=by_)%47QD(}4d>2C}Y>yR8V`
zxN&1>Y@A)qNJi#;oAJ&v#u>I9(Bc8GUw-<K>DtV_ZIwtOK?rt$_T*d^$SL)uE1fxs
zOJHNB=?AE`{s~)C&`Y2)5i+>D%-&czOPvJwB@S$xJ7MFB3ZDxLgETdKh<RYis*jJ{
zD%V@?IRF0PO9E;FEID$0SMC`Rz`af_r%@{%WqG8bp`l*xa9h`N*KiCysDrS1K;yDG
zE~=--7F(JAg3L0cy|Q?h#SGHjx?}}pyYfOy)1PlUHxijFfvB`c&QM-p?G@$IeL&3o
zV*@4}KY=(z;m@SAkdJMl6A}`xKu_szou1hPWpfJ6{`Bjq%*mV{GClXF_QRxx`B-3#
zes_8uy=)1+(@HXm^}l@R4{5pX30pkmq;+e53T~DG>jMa;#V|UMSODsM2f*eF2?OdO
z9zrI5pnae$d<U~cU(rjC{uU!t?ZdsFa#PULX>Z6BqiXi(v2eT%AqJGpc!&pt6NZ3N
zNPcvQ!+d}M7#Z~{*CH(<92HU!DTKX@I%Bh4j>N=-<Ue%ILk9~&{2NF);vWS%<Ahl^
zWMpKRijSbg*&PBm)&M{PyPuc~=k5o|q<7l$-E_f|x)=JRgP_-$QG9H^ArU}n>g0ub
zRZ~;*dv_+zZ)pYUq(f;8){ukThgcL^;-(s?!JS4_j~-LgzWT&%9V@c~NNhxqnlDD~
z#iH)wKVAS7F>slXGt1N?fu%^DAILY9wnaA2zT&&UKr#@pGIlXQ=&EbP2s*Lol;40J
z=^6LR<m40EpWn{*=ezd6E;ev6>WH3m6uR|MihxFFEd3cMPFztg(Lx@Yz^f>u_knj*
zw+s&oq8*ftU>Bd7ntB6weFWnP-<`2F4D1h3%r@#s{>T5=Y+%0+1XY9;;M$)M3J8#C
zSTBWio42V|Y>}sx)6*r<5rT#jk&8J8?XDXCMLF9Noe;nk$7f&GP1@{3=RJba9sXWC
z(mDLhqLsqBk{0y_P%WQKDpNh6EbS3VMT8l0SsE^AOt%a{7Z!QRyT{(qsqw@Z9RcZg
z0VVeE;hTelf(8Tj=B1>hBu>@|P_pSZggy2Or+4|E!HMS+K$d4J_MGiXP>p*UfT)%!
z?QLzF>Jb)m&|CI%W(zn%`{-D@1PROs?iAn;GUy%96=TN-8-ZgpaBu03$ALq`8Cp>@
zMaP(RPdt>8vYq$ZN>fa-2zv^mPKHOiajc7kfw6JsKwX5Voz(v4;_W#r8a=#~9a(9+
zgXPV<-i++Ny&szWD-bRjha!@3mzIybz93{J>KXy=OH4I%7cLE`+`z|<LukGjK#%=5
z^weHBfQ+VkXtM+71bn6Kr%r%p*7l=`dP6`+t|*_u^Hwjmi0Chl#aj*gEFg;alq*Uc
z_N!qeBv_4XU|q=|t=6EfL8dX<kxne2dVosQC@|uB`R#HG>u`r-BC|HH6Hp|%kzW4f
zoMQLyKgZ;|Y61daYCSsK$8<Da@V1*qul;P$AGIDU?VejYKH9g=v;->b`XIAzrMhG6
zQWc|0#uQsqk_BneV90GHi|GUbBB9+W(jO};)*HvKu>vY7@^W&;5yqybbjB@w22I4N
zt$WY~4GUiGV34X%2YzE<PN3KdV|I_okV3E2CSl0cc%(fR_+A_l0&0Pwr{z7AlUsc=
zWj*h&-kTB%(HjAki|!Z#+R@jA!HeMwno>HE<CL>7kewQ@v$C@E7wG8dPNP8NRC$Tt
zvPUWGIrhwuvE>5sGTiWks5?00fUQQb_M>4v7bnprHHyKgg!MD-MQXsFS;WfcY=4v<
zp=U1d4BlMxS@oRg!~B}@f0nAE#Y;NLY<k#cRAADV7Nj)l##mbc5TWdFxWJey<>qxp
z>noF6iY-Kv%#yvO7e1fFubFL&;+BVEat_!iP^iAa#+f3ct=%Sjm$(ZJg&$~<?|_^f
zSu9<|R17j6{$Bef4s9P>giIp1QILHJq|k#L?kfZU8tRRx0DUl{-oy&JjaZ&3LgfPY
z)zT7N^8)gX{1g`3jE00}w-4>xD6_e)Zmjckgz>j`m+pLnDkdf;7XWzhD+sI5`T_NG
zZKT-p&c;ZoZBn~CR0U&|R5<KAusTG<b(>r?H8s@-)+Z4l*iHP^6Fy18$gXX+M@%x5
zymn#&tF<TD+h3ojSR5Gb1*V0$^+((61uXz6*?b#-PsyMG`f9?_!Uy`tyRC&mKR{sb
zKnu=X3ZxOX>Ni4?8T1=)RS%$Bf65*q=IUAw8c(?M4<9n*T>H0?&Sj)VH<t{M=_SC|
zUk%vLCUrAFxfQQ+-7+YL9}j1>BD!}ZR1rUD9Nt2KZZ0=4Fn9yq&0EML`0;ECDuu8s
z?gBE(<ePKg16PS&6FcuCYjY1%GrFRDeA;i`I48d$IRiN`oP<RUoMuo+$PJ12S7<k`
z@Y;QD7l>BuaU}W5$1uCH!kQu<#{!q>1B(P{^z-M>O(gLsu^2wI#nG{9ab7l!Wwa@`
zR>*y&BE{`w-r9xqPJ86;1J!pE^#<7Yi(1FvtT-EQ^DT@H!Y!YV23nMxgwz7F%9D!z
zB_}IOtS2r5b~xefJQL$&k&^db-x{LL%(I0?Z4ncZARZ9{qLjFaNg6O>KjF&PSH|&H
z%Uelt#)G%HLyo0m&JO5RyKAHnoj-r+++`+WNE;-lPoGBaJg7sq(A<)clG61;7cHNG
zw&{}}$dJ!aq0xKO)60#<{&x8_WIC*Bc4)&q?}Tl+RK^0D4<(O{2sEB8>(fvW;I<%P
zn{l){&tfP@<^PcWI%6Ob_mY9sp)^+3Ra;?3yXEAc`CT{D+T8(IH-OiKsrM39ggQrD
zVv?HQV@EDvu?_F}4_jOYd>%Z9aFbDUiAhPPQEQ6@og<hyO>ec<bp}Z?B1#?wJlZTP
z`e+W*V2mAKE2#hQS>7xsP*SIaK<Jxa7lQl+`4cj7UPNb+`Jh2L?St2)+JxEa61bYx
zQ|eY)=uJlQO3ez#>^3@1&i>@XnW?F!cmfm&4b6%j6)wS*YbCj=*^!v;Ah5Co%{i`&
z>sWXg8KtCYm*u0L+#}9*01e6)=S8@lr$$b$a(eLm>^bQ@I1M0HB1ax(XXVU}S5=^{
z!L67X*w!rcTbI^`RB92xQZhT38z*DFFo2e(KvbXU?d9Upyl0J5p8Y9~3b+9%9<<=j
ze1EHvuI3qM!7UdnTOgMrYO+TUybjhR%Z|SLQxI;rf@dasq<b9_co=d#b;yscNnd<c
zt-J@J=$W238_cn^KULIiqI}-uHWiOe?Nii4KlV$85u!gq`?wtzUSP;73&EkCs=3V#
zvM+1I_lOEcCOXsluV1H>9~3hLu00?Uoccm&c4&~j_gQmmEe1`$my$y%Y)#2;%-j1{
zFr-#4)N3UfgSE);Dk9s2vIEnlO0c?@2CaE3Q3=2`)3Dq-_xZHwo-c;x%^tjOafcoB
zgDoyAizx!oUom0tc(sRChsHkmSVm=(dQiCoSdv7WaenDoP{;BrEvrCzh-_yq0u}@3
zV!#asMs9Ak<r<(q2(3d48bj?Rvp-tReS<nNg0(YAm<^XP)`gx>KF>X*B&f_aY8fmW
z*GmpT)E~?6Wp8p?9J*UE2a>TNpH{D;y4BtHK<Jx|5LD^1qTWEVfV-C@alR~9_~`co
zPF87Gt(f0EeG+DdXdkS!^Zxb{W<=M|Q4#k<U-h>`fZXia*)IIdmInamhsVY-j-#Oa
z_v(#d%2Jj*BuuFOl&1#h08XO-*IR9PRpeWO-A#Fe-+49uVBLD9AA~L2>9(t>t+Uyv
zT#%T1XHVhcz5_utY7RDu;VvcS@L2-E2&;B!7oFkb`~!tCKo|NRxN0XV$5Z;z8HQf<
zp{S>xp4E#RypiE!qAtcV-9f1x+aHS-KbrM7I{BpTr*u#w_e_n~><%sEPJK~^o|RS3
z;xA^M@<M}4j5yLdx~_9Kb3>oMN#p2w?*3Rr#9P~OCcRV>39I_8K%u=fZ6Sx*4;Y6|
z!JX)O_Z;oA>){j5zs~!bJ`fdMZM)ZrLV^1&W6!|Z$vIH~T1Ii$fhZL8@DG5`0h1Fk
z4GmFDM_KD8=p6U8wJ8yY7U%|aKwZsC)v;D>Uhw`(KcgATXPqerRmwLmE>3CRAv`%}
z8$shJ8$|34tB(a^-YC{mv&2eg(Hg8jh}Pi!HI=&t#3hi&?i}sT(D=Gs^_Te5`xBFr
zzEsn7X~3=n`fA;F1J%77{D2jvO2F;gfjgzY91H{Q;8j6+1M`#wk^r<i!MWbvs8gp-
zHEm>x`rt#oMJX)0ceb>Mnwc@8;-So%4Cg1;^D*?6K=*H^^QTr^X5IlAU=6y~M38~;
zh=^FXl~(}5o<Jei%kYKI;uTZV(=F@-RD2hq<T-t+21IOEHhX->kYco4BHEH$%WAyx
zcKFK|FSJLBIo76TT;${kr#s__P*=c@(|)`J#DgegHdxZeKsArY`(O}xhnP5^6-5+m
z$UtwQ-qGE-A(k=>ehB1i54K0G$p<8~z<dVDR1_8+9E?aF83QjRS1;ozGe5r;=L#%A
z+2YE|N)WpSa4Exqdq9pwFW(CxbIPwlH1^#xZcxl$y?VuSP-!uk6En6A_swww-6y}T
zOrw!0RGqsiTV=cWj!{YrlfHNQ)Tt*RE~31JJoYB@8ma3+$o>FoQeh!Sc#HA{M1*~U
z;RYVcyGn#6nSc!t!?!zzh~iI~GdSl^3K$XxW!xEC6qN%+8{vi`ZhiDwXLY^>kssk&
z*)E9!VD3<_bk6IaP|eYP3yfuI%O=P;iBKu82spDr?nRsgu-FJnst%!Y!G_DDKx&QD
zE<SZbu({`M0X;%b8N_#K02k8m!muU00LKd(9tv%s$i?xc$<-o<xgP!9*)3P#PxL#Y
zd7-bV1*)n$;lY`{G{uwX`Mx?eJrBcs@nCnlI_z=`Jh-;mNk1rltyripJTh*u#AVtX
z&MxqKmukQ;5Pfuzu&QA*gGq(8^B(qYrfyZaiY~ArcDvw^RLOqH?g4wnFEq2TkYutn
z>oTGO0K0@l@NK1(NdW4SnwpWY*DJY_5)-4d-$6vd`L$r{UY^NJQNz_V@6RwSk{kXW
z6cp4@j84H6av@+kXHL)Suzz~H<=V?`69==nmW4!<4_`mV&47py*2-G_I}wBUpwFPR
zK`U<zq;X<g0Z{n!Eq!b)_#ym$|8|Dr{u7!X)tNRhcv_-8LoNdw!CJX`0)jITT1Yqv
zzJtX~g@hAes`dWL_1GR1h%#-+pMfFG`q1DZ1V3c^=Ku)#idgj&dKUV=Djl8&81xe&
zufJ2;Fz%aTZHJ=pu#$%X^WN|3KJKmWc}k3?TA1d&EugHydMt8$bZ9%<MT<yv&IfCa
z2tTsAY9sTBt!{pZgrqyB9%+sOK0%!gZ>^a!tBcU<iIGYgTqt4)P#%%U<<W`-GFG!n
zugEM^X%ws<sN=&GFCCWfV7qT&sgf(Mb~Lv>rmKHxQ=OD)2gHlbA7N191PTTA&DZ`+
zwIE=5pckhD(j3s{mAm?<L?1lJ({I4fv<#^e!-n5gV>l@uD6syfzHqeI@}$BQg)D`H
z7Z`VK6^E$^c2?{xja$f^pQjX~tmS7_Rn#XwG^6YG;#+X-q_)n;d8_$3@;WYl(#}!K
zjWydtiY69FB+YE4;#LR_9Sf}@NBbL0OfoDL>(6whgzZ5UWeBJ8AA9_mdIuUT_3VwQ
znKDBTZWOfIjmv?0mLr8hb6{NKFWJU|KjaLxGit(PCuS@cY`P%dKT<vGkA)2aSz5i?
zofnSq>RDsJkGeB7%9!N@U+NR6XaSeU=#ULy=dpI(=Ay0ahd`QL;j*{_NaHmS!D%Tw
zH#N7;uZYEc7xlZA$E>;iF=pvwSYhCbxrdT9H9zynBmaRd^Pw-a2;T(uC^}&I5KbK$
z_YdHi0-%F1;{ZXs9O52@bt-EKmy(i%8YkbmARz~?#`UmApnyPn8(U-7u5|AH;aV{K
z9z04Y?|J6xfLFf0tyt(`(;@B&1Z+oh=vxZBjy;FV3fza|?molivL0V#vxZNA27tpr
z>9Y<fa}pt&+uC0JA)CXJw1KL&;-<lpbwR6;K9FAM3K}_8z^3iCpFg(vC-mXwBQ9W#
zxW(r%>kU0I*fMm+PwW5$y$%~WMM;h%p>k>XQianBE(-Cfo&I0}?>dh{c)*4Mjh4j0
z1!!EN$rTM<E>YPQm=Cg`d<H*^q%j)qH8nNq-|zo}Cj`BY1jb!vsNi=y@MWve_%?S}
zv@G;pc!)a|m3ZIJvCQVj3vE?7nkRn5C-gkFE_r{wP0N%-$;@Oq241|Vk_>(`ZqAmV
zASXM%aIpV285)XUNU*kfr|_8twp=_xa?5;Dl|)k;qIYE69%;CB+(ZT872X<VAI||N
zBBHE((Q$bUNux+3Ja_c=CY}|0M3xm*iMR`a<>8kEm54BAEnHYj$luC2+QU{TJq?!D
z378vL>=SBoX*rhScU-iY+vtfo-tnH<Yq~l2YhQn&kD`&VD<`{BMYu{g5Q+^MVjO`i
z0?COsb!&d2rlwzqPZPA0AiDScDPL&+YLA~nK!&#FR<7gc=jSia9{I`LR_Ft7I5xk6
zt{JtUs1xQwyVTs?F7ECw;QiJ6IlaV&nNBQhB3?<zH~uzjKgD^?HY1E<ZFy4rp~?nD
zM4Uw-c4b``>?MTWhZO%xF13Js?F-lyQ|E!#;^^;|GK{d3>w@~z@hnLKcGerP8mL#R
zjd!F_S}wM^2NtWp=+h~8uvPa(z6n1M;A+Fb9ZAx#LT0A>fzZ2_ha4RrtsKA0zVl~y
zw;n%&iw1%4I%ErAAw@vLR7?;(U4}$t5S`=+h68rh^v5^<fC7e-d1iFa!kM*hy$ml-
z!F(yqt8!tZDvBaEjNg(wVoR<>yBr<t%E=zC1B=>fsHLRh{BuKlX``9CU|9QSnSH2|
zTB#eg8Q6Z!O~90axE>%cBAhwr?~Km}wljbW@}iajo2(N?L)qr943O5mb<_C8j(Bg7
zd0!41dwfHym&@fv>YrN|K2V*Wgb*5MOa@iWmX!}3_^6OxcOqB2?D2+&`rfSf9!5<}
z414}$rBD_io%IAxZtHO|NP02VI-78Z)$<Lc-fjy+PYpSwXPOVjH^36!ykkaGW^;pw
zM?Jt{3~a2$))Q9TO8{c0!XrTVE!lBNZJ~Qri_o_~8$#&?i`-e@E!8U=nE|sPHcMFO
zk);sEns|-rpaLv7UOurKzQb&R;T53;P^dG6l!%vUF@<>{gzic7d9ll((b4Z=Q}RZ~
zDd9Z_0i$^b#xFu!!(5h7lQ2DC=W+~`xn18Y%}JT7R|>X1+-)^^09glQ?u_Mq;BSWt
zjOFucW&pMYC4jc3)-D+_K2TrsNQojnloPD<B0mla#jQI_03kXbZVe(1VU!P1g1w~F
z=xK(7gl2t^BprK!o;(A>^=XCQ;+3<rMurrUl9FVjy4je%SFc~UxU_`U;mU(UDmcy*
zwu9<m9y8J_$CW&P{+w%ud|+>5%!ET9W(L3=_?!3Mp3t2;(zx!ATq|nq_Q8!3&cMnX
z&F@5rklkQ~LuNkU-X{*ZcboljWFNqUg;sk(<|gF!Cd1Yy5Bboy0<3TwSniyZE`}zf
zrTaIapmoC2^6SbpWj8zydiuE0+eoL6Gg1^vo;~jbRTt4to_^8UO^T(bevwkyng)GW
z<cz%7R4+`*fH~<Y+m%x&91&35?t(&qaLO=T!Q2a_#}8o48>pXtAQ<4G^Z`CKtn+&v
zRkq!^dpB=r@8ubymMzq)x+lcaVN4_lxd9|42YCbRd2was5Sd<pT)w-Ai>xRXQf=`F
z;)?tRg&6p8fK|4T-nvVy^~%O4CiKS2(a<bIO912u;QXcKT)=<RU}zNx3^fKH0((0Q
ztYi;BX_U%=A&KaAyJ=v>IAC*<@;fbuG-7}Q0dmR@K>Is@HJR(00pUFpUdNW9DPS81
zj~nTAVPU;#F#9p*fvw}N&19{<OR3#*8r25a#PLu+W8Ic*KXs6TmtnG}%(NBS+EXYP
zmpFeHtmn2c!2uTrrwzaqA=0`bw#wqb_LRBQ2?QIO>6UyfQ;u#Gqs*<3#EGkXAm5W9
zRGG)#W>8}jLZRD@!fNp`Edb4V3Sln|j~<b+Xn)x|W=GE}XYK)VM}&At=C^g7f1Jh<
zS%f63F4V$4%qL8lEDZ0-+e%LN?`s)BW(@3%3=HsS@pIFGV0;H4&-VCm#}Cn29H1mC
zSkh@d8etK|3yO?1+Y<H!C)RZcJ1ej4MUA_lre+>e3b_wGtw1(7*3tgEd2ZD$)Ef{Q
zd1VlJV22alQ?i}x0FjfqM|724ELzY_732)$3;jYuF0O%B1ev<%ZI9;dwTjjBjMX|i
zgF_637+tO#wq3Y&oC`hpPndzZ2zTZp!coCU4scZGXz@F-kc9Ifdm~1I%wk^8gVK4=
z8c^;cjv)ESg%4hR`}S=<G7oYGal|<4g3w;fWDFGCxA7`B>|J?T*-M~y=k1M^XfMPv
zeIy6guoYI9-q#cy&n#dsLs<Ty3a8x07K1?5!St|0e`v2-D-pY6Pfri!bn4<c9Af0C
zJTB{>fr$?_g}F9RHd2O4tm8fb7xhWN#qRBVyh@yMrjV!Ud{nvJL8a|tp~amd&(h0V
z1@6ThV6x8At<nTFS9smNwx`At`pL0I7<-@^n4m(&C1&!HSs#i0F&dwQl+FU}bCytH
zj7um-Ex-G3n|=?-khW0me0O;$#1?u}1AEb0g(g85_d>i?GsRc{bIn^I4}K{N1oMq_
ziqyl0Um~XcxyKbSHFZzE@aY!wy$0SBW+PG;lvPhw0?5th^J3O>q^g3h{As8!yG`yk
zJ^{f+$Vd$>PtS&BR6zz(vKWQdT_dNnt<6Dw0(#AJs6_ZC$6kUfY_-Y9%)sh4tRsSh
zw)NS|e2Fr-2kQySng|J2bGRL4b`@y%hOshx>^^8FN=5#Qw}IZjcz%-4HCv}*z=*nq
ztE0KuwhR-Uhh9*P@Vy!T)Q{cfRBigg2yAXh?BA_o0Un^|%HO<u7hEUeR=6jy0!A!t
zr==0j)X{q|t7FCOdElIh&Ivo~4Kt7NO8Z*|-X)N#SZMZ<xilkPB*B1_Os?N!t|!^)
zvy2&RwlR4S=BZK+!LnyKBIveVT2vF<HP6>yUPE&As>b4P&Fw?Tp=8p>fFVmAVnEVP
zRHp_JplFuPyuXnwadT7F$7bmdHQN6-8+Nd!u%rS7K6TG6C;#BMa0m1DO`5>3i{}kR
z0Hu6@Rsw7QyrH_Oy+5jU#_Pd2EnPVAquv*>R1{ZKKTdwc1H{niM)Rd2n;)XP*G{-?
zx(XswY9Ng&e)Mq75EvdF2E_`xt<=88hKAuW&z6we>yC5KxM&H!#USN^YsR4v52k<s
zZ)91q78!#W4uWoayDx1hSxXr`m0-xjT}D^GAz-D}fVN6sUVy7?gh4NxK^><{S|#ws
zV1iTRBx=o$cocT~0^%oz#&86qiuW9Z*Aw3LXO`O{Ii<%(W*!3L{spu`<y>8iRRqE~
zqFX-;hHZlMRj?nQ+tUNY*-~FRX{^_gOT|7&@|+hgT*$IAXKH?VTe^qkvQmmn-MvnM
zUpD(Nh;Z%h-R@=OTVamrU<Ga0l8@p2ZU)Jlj^=u@yGRVVan1ttjJ_oa;3L<Odl_$3
z>x~0LbtUsJrwV@y)Y$%deIR_9l=Rt_4HYL~O}p_wEv}fnE{&io2MD-J+uN}}+iXkJ
zRzUYY0M#{vNM#QZ&7r#RKTgS)@mo+n2W@?x7ME5d*co&Y%O}X>)7^HrO)-)zrZhDi
zTe3R5p5r`@n#G}!!2`Jf2jBSlDJ?bj12+JsEXRzRnRifNn$DcAUi(fFIV~%ue9M`;
zT1FFQD@lPQAUoXKA_a6xc64+E-o!WH!$pQT@+^kgD@q9Pe~dG<K}YTfeZ3!WL}!7M
zjO4bqLQ~{3FTbBG0;|@0gb@P!Kt$KV)^(`m-@bi&17XgKj!IIcbH*?S4qFtL3myZ4
zA>62O-_3`&onWB6Z&G&YQL_ohI}m){LE#~xq+|ka+yn?6Xh$<D&s!r}Wzr2Mut$A^
zW=R~Z2w>JJMZQEpKpTk$8PI)i+`M^dG==H?WP3-4qp875>=tCP(Kc4-kC1qP(O(Va
z3}q?s6Z-*0{pTtGJB?OiYSLgV^ezxd-D!$+P`|`MoCV05_r~A<5-e9C?|L5a#3#J3
zfYrLbGPLC93XHN9F##t*27{c{u<I`XAQa6V^g|ZRVO9ZpNJ8H&Q_w$9*L2;MIQtJ7
zdK<rg&q&#Fq~KLj1ziQa!?cK!1(yl)bfnPukij>F3>Fsjp66+ealCB?Ty7#r|Imwy
zLP}N4)fGg}Fwdx+JgS$UL1Sb{x1!YUJ=HEyJf~3IAgfoHz6&65*e-J2GV4D~O-_Cb
z@}3_Uoq;35OX!^w1L*~J=F8vED1HMZF_sZt5jQj8xu4B+49lT2z5?Df%h(Mh69JF!
z3;w~^h#MZcn|PqffUu7+zsSk~ef9>l6h2B>%hnjmx`j6BCHo1ueL$eSBhz)h59kpe
zyvE&G8u!YElluSh0<1wak5<0?As5n2tw&wAu#BF6d-NU7vSIZ{y*T6_#Bb0xF@UkT
z4Cq%Mw6;XO8qc4w`@=oHfh_h8>@#nH7ZP0@E$N&-e(bd0d~*&-btMevJ(SL=1x~~W
zIcgS$-lKCaN&~wx;sXGuixf;BDS>zAEY$V+!Ccn8?WH>~Nv^VgW;ybnaX_VR%-jQL
z@VPxjoqro!Gb$n0G@4I#Fgcx>nh#G8z?gz5Hke3@AHuu;<uoj3(+;>k`!ZC{A=AKt
zfszo8RXR7WCqDL7;vfbiaA(Hj<af~*uzsx7t9GXFfrsXR7Q}#~g}d1W=l~hAPgDNu
zu>$TRVE=`nL$p2^SFImypoMpx_oo&lN38<IazA&_t&9V#c%*-xS1{EJ*6yj3&mZlf
z6ybf0f*}0(L->M0^@rZ^UU-S$12JVYTS8sCvortkXaYt-o)`b?GYtOEj|BMn-_t_<
zzx8AXmAcW>HsQo4=H})QYw7mNL?D#bYhcm=`)<A0@xGq#C=A}w-Mvc!{)=xwQThN7
zO$7QKCM(3CAR`|Cl9ocCO=M#D`kzLEyX*%LQ+9W};CB4|x*2zVm72!}1<_}kR}<B4
z<U49;U=R<{+;DURR+=XO0uf07ac{y<lSuAhh4L-hQ&98G$p1OU_3O;PExgro2kBs@
z5_zCP{p<6#%VWGU(L9Q1eqmu@ux`_VmkVsJG{E@54*vk&)01ZipR7+c1K^|a+?%7X
z{Qc9<D%5gyvd(!vJ#}m(41)2mSCfVi#iR$!6Rja$=^daB<t)_z8N4Uw%ov)08CGO`
z0IY}ZKxqPUx{3^?tTX7My5<D<i`JkK1q=6Nw+9a$2gk)rj2B}?L3%fIb;!@p0fY07
zovFYbAO&N0vdo7Si_BD@|H7k%>Bi}$pUxK-G`XPLJr86#DEp;Q7)$(qDZ(2sQd1M6
zpkqZ4Yq<0;4fxS7pa`M9;~-Ce0FPUh_Vv!&{^+fp9ZAp}Ds>2812p{pP;@<#%Mw4{
zhS3K+0<7V$|GCCDO+vWNehM8wD2RzC8-ihL{{7;eE>PldZfX8VkAniv{Jw(12Q3$i
zp)WYl*EB-2N<vPKB)TNPkS0B!&%&UBa!F@vYdy3fOF7p48KISyO;I6AijWXp^KQhZ
zW@T2)_~a_YCc!B`P0x&r(<@8Aqu?2EM`3%xOp&_dso%Q7;QD4w^9wX{REv@hnolPp
zCv&}^sIad;V&#LFm~!>q+DzpB!6}_Yjk_)|mu&2UWd>^pNdv?{V0O5~Bqa19LBI?x
zVM%9ma~;qprcQ@2H*EvDSE-JWJec*@97&e13=~=D>$Flshzt}F!2hMJE33uGHI|3n
zhtpHmoHyejuhKaMONYHOd3X$|%pICiSoB-5F#F44(p8HUfH51Sjv=mhxW90xnmRhZ
zgRy3<oqGbAUj%{!ADp_so|FH(sERHyw}DTS#sza~lza|&(C#AjQfn6ua%o@d;X}NU
z66-*?CK_mR>Y(F6h-)ZK6}w=zM9^kIg%CLGq5L>zzSqPjBn(p`GcYu~kM{$6!6hI+
zCplqB@GUGKG#pb)9@yj@Q9SsyO9c;gLwT4iHa!Ozv{T-8Dr!Jwfr%E}>;>wB`RSE?
zDlU@Gn+-(4oDnXq%oAKHdfF3BfBpJ?%VFv(;uFx?ArB$w2A|Y>n3JQkr|buNprWE8
z44k3B9X!WY3(?~Zb8*8uUeI7elWe9dp>`VhiAi;MYDWa&>g8;6jg)BP_h@E*SU1Dk
zw939q48o0<81~Uc?!zoZGd3=n<}SMvp9zc#attM1kN}BO@h-_4EGj&t*#?i4BgvaL
zX(7Go^tUUR>M$@<bPb+j0u#k~uAjmN5T7lo7Lct#G10%tfl;`{(p&B7m?c;>k>MbK
z2!YG@fXPM?latSPAO6Q}*;QakH6a##0yrEo#v+ph2jGd>a<<`yA#*?(Cs8HTlQ0ke
z_e)v1jG``+U|B*!;wK1scDG9_pzA><ctHdz*xQG#&*pXH9xd!e4OM!51$lNvg%j)1
zM%*z9MeVVvh`2IGl>cKBws6K#)kOhK2C$)N2LwNq4Yc1#Q9klOjIc16?j8qx7?{~k
zppFhU<ORUL=D2g$W6xm>9`T~;buX$NvI0CMLDh>)^a-^5>ScDxm;)Gm^M>Tfqsag7
zs$54yQag?M*!+3ViS`vTrsBAvFj|}j95M_@Qh+<I0o+%}Tscfe!}Z2P@<zt10eejK
zcKkDd^ybZ@7UPqZKy#^sCeu0!0nwPf#Dq))BDO+p7=3;YrXx-NYq$RW=JgOe?lb@K
zm>orpy8mQ|>HnQ42>p}X{)>Z1|F>0C`Ls2e_us|hs-V;1^wM<`EAdg5ah4M#B)dET
z|A8*9|NG#h?&HSss>Wf?iTz(*OA!WkbY!+4N<XVmehzu3r6@jcr^#_UK$l0^R3}lz
z^(=LjVxdZ5fOJFp&zZ+(XvMXd?b1GfO$beVaXnScK1e1=EO>Z-$9HAX$lz8)Yiz_~
z#LULpC*RPjfj#%}xK7=BZwIpuHBcz!+c>PqgQ-M9FRY$G{^x&w@PAp4bRmom%5o5B
zq`p1}FRmK9^U&x?rz_8YH{%<uXdL-?3EyYEIB#3PM>L@^u!y@Vx5_lj-?g4|ERFkp
zKj&D%RrAig^&K-=S=m5pb0_s)_nh;`D`#p)vK8i=)3u5EyShG%jEpR^oJ4`lw8Xi4
zys&(MaXG;@PG{)|^73bSnjIsYE_N?z<piL8?!Yr`R+ib>4Vpfz3v=jt9d1#3oLPZ6
zuhxV;7<nr%i|jDW66^&*4oct0$LS#;K2=q;&MAK%xd(IkuD37Z-@fdpQEKzlDZDrR
zeUULdn8`WV$y2cg=0>b?QtaV$l^8DOLmj>8LT{0ko4Ydo&NYnlkiQW4(yIc_He@O%
zMK~|4swO;6FC6emH-9D@TDU&sy3RAvuI_g0*-``?lzkw!8Tm(Y)c{A_G=(>3-Y25}
z3A6u}y9|kc5x3As$xNDs41kujp-y$nL~${LF~1{#fBBQO=SEbv*)u=^Wgz2k_zEA;
z!T?O$Wi5-KK+nyjVbBScw6sQ-NBA;zkQnd|Rk5tyhsT6rqwPC1uv_|hLm3;#>)|z1
z6cfKCKK_bxolh6o>QgF%xeGfm{-gr*x6H{b7&(o6+PNe{Co!S8n9FT3PqKPFG}oZ#
zvs46+t$t>w$1i_wc91D9`ZZ8G4m~ZyhH2S>pK>}QGpEIEd?OaAjb*cmek0mS#zp6l
zgLOq`oOc%fT}L#%pm*#lv>p^_P=B7g4cId+k7e*R_rTt)+8((hN#o(Lx3dx_n()D&
zYlMMJ+vdKRx%oX{@!MkfOJq6GsujrLQ0?g>-X&F!k{>UZ{tCF!Od?5{*Kz~~;OY36
zgpYy+HoG2R;JI_~5|MV*sjYw=hd3Xkf@QgeKMy~1#GnUnG{dvBdX|>{<l@55gQqKj
zxqld(1ye!k+1ZxcpFotjA$7DjSmWsS*LO0{^<qvt6nP!|qLDhfTFhk(+?iA`3>_xJ
zEJj0ZalwcOXyEqX3oah&kfM(X0T}A;&xtM$UC@K4wN&H<ca*HFL%%lWAL>vY8R>O2
z-l*;tAex|Xl$*Cr=|5;=eZ~COzNV5=ule_psPVyZwgHK%ALLdBS%ONImX^6!&DAt|
z-NPOwvt*p|HtSzFF%De-y%eaDgFo-y@pWm?&r`v%zB#Isr!Nk(Rs;I{jnr-y^Yilq
z>Pl*wqw{kzkeVRtAjh-5YM!9LHbC9<Cwb68@_<=@*~CYvG}S7rx37gf#+wT%23sxV
z9z1y9;)dlvC{Ow;rouMM`JH_Ut_I(5St=UT6flB^6-kYswbZiTu}U7Lg0!_<xzTZX
zXJ=(X7^2fcMJL7-Iy<Z_M0J{^mxBp5FXz1dM1*UvIvnNkk)a=Q?a!D)A88btC?cs6
zM#dvW6P{!V%KTFIYtWxxSg5FomIG1b%Z<AO)G4-E=g*(V8n(G+jI$3woFqK(y*0P{
zY>Ci(uS6qG$iwv@*cs&@lY7n?$i&ziFqf$7w=j7Gb`|+xu55d)W=C@~vwUC{OG<01
ztE($-h&gHUc3GO3%y#3fK$+3n+|1klzEh;kxD}aqOYqnOP7TW`<F=<xQTE#H@}Nzp
zahBkNUpuBf`#C>yzNGAK&dH+&J`WzsLNn5}PA-{Sy=PX?2t%WEQu(7+73hD%2aRK$
zpwI+LV)TvO6;QlYwplguZ#YeOI0YNPeqY>n#6dz<vj^_%jookbLP2nO@X)I1lqjnf
zu$=_o0QR#}!#D=9S{c9>44WkRtMcvu_tXg;>Uy@dM`q&R0j<)XQ&#~#Ecbb(IL6~L
zb*5WV)IgZMr{K;x*aIU~=wi{Ju51StyTQW_!JV;a^nga$TuPK(uPis`u&`NQn){Nu
zo_X<ve%`0ouU}`W=jn%fP?>^E47D>mgKJV&Q}?|mc2LW8b<lCtGTc0;>=r!h%?%9P
zu}}vEwJh?)w?X_{gdesG-N@$rCnY&$!p4OM`$e!DKJp#80wpm3Orn%<YS}`@i$i%=
z`5i49M*k!rlT6(*+`lk_{51f~@Jf4pLKpeloAKuV{NTUGdi1AOoQq#VayoS8PEJ=L
zxiVYl7Ig4o)TL&$T(Y3{SaGZN?3$S!?YhxCTjIs{5#6o%K4O36PomP2mG#ft76GJJ
zR8&-Dl%-=RyP?4tTAp?Rl@aP)f)AMRR$br76D+DQ_1-{idTL<NT`{TmMf7vB&n%T0
zF1{~1GcWs0*0poTTt7=t#hTn)4S?ZgzlJ|K^yXx?C@AO6plnxEya#V7hsR!}{n`G(
z!JXL&rd)vm&{5I$oH1a;dW!WjAP0a6a4k6{<yP`zfL$NPIR-#?N>yO^tj<JkPSg|i
zy&O#7^l$UF*{Te)d_-wMh>#EF;NaMt&2#kSY&V9nXC*bYLGgZ+g8x9ysK}oUZ^HRM
ztpNn{|C7#W9!ju?WPkMLTe+2AG<S9e=c;Ztw64s)(Dm-%Fj+Q@8E+i5SDVl6bpvH>
z@+Zse>GJVXUr!9?rt;WS72)FY1r%!b`ttuzk_WOkX3e0_=mY@Mo$(B`nz*C@zKniL
zVL&hlf>L#BJ58qj!Rv6)%MuOK)9_-RgF1r5RN;JScJWAt>=GaX&SD+L(2>4d$Q%AZ
z9?;w;KyFHII0T<xJB-=wUc~oV1hBL;{|g>V*mnO6Rp0o9zv@i`dF2Y2Ao^}JKR-kL
zsl2@0u}D$rnI`f}?+*I6+EVGtE$~c)*c;2Wf3om^>xuDDf&Ik-jY?-*m_?5Tmj>F2
z7x`{fX8*##q)rU~O65H%yP>DP$qEdG6*2z@d+z}j<=J+NV$_(#mZBI@Fh*=35DOq6
zV2R>@fQ2GG3erpH9gIdzR4jvt0s|-l(v{vZ78H;oZKNv#LkB^?q3(4L=KH_@|Mq|W
zy{~=Fx%NKS@iJaX9A@Ty-{-l@TI*g<%VdW%w<gIapkS)cZl$b~IjqxI;Ga5>m3`yt
z4XVz*-gSR5{rtgsxu>Y!1zU_PIe0;_@lHC3PwvntsPFBr{f#snRcD*iN7>HG&uXHU
z&8EB|)*rUR;%LqqKDG|NQL~MAL9>m#YbcZw1vx4QlSivxE%zunzVO_chc4IGJLV+x
zsHI5h%SlQ~TJ0A!`0IGrA898PtgNiq7iF{eB1ljjQF=V5IH<!B6PN}XU-H!`Kbn}C
ze7n^5Ad_yR4l25annDz?j>Vg!DH+pcA$Eah4nY;WV}=_xa~qo34kg9$BOBC0ju%z(
zZm3~w>zn~3^IEqH-SwR-3O?><noNFt-)C!#_lV1tf*{|kZQYt8s_OmKWH)2Bvm64E
zn<$6xKXb9PwA9M^{8xmLs)~w=epyk(qNv=qr}c-`)hk)UI(m9JTDVj+hKPE7mOfI>
zi;W|{^(t`jD}0D$f`Xko&ylEyE@_h9VaM+-JuMmI_9zEI%k1&*U&?gd^QX#4JDaQa
zzF0KL1mf+OxVRFSk4&p|8T+eB6@hRYqK-nsms?^vw;3cWC%wVe%HG%8+nZlp|MkjK
z=|$*~tD?*00xRIrlHIpPNd7(Jw<hlNl2uailS@CQy_M}bgm~VQkYR|+IVuuxIqlM-
z?(|SOb=NWuNH#RZN4M~MPD)zGYwUF`Gkf>Ebd`XB$K*u)&e~9)#VddUFs)GoQoSgx
zU7Zr&Ro{i>etyx)N?u$Q3H8HN#tt+jhEjDV{=Lv)yTcBJJ>D`n-637TV6OHW{7bce
z=l#^Nya8_plWjEcXIvq#4=_l;!+{vy(D3eC?1dL#_F0Ri8-EHkaM785f7MqbJ7<ZN
zxjA{X1qj!{u|X~^38I3nMRKnByf<*tcs92ggsviH{}1?ZON>nMqYUQDFh#}a5C>&^
zN2c*rc`0?pDW}LBn+(ONE_EGL>r%~b??@R@HBLqKHh`+aMYVc%O%_wMC;0M|qPH07
zYHJ5xMf2eff2D1hrV8;Mty{VC*I$@z*xl~5p`iLQ1FfQyZmnZ>qr%e9s!7QH={E_o
z)Bju`7WIsbjBxA9eQpLCdTfRc$AvorM%#4!Ll?I)F#m=y_ss&uTaZ0y-S?o+4wsJm
zxUaU^28Cg%3@60j&bvo13o{0e{aKt2P)fy6wX@#8dmfFVj@zwW5CqTDhsOrOZPc(L
zFcE!QxXvb{N!MBJ;qgQ0B|s^^Y8wGsBG;d}&b85A=@Q2_Z={y`%87-aBjBVpf;-R2
z`Q63B>(?K5r*++sa|^~E2<t7FnWo1$jCI)tXX(=Hp2T2$@ecI1MKC*cd^Bf48z5tE
z0Re&Rdr=G4c>TmkhA7n(U7Db!6-ec*?$|g7c#fE2?2~RB7T-39wiooqGH6(b5sMUo
zIK}<kMc2*8@?Bu(h#B+Ss8vLt3%crK7^Gk*C>O2=cKu)i;^BKhUt101xEQ3(tm@tg
zhXO}@#f+{H#5PkzrPBTel<3=_q@u$6+jc~N-I@Yp6-l^2t>9etZ%qcD?Y`htCz$Jp
zURDU5w%2%7SQ-3CgLeID1Aks+fLT}@ej)B1w^vG8A5Y95?@ko|hQ438h?5Slv(@+E
zwy%_NFmZtry3wghmBB<!!r7?EWne?zq8nkC&o#{-W(^3to8>iN0$zbeyR-woq_;UN
zDJ%I#jQ<k%$(reFc#dW$b!510{=;y$SkItqo-Ci4REHH;Fm&rCxQ`4bCN0~7TP97Z
z({qq(nd*0yOF2Yub~?6tv(=#Z@drOZ1!Itp{dMZEGxOWo7^e<{^HLA@@rwLwXsqMY
z{mw3ty`MV;TF>RcZTe9@P1<`{P(c{i*x_!2$-Ht}Z+P7%$Uki~<wVzLZ{Zi!T}Ln8
zt$z6Z=`rnsaZUFLN@sxulkL`NHCla^J(r9oPevB=c4Mc#h|r*Ic2^V^`~LQmC30J!
z)apL0r>$?l|6J_sOWg4i*trh_t_!f`=i}Qq^IYR^EdVaaH73UC?Bm(iH}S}OIX8b3
zG`#J7xCI)$$LXu^O~<sl4(7uWf)>^1OIF`9z9M7l_dLw0lF|MS1_$lRfj}IAff#G<
zdXqO?CRfs6^zreETDM92=YrbJn9ytxHCRHg*nL$KR|P6X+-~5(lO<NWd~g>e>{=k}
z4G>>|7lfA}f?KDQN)gr*+(b^iheCJBV4|rUTVVSyk4}Rb6$zVc+qm7L@}8+;lfFnm
zGAe9StlvWqkM|z-?&}Kc=)+VIg>^*rG5)JEZ%)rWir4~srubyLV4HM?fq?;<?uN#(
z?)u2Fn2!Rv@S*tG-Sfq@wfIJWPQ>hH5DVj-cp2ZIqa)a^%xHQju<_R?i{pBO6@7hN
z>P`g`sjF%jChzG2*^t4s$uJ~J{=h}Ea|;e{YGbO17hO>Ja0eyI`mm2rH`Z(Fc}_&T
z1A};hT3p-nCH~ag6(z~Zh&J#DyV&wq4&1MNr3_O#Iu@SWn}V@00O<bcOha7iE$SRf
z#8Tc}k&~*^#^u7Vrvrn?FR3ksBkWEL&1zkIb`tXnq0lh5W2&QsqcE0bI(!3zhGN0i
z$&CCDkrHV=;>}k^m&WIb_$(Ose3kVzC_Xkan(+WW47B*BJx`i!I1i*jD|<CQ`Lwt_
z`gIeyfM)e3mjpFd+!TiMP%)$_DDDF>^kD|0-*#`HRz;A-g@GTEp6Ji9OR+)j3c=9>
zGUJ^Qj!=wJ;bIe;5HVGyfcYd}*TCHpSk|r?O)jDfa`BFq$%VF#P@|q38l{`199U&l
z6H=Xj)hhLwqE@~5WI<AyM(*Slwf^7I7~1+zM3Wq<mluTDb_w)fd?>D)RiHLdIwp$r
zQAL=6FgL?@B13Z&-M%A(39v1Wky+1wLCC|L*#@<Zt>dGH>(jtnxS!_((~|Uj3H8=(
z__|8YOts7yB&Hle^rld)?^r5%d#mvQx;ME4-UB66G*VJ)uuf4;u~7h^$n8@14Z;eQ
zmMEBXfc*VBZDLyUx-WyLCpn57IFprAJir~>IN;eb;Q%ugDnNATVafPl5f;Hk=o+5a
zDX90e-ke$J00jXDfZ9uiiC+Jdas@e5n2!FA1d$@73$+LH!}<5!23L7Fk;|-^c>SE2
zo1)b}cj>Ofi3RI1ZM8RU+%_S-Y;H|L>r9xU&*aCUQAC%N!gu9(qoGmWVIf#jXR!-R
zI$qx9zEDzN0#8g#_GLNOB7uj<q%<v6UIlmyjIzIykh?J8Li4ID!9HJAt9+>)J50Dk
z?VAIX<8t$nAUH49U42P#?H2vapQxP0g$^UY{scV8hNy%M9Bkwe;IHm*b_;{K;tu$^
zS;KcSh{_3&=I#oEzb%a)`XvRa5*b#$!6w5XF|+j|Z7ZE@CyP-0tk*<C3*`zb$P$ja
zzLU)a^x~cV`s5{j?~xkRo@$gjXp80aCaVA;ULz^DW<bdE9lO}EA}GzFp?<78)m7-e
z?ee8=OTo_$**Vh#p5*vWb3|Gi{a_+?1CmMwMkShw)jl8JehNZW@<?=y=A!*UDTt}U
z49X41)MVB}_h+H42m)4xISEGtyDIm9?t%m#e9I0D4ws=$493`iY{^YAm50Y)Vgg|p
zWAYsQV{}U9(wfOP3$kEgNjnG;OB_`S(I=>7;Hed0=k;z?T>N)l@WJl5dr7>Q-i!7b
zVliG7lay4hscjH|b$beawPhI1uCr!+x8cx>wHH=1F>5IK?BrH@3^m$DEMb{<m$wf+
zSDb$6)e8gS5Zqu8w89LS9M#-VjvQj+uii_Q9?=^&^uK}y6N9uwWj-+Pn=`vkJpLV9
zUOemV`Bg&&GZO{wP@XuVHadk(Tk_?`ki!c7K~{-Hu(6qq>ARynQLizD;WT3FGK0$4
z5tp!mL5~tl{x%9fYCGTl-iQa~jxk!jnsSO;!jz3GqQ*sPeD&lJEEvr?k55o&Cj4O2
zvdSc*0?GnegO3+(2QE7F9(Rjs5vB=Nr%d<#h;43zZ_x04=xe}SY}TnVNQNg%3MN&!
zA;6@!N3&dXsc!>3yd|eEUGEd(PQ^To*2|Y`>tBZjk91MImK5~NlwsB#xO@=UCF~p=
z8ttFAqXs*~q|h4|QuX~d*AI7``nqyt0&XpcJr=#?*-q3u`VK|w+Z7c@UmJ6K%@l)i
zhfVOQ#nGsEhnD8Iv07ewb!-pox{Pyt`1oX-FqVLObimK@{ViHau;G4z@DeEO4h`ux
zq%wMO5m5`5h2&*1U-Z#(!X1Gl_+H-}yU207xI$;MFt4E^Ey0<bmML+@(8H_VTh@Gh
zv_RLn3^CvuP)rIpD}VZBcgl0GA^)&LOlpqnrf}SWPs*OYt<P8c3*VM?w!FDTr!v-~
zKK7`$_DP}pYOK+CmV|egiaUzkp!lK*TgtSEKEX*oBXKNE)T?~EqbI7urf&$ZRPg#}
zd9}fv!y2ZqKo3O9M%M>qLExHOqgdo-bp7};6|G!sDj#NQm+?eEZ$Se5VZy-ocicFp
zRy1e9>T7y-d6F|@zBBAr*++hJ9F4!O6Yz8W()Zn}mV-#lDMo&Kh7k~&tXKJ|9M9D&
z@R=%J>f}jKyumR&jo1O@msqmL)<oq$<JW7$HYtHq*$|ZS9IDk}$aM=7J*Tj3vii~+
zl54c|oK1OUAj=k#Zbq|v;bAKjvpdl<d8)~eR8Nj(6qKSMFLA9ELI4o^JlwCZ03D)W
zY)0NxMvX^LLLdb!B|&Ems);E`)L~_ALl+AP*j8W<Hc;=o4>N^V*=C-In=oIbh!o?0
zft4^xbgXduOnw6j`G%7d>y)nG^d!BE^uKStz@QZ7+<cqI^2bgg+rETEH9#z*F?sR_
zDN4bX>|2iHXJ)`zx<(;)@`v&H3;9z&ew|u4s=eYX7t<Lr9fMyo@l82+d)_@jnb?M%
zTY|odwNO0s2Yld5Xc5^PoP}o=dFVyhFK)qBi7a|j7BpdLk?^)={(Zl{x^T+MKi^mt
z9{zkfG~Pb%HGN=?U%;Bx>2B{^O|RZY;lsSCSljZA*oe`lWyK9BX3R$$5rz{VP67Kg
zyCCTDf@mRFY5$1c%L0vx1J>2etGtd&I181bp23)!T;C00V&rta%W~3E4-5h;C@wi+
zo%-rPpKXEYc;eGthgtJ-Y1;^<qN>=)xX^#u3j@c+4SrUU`3=Y=H-zh;ZfR&z*PD=7
z_SH&x4<}{^+cp6~5}1EO$25dgVZuci2`EnTMp95^(he`*HGC+po_bn;VJhCK?2uw#
z7d8v?c1bBYI@?&tkG@_9@t$?E9rXdS`A+*h$8FWtcl_2Z{Dz>R&Ts>ly7i0_*JS9+
z<&C(-+Jq~?rzfe1^#Ha&9+=Y6H=H~6$GR&oWP9SPtmXXrG9=&hna_kxG;)~2w2n|N
zvf5Z_hP0o1v4F+IRo|memymA!b>B3WW5C1~{!6z}S5n->qjUCm-CH%_GnRwZ`$Hpo
zO|AUk%sY41QteTO9KS(jCjq-eB{E(6@8yH642nOJ3vOiB@XPH~TWrHjlW7m#!=c;d
zUQvzxxaCBIw4;(|F!jCOKSP*vTQNd+uRl_zzQNT|Wb0CmXa!cyuPBP5)L2321w|2-
z+)z3|3XobjjA5p%G7;KZ;m5o_K3Sv2MBEgl)T325gRY9xdP&02_;^N@np`^)hK2nM
z4`HiAso$Vsq%@)U{^@s)n9J*c>Fi6)`Dw{;N{pr*#IxF%nq}m};D~=RUbg@ydRXj*
z8byb03AKmB$d-n(zh5FfUeC_INi{D)P}8X5@}5moHkO(ghcDi(?PWXHAsY;fOi*uV
zLU>90$jry9GcT6<44y?>ErYw~cvMd>I4CHHmI#$qy770u_0G5#F}&Nk8?8BJVgppA
zn|45hSh_w?pGn;&QP%|7N%|1rr4fv|nVkq~&J;j$M{4!rdRtH%4x{GmKpuGsTc+*c
zn<!N|J3H4Q{S7~p@YxE<TPa4hbSHP>gtqrrcOjG1Swypf;sYB5P;{ZIzb$E{RfcD1
zonB`OR-u$A@-;IbpWg;{M@5k7L>_hgSj_DxkV;WzHdoiynquV=$&`pkM&x(guHEs#
z^D#A4j3V|X6|$6+!8*f2Gq@dsF)u1Itux=AS1m)6$jiBKxb+%|Gb*KA4M_<>0q`(9
zl#j!_%#eZW!081i4Aal@%3UM~C(5^gR#)S^ATE%GFscqqLsJliO);M#({^p&He_<Z
zx>QIeWdHRr;?r#wxl!*?lduP93ZX<B6kj6kM)#&xsdcfFS_H~xiqw6M7CCs_8y_E+
zY#ujqtlXG=KPd4S`T|p<0f2yCL6WNi`k3h}-rOSSQ8iH?Tv)#xQ82H%y851mh;!$h
z6@rq$&YNmo&D^cd)_U)V-wp>m$ZXc4Y@%J8QA3bokDq5Jn_hIAfQ#BOYHiwB>L0}P
zGaoc`cYeiVJnl?tOb|N>kwTM}iru6Sh8I3Ta~g$6Byrt&mR?H*3Zms&@w{FTZHb2c
zN2S&FKVN9@%s`e{8#dKzHXLU(C9**1wYOrAClgVB8}!ItaM;ywZn3Jk->R}hA3bLK
zudX26%K;TPTBxs_hsAbWkFY%*6xfcnI%e%K!%@F;`2O*R$fKKKA!kZAaOg=YA5-o4
z-M0k2x=-LfEbpgoGH);Lk$C>cRuV6`BNm9%k7E51(nhsROib5}<<h^2Xuc2o0-c$N
zogE6k09zsa<Sz*$G3wqTC_3G7!pGr<QYw*0aW2?XYm`<i6@9#()z9}he^KJqgQ5+=
zH`d&&&AhS^KegNLFm@IO5$@cz;s5H~uIZ2OmfL*SDMS6|i3@^S%|0kNMVfZurF<Gs
z5f%cE#<b0`(n!0wAa^6n>8SgJbOBd2GMHcizBe<3;)1{Jz)d~K4Ego(m&AKV&I%WP
zU_JO8rw}?K`{lG@8yJ=$_R_B50?xh#_(1|<SH_*g$NRXb^X@w_=q>r=j{?dBvVRKe
zv0YIDJ~yiBSGEKw2Z-|uhH$0WkKghjdp;*WH0b0!q@6wcqd`Z|#Uh?0I{EY7YDYvX
zkKwAwQEl#d)Gt0Koc-Vvvm4OOlyeK*qQit0t{Faw!9<gnE1JzC8aiLeioF5-PPNKV
zpPjE9Y(}d^CIqHFK34=Z?Z^}cz-K%D;h`lxe2tVnj_Qzf>GmZo7VihsBYW>yWi?@&
zXtm2>!<1TRyaMPv4j-CeFbNF;&f0aE$@_{PA!<5j$I!NIxsU^9Lwy=e&_+qgQr9#L
zp9EyUi>H2NJMWbr4x(M60!<p7ycx3%mHMy9J}vF^F9;jK@;-41=^}Ou-6;I==_a9n
zDEApn85-Tc!l1g9mE(t{$r>6Bo0;VNJ7fr;YDok71aUjS1g2MwxOXgZGF;p&%)%$H
zg~x%iAttv8)xQ*L0!c(WQ-D6=bZ8ttgOq6vfW>Ee@1Khlfa!o%vyh~vMq~-^Yji#w
zkLO-XuLBmb-F>J{cwkG`PbZ)LQ6p3^Bv?QSpU9G~>o^{vx^-WVfLQNc-Rw$0@xk7C
z(tc^8@wp+&uC=d7lCq;VvBHfoc-sWyu14FQ?5g<116DvOGinhd0+f%h>PSgXcS*P(
zA^IWe;V61E_T{v=cnbyXeEYBXf{tNffbD2MJw`Z;nc_rUjnE9jV`{E9+9a*)5FraJ
z0@Z%6elaS=o7Ows?q6*M^O6$R#<3`wfv@r`6mlHnF`K#R!ht3orD4%>?Toayq|uO{
z_OyUaEaPOn(?MblCUm#(ry$GQ#zlyJFB<S5YJk1>PDNXtB;TOz@)<o+sdu5iq@J5l
z4rH!hi9+c8b}`1zOYp2v(*toO1Qfu_It({#0aItxw{^LZr?D8$j%fW9UqzExO)TdR
zQDbRzuK*Vnte)p$CU`bGfm2q+8kPYnC)_nUhTFK>bd_jd4Qn{d;n$X`7B)}6QX%}N
zuZYwuLFMQIZq5kJJKdqNOB<^6!?tkL6)uA5YYNr!z+-K`J*Wmf7#dsiWZ@^+D2*K#
zl6jB!gDGoB<|@|s4x$u>&q;W#C9Mehnkti-9Ckem<QVBO$~N&fR`~PoRQ1opC`@3k
z#FG6+5OInsJ})Sw2jaKUs>%jn_!6LwKk9nV0B<VL2qr{?zEjkLfQp=F<$|IHK*o}h
z{{o53og69=v@DRWLwcn*DS&&7m=Hns_YxDufg9+YYN3xpf~;n<GqMJ1(i#M7(>=^%
zk7--B0MmAaKLDp-x~E_Wp8p*{rWh!^q<}b57_0q;0=UNwL|_@vQbt-|+WWgcW<wm6
z^ld9*E5&ba#RUcE%%a-93;XzU&Uh$_^?yMrS$Nd)j3Ib&nSh+K)n1dwcsG$Ii>Ge)
z{#I>BOQ=iElX|VCs?J3bJA1mjheF44Hx9U$duHF2qg^P@Mn{8sPK-Kkx-(L&9N;~t
zh&BrZ#<<$y#oAJZcpqm#{iB$fA}EKLOb7J1rqmpuyCxkRGzaZe^?`V*=t%!9O)qe^
z+?Q_xbmxzuxDEtzl{<%lJ|RU@*>*)?yt5INJJmJ8Wnrj`h<L%xJcKG7khjPHW}MY)
zF(yQ07*!_?)#7ok2#c<+UmD!axSXQP4WL-Qdp-$)7)$_#zE*7Rzk!3pSya_AO+ZOl
ze%qgZ0AvoK#NFMxyQqkx{~$6zb5wH2jyB4yOTg2{Lj{Xbjl<~(p(H0N0R&uiG8jQz
zgL+GAU`^DKOQ8l0L9y>XRJ2iV7pL5J+EGom>z*}MKmBzxP->1`3s;SBGlWHV!A4}w
zwK%}Qi`W?gE#UcBiYb8XVj1|Xnd4i{GH8XX?DRv!a2rUW-)^jzBV}VXK`7uprL-Mo
z0h}?wL8UR#p+S`8fY_of@*X06qMuDhyX^6-0^o2mgdzdqrrNC$H%?H<Os``ELPup>
zGQ9eNQSBE`qZNcqcr7Y2BiO)2@;^WFr&R=4fl%x)3v+YxG49NW(TwI_mbk}!vMoqm
z+py?)JTIUn@W{Z6fO*^^CuyArzw99jyW5eCJRw{}sPY#6hTKI@f@-xPQ{w^TUnZk_
zOeUle$eQiYI9{|J?>Gs<{3XQ6Bq|OVTYl-87q&xLWT0VvTgczQqIkjN5S=Jn2e`gu
z0&B5s=ex`LPOtW>mbqj6YN+~{dOIFv2{G`1ZFbbPWG?9kj~)j4XGfj%sp1=iDB=;E
zP=jYUKPLj}M@DVh(CSoEoBH(8JlQ_hg)5d8A%4T204)=kU#hWWC8?SRG!lpb=|l%x
z&&0EZ)3&In>9MPp^&Awc3w4)A1>{&FmeEF;;Z&f(O#nfWMX#5jGzN$=^Wq-EUrQ%F
zJ&%1#g=!=$li^7~OIBkfO52@A+@QZOMez&uIBZFu$s+@U%NkHjF@;G4u{k;<xlm47
zv^}F>%7&T|D$v}=yP{L9>yi=0wtzGyVwNsT^JW~xsy^2UnMGHU)t}vS)W|3dI1*uu
zFy-PFO6NaR-tGPW2q3AkWca_2M(#6{f#B8Q@s}IhP{nu>t%1@B@&GA@19YdRsJhXr
zhW3QNLI4Q?548b{f`QEn1qVprmj3aE?MVn4s9p7HL1M%IYb{l?0~4>V`rN<_bo=SK
zOJ1N+WuGhv$d&>%=JRImniostdMjGl$b?~}rznyk6x&^K_O}0xZuTFRBaE8aw$%6I
zzTmP|P(yIlUWg`U{zyQ}hG1*vcWOJ=0;D&FDSG>oni^)|Ri+W}J#<+%1TYg^*~69~
z7C}6B@bMz31-!h_^>A&=&|9_LkGN8dT|($5UTh7`F^{yBm>H62cWm|=E+se;Ae<Yq
zsDv9(a)%xRhfIYK<w=|$X*1x1{jg3J9x*ITvC&vRdO8X-D4a9(@pTwXgm9YxM}a5L
zt|;6{l=qg7{&u)RM*%f{JJA-z4MCO(K!*eJvWx&tVV(9R1Ecy2L{}mHqS6aH72zjk
zIqbn6sJ-q^3QH0KY+xvizq;CiE}usRq$PQa!H=Yj5Uu)g5oa3>uRqV%CVsOO5*OMP
zS!k3~P?~elWnsWNl>0efd|sh^YKP?0*itB&tVc&si#6Yyoq~8&e*#S;gGoZ1`z#=4
zd5n6Y!3{-N@=QEtNu2cZ5-~6kUXJUj)5Tt5<~p8jqHRshzF@5VDS~*JyTJU`VLgGI
zBuGWL=()n^LociWbImn$C@WSRugOLGT`Dw;)^i)stdlq{fu8P(fPcnQC4Amr`WsR`
zIU#r?CSP>xQdT>%E518bDzp{|C)kjb!I3>)hh5pXWI>LftZWkC0axH)sRX?cS61RW
z4Ewbv%tQ=YP9#>Ry|P)V)9}mL#o+xiOu(UK;RGSa&N~;Ra(>u~Rr#pCU)^{77&FzP
zsj*{+Pj{Bsng*jf!upA-Cb9r<^AKXnPQVtb%w2HnC1_#j(cGm@_X`93)+j1~ai(~|
zoV#S_56EIxu&0`qfWEr!9qd%!3lGdQO{j6({#H8wz2AnVa(djjt>&0yjSjYnh*b9*
zqh)MY;LU6UVr1p4YV+yZPqn?cUYg(RHZ}^YlHlCt(4JpTQLT_uW(5pwWP#$xJ<~`N
zk{OK-DnL8aLhzOZ2(e$gMke~%E_CyEt&RN%>NP|-3yn{4#t!0@HT==yY+ztuyy1?E
zK)Em4M%n5zY?TzCgB(QjQaj&?22!AaOg3V+p-;2hdfFxq3rqrY>L;(rcHbvIOPR^R
zCj@y=CL}HatWHDQN5Dtx8U72*5*WshG`H}VES>Ne(|$KMUDu<cb{4qxduw8rpr<$c
zcq7x()9cuJ*(K@n?7xbUiR+yW0hT2Z6$;6j#`G{rw3x~UtC|F`KDG30G@GZn<-X0{
z=Wr=}smD<2`wOGu8aOQH$4L_ZXmORq473TDTaG^nFI{><?|%l@{iqx9UcBPp<GSZe
z{N}X$Gp_rj{w{r6k+fa=j^1`>f!idQZ7A&OXWDO5|JDK^HET_M8+nv{AY(zUXi^UD
z*qDp)divIVRhd^hk#%v~fpGYzPQL29a-92{rl_xpPx}LF2*^r9F_-^RG94jlrWedU
z39W_*kj8n+s;Fd1f<3g=uV)4l4pu^<C#V8tCXFa2C~%r!L{(&JWCSM6UFI^$z?%A8
z_qPPxx{$UXiL|Wpl{;_l;LUFX@%}U5Z#dWXEhz$YF+}1zKaLh;W}~Mw-mhE4i2sX*
zF*G7ke9ALqy*O?E0E^{xD@8)xTK^n!c^7oINN&ym#{K<w6cm~x04IVgmKw_dK`G1t
z7pLjd7qMo_oM{575mYH=x;14;En&ixBZnR@lt|6GA}oUzIyf(42*Pz`Ch@Z*jqvw|
z|BPzo`^Z)GRXhaM>Jow`D{|eiu1J6b@`x#HPYI7r2AAJ)ewFPD%vMoayUFgbWzY^-
z>ZWbzI<{}k_+e=+D)}oEh}dM-Hg#po`f&gxvjF|^MpfY8LI8-roIQ%(5rCIjxt|X}
zWO*IsZFCX$KcdZ&jDJI$<^M0y=3(Z4fi|ycT7%Rlp84jC@Id2cZ$zjaIuULlSLj|j
z{55}t2_kiAEtj7D_Au_DtR_?x?WFMMkx*G^3bbXS--~x63a9T=u}h5}=x}lp;%OlF
zJ<0a)Yj(9NcYd&eZRlx({pEljDmCloXju;xL{#Wfi$4slr6eO6I8Pg)L`>Jm@7Al_
zOnvR@ndsM1sPq1G#_=tQ@2#*h=`Qh5W5~+d9Sa7}NodSU9Mr}D4((((Ksn-cf3`Cs
zIaPXBE+qYOC?@TRMzDVH*XJa@L!!lR!Z;xz;sEh|K|BVg0!#xiD*OH^#BD#~Y%~*?
z%myHP0yCwg@&dbX2vWkVI-fA8*}RZCOK~T3`vD<HfekE1oKB|R_wx2KbB8igG0;`I
zqdOyQDMcz|Z(t1SU<SuXlGGvMD(OS521I<JG{i^7f}!XU0-sBx52uQcVw->DMlBaY
z*GaqbDWk(vsGWdXW*apdB&O0V<6Z?GE?h)9B-H;f$l6H7AmS=}^1Rc7ftyYyLMWiQ
zcTQNH#Zr#rX)@oZnHd_P3GnK73{yck8wPH_O;U2^vJUVf4ypET%emUh#M`L-<6-#D
z%!K91E%lE}9DX*F=p-w8aPOlpzn#zc`rsE|f3e5U|Mc&wcI)<-G`<n}Y~AG^!Eg3z
z+)~=`(~FLPx9=Z=N}2cK>$xj_Zc9G!g~^`ugQqtwI=khB%4a%qO(PR$jp~1Qa+PhW
zX_Bc)Vh?{9*60oymJd_cbngC;qvow%01P;F98+V#R5Zx6-#8W&dY>L4NeX6>CzdT*
z%rfv9w=ov4NTe3~dj#lmggR>At#=sN>JisRZJiI)YaIl(i|6z*@sR0`QTYd<Rt+?J
z@^s}XM%;eHL7e2+MK^={Xo12t5DW-d8wlu!Nofd4A?v1KIvGHtWwb(MNFHzubC&Qz
zJcT*5td1b7WJrQk#lM9+UXs3ZjVAGL`IGGuBw&N(Nzno*D5#R66+YczH*FeWgJ&h4
z-FcT1J~1)dKyKsuO;ER%!Y+U&w@jhw;R0Q3wr0Wc+<`(OK_cZi2&WPv#MRg(JoKKv
zPO0s7$OlP}Of|}OFz!*m*r-H*Z&RMHS9$&Ui76ej_X0Fl%HNl<J(c5Jp_Bf>9yDBw
zdC;FDEBqQNki~;>3pNt9h3@ZFuaD?R{iXoXUGHrmKu+JZc>oue53Ixt@eCA4g2`r0
zD;$a*?0wmj<#T9-W|gx*lKJ$piqL8+ZJ0OzTIKtpXXDtw68SFoK1QC)IaHepaF<ff
z((fOfFD`rX4b#rGAz!ZK;#hiLoH6SJdRAywEUc|xqAY%8<9sM|s=>Cw)r>pRB2?Eq
z<sl}OHyUp!*D#(k!*mtTPbD}lW?k3qaw~x4sIhazW23fFsj^paVS-{|9eOR<UFeQd
zwgF?*N92`3Q<vd3VKmjP&23##a0@?@&~;bU=_N#Nq0ktZAuyhsFo?xQk2*y`8*mhV
z*jlwyn&ZxOIiE?|w4v3m@7*N?r8bSk(&QdZE}bDoUicsaU^0MQ6Z|CirI`Y#x9Sd?
zl6HqOsDyga#Hz0FKBF8Y<7n!F`Vo%i`8g_g9;2&C3=1Q~ID8{56tu@d)SLK%rb`A0
zo8jI-APbihR*pp}6;$$_$O+i4P+tztrp5Zzf~CvO&QPJrC(`B^HlYs|D8p;2$H;D2
zs@*{bxmy>&MA!JwP8WgbMd^!k(7SZ^AMKpCT=6V5tKreH_}H-N57A=vQ5qucXyN<s
zT|8~M@eP4<;Ue<jB_Ry8zFV23a%=~E4&O43XXS6?-~RqxGVFzvmy-5H+iO;G`D2{h
zBTv9%(HLf{eDc*S>f9U&L>SE4F*L0>cgO@PD%CMI*e^Ip;)&B#G7gJtNbN<xcMuu^
z5rePjV|HuOn?Y&?UQbe{<CZZ~7;>Hxg0G`~m1Y#TJ5Vta;!wQV{mpL+h~;lY4Ze4F
zCQA(3wG0riL=$fFot|9%>!UfO6t2n{E6l-%l13GeJ)fFRg3jSx5(wbr9!kc_^*j_)
zBhoNHXb`ZCbx_dTmg&A0sV`lnHL&aB60=1ZL#S<n&3Gedzbw?4T+~dfN;N~tPav&s
zWDHR!s)3RsQ7_7Y^mv`_EFZ{A%~9d&+$!Sp`2cFHDyN*Z6jyd!OpHagHww5}IUi)J
z&|SJdk#PdPn_o8qxT-Iz0VQk9C+r;ui5=Uq+N1Lh$$pr^q)tJ9-Pk>j-nrgf4<8l&
z+5;QXHL$|ePy1#4q%xo0!;4{md4j2vY?>xjv?4|a#1cuX+IG)h&2oH9aq@}cY^weA
z@o>nbIN~o144-KDVU)!RG*buK>qc~_hX|>q>EcKN19=I9i2>GaM0ozK9<7P}xNI?y
zS^9o$MgCZ;lN%V@a2<dF^Pbs|oD1TQNDUE9UW(^^?sha4;x}CgcfqqMqp^@y4`Oo#
zT^p}cyF}UD>;_d^e23HMHhSO_2xh^_hkwyftvil=AyKw4qj{ovtiyZ2(<mLv%CdTS
z`kCD|J*lT%_fc051>@3oD&Dkvd1}{irOq3hc?1Sur<Z}l=L7J@9s<f8nV+vPJ30Bx
zmbWZob))C!VrNmqNR=`~b-((b>fy0Kyb4wZ$5~hAxCWY((MQ4zdb340j&*Y+RUuA8
z3<d*S9Ka3q<pv5D0o(|LI@{&IELEi$eXQdR2uH<+W&J_`XtRK+RmEu!=g49LC)k#x
zx2z+x;#|#GaBMh&Y*q>ZG^}lAs$eD%|8fsQ7<D6xxIW1y{#kd2R0Gg8aY^~vU8CT|
zD_d{p^w>bLrJ50jlrl^ERGbo6wJHTm_7om{@D@4r1e9a^e9+3_D>q4!Vv^>nZJnC#
zx{3yt$aIFE-G)4|LrN+=F8@kxh31u=t$!fLMt5ba3T}m{k@79<pOxo}#DyfJiC>IH
z6u>*xIC)~<t7d?hOzE+6JCX)TTv}y2-7BZtB(}pL=*0+w*+t1X5E1v^Cn5}|y>t-Z
z?f_|#DKf4Mph>E4dSB!D(>w$`TCO(&cj=2wijZ^)7)jH7JRX`MrDtWZOdZ#bS~0Q9
z1$wm%^9Dp94IWi>I`I_%Zt6#QL=>)tAKGx=nXwxs?f8ugGX?#_n@$ksHVyb>yvm4Z
z=_6td=Jz*=GVfEklo27xrxyqs6v(z^cHY})c`?<R2RQ2>s)5>pS;DLH+X8N>T{6Vl
zn4nL1X18tQDl=hxbAql2UG{?RQk|&}%0tXnPb7dOAVmDG^1kGu;(YIXBH1`b>Wh<T
ztGiY<{0oA@FqUJH-E?#F3aMT1h{gIOhg5;ZmLy~7Pj#=m1`WQLw^e3tG0e@ZNpWN+
z(S%8o!9>i@i%COG$l~A7)F@_4CaV;uj*&*wlx+uKm(Up6(}W$6j+{yZ+JeJv{_3{L
zK30x{>CW+J$Rhy=Q6<GB%|^3%|66eRX%<Zo0FcInI{{sySLePp#ekMmi1J0iDF!@V
z-))?Z)kZ%}vPwgMD^OLoyr5c*M(3$*q(|dd+rrEa7xS?52@rEc^VY#Gn098i0Y7;O
zWdE5Kdq#1FGqa;<8oNfX%OObe6Gfho=t%9&RryhK_D9&bmy+lerwewFY!@<BUdfK7
zj+|@3zJ;~-sI~`-t{CWCD1>P{LF#o@d*^ZF2Z~UQ4Fl9~#~q>4OV-(Ul}KO-x6Cy^
zg^@^gAL(j|rh}<>7<RLi7)|BCVwhCbcB!TGmLLdrDFSjE0(f8s=)+F8wsLi@o^0)T
zW2h{wKiCW}me?>x1wI!zR`RP`Br)(ovB#l#8d6LW7gDOQp@&nM2O*gwz;WButI+jq
zr11$1*xv-7WL<5f*L^1a{>PQMIN??esgx<)^081l8JM!GOZCfTiwQQX^8JFqK1lF%
zal<sl1A~3sb;m6RB&skXH54h~y&0LL7>2zel&q-@hRD>RaHZKTv+r@EVk-!|6NYBr
zYV3RICZ}V}Hr!WR``D5Z2&ynAhfcqxe=05dJW2n7U8vAt3vT~L9D=26Y-|iR_@-!5
z?lzntP>Q=W3@BIH#u2Su8;u5GbYK{w*RA+lRlHa$$&f{rIQO~l<a4d))Yla^FFPaM
zqmq$6S4xmkRqhjn<C%B9lUJkcd9Dt)$K~|(u?C|;NwA9rWePJVg4gvD`b3Sm%0&6l
z!I@NTQlux@2g3nPB-Ma=`YTfRqr|}xOTUmZXSL(Yt*T~8hM*-@d2f3UWX=QwN?D*0
zgZuH<J?S6sS_?WOy|$yaihb=yx^Y4w3kwTZBTrJ<Sp}GwK96LW1pKOqXPHE(TjO?~
zZuJCM$kUtFx7xrSnK!2gJB<^})XmZ`2SpEvDShi8tvl=Lk3PY6WYNA@2_zS1cF-KN
zw3uhPXR|*^RgtY_`ckVXOm+ZlQynV1F^@5XMsf$rV(ls^LJ~>3cu73_eyMZ1S^UTz
zXRAA0LOWIadGDg`-F=J{4miN`cS4{OYR9w6y~PQzt}2)q=?TO{)mAD52#6t?kK49U
zF>Al`9zCaG(avLG;~G`f3qj23(j0@$vOyRg`$YUYMrItSESpzGrp8tl?1s|I5uDKw
z4FSkNTvi1Y0p!kqOXFx53Z2&$o#cCWC#Vd~W|MICp3RDDBt;AE&`UtgN;+te5BDel
z0VWIH#w7F@_783i5%Q9RKG4&}0RmY^GIQ<#gsp;1{9vt~@10a=jpGf{Z0|FT(`uml
zVVd}7IkoiB;KAthVvgkbvI>he-|!Uia3axTouz`Vlxk$^;aHjn0yhVikm8HkhBC0l
zgbT7XsEi*DS1y6Iq{DSe99hw7+2fVj!@1QoxYv|$=_@E<rECg#B6UhvV)3a{lJSb3
zqwy(a^__QrSv4KM^cNIRic&VN2y{_E9gHUG*$z*?QEF418CD!_!R$kV;nRfm0jqnO
zrt!iK!Ul-ECHb|g{NmL+e=3rza!BWuJt<>LA)PIfGh^&C>#TNsMDsNqOePUu<PUC9
zssOE=Bg#(VX>_t2PTA(ZFQk0H(%BnGqW-GMY>2h2Tv#+OQDjt*PUVlr1vG&hNsuO;
z(@&Cf>C%jBbOFdQJBgeum})3!$3&`CmGf`4100OlSZe3>#k|EKL~2ZARH<`M2Bvj<
zA3j(tA&^dt0$`g`DAlh~YT<QFOM%6xIyTrANwY*{YR^x1g-vfJ^aWEiAtVu}o<A+R
z1FiB7PhUOGiu0qfA6Ms+N`^v(Lskg1RT~&oyYvDQChe(cUJg!f9Bpt!XVea@Apt$x
zAo4pDv|}>?r~ETRd6+q6G;`-AL>hIDgq@s1wQK7fkNBar?+qYdX>X5ergMV0SQHG|
z@UiOMX`<F2>j%nYJt~Lm{XP1=(krpb6tG!*)htG%UZnf}2Klb{)mF}xR-A^xyqcLO
z0U24(s~SU|y}~+&S-3~UW)Z#Qx5IE(@h3!?I6#7{f^D-m11IV+l}JtVQKaf1r`tw-
zaZmZrc_i_N9@O;GXa=OY9Fq0ZNDR^QG#g3-2c#qB<rwI@Gq}Q!l&=H`6NN}sDv?<<
zyi58<U<0#aJJeRqGv3xUG<1e8Pc=-}rcWwlf@aEeaPUJvxKAqYp}Q>OpBfT3%HLIF
zw6C!vJ}k@>M^q4%y!8u142<TFB-o|xETiuLjy6@dd{wY$=4nE~X75#;vzg#05Fipw
zae{Mvce=c*P`Q<JH{)uwVVu8VjuwCkts_d3b&c<`Yz-tsmmFr%!Y1z=ceIl<P-Xjm
z6rR83s48i!k)vWXew4_&5B*IJ*8esi=)%s8l%!^l!6Y{0(%j||SkeUI@k<cnKt&IY
zD00Wy1);c+5-2{MGcG$rP$0|@&-+km8Xb)i$Z|5y&+;hRFmRJrL1w7E>;j_iEdBnc
zA81Q%bLoA(NMOe%RB$*E;UyI14w#nUz^k_$tF$pHPAf5E+CK&Tc6w&n`QL1j#xBi~
zNNw|br&vp~|D)2A=>5YNCXB!PvdLiYvynC2l78_!4eo|>_?TBg#JeLS3>kr?=P7U{
z3xEIeyJYl%?bnrris<?ue<`-c3{TaOF)b@-5}B&?WPD+g0Y;&?L=$60xw`pYSnY6-
z=~{NTFa9BvR5yg(arwj2zQfYN)?>)H8!g;D{NT2m`6Nx!%n{)m<ACz$v4eO%K4yR>
z<35PXT7%CDZj{F92qa}(x%DKkYNT9Ff@XN*)JJV)``%6S5o+$R-l?r(Wt2hysKN}l
zJR{Q!Z&DJgS>tAVyvO5cxkrN9zx&-M&{N>|8o#<*c3Eeo?@CHwJu9j;h7yXwq;iQ=
zm^p=;Ah>J2C^`g6cn6g-Ll_oLe^x=p$m&PAr!&$7=;~<co3Wg=XwA^L5lx*0Mg4g^
z&%-$sg)22lw|_dZmq^bIjA;zR%aC9QgYdU7Mi+&nPbpK-kQa%scvb;vG5MpHNp9Ef
zS}<V%_Ry$}2G2KK?HcFw0|9<P(-Bk?1JNqR2u0|csZJ#>zGZcpQi^kM+`{_U)`4$%
zqH|R9!bOOL^_GXTByr^TaZ+lK!VrPiW{DenL0!+r)plj$(*3ab%<vfPM>ZKMm}bvd
z+H)cPPR3lNR!>Z!PLOW5PSUd|c4&=NPs)WA{)V1fNi_$PY8Xj1F*zO(ovoUmE8n$s
z;HzNIU2qQw7V=HEMNB@(gq62w3+mN+`7$NAUD58~;YT2XJE;G6$-`U#DB;nE#lBT3
zX#)8ZO#>YyN1>QoFp*PWWq%ojbVS!EZchNF>^#L1<)GJ-QZ(&b+2cE1K64xQ?k7SO
zcyncd$GVUve1Jl6{KF<qXsCobRKyc27v-1aA5lt(LuyRE*u$*s@C1=iQOZ2Ctrp?F
zG$W$zv4UkgrM)QK<VpYzVw;_b;!leGNbP+Qzv>6l2w6`NHr@;DlE59g$vrP<UPzc*
zn~@)YrqMRp3oRUhu}gNoIqgz9)^J_me*$++Eu|QXII^BbN{6wzej?}(Dk6d}$Yuzp
zQjNXO+T^y?Qr(Kmu#I?l@SW&RuOWY+kBkZ>G16pD!L$_-4v7Az2*V~37U{@&9nuwr
zkkWd6g9<Hxa%0qnV5i@v!9r>)s3ju)oBJVJk;Q_rAjL?`qVLC32JJ|mszdCo!}TXY
zKG4Pkj22AN8d8-?>zDyQW-uXvcoMLA8E)2V%MdW!w{fBcP1cdkQSg=mVAl?%l5A_2
z5E!<n2@FA*KjG^yOwAlW`*@51ftA#mlGI#UP76K75IPixOJ&M=Xx2rfw1&!bil_~^
zR06Y57L5mDuFfN+M9*^z+;MMfjmcwVia(<Tm&XsWsw>8EqZVFCVAmRbyf<QWD{&A)
z2I=_^^|o#Hh|wS+6BQz`pD^~{%#ZW`0?<!NGB{q|g!*(Ek}4m-E=_XqX0~oDEDtPc
zC+xlsY-O;x$LAlcYj%zp$5EF7$U_#9z45?`1r!Qs9)j9-1{0l{B8_@V)lH+oF9tq^
z0RxV8axV>8K%Ro1@yGp?_CEI(_%0nOl7b=!$5*G94v*6pd3FS;oAfv|$hT&C@Pb<1
zb9jOb(`*w2aaJRr_3th}@_xG;-0r89m+Blri$Yx$&}dzKGkm9)(EkM))V_YrbF{)S
zl>Vnkv4XxfmpXXL)rj=cy7@lY1crvodV|3!A?}A}$Cj5L1pG;=1e)6;W3~OyKfg%A
zTNDGwyVi<YVnJk~-yi}5W_8lwORcFsLbx+U$x}w7h_F>ipRig6k#89hS(qv%)m9O}
z_v@I7Cu+W&CgAbwD_24=t;JB64?$xRbF5H2K-nFDmfR`p@Jy5s{C`34OHy8to9>q~
zNZ^SbQr#^JEjLg1LGr1*^*rFtTI<4R&FILa3=Z2&d8MqX|5e~D0-wpvMfeQ93JcjA
z*|1TOo}XJvN|>r4hpfiy20Z9(P+UCY3wqJv!=IeD&U~^jU3TO2tQ%PPtU8fzeoYQe
zG{LV}m3PZivUBF07J9OhH>bomU-RL1H?n?-)=u&pZlxkGP?{!S5;X{g+r5HBTD+f#
zJUM3o+OOI>?LA-4BH8tY%Vt&P^K0&Gcf%2GHOe)-ul9()`|zV2AdQ-_JbaEX{n{rq
z@hjy=yf5-U@SQ)|h;7>R$$@Flr%(OYw=jL@Q2VzQ;6J?u|Ko4>f5TgJH#JeCR^j7s
zHj|#CazE2(rtdm<1HNVFx6j_Yf4z$Tg^$h4$>B@aN6~1?w&*+~o2?59C}pfC?e&NH
z6pw@h|E|!Oic6oITClr(!yB=1w99s*s?&brv5}&f&T=VjPGX;|vMTR(t})8aye1w_
z*;TTVN_YBQub%|p`5rhCZW5VeHzz?qYun<*HEv}_Jy2@bLlC|)Pite|2Jf8AD~bR5
zqsSF?pks3G3-Db$_(*TJpPygw)UphLjoP*`W_kR<yafLJyzVr!5naCzx~$pXQfw>O
z%+g23TE1LPYnabt-ha6w{ul1$f4P_bUtLK5$6pAR=IQ<iG|>kFh3YI~sd6C-BY8Ek
zysP1ZL$bDd$$p&b;#jmkFg3f3jI}V(7-8^^XU}1M|M~n}BDQgMYi7|Dxi|u7N+cD|
zO6W-mT|!(I6UfmQkG^x~L`n~DooMX6y3h@{Iu%^|mdFUh@U0xpZY4<Rc!L}%65-P*
z0TGHch`J(w@SIEEJ4|nt98nU~d|S*5Wm2{8I^H0X+3+Z6l;Z?7&3Yq|*b&5~0*2Sd
zn8F~6pO}^x+Awg?O8*Jk{e0QK2@e^dyC@JII21XEaVGfH@xGbZcZ2yV=!2^I&=q}B
zD)A-s*gFo|mH~BS3e#K|^a%};g*1AI)>n*t5~#K(Jv1PHo}sKrfQXJby<x@KFWsaG
zW+$yMgGm$leL`mMTc4?B(EJYcB|HWR!XZlR1Q8|J7C5vQ1Cdm;u;&u*f&wRNWNUhS
z+uqj@eFo(jeX6b987o#)&i@Z_L_S8FPxiv;u^BcpHc3+fBa_a26hZ<qA!8^IxisHM
zPCKLn0jUW@g^u(C4!MhFF=?2Kro3zHsYZi_G&K_vx25HnYta*o5SSYQ#fPM4f#D_9
zB>6bDrh&eab_K#xnm|{Y?CfuwL^k*k=OgR+mIa)a9?(DBN8SS@j3$W$q4I2!5H_{t
z6MhazmfRQ|0C}PWkfCw@?QjIWa+8<8KKv@L@J1}KgwWX`^fd}ZtgbW*d9``C4bE6w
zh;xIaA_Z2pT5~B`cU%-rG`HuHjs^-L&D_ES19fyG=sg%c-&HUB=UxQuh=pTbZFTjn
z!a1k4MmkTsdS?>MLKZ}1w27nTYAhyJK^#l?C1|vWnh~<q#hgBWpace6Kqn?#dF5{4
z6n4|2*6LvwFFfuE|EG3`I-ZRqU+w1HFVw%IDz(V<IczZCO$z2K>ev_;INk;ZjSt1(
z@M-+=I4|V>xm}ii8TB*J%ShlKJ3xhNoIEloP<`_1<jTZ_nY>r$>RpvP>-ywZS2#1t
zpvTVG)JT;si3n&6byZUV&~$1tX%0Ub@N(3IHYzK4b?|J6m?l8|%dtW7i+PGjz7OW|
z{fnAm2_%sPHyEZcaWACr<!AFsZ4Th_nFC{zQc_W%tC4LQI;VkR(O^7~I<YGm;I0U~
z7Q9d6eSB|{JoF#s+`%j0H=`ApQG@->o7uoaCK`s?T_8-dP`8zmVHkx0nr4sDi2Y<G
zB`@(t5KSn}^SIb4qBaK(7E^bMegJha6>%<Cd9K|RTJp9_X4!PhwLCQd-~T{q@_&5%
z{{KumoUM_d-NDp!L$_>P&i$K-JD@u=9bo^xtNu}R(7%jp1D(4hGAI20yxjR3P+{Rz
zLq%@(e~e9%|NYp6<w!23cfTFvufW-I7*J@z!FxN$AfU*7^$+#V0vCHhgFO`DEbp<q
zXMd+=83L|ILCSGHBFVSxh-3#({=p}I=06to|9>G8{*MOdf75$0(%=7Cb`|V&sF&OY
zYaVw@MQ;aKL{0#tc0?tSo^3x{X;d4}Z6*X83<wc}Ro>(vj}b%O<PBPBbz<y;W?$T#
zprrPSN(#*UY$mvlayU(B6ZVh@ZjIo6@&F~dbb_fq1<TVNA6}dugRGK_!>32>W~7LS
zAv_p9J7|SF&MXu^2u&OT11K4+m%rn6s5_!je9boO%FJ1qWDN~vK$|XHgpw`Cd-mOL
ze#kuS3ce7rT9s)6rC({HW5u2Y>ls2&=h&{9?MT<li@)&(gzYaZ*ZW`}u0F_sczk8%
zBuIw-(b<dBtwg)ZGhbC7Vn0Ze0nn|^q>>GUp4jK<EXjUYkdYI?!)Nvc5zvuSH6Ax@
z65o6#03q5xIv)eQbH}dr{7XN9d%HsOv)xBVGEZlao`vc-es<kYZZ_7|*2f;{71^&^
zwQ3kGaQaR)y?q?ZDt{#2Pc6_tdZEf5qkxij-X8C^pJO-BJdm2}laHAO{A%j2iL*&L
zuHn%LouyP?#`=bq!HFd{M#8g=*p$+Aa*j&pHvCu0#*E3u!IlxTn_ru+vySo^_WpVX
zq51r59h`jyS#v|WkAr@Z{Q?w#$gHWEUQim69c8_Swi%AGIehE^9rCF?vaw{<Xtw08
zQJ$~6YLa@M*42O08QKPI&(%fsw0Pw<l3Sqva>qbIHVoV&+1;tc#JO|k&VY<e36AU1
z_efB{i)i|s8LyWi4$(kg$3@xl+SjBntZE6yxdB41*V9JqYEh1NA26%7?WEaHCP9}?
z3cRuuZnyf6<e)?&?F&OUgo6`&KS6N+{0eA1Xz)}Rv}NIj679c1E;0&=Z+OnDq|F&9
z<^9|KJsxES+J(`4`>Z1*hB_IxypcaFW836D=xPEmLPS(eOlKyoeLlU|=LRNX?wm8`
z!14^q7X*lw8HnpuVl6;5hDgy;XM+b%P$eo$O{dvaEmpH`ZnH2MwUGY64142m^Dfd&
z!^F;QiTsBOFi6&#v)cs)(_%OgLcU4RE#!F7sqVozCWxx`<+Hos7nPA0GP5v1>~H^j
zNA>&7IQXUS7+k$`eg|5&0oADLOtXMx_ZImWHc)M#Mg=~j$4c`ObyA{Y3M~7#yK~(8
zEb?xaoXXR>j#JvY(!NSS0mkt7If(=hN2l;t?NNrf(IdM1KYYcfmvl5sU)<I8n`Kqh
zfRXJ7g9(E_@+@uAKf>WzR6&ipkwIYz`^h%({xInXsNRI@KBn-y&(_m89d$^krUTH=
zxwy)(Vlllek@w*!9<l!dA-8Zk{)IejXnJoEs4FNbjp#@N4A4E`^`q3|;eWc~XS*?g
z<VLQ9G{K7Mw6otqwuWdY^i1S1Op*ZFHzUNukZ^;{!=Tf1!p~ZwN2Gv^0f}w!BIy^8
zt^({$-XaVpDaXisxUO`$b10xTnlfcD$<Tw`;b55SPN?Kn?_OgDll*)j%}&bBx?4t~
z5Dt16H70pTLg*uro$&*o`P{!1sQ-fZ{io?LRzp0nq{(33vk#!^%^<yFFrg**DbGW5
zfHNQRpyPxw-!mJLMgIW_0^Mxh;48)SHXPv5-GKvt2;rr0jQD^~sfTV3HD0)&58XAQ
zxY@tZjm2T%w2{b>9Su|;y$VefZ7<-P9Ee+acNKJDC7MV)STcKUGEj>SLfsBHm!2B}
z0`Hc3Cy`VG(oc3%yHgOVeLQ1~Rl&Iw^?h`(X!kAzGZK?CLu-1#7ZI*NU!4LM@92<4
z5Cfp=a+TJe#}_=S<>lLSzgvFstX*sIe+n%UkoUkD`Uak{9{_w9CP|QNNDI;we7!cB
z47=0-S9-E%v#V0Lg6Bpa9|0m+P4x{u=NroL10)Y6Q9aIT_>l-Bo*^zxyOUpxc;{Uk
zk(wTovC!!<xDsmahoM>f5hfVG-@DI<*PGIm3JzOLfyj{t0jZ9l8u1jhWvzd#ONM$Z
z&Ehg*QwQWRFEMU^n^0@aGnGJ81a*GZe1lX~C6B&NtGjW|_t?t08?uiob<@nUzq-P0
z!ay(zR00b07M_mT$HBM65E3N_j^F8obV6L4<1k8x4w4Ju>0q=!coN6e^gb=l(D(FY
zLF+-4hSu|Z8WI8o?{uK9@93W#XRzlG`}yDtcaXJN9qN?rP(xDvLoRU*^tC)a#9Fr=
zQwpp4QE<CvEc?gR+cjI|BQ`xhB-oW^{`?T%heMGLkjeM|_}YVRKQU|4S-#X*g6V2#
zbwN-fHmE7eo5+F%f=-7$RpM`rjt1pc4M6a68{s1(vy28Eyjs5t>omW;{MB_sHh>3u
z7GzbOQlf}sZ`v^bcqqxRY3AH%_!Dr*tTwI(=D^}O#o6Nh)hqv2o4wmmLe`mNCXI!e
za4=8s)hk1}p`cDneB*DE{!FZX<$>5!Nvaz$TR9NMlQg&>@>#O7f;37gGVUS+U||}t
zC3%uoeC?I)gGm&VXbM6rBM<yi3VL0a%rfQ9&S72gKr%T5I|Vzc2|Q)6wEc=ock=T1
z`)<WVlv`ZCa$C0E3V$-M6qqe*zqPsRc+!4Z)6c>=-O~}q=^{kb!AD|=Y-c!W^tr=f
zGa>l;z9|hgwY4;AV{6X^PUL_&|7Z=7Qy`oIw$K8$Vv<h*(K^^xJVhAPOv(ezc`2Y0
zNnoA@t;jG*sxX1#OB8q!%i2)MV+yPo6F2M>glsQC%&{o+AzqLNgaf8~V1gh+uJdV{
zAtyAT@uVdIi7yQa6Us%jAhJ7Am2^LdvmxU%BnS!X<SB4q0nqo|d-!lG-d>+hDItq0
zQrD97509HeCbr;|vhU>hIj6F(V72Sg;(u3NBda@rh<=M)0?BQGv=k&RLTTxa@5PfY
zU_gmC&_r%WO+NfO>%)SPk<6g|Bi%hkb@A@-K9N^Fm`uG%hXPy@ZE+CnbB4VXZz>5Q
z47rx^+6)I$=HcKbDo#k!N%yO%LUS0m^%b7icnE(N9$p^)+}!or^8oMZjp*Jp-oWSa
zU|VV{SkKMunwkLeT3+HuHXDJkwkRg?*{QF3=IK@Tv`jU%1knFmD*xedN=Z4qgB$?E
z0bR9A{52rpRjncX8%pVaWhEviPQsr96fh>s+R2eYUt3$7^eP)Q^<;pBO3+8aqoH#M
zCnZbiOs|=Fzos3NRYTASWfmpFIO@h#Tt4spQRsDDbdWP~%*)q{^kL^bxjCvEW0SK_
zsoMa=BA0Y;ybo-FMvJ9!qCGW@4Ixk5E}$tx<dZ&xrDv2F7bofq8n5TEvRq=CAuzFK
z-kX#BM_N-`Ct+k9Usx)SlMCz7CVH2#v9vFqA6iwfAGTl7wEnPFo>s%ws*dLEv0b+;
zR9e<t@_IRM_ei~L{+gdPA<4dt*%Icm*4~Ix3f}-AnI;HprHLuiJvYc(J+P)`C0T%N
zWYCZZjeZ7ip~hxb7J=+l5t)|uoUAU}Se!vpEST$L>D$T<CuL=QvtZHp6(7ba>}}Mp
zw@(<+wUMo~dL<Nnr5c+277WO5VUSYcB3sMCrf-ByK~Oc5%+gj-G%N4osBG^awH`Fi
z%5?2u_4RL;5xpZ{ZuQO8+frlAx~k6)(L3VLJTr}kt;b$9ubC4r^O>P+6DbaGNVX}e
zOAyBfs6m#9VtNC!?aVgpG8-4x$bx=KbaZiKpCe2&TotXdWXa2<Q4@os$CRvhDmCVC
ze%FJnjpT2_stdRS>C;y$;3;z7UnD-H)+0fRLhsdMO5J2lNOZq&5eYsn*4C`7uk4|0
zPbmY+51LTPj!nHNTHo_voo;s;MrzQK9b!^|H!##TGcO<3eQF$4J072I991f)BwXY8
zvbx^$K$_t8<$Y7%@C6w!!F9o>?uIb=Cw%4UQj_D|E^1VSp<M@ZaKOj<5bI9UV5nDk
zep65tn8F@lB|MZ|1MAn4G>qAXb!u+UBEFx&qh(bryZ;zjAx}m@s1yobCDY+Y44O$w
zb=wm?qRln*y;)WppJo_`yeL-xt~%5M<4H8xMoLfUxn7Wg07}L1?IrPll#XPhxa1kN
zVw+bj`(e4JUML&bzC0N=kUtM;ph%jB0e*7-SkaxA7HJp=-W|$|4GcOtiyU@o00g@)
zh(x%!Dpl;wOJ={>`qd9l(0wW$(V2bvV$ToRxR>9*X&iMoU-7fuj`MP7j67Y1YO4dr
z=1i`-_TZ_=GJRbed(rvJ#S-L{?~Lww&R?Cl&niwc`ckn|oA)xk5ZhYs)SI;{V~@G+
ze;$2&c|uw0-M!D!!fRvSeJ1|HkHvS-A3WGUu|U0l9%HYTgT$0i=*Du#qC6RnYk!vT
z#*LiNed3W?aqqEH=kndXd$)15-!KsKIVj>E1JL>jqbnX08Y0V~0#!VLXTj82UuTFD
zSU=~ZgI(L(+rRkYi~AV`+1UaMmaOI*ERYa8s(I|#y2PJnAmPsXYdSVogl}eI@5zgA
zOF&#Mc9Lb)Nl5nB9@h!}PN`TVW7tF`_uRR2w`83~=I!}fDa5(=^*1;SJI3wO!y7us
zYEx5pnC}n}UcX-Ql)0LjO&&ah3>(G8#P(ZQY)(#2{#r=r3TTG!l_M+LAIcUV)ma>|
z=Iog>gOFpqg*u6U%wmTCkc$OS$NvSb&S_{pBz0bnCUd6tU$NO>Ra~cEf5<ODv&?3L
zp5pelMpNblXC01Zm#MU6T<;a!<@Plnj>lbl#oAmv(J;@_LE~{#j;sCjq`S_`_3T^W
z;Xfb&tm6C0xNwnvzSqh#KkWW`-i;eKe!ywDXMW)01G?X|GIqa5o)9phJ?`!@t*xy%
zbGQD@_X%1H9}U^078<EtwyV1FtyOWo{?4OXS*(qotu<XvBZ*a&l}F)Eb0D`ivN$L?
zRL1dhx&W-Vifm^q>7Tn+*WRcpFJFps<u9nX2O;Y)27-Q2U0vzu(Qj`kug7V`e<CMm
zt@UWwp{4btuNmjkKgO=`1F3WZ$1Bi@;;>2dJ%RCYzSEa4FKTOVmvo;@e=s(+-xl|x
zc=MQ)zvQU^&)}k)<%_a1Znuqd5B~hK$(KrlOcm(5zTDk(W^-u~PD6g^)O9q~rhy;D
z_ihwgkajJ+Ut`(Zl(PZ}X|aQ&<3sDZZ67Bla^&<cpw^fRU5G-WLC$$(_38#rPtT>N
zPk+Ybj<U<N%AT(Wkr5Wd8E4^->(*_xZ~fW-d7Al=F&Fbrlb`lSSN+*=bhU+vjZW}f
z-nuC($a{cegeMq3@6j+(&t=IvKTfx4;Hx&sc=qfW&i!4FQ^2qL`=70@)+o9hwarvj
zCrwujzNUU}-u&{kurPbAGUq|Mp{lAPPWGe|%pr6+fzJI!Y3Y6j<4=<n8&aa;&-WZj
zVn?7s{a(>0H?#ct^9A7Ro?s0;yB-b|gr@&9yM;>4Ms3xm$&n!rcH8Ul5Yum4Prbb6
z7|civyNZK@{jQfqi%H7u-@iZ8rB48z))xM<2R9Ey`~|FV4j<C7lDYY&+}zyo!<)j7
z#ted2dy5Qo1~0(oYh0c14qFeD#VROFR<2#UcBDAVi|b$?Y5pYP&!azn@fk$*mqS8U
zL+82_nyKw_(RWTP`vu>>?n7)`+?;Cfhu3xAwcJcDX01CF(BSTHAU8BL)D2_)gP12-
zj`z!SX#dK=!9jo7!>M?&9t`*#*3tQMsp90}N@ue_Qxc!x7~_kAkIrZ9i7c#7PPeR4
zZoG2!s`PEGLx&dey}>0gS6grn<`u`3nyARMH=8D>skJwnZhO$NeP6%LhPhc@3OF=a
z=j=MSqEjvd&5QZ|xOB<i`bn8kpt!+jhoz(SfB7Z6$QbT0myzpEf(Jzyh_#kmyJ(lZ
z{8sDNKau~;U-^@hUVOWg4D*6yw_I9eR1A-e3+PqQB`ZAGdFJWiKxx~?pX9U~aRTEU
zK7(v0d({e;<g|E)j)OYwE}<cUNqg@+LU6l^YVjS2Q8!S7tnU+ytvAQL9i0STBl+4a
zvwgITC5NGd9ei)l8SRGCp#FE6f*}S;201RzaHNYp&hy@QplfydjT*ODKTO_Lv~3j$
zP;N|&iTN!sFwk+X%jt*_kw7W`g!UVb!9~ro0mi{a%T&TvI@oT%{bPH%5{`eiVvGfg
z6$Pn9?=v;s!1rRDU-#Iei1S;#=WOluuMaUUi>m5+V8s^adjx8$adqj;_r9Ai{qe_c
zWo6|Eqgzu`Q_uEad@1JR@Nw`2A9&ZfA*&wD?cE%wFT)EQ+RsFvjR|{}#8!0)>{>Le
zComBBt#-B5o;`E;4))i&=#_N$_4zFkl;!)_a>i=xSX|A>mB6o*dNjptHh30qW=r{t
zPHRWjn<hWiXT3-(HTm;Kq9VSs=Yz9>`mGR8-%m)`e_@H>mc$35O6O&k;9*#fs@y-Z
zyce6P=<?6@=Rol-u}l+a%69gD`Gw4$uOFDd`mwhmV`-k}SRqbgGN}pNc5d%)nVCEJ
z9)TWS)ti)<2sx-6U$vhU^E1y95v40PG;KLWn(x0l^Fx{3rU1Q1eVcpGr#!%M_Wnzs
zB)q}h_EPH~P`!x&`v+^&QQb`y&B+2Cp^{E6%@T5QtN1Ds^rY>V9Cmd)kXL-C)WlAv
z{k5q>jpu`94To+XAAflb59Hg;KC~X|E(+$heT$qd$oB>XuDzph*@|V$gqv+|;%sq1
z@qa`--660TAmGd!r2Y97yNZ3gt2g5a`L|elt(FzYdB@YOSJo!zEx<LkwV5!yH#Zdc
zDn5Sl<UK}GUw??S#-!ijP0L>6v~xruXy)efZ4wpTr~Kb~yYg@<_pbdALM1Xq#*#{@
zka3$s!!b|E7)>&7Lxvor!4%R-5h_PWA&P86gE>wmNv55!MaB}MsNVZ=-uHXo_s{RT
zzU$kUTvus(_VfJKZ>@XX>t5@(pA#{mHP=LZ-?k%zt_MHVCEahHP*+{!jo^S^7X?{$
zbjNm6<;cRKB5v{DXLz<dR<VnVi|gs@m)~Kx_S1j2wm0h5=ddHsd+%FYTLX9UZ{D^o
zzbqxmAWO%~YkP0_uiuHM$n#^sS9Ttq>ZwTz`5pG&djIHeizqDElzC;SMCDrF)wi}5
z0PV#^%nBUx+jQKa*4$e`)!eweG*zU)LKWZP!YtDEc`doHucqt`a|*9NUB^XqbtC7?
z9PbHw2P=B@NpIXpcCSmf-OkUiL%oaXrkB2}larI6frr+fW?M6})9~(=*4EdIej(E`
zFgL#>B|MdIc@~<{6h-$dP5xFg{FrdFH8p87O3D1gA8=Nnz;N?#Mn?5$!F%2J4c%Pp
zH*TySbdNl9rr4)ncVB+kgzLeKsVBXbB3te%6vpHX?yWj{^l1Fa&^${S%%*ud@qxwO
zLizkIyN4bztJW#0$Xf5|vU?jF%SkeS0&QHYR@H;ulu|f)I&0R}-u}X^Tem{m<kqfJ
zICeowjhSiKzSh@OLDklnd<L?2)4L5{&Og8STUt}=efT)I^CwQPX;0#%>gnm{8d~hW
zI~;=3B1XNME)59sAO3o!{r!7!bZTgyJgEe9;sd0BXh_D#{{DWPv#Fx1t7|V%rZFx%
z27rH#L@t#lCu%U>Hx?pN$!*(Kf`->IG`w&`jel|zdu07L<rfaMcrO{x<mKfrrlrL}
zGnt7$(ed&1AR{-besAl&ulD`PBp0sU+EH<)V2X{ERZ>=V6%M(rgVxd1)HGsr)D6x3
zY@q$tlA-m`Q=*C9?Pcf{z+ks)9{uzw5#k0(Wo1De)@lHk0qgHB^G-xx&w6QTHiWyt
zQ@JUsU*C}Fz>=qKOc3oX^A<0(FOQ6fAV?~At*Y-Wsj&2&^}X&k#-7qoANw{JzhLP+
zV2uZvK9wp7*Qn4e)2PJD!}H^@56g~!A3<+O<#L3p`R6OuW%+;ptoY2;zt5G&_q{ot
z)xTj0{(ebWjQ_{6@|U)p{QU3Z>6`c{K{5aHz9ZP~|9Y36o_ziP>T*)M(T;xJJnyV|
z9_Y1A>DR*?9j~W^nVBWEwUsr3`n_o>GllH+8r^$BwRLO>Sq(54t%VMhDMj-9u}2I%
zO(bZsV3@POk`7SEI_?F!m5Y2Bb=Toa5AZad{^QCOwrukl)G;b4d}U2dO{dNC^o|^n
zV$tylR3Sh4AhRr8l;AP*JBc%xSlHOE(LvpimzWvv;&pa*j*f|`MnnKO95ZO+XJN7}
zc-z%~&Cp3kl-agG0ADz^Q+m$M&zME$Da0{Fur;=Y3vG}5hPhynKHy*28b@YB)@=xs
zJ>ay$x9@^GV{RO<nSJ%@<wUxA@ERB@Ep3Bu0`?KaJgjONT;O9<D{HA~<XxeUK*gtz
zkXKOP5fmJryhuLzXH6=RSEHZRugox+(}KW14$P8E&qL1?t#D1W{BVVbhu1YWGBAT{
zlB@Dyb91wNSNW+YNHwpG(>J;$tQ76DEwGl?`lwP1-uPzQ$j-3Epa)tdKO8YbOPu%&
ziO0UlJu;-x_=@bo((ii?-&_oh)${8b>%>DpGx2fipE(S-k!H2$f`W~oF+}(j>H>P_
zKDOnq2Hp}DDFZSkv=P-V0wp3w@rjAOUNpMgq00zL2n)Y+=S~qk4x#Pqdp|9U`sCwB
zR~A;*%-grw$sLQQNEcl)IMt+GPYTxXDDK8cw@3;0f!zR%IXP@_N-9C0C2$lkmx+nV
z)cobkeBgYa#q%g@p+$?0&@(V_!Ju4eN??GRC~lu1NW$z@p2eo7lap(bP5@o<&xB~8
zW*c?!B6naE<D{6F*gRY~6A=>=vsirw&-iv~zq0lEKc!vh!7B*PbIgv8j_fN}>cJz)
zh5rn-w6yF41(Ox@99thw;MJQ<B|ET#L<HnxsTvJ51_K1Lf3Tt6pgrtkAiWX*($8<#
z>nRjOnF&OmXq=l>|DqKc5Koc<xfA#5+WzfUC6tt`LtpM3ZSDkrnrSJ6FPucq@nGac
zsV!7OtQ;I1A75v$sCx9M6$uM0xDUn22CNWuZqlKWVPs+=$Vl1g%#38vVT3(00(IXF
zl9EiwEOd~^@Q8~i8l@>6%*ZMC`+^7@(}Je(mZdyps(e9QQ&V%meG`U!dpp5X(!7z+
z3{1z@uN4yFz%GI)EKQ@Kx8nX+TOuMdl6etnjy@g<iSoI*Bb;l`_d|F@k=eRc_Oa5A
z9VX9nv805lrR7;H(b=<SpLTQzlAs0VZ)RrZEGKJ6`_jGB8mJgK@A@rUIMDE}Yii1k
zsfavl)<mMe;}A&F_=J#M0SPZ}XeFXI@Yim@+gAv=2;0u6o2(krDL{Y(VR0HNK;3L}
z$z8knuz`fU-aq!V5^8r-e0$F+5kKYuySMa<g@t1QO#2ZhrJekKd=Ef35a|Eip@kTk
zxa#L7b93|bYxlCJt>D`6sqv#_m^;G0V#R|yiw#^NDiS|a$jScQB4@UaJ<sUaYifFt
zmU7)$SeV=|Zqh8+G^t3-nIZ`7y@rM*-vc&o+&D0%uCC4!Z(!ja?(CdN`vuS^-L`rA
zb^%0!9b(5Pd4z<z`by#WdFPzUed~{WM^4L=xNqM+o|%wG?H?2X?=_t9Nsci9TDf-#
z$8+=YZXme9m$3dZl0w;pBi-GzhM#svO_kd|I2J!2>gicfFaLF{!-SAP<HI`XIg0s#
zxArU4GK|qKdab_XQB~F17ir3Avb8@t^o||N=_*-VTs(`;q!$CK#pmzdTg3>?&%?ql
ztVm8yw!T67)>I?%jNpL)zr}fKbz@`WAocX=)4Eh@as6}&TtjtBON+}ml}e4YEhwB*
zW6TNtmI(BgJLEnZ>hL)9AlWN%rUSDdyIi`Af<e)<61qM<K3R=Q0U70E!g48;z(PXX
zR3<snIQPNiS#tX)4N?)yl32j}aKT(K!QL?u2(oh~1lcT>;LlsP>PCv;ThR(bnTPO^
ziGaGb5hvbw`O2xdlOvfs?;nlp`Vkx)9G#dLWsp_8_4ho#qYgqC(4u4S(=1Y89ft|H
zLTSq0Ub`Nl4$Htu9OX)<0_zQGYQosJN{w!YZXFFMha-vgY+uaA$xgNCB6&Xf{?RFA
zbdZPlb(gxYRq?smB~Pyk=EdzFYr}DutRN$9THlpzcz`E^2IP|J+oq>fDepnJko_Lj
z(_iZ<PqaxBMZuEAId>ESKP)F;`AFIvj9?#D_asSM8aa$;oun5pqH0dp4F6eWbWe86
zmTGXWYy@IQBbhiMVtt$q!I$NM16e?|D5B~keaJU1gO|jI8JKuf8ThFI!PPVNg$-~M
zl6bb8UOYGPK`vjt8n@!rhvDHYaO?APPkzPI>GqXhki<pNiZU|&4#VDN1*#On=#_yc
zp%c)Gj7|L2)KpxVKSaAzxROguJsO4Sm~eFe3~!c|J!hPI#`<nOdW7DGbadkx2|xO9
z3dLsE_B;Gap512=-S9bGOKk^-hZ!i6Eh8>px#H4tT|=qS{S{xw`C_Cmq<2HJx=uhq
zY0yva_mAVDA0#m_F{+k~#-c`pwlqa|fyl_ni>awGl)%a8fb!t!W7O8iFf;_b-38W%
z!I7Z%`n=5!ta%VBeM|&)RV_+A!h>dTL+~gBvH0UhamcJ2pP=CGyX~>xL@6f?(`}`c
zmCGELhavt*urp(z??%l)2qM^i@uD{F=k_=U{fW!N!oqASPZrp;L`rnhIS#^Yefi=`
z6n*=Zy=C-y#y2KyAD<%Hi{K@AbS)`Ibf=--%!11&zEy}_l#}D!P77#Re%sU0DLXv;
zG6Q>^=2JaHGcsy5wX)iXxi`E6vhzbZ%ZlnT5uz@XOI*|A3rf1S&WL3L)UUngcd$ot
z3kjW&TcyLzQeeF`XBv+Gn#Rk67Q0kz55y<;?EIaHa0U*LZhJL0TZ2`vXOJxgeQO!+
zx_OqIa75B=6(E{}=wQ1%wkPUXIIW9s9r1T@9Qq=sxd*qWd+=at6Zt0FPXY^I3Y!M0
zQ2G|zY=WodCD@smm~?}K%j-KK{$V9*FodG_&@U)J>O*lFj;E67@nQGS#{@ghAn@4r
zsR7{Te(SW`_VYQnALP9J6y{cHs~!fEh!-;<f!CL(Jb;dCV1b*wP(<mO>BgaRETB+`
z2x!bHB7b4uzj|~k5C)Tl`k)G$+8wg?#pkuV9XWn{8G$j>8&R+77^LDY^PodVOdZXK
z#w+A=tnVfzC$kc3<&Q*n_0ae^SusL*eS+{J>N(qTDBiv5CGH(B2OncEkSKwO00mOX
zxYAvURDjH>h^OGtgf+r?tAY{nca^sU)y;r4_>8({iu!gbDV74;Qs#(|ERWL_TNJ_k
zFC%PdK9Vph*@mC3d&HTDm)AAMUe&}FHY+2dA+SB9Y(E3ek-C6k1rJ9`3kwwjhnEuw
zDg(b%b6q9Ku%St!Zz;0|KRcBFyJu{@y}cm~5FjpIycmJ7HUPzU9lVgw^pH{7>+W*D
zO=97TRQpFE{Ky{~u>h^bhxeqt-kD9K+0er<JaC)XIspMTB)3u1m)9SE+y3K1%M=Dv
zxlh?7@85WmxS>L=8KCDD7Z(e(#YRUHrgOCs8uS29h%T0$IB^2`hE#?B2s5%x1ji}o
zlpR(D)&O1&F1FZ_ij^%ii%)3fR6Lcgo2iX0%=G3Hjb-#@L1qG=qsDDZnfdSEV^`xI
zf=-=^D_%D=!xnM>zI<G<-N2ntr_d?=%ER)<Qm!)<{rJ7DJz!{Zapg+Un!@zCG!olh
zvtALnTw6~sN2n#&wdL8vvu9bzoncdShzZa73>wQ~7nqkVlM5KN#7w@{@x6v;FluWL
zy|*&(UOmsLhve?3y#v&XYn0*Rgp%T~kq!3QH3tI%v>1LKDeF2eMMtxHysO~Xes`Oi
zy#MGImEFseRL^rA$N&R=dzQxNCbMUi?l;1K_F%rpe?p1zbX>EReLum@$$38|v8+rv
z;(T1ZP~dg<Qd>B*xRJH)`F`DN4t(gz+`YW=u&=;@j?~KO4OrJdhvGKC8%W70C|J-Z
zyG!pnoH%>#oHGhAxAXFJkPM>fn>0DF@aUA|#Yr-anCc{(BL}qrbzvzOU8*c#(J2}U
zI;Y!zIrLriwFJTAZ<4bvP?EP%0~~hEmXe+<O)ap%+0*cAgO~Skd6s?hJ?Cp{_OP<B
zME5*2_GFBz`2~itSy@>T5deu4MF%pt_==M`-9FP3C-PJR??gpKy%Vis!$$h}CXW|I
z`O_eFEv>AaQLH6MUO)5Z&8Oq;yu7^iHgHQ9Fr%<NO|!-kn3AUq2d5mFveDu8`7*dK
zYT93#VSKeg%_lB4Hs81-siNY>)UzkZPn1HcrqatF_V!L18U>S4i|XY7`^>j3<wS*o
z_WOd8znUmOIg3t&1pBjR&x$DG$K87#7SJAfzt@yAa2I=fVd#qO<F9qcKk-Svz;c$(
zhqZTp3?gsl((0v~xgfzlJ0vQq?3Et0qVPdbP>{ryE&CQ1P*N`o3ibEj1;U)%;#F1Z
zB%=!`Lfy>9V7<69PAS!IlCGNTFGKlbYHv02|5S}h5NTWiySi)-G+gjL<?SISA8)Y1
zxCJE<uUS{e1Fp^PS^4=ZQKZDwYP&KnNOSake0BoCbF5yS7$W-3{M!1uAhGe88UNhu
zp3X%6!sd>H7Z?=yuyq@fq#3~>NE0U-_L3I=EOlw~OM>CU;d_nXO6!LKLOqO!flaER
zBa!VL^`dV%_QHjN0YXo~I`oZ<#9-chKlm^{@7dC*{n<6hy_kq~%3gJV=tg3>9b+WL
zUpY<XX9Oo97jyZtR)>4%(PqB55!#=HD*^WE^b5llr}m(j&@o<&w0K}*4$$IZdoWGc
z!f{=L-!og={E6n|tu`gDNxOqp|42++jiMeup>1m`eE-3N3FE(7T1@C(9+}&Z4~=pD
z{P}Z0y?>K&x({-k!|LmQqQ%JPuoUU$2MtLs!FL0|TXXY%GWw%_q<`p;1T<0`CvQ*!
zx5L%A_*K9_k!tHeotuQ{4u4U+YzjB8LLFd1|G~kjAoo4K8qn-KYiWrBUw~}(6<Ux&
z?!N7D`#Rn=Kn~(6KK4SGdZfrWw-zpuM^sd{+*pcW=iuNOUX`I9EU)B6-~7_V-p(!#
z9jtw5VP1xRag`H2(4Jh4f>7N=O3GT4OHig`)^=DvgY0Vxa?lNmiu^?1hY#l<>1@kU
zx6jNgT*VlX0KpnJKJG!Cw1(xwEr((h0{f~}r6+WXbBYdCGgB!tGMreXdT1V~fbUbm
z8OBinK_aTYEGSGUckL5hyLPiz%MGpf1JV9?hBBCbAWs}ZoyX`(uKnQ95cBN?t+L}&
z#!D71O<K5O5=5e$=rs@B%*PoN5>f*~3^;ineLADEhrYrbBve(qGSw4BDdz#FkeV7)
zy`!VBLgriuAH<u2K~1+eoA0(E0o#N+7KfIWmVJdk2et($)CXH1BN2y=PDK0N-e$BN
z4WRKi0xllI)5t)Ku)6^_g?zGX>*@n1TK{C#{4rxe1c=TeCH*`*drmhq;_6kwnwlDr
zAYP>2;3F_~)=2rEg@lB>{2KhPf%jyMikX=i6`4NlMKIhr6HJGWHmdz4Ztczge7;CE
zAu18ne&;U?;Qx7gnzzLSX(5DE(ZEBHC$(=m>UX8DuP;Q4UH0I8797y<d1~skZYFn&
zm9{oPCidutx=Ewabkt;$IdOY{kowD(Ez30BTsL=T;nR8_>&JyIwB8k*oOR*~w-fr8
z=R$_2;^|BwpSG{o($kg%#?pKaF^vi$dK_q)&dK(+sFgyIB_$;xWsYgDA9~u6wY?Df
z4=_ESQ4%E|n6{Z24}^24!9?|l<f1%+gO3xTL3%kZM+R$P3y*~2Ezkowq%&7v?(DR=
zm78l<HT%9jfRzA#=NDX~(Ad_-i?=u8H+uPfFMM^mqXAq_J(saXXlUrg8#fX$j;0|q
z)ZQ80PxBaBG4_JPGZE&h9Q5+D<I$9gTK;bpHJBFw=734tUgFZ|5U29CTdwnFB(V{F
zZ8>>&<uTyFfQ{u`>ex{MmQ+ySCF_`|PDOwdiF*BYW~L5uhcn2HTmb&a@(q@aT(j*C
z6@3st)q+!}&W>97Pjt5zs;dP~UVybrN=Ojd+1W@UmdD1%dKfqM^sIxFqY-Q;q*yy{
zB4fZp>3Dc-v(C3U4d4Z&(9NH97ie8_>(&)QLPDTk&+X5Sym3PuBSvFzCQ%>QcG3d3
z<+Snze1PXeXiBF9azO$Bn81Q!R{!vD6t-?C+R+gG*)WSrvGR3ba!!R3c!Z>@SEJyC
z>+0&BqNuw+dkqZArYcCT(CJA+QPIv#m!By@R1kaV(wcF4IWNwUmj=-%{ni`a920?6
zL3<M*ZW}<98RoP;2@l6)Mkd^-{@mod{tqA8rM^`}!say>Kkp_2ER0PvQ12z4LYfeZ
zDjH37i4^daK6mJZl^|(Z>r?ac^Z5krki3ZCIzvOz^xy7vFwhdOi<<jGuXnNY^Fe`8
zq~Fvtt|RfWp5B}M60di{Xx?9L8g52=e;4~NtE7YvV4+N9f`jCGn|Mi<j)%wWfrX=E
z0<tab+_ruMuprE|c_(_Q2kHv~RGhAX0h>Xh*u<pIQrKA{5lMT<0L_fCSS-X&zac~A
zJq%4%RaF)LkKE9Xo<U8!oFYPga$lx||Kf!I`){O4{p<CA$PknFPrm-22<g8sMqcjM
ZYEd@Top!AW7>`d7`}B?V3Uv=f{sVyjd=~%!

diff --git a/reference/plot-21.png b/reference/plot-21.png
deleted file mode 100644
index 6e46df4672cb09d6b564affde4ed3c3ebf52e460..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 25154
zcmeIb2UL~kx-GoasELXM3n&Ogj1_^P0xHcYNE87(9Slg3Dk8l`qoN2TqM#r}jRk4a
z`vN0HM4EIKu#hgW=zYOjcfN}xJNxW&?%Cu1+5b3q+|3w!;MP^X@BQBQdFC_ce4gAq
zuCBthjDH!0Lg6~{ozh7PWwA4bvdHN3CHNa!(4!&zT4JQCqJ-c0nLd=|g<oG>{7&1R
zLJ_<{{<k2YbJ&DJ`HFHx>D$xJ5Br<l9-cjuqc}9EcYpa+y@hK(t6n3vyTboR{wo`=
zPae9o_x#bEV_d2Qr`B)%VqEiqk&CQrVs-YtpzzacPRgBJcjfxyv!?6sY~RlJ?N2;y
z(!t%+b00t6ENSvK>$2k-oa~c|uWL-IGhk+!4f2(_>y<YuH&{sva4zoa=jj!TIe&U(
z{8={V4>wm0EhK;V`e@#A^82T43na+z1CRfT7aTueIM=KBBqnC(yLa!5y52vyCcf3h
zGaR3v7xyf=VsTzv43(;2VQINZL?pD7H$7x*+J#(B7XNPN1>`bTZEodU@OO8;`Ihw*
z3fSJpz2JWQ<tOS71~0-Z4jBE{7izoU{H_Q0cblo}U9MENoBDV|CC#S!adx(x>sSwu
zFtfg>?e^|6A6=V_%?2)=_uiT&iSgUF7y3y$^j7cfi%DwpUbNB&7e1A@RN>1aVJ~<0
zK{vXu4Bb6RxqX9Elii84XL(IbOoodTS&||my}i9gz12~66AflX3e5i1_wL<GV6&J-
z)7=rm61lUNuD0*gdi9lf&e(@_doP~5+*`e!Ttx(H?0NgoDnd7pZK+BWu0GqY?>XIR
z*JM{zuvB5_hZiw1w-<5CeM6lpkSKirVE2=_IGe1ZChh6;0J5<5ih~b!n@PLI6%Ldb
z2gvib7X<`dUAceT%a<=7W6i_e1~Rhs+t-^t{8@I8kw&NL7YW(Fm&`DJt{&ECo_lP|
z>T*T4yPoT(r;?X@DtbOVRXu&?Oo58xOeN>;$yIR)NcZ<Y?Kq#*pr?;pP4JwXF%8hp
za%AupC7)J^pw&(}k%wTuG%-0!)6hdIYqG`7Y|~?9!(3c(h@d?dq~X@BTc$<)vW`}c
zjK&l-e<`c_S9)wdLw8ALSE#KibL>-9M>_tmopfL^zQ-!{*|RNfT@TEp8`wi{xXa_T
zlb$In#uxbVnFYv=yfuxFbsB7G!0Qu~jk^v}2im8G+wDC($D<9UT&b~xCBI%<o|NM@
zX)~l(E;o9tC2sW7^N2D(=4)=xR|Q34x~V&z^z7{H9Oq`nX=72*(Iu?eZbjR%{nBhK
z+B@~I{V5HjsU6)9iqEI6b8F6YDD`Isc+DF78ixc9cBe|Qn#rzgT)NwtR`KqhsBxb6
zf?@2+4(s=-;u&8b_TSL(c)kB*lfA}<%N5@a4`dCdn)j0@HT;O%V;irSPWAcR=(x-s
zHo0>XFXH0^0yPx&4P^GkZDA@h`;+WG9N8-6#-g(((kjzz>Gx%wY?3_qBn&p(*l_7P
zx#9fPr4!BCv39~~MOd@;yhYr%u_`j@ilIdfp*Ms*oi7hG?Gq6(@R}RfdUftwU;nH@
zD)SO!ce&i|9XocUe%fvF?(W{Xu_}Qa|I%3fENNWrdZ!i_v*n@^58V~hu6efL*1RWW
z7T5}e$hlmK(M=N<5pkqhr%g4Kd-D47iroqc33+>d*U!c0jj((P=Zo*$-)C>?-YIH$
z(_c_ldVfoMfTED5o}IEy7M{^?&y#?{-kRu5fyGK{YHIHz6crVi4dq=k*;p~Vwwqg1
zw|zD{Q<p2$R~zg5`s&i6Y?o0ZtT_4jH`lp_t50btM<{q)@+g-b%w8vBqxb&4<R(0%
zwylcOM`E-RM0)%Bo*;U(xQ~`jHFYZl_bIwf^gHAD3CK@AR`xY|`_smZ&JCFkeY`lU
zDm(Qv_uz<%1QzEvrdS$J^fySJ%kyrs+jRK04_4uMgC%YKG&O)`*_MCv#HmxOmu|Rt
zfP7<%$7D0P=~B~8p{0d9DjR-Iv%h!uZoV>~{#_cCkxC1Z--3VWC=Rrz@#$UsHgHaH
z<mZ5V`N^i$%lX7_;{Yjb3A~w^nd#ge=F~P^Bph&vTYgX7<U{+ioA*RD2lDPE`_($z
zET79J0dLTLahPJZ?{w=@#rw?9$A#HbMPdBWI1#ibDgjN-Hv}EE0)?e=r@KParaqs<
znXf&UTN=F3$ESZWpWaWpp5I%R1WVAno_x%49e+n1EtBP%C_lfeAy9q)_PR-5!&##z
zb4$2|9`;S%)Np@XY#ALNU+ym?zrT(hlT>DrY^%^Qsxa_&6&KIX^4M-kDJdzk4g*0m
zOgcMU)>d0v8)x%bF-w9vWfaF2a_g_`8QpE)^}atwJBhzJ_lSpwhhu+zLWX4SXxQMM
z#<ZVQ_axWahn9D9;m*vx-nUvGvaF3MPT}6$>K(@4hlA<AR`hH~s<2DP&xmwZ$?S~o
zaDlDi&PFy&S^BhkQg=^7l9?X%^NG`^t6N#kPIUP)xzYDx)Jbf9CB$<6e#6tJ*U;w%
zoa||*o*ldJiY^+Ty-GuI?y`~+LfS@LW5SmB*!1+ho-<VuT|o=d+x=xnW=_V%cH%~T
zRtoOV&zWoyc>46I^(89)Br8Kp#$qE;Rh?&t+EVw`QK#B`1w1_c_jTRhJJa(-0Jn+o
zG25e}_|BR=bVJeV!YAvODdDt(inh4PJZ7kO8kZmmvJ(NGV^6|#M8~@o*)DN&8RZ`0
zuGn<_lA8JsXX50Y2AZ^TTpa@Ea#`c?G6Etig%#xL2Cy?XU@_Dt=yH)d8JA3|Yig{k
zy3G`s+HvyV|N3JS|0}g9qy)S6!p>ALHj^&n_TK+pw?`9C_d|PiU(YSmW2Ls9yX&YE
zjg|t8G`uW*yjItq_B_JV)u4<Xb?SM&ZSI_iNLy@L%Pl_vnJ$wicZ=BW)QWO~4~AG<
z@PvX5Trb4UrNn!PP0+pA_Ynx>*t4Uh*JE{Qdro~L(zZKd`k1f(zNyx$Y4tV(ofoby
z;j(aik954LPH5Np?~;-v)50sYbak`TB9NQ98bizX+mv-#mX}2+<`xuL$Gf)_36lJl
zGP5t$@!j_49gS7bH3H5=pOvZ~tjlFNRM*w1DCM`lzG_@9KWXLIkIO71amA3?SZCBz
z@hG7?-0iT7i_6`6_pFD<8!h8I{Vy;af}E5r<^L{(&EOI72Pm?fk<*lw^84a*h3K<g
zGVSfcZ0F0LK5Qm&0Uu@aK13?|^@G?@q#oz#&R~05sHwE{bL<#}*^k`aschC{u1rJj
z=G|ZV`m(1#AMdGrqU^B8tR>qeV?wfM!|F3<&YU=NW-XHa32p85ot>Q-@uiFM5{!$w
z{4EC>9S<Wx`S&(wIkgm7sTcVRra2#Jxyr3je=c{>W6&*QBLC*<&<(p0Kg80SR)xtr
z`Qug<;&NFt1hwomq{>-XSUgTmb@Yudt*Nf=&#y`LuXX(+SlBLXk(M^f_tC;uY4T#v
zkpyDR^V3;P%_2Nuene5<f#$5DH{6O%t?P8Xl(J4eR~J)YBQ1qn#_OiV6;n$~3!58Q
zT*c>iWvq|n8C>i_g0VwiZNYUe0TP7~2?Lppa+LRYLA#DUSBuYVQ<31AvuA_qhIK4v
zr-m~o?r9oil(ITYg1c+G#kOz1iJ+8PJWO!T)Q4c|7<GkZnx-Yvi~AqxWx?Ix;36y}
z2lq5KHfklCpOC}h8onmszVYbMqaq?k<GnR@y;08#z1Wjpw)V5N-&)vz%jh~Ar}x0p
zslVQCX?)~($EL4sL}lHCPubUhf1}eNa&o3GzUa#0<*D^vbdSl-M)Gj+MKb$q4cRkw
z>20?}FVPO&Sn2nv>UoiXZ8i@=tKwtik6jOr;n_E7vRQMZ?#g0yO_^!d!%YJgoHJ3e
zYhlYmQE_qqmK?XDn_I)zkyH|6n5z(%(G}9fS}u|O4M`AU{@Zu#Xf%qtR%hsyqul-O
z{4U*{j90)ygXhPmHBX;TF3@=$pFQ%9ccY<duC&)9J>B}BnfF85+S;ZHWvD-M^2j9y
zpl#q*jWTPm=@LVmOED3%9a}Rd5G93@OiORAR^PjEW`ZF=Hk_Vbwwie+@^4tqk4sWI
zo|=@Fes0^>?WpO@S>qs{A!<1Zf8v=vkN5V!dUJ^mB-ow$$*KB;N@t^A?2XKnSi{!p
z=Nb`$F^LZI;i|tOX}MOhzvU;7W2IevLoFBIaAl;}83G(S_r<1(wB<-$N%MR}ua(Y?
zU%ge`%5jO6gTpv)Qn}ON6DJU!Pj{r&#b{UcOX4<<CV4EIZc442kNe@?lO7hpNf{F(
zoh4}vY=8;rI+8eb)_CgU<1;2AzrFfcHCS9$+oW&mS3c@bmKqQDN=n-_Y2{fss@qRW
zO1UvoG*|HN*(TDWYvx!Lk_8BEAE%+ZkXC9IoO)jEBOa%n$~)`ouE|bruc<>xb21TO
zP4_3?*wb3VWf0jSU2sqmNjU7dqpGteKMoI2$hGHDlj9-P?JL&r`a08TFq|f|@6!1|
ziResh9HV!4w`VX)-?U7Il5D+FeyqaBM=QZddA2W?W$9}el4@T0c&eN|TYffLj6Nx;
zj3s%W5}z~CSf6fZ+~nF9n=u&X*m$ATKf&e8jOmlyT+=o2o@?2I?73XQ7s+;8!!DoX
zM07EOpHfp~2hukQd5l$>7Fqdq*N&3Ha*s+L?Rp$GEslxG?e)92n7+TyQxYWR3n1a+
zY%5q%6RlZf>p2=aRbQmQTT@fR!-L1bvuNqcCes+CHLC>wPgNR<f=dN2A1BB$BRI}v
z*8{Eei{IZCkg@R%sO9G0WA5wSH}~wti*3);LU{`c2Ia$do4)rk$Z;)DlWAT9@X{pm
zHZCgaCO~FU@`b84<W66z2eS}qxcTKc66mv@srG>K=1_8?2>w~{DWOOxtAOon4&Y>b
zi4h%7-A|oJi7!AVUH{<N77-Ctdd95TX-<vg!?S6`lZ&LLxVaCi5(Rj#0NFeP`|+yw
z&R|3Ta?gn*^3pW(_Xv7U<MaDT=fb1q^=q3+xMR;vro9+Mf%K}t^--wor(Xiw9{Y;l
zvW(A>1`fuqS*PIc9N3OvB1({esp3IHa}VZ(Uxs~m;RzMNwZJzg^z{V^Y}32=dhs=J
zv8{&KZfn=BwX4-O>ojfgWYXwmjL>hsKj>Sy7t8p>i_IFw#Ui1M*BJ`qw0k_6%bo3%
zD8lh+FZW`Ns-I?Q037?4dkolT9CxUV(bmmQzi27b;%uzv{P9|;KP{x35jVpDduqF{
zrumUrNQ%I9QWmFuXPg*~Qh3yr{l{pQaT!|%a@j22SmXWpD+1}=TNeqNpWwPCZC(+n
zmF;YId&9oOf+8TveUi2<*=M7pQU;7I$`skN)>KW@rikB`mBX7iZyxRl)Hsaz7Rq4G
z4SFfsWPW;vO!~&LuHi1tHaDk%S&xK8kaIhMfy3$3?l$rS&K$5Om~E-B`$akA`vjwc
zTlW<_T;(R}jU|C!oJUJ#o+KvjB8jpH#p3!+D*m@MJVydg=rktMcB8u8JKYh;3Gx@J
zo)ut)j4K~2SK_PI?J+;TLO^<Fp}(L2>K+YI!|z!PDsx~M1z!=$(Dp|+gl=t(aNiZ~
z()EDcC<$gO_xD7S%K7#>Ns+_1*YO}*+Kh+j^eH^+`03Efo8j_qg}Sym_n0j}!mso<
zS^*xmk!V0_41bgmyMgLfBh2{OG-nn7U$h5C%$|RULs}WF8DE?JBUa=Pxy|7w`|<+o
zv}RrnxnZSj)=Xa|w#zzcE3Jp_BgK+<5-(o9yp>{EXH!@I$2y(5m59Q%LK$mDp}d;g
zS;_gs|Bg#h&6)r(+=trOai;IZ6aO7!Pm|xqIp&ZzX6^rv9L$b)rb!J4`8bqF-o);!
zeB@<comgl<etRo@@sDK5U-g2g*Tm0h9>*bZaB#RLzUW+V6;@-LsHm@x4=|fDApZ}d
z3n|X*C@c!75=2R$+Un}+HR&SW`WgX3@{Utl{|T4#S6$x!4=eE3eZ@WAOv>9Er@Cit
zAE|=c8n5`NeYrm9uKsWSg)2lJpjk~_U6CPcB>29Jt-<VYfY<HKszFK@E?nqGd?DNq
zDQ&>g#S}}vjE*)&@*94?yFCBpndrxcxt_yO8Xi0R6=r(2ZQp+3i700i)+-QJQMX6L
zv5*A9-04zV{$@h}VVS0jhx!r9xdtK>y*%~YhYL{8r#95Z8|IQ)0?}^pv>)=-2~Ewk
zQ^pm^|8X@Ad<%f145~><DaIgYKj7~&J)XR{=PCh0vhG$RyPcf??g+44Lr9T+1dYl^
zhi(uqMcJ3PfZKEEDhZV2d+nHvG?f!4DtU`QA#K^B{%{>q(BS|fLF2;PH%CTBv{J3K
z^+5X#_eRGTI@Ia&A&2j*sNvk+9`R`@XTWt%F*-KTEQ_uE7@6{@s;V)*-%m6yUF4Rp
zuW`YxwF%C{ZPxx_($+etT#A0$c&KeOg6$4Kq=bY#*59x@F0=O~R_-QAIzT|nFM<lj
zkD97NtMyS7GT?Hyiip(4>1P#CfdIJ`XD=!#A?2Hf8p}HNuiN_Y;z1=PaDUJ5iE^mW
zdQZmw*4WtC($XddEp3pONgf$kOCFw#FP$w352U0dho~B>wV8ZRp8pX~u9oY7xa5Wj
zG&x~tD9oO)W%CNkIyOBn2aUC<Eobp<)uTuGH$2yjH{jtJMp>0&^P0S!A{*KKSvFxZ
z;=!>I76f!ozT0m5+&PAA5#n^F8XH(dC13GS({(FXUvp4BSO^*Z&*F7bc93fQaM1%i
zc=w(|!WN&NQYaPIQF)#1Sw&g)GJPsScz{tIbqW`^@a&?z2@7yfN=hCcbG$Ur7y;9x
z-9@jL3U9z}(Cv?ojrH@Hto>yDzUv2@ErA#Iw3eRV#YN4x4g@4R9I%)Y`D!+N$&g0q
zjU37^k8sV*fh^l&q0(X^Ez4blcWSR9R9BPRU=AUc3jAbR(hDR8pFSPbpLyosERg7O
zuBpkZWA=N_JtwC)t{$?9`{kFf3F0Od@v-~gY~<ak^`(aV_~{LRY8eXpnU`<+_@uSC
zMpxR^#v2Or3|@P!wm2Q7p6!UT6?QmDmyDbhU<<wJRo?pg*w;2Ue<f>$B|^}wA0O+f
zG`<mFT^EtFZO4vVgbNfA3HM@6i%mETw#bW!0Dds6gB<J6Hzb<y@FdVZnJGs1<{!#4
z{iWGFKf8Twtj%;}gy;r472ny+HSqw)TS;kyT|Q6jsSV_XdFtkM3v8uBOJhrzl>wfY
zeLa+wmF-Y<D=FpAP118q6Z-1n15|`P_K1l1@9%waHf{}h+d1ZD!J#-jKU59jB+aL3
zZF-TKD$Ogdw=P)UQ9B#_0)d+-r<%K_G`wc2PywtVRSMQuMyBxAT2Z49zZ}6C=K~_W
zjntglAjLUY$)``?YIrk!1XHj)gk}!uKZUSCN1gq|mQLjw@cQ&Ln7o;5fMyoq(T2n6
zk0crPGAT22KlnRGixnY`y${)9m%o5UA;f>l9kF!VHVN%T+p@!4ighjs^_kntyN>w~
z_J+U*XWT;@VpPC*UtJL@Um-&5tq_vi>%X_nXPcPy>@McypfK4S#~$q3+*KAPFql26
zKGs(!K|scv8{7eR?%a{i79s269i-~I#C8Wjk%uW{6WMnw49;nqEs;3zk@0>-%P)2F
zT1EZ6e_=lU>$1_mFOu^}|CQkB|M7=)np3F}{>?67ry!+u?uBle%{B1_k&)7Dw-j(u
zNGHp+xF|H$JlYbR?PUZE`MrpW+K$7$xCVKBo0x}$(2MA3-<g@2g_@te>I3`P^wD^6
zw)Ht=yn_aiE203!c{4xrzC!ey#n}SVRzr$Z%Cd823k~{6CH=3GHaQ>tuT@X~{S|+e
z4*fr9CC?_dJpSDa@UN8K|C?Jrx?Df6>}6b>KM`ULz)Vb)7_wGHpN+jGuAiv{uArUJ
zJrI*$C>xW+$f?9rEY6U`s4yO-QBbKOEC_1B`r=y7H__s^h2k{p=lkoR0K6?In_8>@
zY2en))mx?ct3m!NibG0)kVfkV<S;xIF1PPpNr}_TGMsxHa9Wh4d1I74E3E>c^`@)&
z+hI<eAVM}629Xw~Y!;-aytNDZUBN3PyeZlue>Dw&9Aef6LURMmC;|!b7Zel_@s6Ww
z+m}iaS|lARDAD=ryQZWpYxBXl1H$HmS2G4nb;7_vh|p?~jdmG$jMt<?jWvP-xHx@3
zxh?N!n{$>7+GH_!=^$#2Dk8tFgtko3{Nu-uw@F9@5Z!Wfu-+;j9wK;yPWB9Ke-1ir
ze0*H;{M(<VATA4oP9pF~-fiMG(SAX28x1sNB)CtHq*C{AzQq)9OS-$_5VxO?4@v-t
zWDQ1!(xH9;^Ls@^3U9AZDop@bzfCw`5fP$Jy}O4{NugNfItmWiP~!$_`+5lLU+5>W
zN#wCcgaVh47qL4i<tLRZ(2lO<ytqFd2*tW=c6t<epW;+?WjHiOF9VCKo0p!z_lmG&
z^Q_sH@>5ttiR__+Pne2RuelGS64KQJk!t-MH6otSn(8jE)xkljefKGm`W9*IOLf)N
z`BHUyt9f|%u$QEns3HV;*emP=61T4*6e7a+;j67=`JvVe2zL9+4)`osv?Mv*5%5^O
zyDSWKa31q|@e#_W3&E;;;MzzZc_(UI^YRP<tNAw+XFpocRM*xX&QDd9v9cLG;56DP
zL>7S3!rj=sq6Ldqr)RG=JK&Vdk&^dyKdd^Jt0pYlki63N5QJiTMq689(h!j3)O$09
zsvZ`U{(_nDzEDv?iub}HH~v@Aec*ovY&+B<&Ml6#w$91!pD@=X*x9cyE=xSzwKs~e
zR-_KVhw%5)o~j1>_z-HGU(mKnf(Z%HPrUp~9wG*<-g3<RvKI)G82zl&&A|v_0;A*`
zqJPg&J`mD5>8#;_cUtq#%2(T)JVuJQ8n@>CAZgc@zs0QC@s9%FFnh2g(WC^L2s{HN
zUcVJ(QTm@{hu+8{9u@+Y)ymTSe0rcB4)_(T2JO7?svKfEbUvR={%>vP-{SI<dmc_`
zJ4!o0LanXb*YQ(&e_}}pm(wN@+)~@!Q|c&an1i`qp>uy<70mJy0hjXA(j!Z=Lw{|C
zCaXdMgf4yNtnaVQY6^3oG*D=tfRZLU01>rm{~^({F+N8Ew<zoD>#x|b_ZuR7>R=u4
z^XkoSR9XKEQn0&phw<?YmcF&NC~XFdFNu09t8|j64)^z5IC-_GAN$vCyhc++S-C*H
zMth!zl9gJbChJhfM%o7n#?7^lO7<jh#+iXm1(10Ycci=*;(!;e6xIzX+P`c^XsPr}
zVzL@j;6tt#I~}Sxp*T1cCNz0MaY#Z$16MrdSjw?=QOs43l((u~Tb~_UzQTZVPUXSt
z<^8?yTNk6oc<?%OjDOxcm9^_X@^t^_W~vdPq1x)|4!7?(tHggiWN?8S>_NoQgX@)W
zog4@o{J4PPEqL0!<vC8vkU}zxO_d%#VHmmbfv|#J&nk1J7It&XtM3mHQlk>u&fc#3
z#zt6q5HKSi{N!`J^$l7UwfKr>ZUdiz?IDvptwyR@qNx@V(jSXgHu;Hj369z>+|FAy
z!Z#oq-6Y(RBuM9Z5rlB%mDvA8tof2yo2VKb@7Sx73-_Kn;G^5XH~XV{lWuWt_amCi
zW8)%!9v-_oeWwCa&~q5&v5G?m?>GqU)m0r2#u$+n!1@azT-(U8rrJsT#0i5YYh3-}
z1o=L)U5!3{cp9ru+<24&SszR#igt>L`SENzoKNWG;kJU|=0Q&xVp^#uD*-Q6LY;wI
zfnAXxA1U3zOr)LB=G<SgMUuCZJyDj9h3wf*YJ1`7w<?$D5t7D-Ig$|cVW@EvC#A{D
zy|*O0aE=e$TmcYP<Xy*mQwo;XU0<<&6>?63X=%t*mrU+@_*Z;sHd(Q;s-Ka!eRnW)
zaW=Ss;WxspZ%{PHPeF7Pf-VS}<1&u84ZJW~$g1zd<z-pm<?_=&tN8dJO%4}GcuGFg
zh-f)$!U^7AKRL6gg*digYP!uYwSyyHM|gQ=L*VD(k#wILawZXEdZbg*b9Rb^0wtx>
zXU`Uz1nctxX_xxJlHvYbs$~AMPF}EFk$I`}-P9)yFMi_KA&Fpej)X_(kvv2?9fNhJ
zSW<oO#~<~WU;!-sc5w>$J3*7)P91N5mb7k67Ng3*LX$o_Qqun8YIWkKNH%|1%7oCA
zUHdt?{R*KD`Bo3mftMH(5H*MI#k+4_DL0fS$ls@GICGNvj&p0~PT2et44RC1R)69;
zdDqK0$&6|(*w6w-PIJ)30cPN&_hgNxp&!z&J2z*u<(<7XDj#`~!-Pj#cX%wZRmw*|
zw`XAgWsh^1vet+Zg{w~36hw3AfyVI=d_d`hA`1LVb41wh&f>5n-j1S!3kJbM*)#0R
z<<lBi*t(8!@~}&}b1a6G3Y$*c$5^F`Jp*!JaszA8k`-@LaKyp~FOmDCG{$Bx&CX@d
z4capRpZ3ta(wZGhQiwmK<Ii6KLY1}8*>3a=-`cfO&O148RV#7J%PJCb>u7{9XO-$@
zZEYRbjGE@pUoDS7eu=SoIqx1C+t!Pz8#f30jtISz^8v}aIH)Rq{P!#kvE552a*;Xh
z89#4Q>H70mZ-4%96@Uo1Jy>V*2~P^UBiF<nHr)nLs!sEM8P*(~E&u1wB@Cn@d%E17
zMk3L-SC<qJQ4VmCugxi!sZaPcuuUrY>2~dx<_5Ay%X*qJ?5QQ3tzw>bRM)uTl!m+%
z3#zFob>v<=Xs6L~MPWn(>tyyM7Hn_(?cF+0IHVJK3cH<UrUTeh0aJa3?976YY+V7U
zV$jV+CE<Rl-9fB|l*nd_d+W192@n->A9_vB`A&oEy+Pvo{Gg<fYSu&ID~+w0-z5{4
zHQr8v*x2<D42>Ze+WJ;;uC9NTj`hv?0#PT}o3oAnCa9z|@@EqWhuG)b@v{o&UI!Li
zmQLfk8g({vJ|I~%cGZwLES(C|#TtcBt#plH493snwF_IYq%y2dL)K>Sx}nUL?{5h5
z67Lr1?{4O0&a1Lwuw`VLFP{=`YKAGNjkuvelyxyVkdTs<ot=B0;l=9qy2My2<a{8V
zb2}ABaa726)(y~ey{I>ql8<U-<CHPrqKVT#dGZd)68M=?B)cL&D9kb*>^J=lXX%E#
z@b++UT^SSbnS}rOrA!c=zfsQuGU9}JbdTViPH&%WrIX%7HzQAft)?DN;1;!8#48m?
zo#`b4D{<3cfkWh|fPK^AUy!el)PIL<U9P}X!hW`K14JibiB~wH)O1vT6?w&hth<w&
zIa!HfF5nXJnmJab(fxb+`w7RL7<H&OIPofSL6A9l4~5u($?vaDxEMZuw2{CRILOF%
zC@J-)(LJs0bjUw@JI2;5B%qL6Vd|WFpDz6t*~7GcTE}$(8y{qn!%F$szRjTj0^hIa
ze4tD0oo6=nuQn3W50<tIse_2HZEkcX2-Vwg$Sqm9p`QP@2QM3&xMc9c^RS@hPB*v&
zcr_HJ&#&Q=aew*o&5iCxu%H%xzeR<oc*F!o9}C+rB_eWtr650Xdl^G%JaO{mhjVOV
z?XBaTuL6_xkeZ1TppJ?fgg-1J>j-SRml%pbSJUF02OZgrMP=r%*63&?88l8GuDEmT
zQp1-M6Exeui)Apu+B2YWF!nowCvl16?DNPuyq$*7qL)RbvLd!xoR7K=AgbpU;W1I{
zw2v$Ug>e>Fu_-`qSDj7qo#mnBQ-X5N7ymfV*ggKpJ$2!*eJ0J>W^@MWF_Wv*OaA<3
za^7J#1-kfkxx{Uv*5s^~dZqLRaf?Q(D-n^$V1Y=&`g4Sfg?tFYUP46VY^=^&BA^rZ
z%L!fG(7sU`oh4luAe>upXT>DE2adUSIdJrx=GjL6tNt7I?Xtmz?1kx;$T4<s4%;Mn
z^`$s^&9(~{!DR6F;blFHq{M(K38&&=Cds65+vXOKf<ecGvjoYyvAeE11^(uA5X7!>
z+o_%SH7|U)2nigh1YeIJ9AB|6uH+rqZOF}P$U8<v8-g7uO-!Bhz+S1Er~JOh^?xS=
zJ3Ve*hZP0V>D)?TFNjp#3G?7=f@C3mr)FYHkaf?V-|3MiSahLFZ#RNWuj^p}0RZ3M
zB(_QF<j(=_U*lUxeuqOy&66ZmpU8z9i{m@zmK!>-V%;}i4L~Q`CRXb9`$K+hkeG^`
zlP}>p?7J6oemnl%7b)Ygbg0<r`4<&$*UA&%A%!cE^pup2MJUL7!kEOHIG;0rS+l3r
z&u8yY-XeQO)<9<T#cQ0^`}$eyg2Ar7fwKFOPrUEVvL^=Q)k_k~gr_DS{O~<R`OBbR
zHg6QE`cm(n$f+Uk?-_9}Z}m!eE#<#XiT#!);kT0Z%thx!L#f=4&rHnk-n`JgYWHTY
zgPYH4t?^r>q`v6w1@*~6?qIW}4@Tw6<?Ktd>G5^h29xhSn(M}MA~M2@LIWmAor>t9
zb5v*uc1qD)Yp!{byjIjDk9M)SnVkF0F`TgSl~|6B%?D{b)mV!pm(i~KP%PcCYs%T#
zA33OxQYz}4rwd=7l$4aU!a%)ofFIn{Nfq;n_p<I{ZfDv{814VC=f}i!h#p>}3{|h=
zQpU;P1(XAufZX+aSM68Rk3s~$Qxh2X;3V#RGL>cnFHe_*U`a`dF?a>ZOC4{CRV#Du
zvrphhkSKyPp&jwt<t<oxH}gmJq^Rn}6iRXLk-B41+nw{=T3ou<W|nkV)2TTH6#^9R
z5B1qD_ED-7O3#Si?zMHWR5f~76$hL5e1YnBG3pil(;AuTec{U=FQ9}y+N!_%X`vLS
zg8s)wkC&~ML#%d1mN5!$-b@;?p}%I%<3{b*i?H`|WQZ)=96P^5TOF2Rhr(V%h!+^Y
z{ZZF-dbQwB>vo$S!#WV%@wBF9_oN^o6kL+pJ!qoAPiD%MrP(Cs^ta^5sw>Wo0qu9o
z($^#W7T?I?;n5B%K=mg8M&Wx-J#eu#&w1$e(rfd@m-X=>@3go$aZm~A#6U>um|;92
z<#+=(?!c1HHioU28w{VhH&KWd5b=`p;l6wl#LLH*`8Bx&Z=0x?3PBd!)^8<Bhv1#H
zkGPoj2!S&l{xp=_C^l*vu<o!gU&X}|pycTW+DI(bi$-e+qO(9a==RJ<`^;FpVma^5
zx3)4)lLJDqf=)A3gj|04^KKD{dz1xtB`+;^rZ;+;fvh>-#6qdCUMtrc2A$N?4c1bz
zROs?d0eN1|HFlAIg1+p@Z1a`yhjPECVS@A|%vLGRonkSG<L26L34*o*TVr;(s0fa2
zmX8l%&Y$@2-@$R{5a6aob$21wc^rGgd;nf!*tN-H<KZE8)b?Kj)puN;KmN^UOEkW<
ze4;1LH`-O|ySr@fCQqCU5K6#z#-Oj%gzQmo6oGdW&d`hd+dbx9m1#@4@l>7|X^5NW
z$1hfgSjiywgC6tbA!+yt=&6EjoJEbK?U)kGcEcV)nX#IKva+!-q?*!7v(Z;lS|PZ9
zPYX(tLKbt-FlHXvWad0xoXv%H7Eimk8*k)qys_cK`ZpZRGxoW%Hnul@yxaQj$~wcN
zv>ti@VOxBBh`-8g#XO3NT`SgFnz$1nSU5j-j+R>hf@Pv1tq7{{I?K8^uy7?3xpFA+
z?*5Jrl#iT(HES(A<~O01CKsdv82b=^n}|r+{=S%Xik|M0M>%+@$MF`>0rE*8w-X#h
z`2;_r8E5#>7LD+SbOV?ZB;Rqau_p-UMjXhst2{!Ohl8p(W%h*+zkpe^O`gM@-J$=E
z4h(Se;MJ1Pl?Nu>GT78?Oz0q6Ht_q-KOPxgL%goj3`yg{jx_rWy(bI-h11qZr+?lc
zWE*v*k-nQ}#^}?~Z?zyxM#+rF{)Jro-nOo$W{!S9ZD$R9KVftsNh764^N7ADl^k`r
z+ou_|J2NVV{OQE(`?rVB?TqsD{p<ueZM-!4r4q4e2Fz*C|Lzrl{G_eP$H!|r*y|=b
za<Zk5Z&KaEIpVlCJSoDnh0A69`CUY#C2ciIN>(k|`-qE!D7Nq!b<oHgiABXCe)D>w
zS4%6-<AcbpeP?c~YsfN^(>sH7+X$RQ=g~$5l$8T1>~Tv*3uppK6nRE>IS&p;aCT*j
zJn_sD50iVxCgBT-r4P))*K^opa2^G}6(F=7@33pLKdIT?BB(~vtn8sylBrtu*auZ2
zdD}X_5ol|@%Iyyb!E4W{3A}A$krq)T8P&z<|7@bo;ax=llZ;3yjnIFPn*$gsjB(cW
z;pXmdOo^^&=3Lj%{>6q@oj`$1Gj^6NYnJCcEOs<L4<$o5y(QDZ@=?}&A>f5~%fqcd
zO5>IvJN~!7HMx0xpM4jKuWNh;>%j4dkA2Dc6eS*^#8h-;Iis&lAc8&H-EL{f)FB2t
z7M;!X8vcRX{VN#3d2?K)6gysh|7O{5&^_(+khfF54-Y~|Qe8Q*LZ^3ySn4=rFHPcX
z?PT@W!SjGU(tZsV6;LClf{1&d1bnc|2Azv!mo;zZT!Y^$B~EQ};PL*;OKrg`j&h%l
zj*DwCLnX{$b5%#T$}H!+V9Z?bsJ^9z#mE9%V-)(q^avAemFt`jZFTr39Cw_f|Bsfu
z-ZL-!!LtaNn7DNpAkUW$*uIAwNb|)V&gEsM=mnW*yK%?8cu_B!FZScgyHzH{p@fnn
zs-&m3%z5njQ}G3q@2a`CQq!t8Uk*vscxceO3Z}do`em;u)t*{VL>~r>#eTB{zE~+%
z`3hZh2j`7O2Lv3pwaT}8FIskn^9U){=jV&kf2pbOcPRQl+5(VsRR4gAi8e1*JAqJD
zaIOlIeh#FpP5$TvBqp%xmkOjKjG$N21e5>}M+l_sZu!BRq>};?ldt=Hk<SQp&&wjN
z?Ox5?%<}TSzCK-tS@^NqVdrg*o<jyOg3V?a=e7_X4eLE-#<pv5+?bSWcii)Ip%Y;j
zTM_<$rELEzLfR3YHFm!vE!Ent!6cYiAs;%_Y7@h<3i{SaE88$>qaj_G_oXb)>Nz%;
zC@XMwp?50GU-G>#p0vUF`Ob?yQX-KyQ;#zLE2v>aej**Hv*p}yyr%cQjBbM?)$iJJ
zzCzO61ka6_Ij0MeB3NTAoQT~?T0g+9;)@59a*1J&I+=q?idf3kV>z`_q?D^wo-U*b
zg3o+>P^ii@n*RFZ7odYsPF~j0=j`>^8skMR=(j<23?Cc%WV0)!iMYnGdOH06qI<La
z*^CsR7@w)!FViH6pW4S~Ud6NA^*ISgq$BV`Vgxb$g6(bk!D0brTM*<Pf@4Q5{)U6A
z@O7Nr+yn9=(3*cqCZM?5l5Cv9;kTOt_B50ec>t^xBW4Z7+vrl*5=zCcx?}p90s4<0
zX~x$jGUWS)q@Mu<=IWoVH8$Kq?jq8+H^FI0=wjXi+#D8Groj~Y63?*Cx~A%8ph{v$
z8bt4u+Q1+v6l9Z(FlcGVkHu9{C^}ssQRz##_^Stu-K_FtorhXMc;|Y01orjCak?Q7
z`|;OU83Njon7w6Eb3PC-ec?Vc=V#TNb?UbLK=Gd7p!81(b<QN6#tS9_&m5-7!3>Q{
zigU_|uIlB4E%R6*fuSydpYiHbn?#HN{Fwdp40|&<P~&)>Xp(ZUu%3T1l~^htA3{7|
z%FrWy#KfosVi~ru@84X%k9r(ny%6GBd*s!XJM5lv-o<=pq8)TE)5pI)#Zr?P@ZNbD
z!u#i@9`!LZlX(3h&JNhmg1Rz>rsy{z4;3UGEHzBsiHD?guBJ`>)8Q?v;81OdPDjSa
z^3Zn)#5hU%WSW4N6E1%IwLba6DZ-&VJN94|D0Q2$kCvS4)BhH+9OgbS!ixlZjfYxa
z+ksdtM3T~qe93vmt$3CA*Ykn<FVtH8>xS{au=SmD{r`<W<j0CCC>ir}0U&$-pQ}gz
zWA5nRck!3ZP55Uv50<+3Oz%S1D}b4Y#e+p1)}w962#r=x#UE8=>mbB<pCA9l@>%ug
z&ayXAsz+M>5kP~z(^mI4W$Z?yjjBG<qhEcZi7K$lo?5d-l*VYV1NVMK259t%v6Lg1
z@^3Kme+6Rt_eGM;`ZpEluZOp%<4^Kctc5;V0G5{zOyZys=n+X+lAO!kT2_-8Aspsz
z{?r^EpYnEsB5Mar^I^b4mKm`cnUn+-papoHrh$tIYVhYj&bxE=#HmPBzOtsJ5|x|u
zr4n9Ys%4VOH`-kufX*f<GOqzW(R!p~xq4m|-t$rg6dFigG>R9%*<_4)8>A08Lr<Kv
zOA+u$#+?w~HfgPkQsvMt^(p~e&Ph9c;q_1ie}Q2o9ut$0)9_`J=xloupwJkVGGE9Y
z#w%f4F`67`o>!R<(+C8HPu;ogrg}6dHAkkn_8+Q6zvem~o<rAHklu>JG}7M1A@hNk
zfa<1M{NUukgA|#<UZ5Y+w+CkG4Ab6B;8Mbf{pVN`l&0st8--0!qa<yb(pR2dPC1aU
z!MXJZ65_0OIDlVGbp0T8%7UO<4vhuaej!oq&4Dy3m0Ii_{K0MPcEDZOM7+fZ^9JWq
zt5I_vKPNBU%Bbl}IYm!eX=@DRIzlULzRn>mLE{86!HfiQ-(AonhZzdDK!EY^fRIPg
zeiMyOJ+3GzM%i7l^PbdInx9pSqCX_7xoD!^dy>qH@$m^Z^vo!hmUSAq?Nhh#O4mS(
z;5X+EG+|1G+CV&H4lo|cNhc|dmI&3vM9;&63WC<>)wF;z6vCQgb_Z{3qI$T2F%3o!
zRk#_(STGb)i{kv{{)R`{t%e+a4En{(KZD7YFxz-()`dt}@>`FRraz_ptd{Y(zdb0c
zyiic1=zsubN3q06@&Y&>hVu;19VUsoKR@?PB7~HE+j%MB-0LfZ1@;VkgmD)6WDX6Q
z0U*f;llIb^zqO@kiuZjCSYw!Z`NI2$`;sa>$Y2f9iMc~mRCdo{RB=G4jxV?=jd35b
zTWnkUHv$FvnNR(c604i$`j-E|7neIWTFNUxRp2zy{^#J)AG>Siqy(%uTrl4pC;{?$
zy0w+jMt*OE|K9w8H(%t~{#8R0JyP>saVNLnukrIY?eKqcoa989)nG)i?oY)RMHWTD
zqe+^6ecmR%H<N^*0PcN+h9edCfU+vEByp#x0}ti+ieSA<M@VBP>IR71MMEFosOs@h
z4%F?&Oq6su*r3N@&WtaLVru@{1r@yiAgKPgi;}UkB=IC#5DQ+RbA$9i`oLYbR|JoD
z{+p7(muh7>LW#{DkG1t9UAdfZ`u&4{qh|l>@Jw!<Lb-;G&qpLxQso3m7z)wSYy%dp
z-1`+F{Zs@UPLlx{q-X79@E4ThyOPYx{YlSLd_*S-DHIazuimWmCA~E8Lh^x{R$9BT
zg7^1pzxE2ThByFMW6ptPYBTXXp~YkH7nB>BL6{NY8R0!S&S4iuf^*S$#2Vs@jJ97Y
zL1{i1WFiS=g7DIWAmfPsp_nM~;>8Oe*c<C{WV10V1ZQQT(-R6sDgQZ?+r4N>7HLbd
zjMC3?EFcq_$mN2;pBLUb@18oUx0xcwGqkWm3^P==pbc0N`bAC~G3i0so26p6uS4`e
z)3!M7NAuSm3n-cX*V$FF!6nLcsz*C9TH%%opvMb99F6ei`)UEJcA>Y;_oX3Hh}dXX
z&G{A8{cqKN0?58sFE<^i+uFJ~Qn!9}FBxyb+K6iOKmblpq%L8}!Lc?ZZnmLLcM(mA
zM7{5-R%^&1P{Fq`-oMcMGp(z9-ZqcC&<hRvS&qoBQ^gvds}x~`FhUhar2f6LLvJ{b
z0xtfb%wq_^mdGKb8cc`%X_H7BI*oVODN~XUi;p_RP%dv0)r`9h9Z@vgbLNxBL=GWW
z(F2#S+cFZg>NE;}DKZC!ND3<XmfTPLe9-b&E@6*$s(8-xB;-u7#b=S@2n3f^2|G_Z
z<k8#KUc5D;NpvU&^B;^r9c_+TNr}A7za-o5!7h`dq(4lUwXh-uP@#thVVIZLNaZjj
zi8H#(i8?B13ASGT1BMt>MX)YEy!_z^y5j`1Nj7bQxh(-pbsz2F?yS)Wr<~7zqHt?3
zid5lvO<>yv>rkHs8U$&Aa$JY&4Go3$-k!*-ZK8(LN6A0~l0}Jl^+oie$P`@p*gE1Y
zvK?&6$@&qUUYFHCehia=2_hmk=^w8{&k-eSrj)n*KtO^;&CA(|AC?Mf{dyF;Hx8VC
z{)-b&jF3%+u`!=jcTy@AB1Sg<3YN>b?BTuy7&QR&OAMSst!SxY1QZy;!iUH}tlrpe
zC+{tzR6NBwO2@9iBdXQ4b>NdYz*iRy?eV}yu<iLU50eJ|(o_M^Y<ic9H;2KYbs&Tc
zNFc<U95{N8?1M$u>%BuUzv7LZ@)1uhz}CH8fLwuA;`t+FqD%rbaKt!CaPnlY7!%gZ
z^!+-Gt4k>6*L)?Yds*X#Y$;f;Q;9b8SW|*$;Q`INzr{Ne&y&M`3-EuQ5l8;vzm70}
zxJQ1EN&h>9=xs_GP{Gl|b0$LU?<a#-^XMl%^Vh*0upSz~qq=guMmt`^X;!{s_x&tN
zMb78W3%u1G$dHYz*Wj{zQB`D!nGv3|kn5A{6-pA(+Hw<wm->r1hfE_i!J46m2;W+U
z1*d)Nu=j%CEPSpZan8ZdvTyb|<LEelnn}~Z)cS=Kc_cMT&mLkWAyx$8ok+o39S52k
zqNLXB`tHYr^}B`pdV6mv1Wg;dKXvt+vsn<?O=?4mcQ|?Q^Rpqq7Lm?Rq6`q8a&uhF
z1*GVnOWz;SjENE*o;fg@twM=(vqP|}*iJ?;V4g~Vj}I~K5tYQphm24;strx@zz1yA
z1J9?Rm#!u!&Bq6mg$gmHr~wKtrna=9Bk<IARj~Ho1(6G1gUN6X=t~42NqirBA=r?)
z0U{y;*m%4FhGab#9F)izJBi!@5rxFry=dxxjD|ptyNtMw03etfa?!g_0xr?FyrdD_
z+#}qmrezN^>g>eit*Hr_gly%+gC~dmZp}^5B(|u9tRcb?oS%aZ{Xt^7ZREC`vfBkK
z;RW6crmdaMo}(M%)Ra<r#cHZ6%EDxgLB(+<wh_n;NO@C+I_$?`RXI*JLwnWkF}Pa@
z5+Y3mWLO)~?kDyQfV=XMfVxWjYCYf=$st$Y*LgSn)3Z<`p#Rd^hWo^vf&N}1{g7*L
zuFGm!OY#D`d1c)6EDa)n5c47MwquCK;cV0x{u>WnS0d4#%sGI6P7!X^3KQ0kIZx~r
zAJ2_?1)w(|AI=y-VN5DCf+3D`LFCp}M@%Wr!suvgKUZ4%ys$CNNj%v}JgwVZc7)&Q
zyh5s3Q>y#*%N-v2J{ij?$p@EEj-Q8M`c_25*QXy6vL~xMg4xua=<{}FaxkK%wpN-}
z)K=g->M=WA?o~$Ice9i6>|Xd4CdeYX2@)n-rh)pgu3KR;BgfxuxPWhzNg_IL-%Obk
zwdy?m`NK;`S#lBhs~7~sO2?~9xyvvlqpHKQ@}!y@84^d@Lwvhu(fBFtqD~CS#8&9x
z?=l)ne`sq6RO~U46k+Xn8!EmTiZ9EWK_|2>G+^LS<9S08K3VYCrZUGHa@?GvyFhZU
zPQ-#<o|>$92%y=;oT`4VP-7nIUpBYX5(8WkcaqN7>NdIkU}2N6lF3?LmR+Ydb|2I&
z8hAUMIw>{K<8D=+9{V8cnx|o!)4)QC7y$&wFWYyQmD;iybe6Rz^U88@90NSTBBBVM
zopge+ypXJkOokWeP`h8h*h*Y#vM;T3Nhzm*5pjDm*krI5!Xn)VrmR9TO$vh*Es~p&
z>DvL(8OOzln%hnsRuaCI24w2ZL&%=Y+3b(aZ@Xds)q{Pxbb>T%2+y`ICA(J4-=#`T
zrU1j~sykS?-9j>#1Cq9NLcBk6&DJ&;b$fO>jU2wrYJz}FBj00fCp3p)S+|M<w+`9|
z&-A63DoCAvn^6&xu-t=st-4_}p<=KvZYF~ZASi0JDPjZ}8wXu5NLgtjA;7~bV$Rl$
zMjLU;mwNs&?A&m!(?By!BHB2yXAKP_Ft_SF@m^2VrctHnT_j8>%<aS=##}E5E@HLx
zAX(>4A_ABlHUaU)G@d$RLEgpyKNX68=ChbJm8)uvQB&GV=V2h%Plb9G1I|A>DyleW
zr}iB|Y{O}4xW`0;V@)!0Vvfarm`W2dSY|&BWIE>(YHLm5u6^%0*Jo%(Mx`Y=4Glz1
z!&v5+iYf>*NIm0I$-tXv1{~<&IWU|~+s{O>rz85w;t$5Zc=0yXrn#}LPs_j{ycEHg
z&1}Z@B2$2@04k=X$+pfih(PzZ+4FE$iwWZQW?~S)Loa>!Z<0{fn~-rL(&+AR7v`7|
zrPM8x$5*!azsEMD&9c#bb;ni~7MK1q*}?ZX5rJ5@j_wH@w<LfsN%{an+l2&HSJx?H
zhU_lB#$Aq&%F<#ZB(H@8Sg%ienxPvr>7ddjPN~>~oPYc@Tw=xSHy$uQGy8La|3{cI
z){WKY6}hfnf$O9nw`r3Mg{p6ap)h<?qd$q=U>0F5dBFOG*aF*J&$&btdV3nnF_3LM
z=ZZg>VCk2t3-buIl>C)r&1Hy9-IL{DA`0xS1|Eut=t2++v=+FoPuPej*5sy_i>F66
zjJnR<C!X~u9%nQ)&0t0>Nz+ALtzf;3C%d+$#-wP2{V`si2FUCU=AkD)%NBgV)UHjP
zX5&om@%6ESGpeDhG&Iyo69-=neP{?w*0yfCN#kjIBKR%oKW9_~B`kNKTF>3S;8**;
zDD^h&Lz7q`GxLc(AH_Tb<~#4lGh|K^Yz5e1q;_ETqpGcftB~0;sZ-N^MpJ(+QSAAf
zWyl~5^h^K{MLiF<u*?v=RC7j?jEWIxc51xvyfckxj73MbhE<zq9a_2X>m@7JeaS<n
zGFjSlL0Bux_4gW;@mfu$`OunF4iSX~9Tug21GEHV6*+($0CLja6MPTA$L<Z6jFNVP
zWpH>}(<Rj;W6yqLVPE*vr8DDU${?PC*-5fdbmeYLka;T_9=)d^9|46XZ!=C-F)N(|
zlK8E%Bjv&mi<U|^q!KgQ*i$l}2tD{A9z)vpqfwIA9xgBS(~lVyyIb}!pupA8sixeT
z(zB8z@k&AilWC+#@5sEHf=d`xrYIl-;JlI*5fRUw9XT5-u@Cv*baL>EJH9!G`})-2
z%_#QZx=Qz&y5VJ>s4pJuiwSdb=pBO~DqZ&qYS-)pm@NU1Qd-+0;i&`Bg_!`9p^WFK
z7^Wk8@T@O5GClw^NYm;thUp%ih2v6TMe8Di)kxj3myuj$M4CR3Sds4EdNH&$To6Vv
ztf_gxZdp`P7Yhvzqq$Fjt(9RJAd>^8;}tw+lpws0#Mk_CBygk%Zr(@5o}r-{e%m|r
z%x(IO!fbUf-<#ILInkVJTH-x$027bbl#(%xvlH-*(9UOpjf2C9xNy!kr(|}otb2wQ
zio;4uXk)T`xAQ)km{9EDHgL(;nPJn_#TncY;yHBMzBX1jn67|?OLL#?n*(lj2xMX=
zL<HRv-s46iL$1Nr7KEZe#!7{$R#WD7JWh=D!dD@g)7#ShrlCpmli74cGDYG1B|&<y
zfQXV>-;m3YjYsY1cgj3WYEt03735Lu6OO3IipEp~%>NMZhz(gT{CD?i>G1R)?v(^L
zx)A!D44j@w&F%;M8ex!OP`pe=kA76ahsMlX+FoyDBbQ66!@_ujS^wnwx6Kc0tDAqe
z&q5c{sFzVhNLdK8U|B=s{)lD!_(BA-?mjYlH!At~)u1IM(u3@%6SL?kxdo4PS8hD|
zI{R#ruN!i`xkrX9qxd{?>rU3S>y~<PS_zL|7)xcZc0rGySgLJ$swn_Ys8t$B1XE^I
ze0%_Aqk_FcL8G%_a=d6jyCfNqkqTy^0L8&Nz%?=ojm&-NPg%-S1Uk0v88s^q)n->3
znI}4H`5`2}wrRXUr{Ljou6w=iS#9n+Ynl9%Z9*hIt|7A^X%2c|WuMDezm!+~>aqhD
zd&L}P`C8S46kb~w{ggf#r=MknVh5)dW?|F8nJhSzLm7S8zs0sP6Rp%P;wK_SBM%RK
zy3DB!IK`K=B7v^tgRJS+Y%+!VJT+>tdXEhI2DflHJsd2W+1;uaMrGHo-*TXec@(a(
zRB-ITtQN9<&a-^$DBIF85s~->l(HRVw6ODhonFJ`DKETI&#qvta&#;N{%OX9Tz*V+
zAIb%Hc-Y{jg6A@2>O9lG2jGeI{V6h%%=q<{#cHW@WHN5<DJJHkS)|g@nj{8JrKN|8
zFG+8`vV4S|HrD#A&NAKcQ@E6(z$lqX6^AlhdOYk(Z5f+z5O26`aU`A5{o-q&YkX@T
zAODVy>D12M&wd(BI38h*c(RC6un8R7vH^52SrdDPh=>9rS6Xje9CCF58Gy|bzc`SL
zHN16_hIl$kBNc&FV{+8C_PZZow|GDghr#h7!22*d{<_}$va`ep3=Q#fa^4OO$aYeT
zi~t>>k@>KsUkb*Lk#<s|m0{NIeR(%WB_%Sc=NV>O8|A8Ouv2;;CM#p8ixniJf5`w>
zGD!?ZX0#r|<zM;CmE9Z9+L7x4D{TQd?+~wVP6wHgvXso;+<@_b`dzOru2lz3dzDJn
zb);=YC4mJe4F~yPd`A_ry?t>J9AgT(pA!d%jY_Z8iB@BEq{KVicXpWRCbZO14%m^p
z@Jbe(jiWTuS`{6@>_%tD>_nJTi*`lG1Su5IF^wXhOsOrz0wE4}3QtL+4tji>_5c(}
zTK`%(kI5{$A88OHtT7mVJRHu1a*VVUCn0>Lp95O03U4TkJ(`^)d>|=46w6?NZ_A?3
z*?6)Aotk(6kj+!eNuWc2s{7O%VSiF3!Bl{LFHu67kvY#kbwc+YiUggKGg<%^=Gg!v
z<ZelL6Uh#vwxA-($DFk<!Hi3yK$4!M*QA)R0zizBBifmp`|-L!8)<09MIzZv(@Wq;
zZbJ*285uWD44zkNb>C+UxnA2(Pt7j8_Aubutd(ftALsLUsI>LF_(yJltOMgikAuka
z<g(h)V|*%kD~%549&<J(VJY>g6=)XH$=L(wRarCxFMpWV@zbZk(zj#Si_Mu<MO+@J
z6M>K>;;AL&OS_t_>#0ynKOb_+6h5~+Ml1*b|73b<N;EqzPHdD-j`HtKk7YY-B?`iq
z$7{YAwJx#LdpoW5wmkJ<Zu|o8HZXUT#SRc<I?^Q2ye2J3*!AUxQwPDJHD^kJV$rH-
zoFD`dp*+Y+Ib)(=%l3`t+sV{Evf0uJ3WvAp(#XC!3S|M6{5Jx;NCP5n!5G}X9kJPx
z@l+%yjKQ;Vs2vg$nKx+M`R*Q(mN*z6o|H<=*djv;+sT-GGVO-6O`jddR%*MWDt?pD
zju_)e>O<1?4jb#`*&?r56;g4NDmyu=h4eh!*{n+1)5(DD!vKPW{2(<n8SX=fMszaO
z4zRK^Q_I+>7mDDM-JpohMIt;A8Ekft_yUKWye6IOaBA{F#I7d;kH57Po6s{qwy`Wt
zEZ%NoxA8_r+Oj$763VTu!D~k=-+{HUMzh@PMlxp1$C>fb2&}9Gs4j|rsO&#G-0t4r
okwgFV+y3-l82JCl9#ylKzRH$6ePYvbj^J=aSzRga;JGXR3l4d2tN;K2

diff --git a/reference/plot-22.png b/reference/plot-22.png
deleted file mode 100644
index 757b104c2d0eaa75e4a4c6788f0efd5f3df499fd..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 28723
zcmeFa2UwKn)-{YFYN8@(Rs@c5Y@lF7KtMo4)C>yPDAGowfYecX2V>MED(E1hR6Ej(
zbcUiB5s)HP+JH(QVCXQ^Z`}_jG3UJR`=9SR=l`$wzpj%j=Rk0nd7kIK@4eSvYwhhX
z2Q>EzEm*lgKtMo<{{44{1O(<f3kV4Q@X0*<r2fX8K75(?!@j-W;Vb^(t7UrO%crNm
zKWZl+AbOGfZ;o$kzlngr-vsF2?K<Kd+}-GQ^M*r+YTu}g%$8MO&0GESnf>`4SDnAx
z_Uih!;@I!c6w7T^y7kasC)0fL(dn8)ao_G!O4GfOaqDel(OZ6bmeh`Qzn*gSH?auJ
z6cc#$=C-%Y%N;qYm-Bb6xOvvhJ?&x*H;&gZIy%h>+so%%=CyN2rTmLpZK=!K5S*dE
zjQTG3!smS-k?+2`?oIvV_U(_^)St{<+dP+ivth#=Y4UaFu1}rkkZ%^K=Po2)4;=W&
zn0)pA@r!@@Eo3V*1TvTfthMIW)-et%hT?afoA-d(luc)~wY4QBCs(C?{d!1f#a7zz
z@GzaV`_j^?w3V-i{M8>U5>?zrPqe6t++@s7d%on!!?-(lR%>Wz&{;JMFXf^eo1@O`
zj_<UzvU+*Lz~&b1Xtuk{4QYn)yI+r0$e(V`9pW{q(i81F%af9lD$~Apt&2aNCpV@q
z`LEy6zu}?$8{Z=GPW4=?=-)0aGi!T!D}82qqN1T8_|?6=?RnzUrxz=B+^(#tVU{`j
zs*Il!bM3e@$uITNw6n9DCoC46a3a60FHg8sDzm%1GkLVZPD?7S`9}vIFOEyQNqC53
zm!lC`Z2_ishxcfIzIbVB&QwoQYO1b}^2qBGF5L<FwiO!ktqpdi8YfPi2*@|cus7G-
z{^qAtQLoMi0bAWhx?hGWE4|D2Exo?+aP>;DfTC6`c+8+!t8IGGF!k2a`;H58Op0!p
z6kgwG&SaK;lfTOEgs-yPl`B_{#vj{NqR<|^W4Q8gdh?G9x8gNYzDFhew;q4`)BfP?
z-`v}CMg8#M!(Cln<>lopm%%oZ@uALJN=`Ph+2dWSye9iYeC!H0hPV8@#^=%V{Ex}!
zbsRZ9+MM;~f=GFF^$kJc9b481pZ<1ym>X~?;`--n_O!4EOZItqc;HF;onN@j(dFgl
z)P}8Qp)RJ+eZ|A>s9k)RovrfV&_-O{tu4pD%^nQl(de=1ZGjH4l`&D0-Ksj|LvHU}
zaO=ApMXfIbWhLphcuLtlliqU#B_vPr(%VBL8=ZWWM;<kJ)h3!~i%<vfnA%rcn~m7c
zXN<k3hG_*auC1x5Gi&k~)nM)D%N5klWaejiW<E}xJ(F4+TEy7WIg_@mldt2CTYA(U
zWm#nnSJPSHVV)TlqaKs}`F`xrB}REoITfC%jgAI3_|qL&TyqPH!ilDt=~dVTwGK5#
zhb8<Yb%e<U$Gb?L%+176p)oI{Sw2?uwl+7X8yFab?>eoUoRoADkCo1<&{VPx@d$J6
zicoiZ^Yr}DwA06X)86@97FCj~c&M?jDchsQW$Nv3mj<u-y5FxDT*h!eGlHFAb42SE
zPWS!lwL9M|5|N9>*E`#`ZM!L7rh%t8U8>UYaE{=-leyk=lBEO}n3Ub#l`M5u!*I1A
zr@FeD#^tQLbK-VAhm(D94%5xejqHbQ^73Zo_vy7R?IAUFO#C9hOE)L*%)93gA|f_o
zaf$+@OV0_b+zgxQHsj+ETpcV`RXzN0|52W@Zj65NlVJqEc$<7tC&QS1-eOKw+Pc~f
zI|eTl?@DejR=RU_4pS{p#3YV$(aE~5u1=ySBSWakBF#|B3{Syj?9D|P?7m><x4(pp
zh|1eE)cXjn+P-DJxclinzG5MVwUlM;+Flk+@&{#pc=oeqeneDMksD_)RI{|ErY5eR
z-w}e{Dq-L{HY#LP(O<M<a9)P#FneP_ktLfcpqF_l;s~F&aB$x1N`A=lj5wV^(-TX>
zmu_tq-Qj$ib#i)=A2M<@;fLKU`KQf7Q}d5V1=u~AyR3!d#w#A{e>WJz^sqcPS|+Ms
zd-C}ekpLWsRvcP}S+3HlvfB-(^Mpb|8Hp$9fk8p?j~4%W0*~6iXU7dGqb_OFjvrU6
zT{sfT5A}=p#3NL@xL9N0xpd;CWzw-!kUe$kdz>_ue99Sv6sx0GBtPm1Zu;fUgoP|q
z&NMSgh*77n7B!>R@*u#tYuk`M4le8DMN!Asx80%_JMl8Ov6n^UrLY9GkNk7#TP-Rd
z(F!Ev*og<<k=-Xin=9lT6c{KWDZ+4mb3S>59AVbUoT(A+$e3teqw;8j%;9j&#*Tob
zwFNSzJN#(=%t<OO$?d^5!O5ljty@cv%^Imo4PoH%*ktqDXa#fTuPTd4K>90DotbRF
zey{ZyZ5+(vdBjP)>BRy`-mx>2HWwrp(Bbp#JT~37^5EVFM<iNHlz1;MuaF@zg?2RQ
z)WL_EA%WUaIwlQiXDl1Jjw9W*w|2PLKGX<GdEQ_3rnk|lDTD;l=*J~pwrQCs3s{D(
zoe%i+P17mQi<)#|4R0yfwY3aYXlb#!lWUB*?q^q$&)liIWMHjjfrzyJ#Pn!RSj6FQ
z)9060-1_Fk^OfzX^){KoXP&RH%u311Qnt3XKALP%wK6}<YntDHo1n8)r~7>s+HRgu
z%$nJ~%;>sxYGcZPUAkdO^wGy#Evut7<H+HVLA)IHlp`PJ9lOCQH;8s3->29l!05&;
z88g$8U<H>}f8DLFLmjIeUmsU!xvC~-VR_y*y`ghqgH9Ol5e`$%WUI7{SxNBA5GDE2
z9fSMH%^*sitTA#be)h|kGNvzYURQr06#FUpjNA7P@nh&~gV<~L-MF8eoLp^EmN!pm
z<q?-#$L#C|6?0AtaJHO$akVhO?b6cCk!NJ%J0<<qS18(_Vx6pt)IN&ncSZ6?4ThSU
zTI+*?xSVO%8=DPu-u?Re01}Ztby8pLUiQaH)zB3Zk+V7%x4Bzv7P|x~{TngMzcD@k
z3&`m|SMiTPl%Sn54`dE=E30UNCNEr&4{t|yJtP%isVDT6no(m=lYvdbZ%bvPKmIj5
zJDbBkE7dXPq@kvE&PC9we8;`p6Tfa}e$1YGgu%#pE`4K2^TDBr^4{emABAwYGZEtt
z=c(Vm`BFqL$H%Ltva0HqeD1npnFSefKM(G)II<JTMa$CCGN$PEMWL13Hh4^oXs|XJ
zWo*MCm5u#G&O`0OB02yB>wbEA`tz#XhjB+FOghT$&{-)dDgS_DZ``=C$si^A`I0&I
zCA?>49?607c^8E0n=LmB{`+zGe|FP9?xz18hx6Zq;0JtaD2N*`cYd+6IMA>AM17Mc
zFjMPm0O}k0Vs4w{8|@F5Af0ZLlXG-oKiC)0_S2V|&h`BP#;zM5o0XPy*FKi<oE+PK
zYwLrG+bMOjRFbuS7(a)XXi`MrH?V*E%@aNa3Hf3{&!0cvq!<6U_Fy~BP=G?i(@*B!
z+G?U6kiW@aoq*$Yn!z^@-rGY!nS+Caxw&}(+huU8$M_J*Cb|p{2OrhxF{k_MsQgs&
zm7_36#3pSu&XEEzuiev6*WYL4cy2_B_~wW2kdoObV~~mz2<{BXZ?G-eY}a<n%ye?B
zpX87(E*EgN{PpXbpMJTnMX8Ce$!ZC-j0K3TS32}1`G}on3s&^~BPh3h_W52Kz4z~M
zdC+$n<acu0qOQ*25llPIjayIO<FhKgT4ka)lgVm1J74Tkorh9y)I_z1y`?tw5%GM9
zlerTEMbhi(7lcHrJBvzEIIdh;L4}s8QSyY!^w>K(=abVDqr4n?CACym*$eorWURwF
z7nv<aK~;0C_U+pT(5Wb7lo|Qp!P<&&fQDelI&)?Hih<T5J6xutV#%lLgRhQek7gOg
zhWuc0%Gu6!F%T#!stZ*+zp1Rs;%Bv@9zKdF=8UIOEYk5KP^I~3XGK1Hj+}Yu!7`8C
z!Sd%9IRoGOu|0x*@=?4H-qX_~vZOBss~c=rLPvGDP9yMoh!ZD$ucZYRE+&;z*SUSK
zxW|wpPmNcR%52gP5IFx%;nT*)<G&^Dtf~4-n0$EEU|Y0tZJKLoeX_ZwWmJ+<Y)4nD
zZF+ELMR@z0i{d3eefp*C+h5jKvuD@-{%uL;`KU%|w^Y46!##AO^{%yCX?9T4<zczV
zwn`u6xRHJ7&NXP#f9F-u6%m;7`iD!CZSpytLz5`$6P^3t#=3vJz^Xa-EIvdFI8~r?
zjHP<4hX4@%N5ytuR3i{;dtSu#6k#u07=CUwu4rRMFkWj>Kuy}$-X-r5_rr(?bHIv|
zxk~mOWwe5Q(pi<A#wiPzZu*|p9iO*Y#NHQ}&&a9K9|Z>~TDgV}2E~CZSFYR&G^AN7
zeKhfY;@Xe1bn}&+>g(#N_2w*&Hp+K-3>ssl&PTtM^<U!Jrdx4#$|J8lESb$K53f#q
zyiHbCdxYlk=KQkiI<ww}bUJHlq&}|xozo&Gly0T_LlhmdqP*QQW*=E9HkttnVe|6)
zv;u(2iIhP}o#+?E@=d`sg=TNT3|U8@L4JE!SY~gAWF{}$4f}=XV1DUhXfy@Zj#=Y0
ztSdOUSJWT)8K+AIVD5YDKGw-scX!`Xbg<Bg(Wh?l+0UOdGyg}>TRoj6a(MQPIW4gj
z=CoW12=thlXezCYI4pDG`Q?{4zd3Pbu--b=sAT)Q%gM&5=0cGTQ+p!%CkIR0doy{O
ztjT^~ua~I(D>|p7Gdk{CW%i_KN6o^@%mpI2($@(m>}pK{Oi-Bvy=oc*yIvMkRz#7Z
zw;0&fCMMe>$~K){5}@zkeyvn#_CW63`@q%=kTm-5A?X>$Um)o+^S_6r@|>k?Y^s&I
z9|vV$xNrfhvzAVZ^VxKra7R<ACkbajx%ciRsmLzpHld(%7NtS4E9I<S&tn+8R-Pr9
z<edKR>9YT%>HocHdi^|WP*gM1yqUIFd&P_I`Kknh8(Oz;nbdyPB5{=+^~@X(NANb?
zDfKqBo)bNXb=R8_vRI%aHgYbz)mDL!a8l_lRE2mn-PR-^mRq07RGFwPZM*NMS%^YD
z1i?CnMSe%XHTjCMA~`Im1~+5B9VT?j<lALAdX{NQ$8x<FE#J~?oo}G8Z(Fc7;AEJ`
zNI}P|dv=4x^5I!o9w&+@inn7}&Ur2qu>c`F&`(Q6p;UR)uyqXiG+4`PyrR}?X6i8l
zjoDb=n-$wnH?ORbVr1_?uzT^iR3!+RPFmO4#bkOYT(z|`Jgg*NWhBAsaj{(b*xO5k
z3BDdLf8``Uii?YL_>?WEI9<y;bt>NfQ~O={E{MTlB-|k<zeED^$+?6a@d1*42<oCp
z#P-du`J&emos{TXQ5jCw=S&$8W)Y9l{oDdlddDJsxsPTx72n-`iA571%bjnF$7@~Y
zoORT|Am}Q{vtw&kIW?R~mYU<*psj19UDf|mL~edns66i{A>$`X01Tkt#Q1N7y>1G7
zbf}4lZeJ`Sw}~-55MWnMUr&|#E6jqgp+dK+yqQ&nBugT-e!wo_`X4TBmS|eJ`2%c!
zZoY7Av?J5TCiZ>x3Pw$R;Q6u_`vRn52u%_d7gu_F*EtqDp|Qa-E(_JGfMUDT4;|V1
z#X;a#too8gR`qePF%&v(pNq^s|E-ack;>2=;npYf#a|+)Xjdh;T^bp2Jl<q_``z2|
zloYv{k~<4IicD_@?wFflx}hBN_QIDUx4(a1#mH@MRvv3U7p3>S9(=i2&U~_Mc3yL_
zoQ=kRu=f_{Qct~JUaB%-Wy9`&{xVEel`sMMU=cu)=PZ3deof#johm^Nl41J^@d-T^
zRpWzONP5n1njT5zzcf>M{5%2Fee1QoVg;&`d8$V;SCL=yHje5t*hY9>tixWy2zF!d
zSOy!UoqkjQw2^icobX7K*G#m2a%5dA`3->~eQjZ9Cay$%;+*(o^U8M5;V8yhIzgs*
zmBhXqn+np-ykpS@+e=HaZ}od-6%5yy$oXu$_L7*Srm0@1xHG-Dfiv?KuevEyvV9H2
zgB9CO9auZ0!Lp1^fB4}ig?Wy5s%Iu7B$RvdTLT`SdUe-*dN522g<@)>d3cCK5B7Q7
zhrgg3YnWdB&TERx@S}<IU+k^k&WU5>s82AxCJd5`C{Wyy94XdkgQMU-LKfN_H#kLS
zfxL-+m18KSTX$}rP^6;*b+=jMxq`JxE%yZIk>~;OwUwUOlTsi1YQ|sJI3h_H9KA&V
zMMRZZYAZJ@gV>Sik@oDr<T6rgX4jKkV+uf^@aCs4%gf5*_smpDy7{&Eg-xFP<62f$
zH$8A8knK(0MKf@-E-lygCgND@mV~&p-C}aRs*jqjbvYr;Dezmb`{%MrI!}!B1YH#O
z+|F}rn&7bMMn;~4K}~GynMSwX^!0;n#rs&n;G8Vkr-jI^csoRmFF<-<7Bkf<?NtC;
zHH_>RP-;B=yn4^6@m8<tb}ts~%k`Swzd$a-Kdv!ipE~>9M-kle;u@FijESMj9_kUg
zW<@ezQ7_6>y4NM?T&l>8hB|<Kpfb@dmPy6UV|Kd3-lGJQqUL7BUQ8CTrLH8mtz;?n
zjlkTh6VgYyK2=dp6DZx^sS9{(Ji(&saKE^{*U0|WZf)@j_MKGjFD`-tS-XmL`{d;1
zb?eCOTIf;mVrPO?<pNH4sH;S)t;>}YQH&cmXkPpx#v$F*kwxiX-=wpK4tb8{iR<`)
z&pa-2crm%;TvL;KA91kUM^be4f(TGHc<0e7$*8?KyH({RWt;A<c3JaDHi537*z^Oa
zaC~67Nb--PI6*G;OND@O4C?CZ{b(dB0IJuxT1Oly1eLKB$kQsl^@hk{>LKKsn%_%3
z<s3CWKCYgdiDY2?l1PsN0?~PU=CUmZ@l<u(zNdateYa~4uiAiV)U%p;#(I}D-mbNz
z{>VP;cd;)rdv~4%>a6s^5+ysL-w;fGNjm!`@Dqm{>ON2qDlrSabUUwkkN&bg^abKm
zFbMsG^%tlR^Ec(s`lOB~(%gEIb&c_Vg$Y@s93?kimeWcRkth!(@`2vPWZ#F)Ot$-<
zb^=U$!Sfu=yuAFIyLGA5rM>Rfo?2@;!w;J|YR+(Kba+ZhP2OUMu;BQ}M@jiUq6R&0
zsMn5V;^eQ{>l2;C<#MBxcThLvMYaZt(TJn~5QQJU602$Pzyp9&+YN#({+W)&V?j<1
zI{MgjEnQQw^FDj1d|y>IMY0IcS4Ju~ll$*~m)|~Lb!-<Qd`W@Df930WU2U-f4S`=X
z;)s6xumLm6A$i|yX7wi#0Qw4%H7D7&t%dGAmJIH|Fe<USwbW$~scSdgrM~la+Eb<5
zH{I)$6E$*`x}@mcBr1&5hS#Z4Zx_Eb<rMV?|A>AfyWILh?aS$?UFw^;BrNGBP!zH^
zCkpq~8o&sJR3emZJgd=`$s)QsAx)BqTQP{z&IMIPBEnu_0(R_uNUv2=)SchfjdIPM
zXg(+G7}50f{K7&1IQO-nc8whI=nT*Nw#59@BiG+!Sg~=Uocmaxp<S`8!cdu--Efs|
z`-?Tc`_m0$wP#;*{_taOdAj<8LsIO%r?dhHXDksWJK@G`%>PPVLy$wOJ(Djg*;6E)
z!*)DR-R)QViv>A86TKuUg0Le$yRBy-)`tk3Se`O}R`IHljn3Iu+LgxZou1$|^>CL-
z8waBZXv2SLp;jfri?G*N9!aW)7>uT#{=!z<rFhNyF!!y(Dr3h`VPtOtu0c>UBXv;B
zU>%8X%TQY?P6Rt~Txt<9GK?rGF4q7N7$}c-C=BwnAt1j3zr{~7UnwQ??e7YxY%d>_
zN-Kaxk%oNZb-Pk!c@UC~EA2RJ7ln!L$2Izc<JV`U&QwZY+bMfQxw$7V{6OOGUVxf7
zhQ5J8rbi>Gg@<14bxNQ4=P8!dE`Y95{AMCAd%`rp$dND%A-IfGgA>8Fg&P&#{=6o`
zfr^hiovPf0IT;xl`-9{*lLXz>m0?^WB;Rlq71%20=Cf2<L3RU=kHh(Odx(=ULJ=YT
zHyNhI8w?Z4LSgWwj9uT8`R$E7Pp5f{mX|g9R>qW48$13^WcH7vtbI~Pd2(|}4&Hk5
z`;;cniSEQADN>22_8yc9kbxvYBb*V+RKa<RUVOfH)wLj-jEzKap6(Z)NicXvWxua>
z;Y^fzjy$ds8?Yt|NOXSsR@;^r$s@Jc(NwSk@$%WM_aN&KUhec1yD6BqLe5H$gwQ%O
zZcM5sYsb`31y9c?su5c`eYo0S5XhMjQ9Q4i3A>Iv7frFvHE^PN!<#p6CYrgDa?U#+
zC7ZJ9&%fH2AEN>qxS})6p;<#TECCd6%<;7X+_n{S8_hU7M{mjh&Y%QH83nc!f97O8
zKFSMgxe}ig6VoFqluy{O|4Je?;E7)+`7)?!(loqX8f8E?XeaHc0F9=l237}=u}%}Z
zg8PMlV(VW&i>H@sAQkW!F@_ew{>)k}IVEKs9idjKr{bG3Y*w$mA8{a}wnQ<eEQs>=
z@Zm$OC)x63`{RLzwc>!N5h09ld-;~s+PUV42{|)7hN}!6s<d%6`v%|zK<$uaYs1SL
zEkO{z`Oa5W#lR-rB=x!VDd&6LRTF)&8u`nz93~l!Zsove97jUAC;)?!Bo~YuC&hVN
zX;>I!B6~$mxB^rLZ<X4+diFIxA<Ct;g=s|<_@lyQUrOr6lITHjGk0w6-F901#=*-!
z_R27CPoQQXkhv=JbE}`l7CM_12l{JybRG7G5<Xm}Cfs`GqF5p7k<cbkK>U{Vs<o{N
z9}-A2MW;@kdQBUV`bQ<SKF4b&7<h122_?BJbc3J9_h9T>OM8D^B3V6F{*hj$596}4
z=0`zWOE+|=K%p&VpY4sRV%N5O#6bcv8E7f6Ts6_LS2Yldqa@T)-yH+r6U`CqXgFBA
znIWy2>olXqE*BSz2GU&Mgivf`I6e7sE}can#0eF3?JFx~B>=CAfCUe6DWv3Wd>gU6
z{J}mu6f8QP(+%w^0bwY-m~Qk+R3(GbGXv73?DC`Ki+i~*R-NnvClNC$20u(EN=SWu
z{eh<orB5Pqdl-bJ+qJKwkDwa8!r!hv{8FZrMh{R~LxOAW_z@4gx~SNU$bm1CAe*+9
z@00d0NDE^3f1AiaZRuZ?31q<UVS3d+aKj$jAGmEJ-4+lO76v>XTU3|p>5I2Fdkc-P
zs*>}Xo>cFEWkRVYjY>$j<sTuTG|fyk`H$cPNlIsSzJw5|YhD_x@VIDHq3hATLlXYB
z2r@+L1SL#oZPHKL0CBfCQafs^O+#98uY(e7Bl|;lD8N*}9hj;%gL5by*?JJ(6YDDh
zIWdOmTOqMWbVp>?7`wG9^IWv5dMR^~RW?C3UD}cKnvtBIw5)IQEo;oM@A&_mzFJ>G
z-a2f&GQ7GYh$f6DymDfJ7x3^)gd9D7IvfDi(MNB|TIz($@^I6H?5<Jy8}(FO5#Non
zSaJpW49^iA2M4Cq*TyLCQ+s6{yI&hR)|x;AHYKJKl+$bP?e#4=eo^!Wkj4Fdql_R#
z;$bmqSO9`8;n0u<`h0myW<3g>Vj_0?OX`|}gCm9wnKN&nL6|npk_@B{a7HAy49K;u
zOb(mIv<m2pb;_~~nL>IrH1`!%N~s?e0X3dI^gz0GA7lJbKpd615&mL_4RbuFZ1aTV
z?K6F7gkWE%70Q0D&LB7A9eQY3*_qtG`h;%Ic$MpflY4^x`<S7Z3M9rDd%8@bXJ&L}
znnAoJniCx49)4X-c1rbZWT9TA)E%#u-G7C7#B-XQ6;gh~C}T;2O#5igj0dUhSDktG
zIlp<ns$Qp4S46leJQ>k{R!yVL8Z7atj=kgrNz_X>06U~3NGtZt{G#Rl^)h4U<674#
z?1ZYxV=j{gfuQ|Y$=DYtBmJNP*w^s{#m9E`VyOsfF+bIDEqVpH!P_-muZ|3R3od>E
z)EWdOLROmH_w7<CLjd4lQb7e-rxYL}zw%3^&~8T+=bFi51FMAHj!2+x-PC6uIp}LO
zTfOWo<z8FD2L1Xd3DiC#Ih%Zj06PLD4H4|Y()^7iy>^w+^x+Xzc$p-pz}3`#Sn*{|
zk8>q^K+m{jP34UJCaqJ41pL915aa|gv(w8t&~d8ky?{BRl9;FuK75O*DV`U@Tv%*}
zbJkQU)DiUwR2dO}K1@Ll;EeqIID?tK2ZzjKYGy&F4WQyhVR2*g99#jnr>U+^-&`I{
z^q)2293GIFuDNRevoE<lGc`HW$>ki>x=}_{9keq}8z8*no^NatD!Xpn9|Q!6o;AKZ
z5<e`c*P}E=c?gdN)_k(tOm$znp|NGtKB;t?uHl*8tovGFD*R{bRStew>63>>C>=iE
zk};uA{8yxM7FHfQAm%yt2Bh+uiJnvj0rToPMN*lY{!((QX)RK88EBbAawxz>!&C+F
zX`oZfE2Cmsb@Gj{9b8hQ-)*Fp3I)1uU##PJXLx(T`Y^*2TAWnH@s4|g0XdUvo5rhC
zT_YEf*MaERtJ4=NheS77<29vA%n<EiUMkYrV?O|J?zz1D8u4Ky>QbOppt0uO9G=D0
z_;AsqgZGH##p$gh^>IPx#p>Tz1)=}ox3oT;J8^~SzNWaixGLI22>zQABs$KC_W-GJ
z-h8&|48Qbom5xE{^~0L8($bHi$Y><_67(en`aJRq6%fnntxo5<!69@%HdczTmZ8w&
zOrHMqDG6dXHXSYRNr3w&kjQSJVdl%=A`RX9e<PkHEzbczq0JO{y<KDZx@;i=L0lWi
zogx^p{D`4yw9$!yr~&Au>}v0`{q4`h2u2F)#A3O6BHhth2wiRGml?nKQftRG;#i@O
zSc{)l>^CFwIs9U18nOn~$yIyYw^7fRV||X42o_N>L?xtXKXMtkGS%-<{u5X=2_~|q
zS`U9*aLX+87NHw74<D|0ex0gk;*Y*=-n>$ACCpNA=iESg*i2MS0%Sm3#f)YUGhf(5
zL%S_@PJ5Luqqx>AY;yzkyB&A0=s%gzA@59_Er_&}0|7bvQC@_iRv<!W9D6{WC9gA;
zJ<-T3fkjUSlqE8x#SLmz<7*#T<klZV0k30VZ6f}AG3&pXN{j-|?f<~*E2qq7vUcEH
zrE_C52f|gSHWHBu2ZJXIEvBP@MZK@<6%Z)PHtAi(nc6D6qi<(w+>%-NYp((LOQILf
zl;zUR`bQMx;Cdy%n^+@8dZ=HQGr$S=ntUg&`?T@en!Rl(f^k;5Szwua45%bxRgF59
z$SM9K!!?XB!f8w_rGA$d5jEmHGwL;?Yfg&iDlswgWA7XkjhT28wt}-!)kP{oHuEAw
zbHz|z_&ef(CI9h5u|Cf?3mM<T8t6xZbCmgE>7EKWPVR@8FXK?u5;h{rgW9(OtZhTj
zZo73S>?Lcy3T)P`?Pq>itTNWT%I@ViE;X4Ph*n7@ma6@~XbpWwr!F>2)AwUGYTfkC
zhd~3+zSM%Ii@+}x*RnaFNcrFw<TX3v4i9^$g3|F@r%{NlaH!`F^#nQy@<N%DxhKsH
zm9i=D-Fr#-FH{Bvk4CU-LddTCgFNU0d7vh&e8ikQ<+zbqW@gd|YlBpxTZaG;_2Cci
z`6xz~8^fynm!|~6DK?4t<Sc}#Gnd<hDzp&^Z%}-uU!2Y=^_m_u2k$(xnY!8U%MbL-
zVc@6|E9WgaYyIhA#*8dsh8}8#g~I!|9l=oje`YKu>n!6)2<iMg#zNS+(?xLp;{AZc
z#N^^&*u$x;JOVX&yJATY9I(~#UwAuw2-LqW2?8d_!KsK;rZHdD)dj-VLKt*7v!g%e
z6DJo~E=!`gqE2Fb>Y^L1ckrDBnDGya95`PP?_W1tMv3+Hf+egHW2K%lH$QqpWRYt)
z1UL|1Km7LV?{0C2$sbPg2U=xg=iRm=f_9sbeVB`2qi&ST<8S6Ljh1k!r_?M;Yo^Zj
ze?_TCA64qS?<dz~asb&OE=q5<TIP{-1_2)yAMQMc^z5&91mgaPh6+ivs*^u5G6I5Q
zA(1CqUju!WqnxprrT5CBuxuWvlrt-tpkeuiy4froZYjd_#iqUeb=$ThS-+F!rqV}6
zRNS43d6=wX`gm8A9cbH!hEg?=v#)s?aTK0C)E5Nmuv#GM7hI9&=oQ&gcemid!#PY=
zH$t-i2q+-q%x`CZ{IG2fGvC9B3Hu}z81l0Nn`DKZ!D$~guRk~I_veCPacppMnwc7n
zLupI?`K>&DTB@;v!{IzoLGnol9<UoISg-5jWJT5H@dPSN;H-jbvSG->1G~y&=v9>z
zb^dSHMDmvk!R0~rB9Q>#n_|DeiASJ}Nuozt!R<cGV9d<aW{z%%Djf7|ZbBD?;D%$X
zKbH7T;*s+c+OL7X?x}G8WgCs9UtaxeTf#aa^SYZ=%YVtyd8{u@XUm%(K2fzUVq4%L
z_MxC7(Wef4cOtNS!|FX1KmWWze#M0qrX@V#qR*c$dU?ELOLza^QAx|JU<d`OFLtcA
z%YNM6Sc_+}6*dC$xL0gBwlk)Py7P`vJPps5HR2J&vmw|1Y3L`^pZw7d@hFh_s!Wtf
zXT3P1(I^xA)@zkaa7MEit=#rVTAd0b8=?c}v+?L$kOYHhgoCl28m?hFO~MK4r~=Hv
z9IY$6qc(^H*e37!PTt}@gv!w#+`O_kXJ(ot86v$w27la`0LN+1XT9qR5FZIxF+;yY
z^82UV6zP)d1_nd>InTXaDOJhw-4*$Dbg7)l0jt`?wcP5MBG~O&G!)n`;d8b8^K-nN
z`rrQM-wy(!t;fd*pqK&-u>2Em4P0veW%6_zY*zjEY6P=tN3#c9HWKe(_9-00HY@3>
ziW&qhy3N5MjxlmoZE?h9ufB8Ql^L_E?HPsL2Nywj+zdf^EOxsxueNs>d_>_Ga&3&m
zELAwA1J-Y9v?*sRC+*A$lSvSXH2Pu`3_WdV5h!%VrTpby0sXa~lZKDWGo4}68|kD{
z>}ksxD3tW?r>_q=!ks-5<1ttPvk(Uz6L&ae+CZ)p^`th1qBVlDwWz7(IS!_BYJ+G5
z?fmgrs^YmDt8S@yxbC^K(jQN~6(H~>?9`I^DnMGiEi+)Z=mT}eY-Hs}f8GGES#<m9
zW1%pDud*-O700+AD@EuCr`KLi#2pWxHLE2uDz>iotR=OXx($nY%^JpTWp#wCBfydD
zl_CR9xw&(FW#x5<C~kC?Qx?}K>mX@464;qjC5aPGh&fRu$8b9z7=Ae!J7)i==HBL&
z6qp7e(jXkHGW8+$hkZZdlzQ~q5lUvzGH1k+Kd3r0xC7l9#EGMAO}(o7JG^QXYeQ!x
zTh_GUDhr9Bi->!|)6K$LQP#5&j$}7dY13PDctdj*vh!o}iic#6iqDMYSKx{f2uY!e
zI=v`+qBrPE&F!nn6CgB|eia3oce*a`TStd+TvLVLGC9>W;~{6$aC2rnZl-V<>KxRv
z>*!QHxJI1+Js1H$(viVH*c^aex;}L_SY32o1)g2c$cUiLoXMh`mmnerE-@y*mTx$%
z=A$y+K`S8nkusodGd~3D)eQnr!OHyX;YJ-zM>UYo3|Z_N!uFE2sg5-YQtC}xLtHN`
ztJm?X0T!)22j#tFK&X&ny%)C>&B)1?f6w;IyUI2<R>NyF00vo_6z6TlL4$;hstB95
zzp!Wp%g4v30ksm@N#qg;bq2p(MW_UoT5ew*(aX(@IV(8dw5u|LgxANYh$c&CCbh`H
zU3EHl4%PhB3Z}B>QaQCw3+`!s#OYb^D5?mASKvZv_7)=H5`PppxwV7`VgyhMQM@j9
z(0f$(3Z5s?E6F1rb(k$~+>$S3r*)t{4**P*Bq5=tz;BS;10X`zr9-nHM2;<x8wVk6
z9i4!ALVVYY&WxnyP}F*TmK-#x9bm%wSDMDI{23qIFI|}6<*1Ud_2b-ywy^M6Ia$HZ
zt@KkSJrOrZx^U3vvM=VnY?_K#%K??)-w%&oNsmvX^Sdi_7AV{u=qo75aNrV`A+4a$
zUs49U@XycLLAA%25X+ZK=7}0ZI}P&47fPl5jgZb@If(EJ02EW*O5M?r0dig##LvPs
z2)|n?;n#Jz%YEpcxQ0<9H$WM7cZqi-4tyAB>fs0L<f_7s!J_e8O5IZ2vLYfPwt7N_
z{Ucmx4|-F1^D>%0^j4@2yjVkL<t<jS%H%`%*(!!h`S4yTP}~n{U*etDc(|833ylyq
znS4@9n7G4%Q1wCJ<Q+8&(OwaTa>E7Ti_|D)2-m2$?1=8$I2q$KvaSI7)RR9Ri)^>6
zx&wiHq^`qLcR|J`M!CWSav+pm=77{?u~U}uaMC_VH%iX)p_;D`b^dz3sC^lCSwvJ3
zsmDY|nv7tsL}AFHrFTU}rMNZE7B5P=gG$bRHDXC4IF{{jl@Zm2GjJo-ekDgvR<;#I
zpE0Y&i07)yd{4+Qht+4H3w#ERoxvx36dvZ+9L;pJZYv21g8E5=B0%aaJFq^)`JxWA
zS;a-KVzg1Foa$7k)*umBh!%%tl|k%6WiNh!SMB=W9jP@Rag^UggohTQyu68nS7-?f
zMloSeQI@t700-t-+mdLMUc<Ux$$8v*;}cr5{@Aix$PxjDwsfMs-mkcwq|;6U04hbz
zy?ghTK!tu8tRMqH$EE+bWt8U(P?Nwj84#0^U9h_lW1AgvQY(%>#nBN#OUsf1TOCBu
zB6I}s4n0G>y0=6DA;%1#We4@Y4N`UlTLlyX(btGpMC=H}^nhNrF2||8q7FxV6uTZ3
zh5@hD_Sm^ats~pnyQFg1VNT{fP>_(A=q!-zy{FwlZ7RIGEW@I~f8?H5O^IhiOt;;+
znZvz}kOS;3r|Xg}AEOP3);X|?YItI`z_%2$>-=xCF2IhaaU872rWz*hWxM7(aDxLF
zo(h!U0zlxTXa<sn&Vs|~;l(A6r5ruVRZ^BSd`g)3M8XZFfM}{OK3Gf1PZ=2{GA&pN
zqO9^!Ivcw5Bvoi~s~sS9g|!UlCdrVRgq4(%@(>b3QCYH%#nB2}pNw@~l2i?04{xMq
z^;j!Ey7Bb}e|7)<zBhRj@n3B4v7R9Gr-$CBxg{>XKX!GBJPOS|+qEQt>H1Zl80d8W
zo+XMh@IgjaPgI7hRK>U$#MQoj{kk>SI<EqkV=U}!X>=HHUNbJ$`sZCU?M)();Y{zW
zTcaDOH^TcFe?IYW?;K_f;kxBH&L?18ejURX5{mQBawx4Yv2t5~M%U@{nnRl{Bf*To
zW0kqXn866cn>RjUeGP}Qzx<LJPMLCjc50W%F=m_c0(`=4*YSCgHM$-Zev^rN=kUPo
zrROab!qz|8s`|{=bLgHF@6cwLo558O=7{)1BDutyL{CKilxnxCOWIgq`NWdhlx4sT
zyZ|Ku6-Z{$JA&>=>r{~cp|8-TRh)ZP(2qUB%<d8X#JdD&$J2jQm*<zudFl4m5v``3
z!$U$x;^M|l;l9Z9u<=bjHKtw9F>*zrF>;bGW!<>&9ckIJm9B@n{|()?`Rx1=b)nsl
zDEYsm$s*h72}vDGBy~>y_Un>l<BT6*6!TXS$;>E$E8367bIw<Cjq%CKqPFR6uPB#X
zCD>pKXI%Gz34sLQN5dseEY6p=uDeb&SGp~X-9!4KZf+W6WAB`)p3oBKt|c9oP>#hQ
zND@d96B9#cRYmEx!(Ca3HXxD|G{tosQLmT*CJ-HDG%%37fk0%{J!NMnY$;_zCXlR6
zC>5f3J39Z(z|3>JBFxcp>UiFT);Qn}JG2m)z(kG%RN|N~3>|}@hLF(uSFf%C&Hz4}
zTs#3ukPoaf)zh}FfKXRp+>aO&JR5}nN-&*`K}Vu=KO%jrt$_0-NX=y;MY=?_kZ*Fg
zBS8**%;>H%rh;6Zd3Y<4PQ!&2xjy(9qil}0UZcV68%;(c9FbKeUfe4p@;b&!ARq0Z
z35O$dRR<LP{P`MN^y9XI&ftzHga_NIYbvx}*YuS|<b{u~;oa{S)Ha7#lRc8Gp5}&S
zY=fyIrF|25L_4J)@zttHuj-NK#5r|7ll)X5>pp&g%1m$t)wsG)Kb+P=_(C_E&U3L1
z6tpOFL&gqus6Oi74^(53Md43@nQ;6rn$0BM5${xW5mh4gdEuo#X!N6uL++qMO5vV9
z!re!`^VR!zsuyOXa8>{qwF9N!Mi$oqX~Z4InSxAQ8QMcPft(I+_gx|Lu=>4+VjmJY
zSt!*0+S3j2YZKpMqY7MBH5oMW&fcC8Z3YyBC=hQ}Ik}hvLsMGyENm<L1rIlWmEHcB
zfLU+Bm;bHH;*VAkS+5s$#2N=Rk2k`f=@XtgkY_>N!gw6s4dgnp=sQlhzwb92twfOo
zXL{gfgA{ug_Pu*+h!q=kl4Kl_`&OVzh}kzwG}Yh%so?e-1NkS!&;0HMNQL9jz3XsT
zp%X|H!gleiQ?vaqTnPNV3LCCft0u(>@Xcnj{?q)h83iI`AC?N3_?;~1eDA&>RROJl
zC>!`DN_rM|ZMd!?<m}f1t4Hi*q8wyP&OVM~i2sd_O^sfl`FMHi)Ls7ipj40p3a9Y3
z7@PV*V@*5UGZVdHz;cMg<Yzk*wh`!|sHm7B3w2hO=rRLPW$q4)r<|aE$uSm?638Eq
zDpG|B$ji^2uX5mzLQSRV?j4ZQq)B)w$jyul73o<fVT*<5e4adXBl7s|HGvC=j(J&J
zRpp|X>nWHI4sf?pA26PU#UN<cB8%Dv0wMA1sSNz*(1J+p3;PF#Qbd%cm!o0<YZxv?
z!SOQb;Vw~jT=uwyP1>6agjO^^b?Qa)lXEi8ea&Dmh_jzRUXv4wF2f;H>YfFrbmHc+
ziE10f8G7jc_RFvrthM{fU%g6bgfYZxM;OEg8F&>?k~ek6yuZj8<J!B&k+Vr(r(>K3
zOc*4^Bh(R_6`;OyXGaGiw>^2Ip3?MzC&Hfb@`baTB><ZxmXV`Eq@YQ@Fg7B}lR<fu
z(H<#5HR$n2VW(Q!G}YKNYndam_6jgcG5hXbn1%$M6t~aY*Z~wojk|`2npcg82TtkE
z(Mmve*qCGqK4FoLjE6xsb*mdRSWnK)Z)eVoFeROMq!EO$79S$nX>QK6rHMO&+QTCf
zJ@(T}bPW%#C=F3c{Sa{z>2ZZPqIhPoDyl8bQi^{&){a|&%Bk}lU_DdncJJ^P(5pWH
zq~4?Gtcbd-8)uwV#J8-8jk&yDE7b6h&4PtSZ@a6BJ9v&>+vyZxo%#{OKXw_9xRMLU
z7Ba@Zh4G-NwE>O$ZKNm*Z>RP{yd8-kXrw1rMt@Z;tU=DOK@N4Aw087=N6dX~2tPKY
zJAsWT8T~Q{`Us(${{-4Y&+k%oDdckP6owueoUv`oS|`clFGf96Bl3W{Ak-Z>=j7?I
zqL6NENg%VA0E?Nyqd__RAQ!}CTPXAo?%dH$G$pV1_Kea!TtzBpqH7t$ixI7t@O6|<
zj5;ja(2WvdO~HDPbQaHs*da@|mXunkhwIUC9<S_3?G%egy6>af?a|>vQFf4=b>!vw
zaFRSur>=<|+ve1#GBY_yzy=xnK-{|xt&{LFXt$}tg!5qu<OU7)$n&tVXX_mmX0SnH
zF1MG27<{-C#I<8GnUb1i9WaO*`$gdxZ0mnvcEqRqF8cSQR(4eIf)6d`au_!IAI}n;
z5+G+W_fk!dEB{bLW#{Zwf*%p&ppMag9wg7?v`%8b6gmTKe0XsaK2Gd=Ta8n!YNgsq
zLXNuZiypE+-dVQN^@*^1vtT;RwJZww`x8fg_@H!VcVgif6jHC4qRs#qXDWaYYfN_L
z5Y=+X;nM@#0oLAQHu7b6cM$d*>Ape)Bi0IJZ+XU4m+oMMmWM@VmSgJpbNTFROoV+p
zix_X5yhfcYjWl!6Q`!_Wb>(cA(DE&}asHmZl0TsAAy`#0`$h7HCL{~e%ujT4T&ei?
zEO#K364mK<s%<i1{Qbnli-sI*cSYjF69dOu=t3|B5THKScKiWU+@2Asc^3J3Vs{9{
z{NSoRs^sC3Z0NDG6hp%dN<QM-px%{*ru4P^vWqH}t&_xGKnf4HdtbfZfi<3`Go{nr
zr2{xLNw({mW-5JWg<FW<pJ>(`2`ahYZtC(^a!w6|5HsD%m|hEz<TA}~AB5@TKFX4z
zj#n$>Z43hg16>CPs9iUuA2*L}0FXinj$4jD4NY4`{vcQ6+c@j@$~o_!pCa=Q^4Whn
zYyOp?JpabW8F?peVMhdrMB-AXoad&+fn+=jvEdSV9<?OzmBTXlE284EUBkJ?r1S;<
z;+QE#?ZB7?qJ1i4#tXn<#j95bU93CHACR^Ig60iD-9XQ5IKf3TAcYBfY|{-LX6eq5
zjaIpt!1kp3KvYyzfgAmpc!SX1+V<Nr#!E8lfEez&qT)zv>*!bp$<cPtSC|bA#ZeT0
z{NeMB=SG2|_oGi%6O{w%a|P1Xf$D`c)BLtX_dX`RpgXJ(2Bnd)3<rnvQ>7F=`_=cW
z{a(N@g7SJT9n%!j^ZJoNb$f3U4kavZA-T0O5p>zbg3>@p(`13hO{q{Ib4RE$yl2pX
z3g513t4EtX(3#*z!|^4$TeuzJLTO;H%<z#!XeW+3(zHVIJMn(~?(&42D#K`2F)Se?
zWO6(`ST%XoC6FaH(90pkq1uoVtmo)Cfy40c$30N<ltAJV0Rzt{7=&bRk`JPad*?nz
z*TC3C^rNs_H#2KHSLXH-)8$0Q#~W7t8Nv|Bvc4YCz}yW4Xd0c2{jp0CGFAc-T+%mU
z-$+n|)^v*&DcML%6yz(ybHmg&P!;DXOKbpkn7CqsDH=*K_mY!kNjeG0$8?F8A=RIB
zt7OgnJy`x3GjV6IAI5aYOtd-@@?dLh6qlour)=Z|l&17j9IgjKFq1up#8QY}AyPx5
zJuu^ajb;A;ZM?`wlqS+`TLO0gi@mhYvAwsEV<_*^bH^I9Nf09U4OAZ1>e<%Zh@6WC
z<5YW>$zrCb9mwMP9!tT-@Kma*|7*z2yC~e$W4CZcpAYP_^Z@__hjigyVkf7OM&=3k
z#+Y^cZ(c<QVhifQ+D<mqOk#U75@V#=7?~)2dOHRAth@%}aXMzXjXx=lj8vy5#-^m?
zp6?%{mvFa20frDw7W<tjuVc_5bsVFdjMHb0I9f}#HiJ(f>@%wckh$s2817O+c%u^f
zM#kM6TavBCQ!eB|`i|_Pu#-30Fd8ga%w(`eq<AL&G%|Ci0JRHMC_`}}Cbw|6Mh+}L
zfUtX{vd7*A#$?(wW;EDWXpr`lRJZR9Ao#$4a9wrcF(jW=4wVP_kfVnKr<}+c5a6UO
zNT1Ifs1tlO4dP#MnUVcL@=2uS%4w3gu;wl^EbC0i`NjMvp1szUMx)eOqU)AC(>Op@
z3$zT#x!w{6wb73SQ2n4_u}P~NX_X~PEHN-pF$c3kIH+-*#;+!E^quhN(lP9bnBOsc
zVLP$TRG@?;)3GoV>P04JIF<}Q0f(bJWl7~Z+D2zEmuvQ1egWfR0922-`gZEHK`Xb&
z?nO{|-_m-ozMZ<g?%1cjPpH%0w`cu-r#JsmT>d9({V&N{|H6vM4mr#@5|^Ew4S?rq
ziTP6=Mjw6U15)z+!JVZP_v9Mv1mpR|z_-%GoX$YQ{JmLL@O}KyeU|-s-_ir1(y8-L
z&?pjTw$z*jduMsO59m9H`Sm?^bJ=q%XNjt8DZ%lz|5d1TenC<8f7ZbN!hQZ5F~WbY
z;{Tl$5%MtFsTBi&K?F%S#8|SH)hFLR`$Yk8iZr-Bz%JFQ?unV@N>j)fzyAl-<eivN
zYN`cQIFD(cZ@de1C3H*K%@I_OExHI>iGfViso@$i4U)lrWGn@GfdKIS&?>%?qvT?E
zNkT!GegpkV6LkTg3Yj%UbdLx~C?rXP_fk%=VOCLGRM{=e;6jTuMIJ(dvAg&VZC53v
zJC}x;A8*8t{(fB@>Oda^Z{nqN7RX9R#ddg3_OW(=ICVSO$cd!*n%k1+-Cxwrc86dh
z8cK4d5XikZX=H}j<Gs}}CK}*Cl!VT%M|wtgZnwSAvC&1Vw#WaMb=x24O0eRu=t`|d
zKgOUakO<qQ?rkzgv4h0FiB{2irvdpnjy14+{8JE0>pg@b*Kf|WLmSJ;UklTFW(0kV
zXkv7S$Vy{LH*RQZ2r_2eWo}nSYSn2iazij~y#ztj*xHYEGl4R1Bo^+q91LXZMFCW7
zm^0^L_<8r#>~gT4P0;Kss++cSb#%P$9Q`KOlFX))cOY#HnBV0q`C}FXXtY%~6oV{4
z6yH#Z{ba-wHmye|H4MDI44$^;j<#mr`zy4n79J525^BsooU)!<o%oQzK@54|4FEcG
z6EQMF%C|xl1%Yn1rkr(MOYy3V<MrZ~ox^@I!Z5J<x$N<Ijxl%c-1(x<rZ#bR(@uFn
zd7EXVX>$~PBBLTedTX08oUmW;H+k>SkJ*ix$qv$F(%$T#4wH`Ix0j^X(HDd5QBM40
zXZd4i0j9-G4l$u(gp`n3W2_cYh!=misZe=?5nE5ULA(|9VM2l-=#<rhDLJp#a7T0~
z%|gIv<4Hh-ze*P2iOc>THlZyoD=&St$;%6y?iv7L+C%P0%PjV?@P}hKJ?_$LBMJ~h
zm2hzFCRj)}d%z!!9X+SJCz@tFV`0lCq5|ggm={;~bfz~Z1A2i8A3h~C;tBaD#%dzL
zw0NcP(Q`Jh;ck);K5CHV>L9Kw=#~SAr;AT6#FIqFo;qsfur8#2V4cLdsf|CbCa(Lj
zTMmpp-B6auFjiCYHp}sdTkjZfYvM+)Cd+%nGuk_nE+iX$!Z!#*??{&m8Q((+BB)TT
zU%p)529>Xfu=j-*CzaEGf98Er3q0oYL_irdEpK?Tu*z<}^pSi~_5o$5zHf(Tf4~#b
z(&}ygV}2Xc{{QdTh!n!dM&Z}VmO5)(JzIRy=;?+d%*BoIC==D=I#zUqf&*SlM>HK0
zvR8v~XQ&uUM84m<_hR;037k9a=h<2(bk(sx8ftb6D2)^032fgM`T*$xBGX;QF)?={
zYi5d9N?Ned+(!nB#TA{+G6s`OrYg}2$Y2E^>>+Y$CJMAXS)@S{Gd#6amC0tJmY(W?
z1Od-FiM|&8>wJ~!{3TB~Jd8RjC3XN>0qT`oM7U%w{2|Wt5+w)y-wghJzsCg}lc1|X
zELummh3;sIv6EB~A`G|{hpm9>1GA~D7y*KAS<qC8SBcIdUL?4}x-ra`mx{142^+E#
z>m-z5Vg;rypA4-cLkU?YyI&`eK^7C_u}on;#tZ0iS81djbbwf(@M&%`zYhnjoweTo
z4<nBWyenf6Bf@{5R6SFxCrEQJ;r&(l1zL$fZbZEzRW)f$gCRJn$nH<dgYlm7h>Tsx
zIfbX~2j*WM$qqc6R`zfVbJ&4?z%IMwcn!jW7>v*E@DP$YMI>TgMz0t~FWD1IJ5+=r
zhhO6NCWD|~uga&BCP~Up1xpl6B3H@qs}P>M8KGTBDyy(_8#p@OIYi}T)FCu5jXHdO
zz+Ovv_y}pywYRtLK5YfCkUlw3L<rHI;c7_rq)$gFgmj^qqDBw7kWX3&W7z|RA)zzq
z`d2^&+2yFww+9nE2G2{ouZHnLgGCzqiMxLyP2$mJMti&^GAn<woth8eq|oj=Gv-Tl
zer3HSLopc7unvBqj)5^`UE4;78k5RL2kWyq6{U)~+@aM5>F1WYp86~5AX9%p0=*<v
zHSt!E39rP6UHjO~$*mYs8`Cq}i2sCC+OS=RLt27{OpIRS%1P-r1FKaqt}?U8K?Mue
zc3Gz6(Qk||Iaf*F51E99F1eTJ5gE@|)v<)E1gTL$9#IWsPnTt1#gcAA1U(9hh44~D
z<)J!DNWDv1AQ%A)-JTI;Xc8T*XE^8^Y13f39)@Cd6(kI4CU{A@sW3oj6Y=v-zY(8a
zjmd+y#xW{ff$^&~wYcq$f0l{L$ZEqi1QC}HkyA+qL{beuBJm(0;TzIUH#6ZiBTc8o
z1`sAi4qaVI+q+{)w*WD^5doR>i+C6<9$LU?FH^>p0?agf95!X;14f^;Ea6mQ4T=CJ
zLh^)FchCo5$Qd@d6~7nTL%cA_yn>P@(EN$onMh|aOcGJ|A)b7cZd?fhQ%<8x@_4jh
zhnn&Fsk34m0*d@IL;jv8D~_FjkPu@#1(~hV%II}HX8X32VJI*<r{%|akl7bx!0%YT
zYIdVm<{j03hsBvXQUPSdHH{3JgDny^K3d#V*~j05*Ox)N9jyQ@uu+ONW}+kO(8YF|
z^2Xzudbx)p9|||=eatQsgvLbtdw%x^p7WzpJo9GI!c^UzQQtdkP|V3b@G-Wp3rx3m
z7_8GW3T@67vR7L>=mW8sH)I|H83~tO>LCTQpxaPgeE68+#a=uJ!j`b?`7?VVOE|O0
z-09aR9G-BZu@`Ogo6#>DIeS@@?ub<C-hD#62$69B(nQXdv*9Z4ClcmM59R&=S_4FU
znI3ls09X${<`RySZoJSR(oSV7n+)e615yZ|Plz_C-p*uQ3AfZ+`9mCp$!lb;HiiZh
z-6?UHA4c>F8FaX?5pjBk*hKWD^h0frEYe*<8XT0AtHF;*dvxw|!Sq05{!ZbkD{*#@
zO5mjpBEgP0@21~MPZQ_Z0YGZ95d6XR_Pwg(2jQF1tBst)d4wQljOGL~ahQzs=1rFH
zPQt9D9rLQOR55-%)v2WF=}y}68xXgN8PbnNt{;Z#(N^@Hx@oI!;9V(|&QTZ7{^nEX
zJM&3kY8n14RmPxc5$4troBF%;2f#qK0fy{nO~Mu2so8Ebalah~UgCr!v%`sB6cJ-P
zKymlM&1GLA-je3J0n$=IM#@znMbk(+DFOM2zG#zxrk~Na-&psh1(+`*u?eF8LI`(9
z|8$J2(FUB910irNs}c)<O-AQK;vK0MC7=}NmW>Yk;z40&HCHmU1=0xoQM$e3B1(={
z*`%e<vp6#|k{>PTMuz^7k!7oFU#upLSe~3hBNG7u;W+Yl1Pu6vQRAF(=7PywKsuR4
z0pD#a1k(D}_9@QTB&$OWTX+q&DBW>da$$Y;6DNE7p$*MF9UU(S<4sCH^snE76*hd3
zlKc36ss(U_Z^irupc8V$$mjwRjV5uhO3-BIk4VaM+#H`troIud0f-nid6i^Vs`(#g
zHxmpSZ#3q@0a%I*XZ*O1^yZM|<HL5Ptu1<hhNwA8;#~KTd?G1_2=$IAU3Hj!e9Gb&
zCKyMON$v`heR-K4hv`nL+?@;(Il$z}pj8jXur{)bC<_l>Lq=_*T_6l&$2Sut0inID
zlUZCnK3J48^hPAedztI)J8zPg)@2AY%~Cx4dJ2q0vBG3t6)|O#<}$CLy<U~A4JDq9
zjxVmRCUYD}Ul*CmPVg)7n4%zUjQuonn4G@>OAYYvFUd|P!<gZi@-VEmRE;Oi$9B1M
z=C=?#8`t)>E)IL69dylB(w9hl8RQ6(V1yxn<v0)X#hm{}&Q=)l09_EVc|ojf8HTU#
zksd}eYOlaDjGA(YrcPkAmjy^I*ST>y*#=Isj-!tkcMFlBM(|9oBhD3KfE;`0+m1N*
z0&ad{izO+C^y`o4DUW}aqxP|K{IJv&G}}2Un4#oDh&bmo+S>?i0(y%g5thkNE%#o#
zu=*Zl`+SU5`P4hLH%id0FePYM7=Q5sLJ80KvH1*UEd@Got6L$Z|EL_<nd%;*HcZsd
z<T#Jb^nnM`9veyL6q&V)TAmE-8ez^js*Jz#eF%dEnJp@=yU-&Fdda)zSBe;uzcGm6
zq=Xum1_N_lr$X{%w4evdJ5<|?3Q!nJ#`ToQElR#C+;kH)f%29=3c1$l4ANG~YRMH8
zZiun8^0+qzr*tU}=~_Ys_fOk=xo$tPU!VmbZs?TgqiWC^fopuWe@&doWI7|JGRe`&
ze8l}>D#`>tX$uQHYE@f6)c$pJz_sMO$(ScYctU~jcg+s*hsz|p%TkRo;IOy~;J6*u
zSz%r^Z_2ZZjNBxvMX(U2D7cIN>XL_om1^}M51EXtLL)+DCz~@yjyTjLv?flkhn-^P
z2K)%?Tza_Esh-@_fzVMl(e6)&xA**VG1~gU`-Ft@v=ml85|W77t|4uvi8GadM0@7U
zDlem0PCfgO9P{gKenv?L&#a{{?P5hZ$=2`GliY1!mt1?f;j7fiylVXti*GE-rR@&a
zG?cX_ZQ`pPk}1A+zv1kqA3jh0?A)&glLyj%Zk(9PJCm~^V4=LHp{9zsS4a^**p3qr
zV&_!K^(<+ZxyK0l_16XTLwok@c{ahD*rnvuYiJEq+-tuiIxGL(HJaG!XE$YT?N|T)
z`!c8mIMWhUH8oMu(fdtIOe8jKx=wrW@S%jH<UVZc7501TUfsNZEPiVDV2$&!`=>(%
z)oPU5zfCJkaWZOJukCHnc&sMxR3lg5-fz57E+p@A*iR$i7@k|&pOm!C#LTRy)7}CA
z_4w#F`hL;pqg>)$(-zN&eByle_?BDy3kwU&K`%G}gLpo^=H%cIk(6|#d^l9}+9w}>
zBq1SDj#GR+Fffutgzr>N1qB72yCWm`a82O;b$%=sE6Ju&5zLtm6zV4{%%dVBX>_F1
z8i4L0Y`UK&6SuMaz8V@?x40F_zOb}3pxCq`+hbx+i}TKHxRq<%kht3k2?vqdS9|?U
z)voO_%^Z+%h%?uI%J`wvj<W05uV3ZveP=v>x%t_hJH>qW=KG4*KzMK%?$V~0Bk5oF
z^Sg(taQD#1Z%=6^I?BZ687!P5y=A|Gf`X!b$5qYH9T8(=&dP8zChBd%6xLS=5wQR+
z=CBj3UN56`=Iu{pteB^-!O+xg2gMLq&~O(gchy%{KLl}R0XgeeUteJ#X4fFU#WC}o
z{3zxvVBMCVoe|jLZu(eZ`x<K}r_=_U)WzZ|(E|gv{s95xdDo{{AB>6{H4O=JoD4V<
zDtM>1$5CO`Vl1;Ik^&~%2VB2-bHn4>O<|?sa7f^Sj9B*(pd|{cK05J^mouYdZXU5!
zMrO&fW$Q2f_I0PL?<h(HbG$4vF0KY0(hm_98{>OU_d0kd=;-JyU3F1eN=g&svhmjj
zJV|q2d4v0z$E}af?)+QbZh<}MrJ0KjxV9w&md(ic_$@8>+7Q7GpE+|Tr!Nv<ce~2n
zf|CoBrZvu-=@(K{Ll)^mXYi*6Ok3N;6fCjLy3~*60Snzh+=zq)iO=UBjJh`(7`($J
zQC(gAuFm`|f!j;Zp2bKN(yVP^X(<w~VQTtt!GZ;2*FH}D>Bfzv%Ba$g*T1%sgJ;8F
z>Oz7@N!H>m)tVuS*z4;|OieFzCoR!1yx4I`jGg<*X~mGb$7a5ZMHTPknZkX=2$Z$F
z0UyUfOcJ!$%lS!2Zck}(&MSnn-8fxGTwKzzS6lj>TL$^tJjU%G+`oScuKx@ef71Eh
ztDm9os@1lj)pmF0R|2`4tsr>TWx1tlhN;8=M5?3ZuB*Gd5^%i5UNKMek>KPqELYF#
z1U0yBV^AU}x$F?bi#p*7EQbW1eEQ9|CMQpxdiqn}HMRSrin)53>Fr$Srw%`7cg%`d
z(6r{so))*Fv@N(b()P?wl_Edi{r}asm)D@u!mh`^k+HEp)D(l|?d!gqHLlJq`{d)w
zii$6Lh{x{OXG6218@@lcbuRWz(xvy;F7Tbtdj}(4cHsSAarDQxxdPbwo0eX}zfwjS
zG&;Q9NS;A0r?+o&892FFKk0i+a|meS<s7fa97fXrq^_lfAaEEP3v**+9qjFYQZ~ik
zR?v47<ByNmJK`xLf)^GQ9sc2myU{vPUwrWeh>xvsV!AyEWL8&H+|yE>!g{;dZeNwC
zACm#g<Tiv?5`)mndma<7zIftAtOc*0E@$IQt^fil8FxcxNg1;rp^sJ{Z(#o-1%G?Y
zP4#{IOyq*Jjf`S|oAp$};ux6-cU_p`MGk#>7{hBy`<GvSVcn06lqiOy`ebIQ>Wm>i
z9NT5lgczoLZVD=1>fTscc^{!xR$l&MqYc{D#@m9JPg7t^*g5oy>J?r7V~CQ|;^n>S
z<p^EF@DxWzM36WsyrNf~Rm0WR(W#ENaDgv7*_}VIE3YwV?niULYVFD+4e-n7Q>@FD
zj}207{2Og*KqlJr2Tb@=)6JiK`pq{rIXW0za<sO#78iBo;K6ItlY?4Qs2saGJFnIE
zRAY|E8oHv_G}{xUn3wj&{CBSa=a;*O1O*{nOJG|UAa9_BV8As?5z!C>8C>#G;5NWD
zUZW|5XyVc}AuA6|Og$MLL``*y$H@{LSp_01ckJnB#0Q!yFuwTJ7V~nN?6z&EU{al`
zBi*}*_QKgJo|9=eFe4$(2=m=oetv6>c;_eJb{evjdHtCXEWTwJFZMOv6>U1nU};w8
zfQVBp4plZ_4dE7w-ke_KVt`M$tTnqoFEnE~^CCR8awsO}iU3nSA^(Bh5e1K$5o-vA
z#a=@-Cnu*Z4dfb<LE!(Kf_D51FHIlQBJSS(`Yo9OacpumwcRfqgGWX%mMbzDtC#RY
zuH!mC`o@hLS@%Kyt$Di=ZD^^@11N({w%+=>j$@H*fjej2XR~)a2oJZ!&__kLk!ZFt
z^BMU&nEp2Y37J;xz53@XS42-6@s_h$CsXV5dh;!k)mf9CUe9DFF=*{cE9Z{b*T@m!
zfq^Syjy~Sw@9&SlQS02+ty}SGlhpcT>-rQCA6%k${}pP|;PM~ujNG`Aq-#wZ*9`k6
z_+10cBqTJ(8fQuP=HDwXKNQ2rF;)rFlFpe-L=#giDVk~OWOX*Ip>A@~*HThaX@&*{
zO5IbpicQz9Th|4rt?rpK323L$*3;WS*MDq!WrtX4yf3+dSF`bH`7`{DETV=SBx@mM
z*S+8F-+vjw{Kq&|gz?WCRdL6A#Hf9F>@%GxZ7mi4eOhNjx*fD5->>Vk6B7@?d%8$u
zk&~n2ya0fP8KXP+mE|&2{O~@Dsih)MT7i<yhr6oAaUc+h)^v_<&%*2hMW6soi%>91
zw_W=>J1ffwZVeq**9@Ok3U^UQ9(!Hi*cj8*Wq|{7Y@)KV@*7uOy!BY>?}1G%;QICZ
zSZOe2|7V_)LV|)80h=j|x5dT9zyH1s>|_H&T=bg-a8tmklP4wPHiBGL3Q1f%%N8tf
z07mO&;Fc=oTkn7;zg+B-wKf5^emA@YmT79+Z`)>o4rO$ahj`opxDRm3_19P5-Q7KP
z?p)p7clChvIgp%wdMVI)V3lz*XWIj_w9S#gLSxUKxb@%O-(P>{PRw24l8CO(&WVQ~
zUI4BnD*5pt(bUXL3z+V@-h&pQIHz*;e7OQCqk++3=*SPgD-I}QpaB`%A3n?#{~0U%
VYBt?jb^*LKz|+;wWt~$(6957zeW?Hd

diff --git a/reference/plot-23.png b/reference/plot-23.png
deleted file mode 100644
index 231b81794fc81297088edc9e1db6be87f0f1fc78..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 19541
zcmeIa2~bnn)<23P-R;2b_S%Xv*m?kQ00j}5^{NerQBgp~piBxvFbp9i0o2>lwg`HH
z%n)rwnt3KdLIT=?kx6BUOaUc;5<(ydA%wtxC)nNIuWr@9UVX3XRlRDeFnG>B`>eh8
zntyBO<|$kA4c~10Movy{18i}^UQTXJgq+-}3tz1TS9+5kOoETK7p%-rfKTuvI!=uS
zA72MsIETo|DaT5GS0oHi`N+xbkb_SgI~$QP-j7f8@Hx+)8pulh$IdP2i<@sHuVlZz
zFu8IJIs1I)^o@r($C*=avU$w1$oMcErNc0C&x}j{ua%|#g&n_@euew`+mny7eszR*
zmsDKa=}~veVCPB%*1-uyUb7wN4Vl)yS<ELmGjgS`34gq4di;E$M1G*Md|<VVz`y^I
zle<{!pQ|lp#B8!&Kv#zqaYKn&0g6G7n7*>~O8fhvMQ7_(R>T!DRy8Not&{%o%6_Hn
zk)u{$(_}aHt4*$y{!x+m_kh3oASd^?Jp8Q}Ko9=14|)|HTMjrqd-lv+Vg34oB5vdn
zCfaYOExIsoY*SORwi*7}jSWV^(1{NpKGby;y4qO@-qaL1sf%F5PGRFlI=3&Rt=4CL
zDkm#)?AtJR2BnBgLKb!~U2FO${Cv}>?H`S`m@JZUK`+mPGHCWssSbL^yG*S{=d|P8
z92{63VUi+9L%rG@IZnzr$C@z@AAfs=r<s$1X-jQYR#a5%EAz1ev(ALMo9Qw85v6;a
zxPB^jsRYa%JT?W!y)?nsvh`{`_yLsu@Lm&S!2?Ms_NbkI-ahI<4d<0X<b3XY;d?FV
zt4kP_J0~3-^s+dTRjXFnnNHTN;+JBEU)HZuFb@8W#Z@M8Ue**vhXe-)ORPtO2Ivws
zI4hH51ZIB`L@_$$(D?vnC*QOnoJRZwbM^K0$&jM)#FHU5k*Ss(ho$!OA)KE(i@#l2
zxwbvJcsrUlH3MVj-o=uuw?^wrDX3$gEmswAc?NB`*4oC##w8RImyn?P1UEgROcMWS
zZZ2PSxmz#|#ZItm#fw|qg6nUeu`M*rY-J(F2br4yTlGoaGh?0PVs0dk((A5=^UP^C
z&}9B}{P^)$;zS5p>|D<K0j8rII*0Uw`XajDFJ*(*?(0a?3mD6$uYIf9Rg6N>L{FkW
zY`Jx1Y2RZ<<9e#TO>ln_M8)#y?M6RoQ^(78=}sk1I6Z#e59N|*g`&Rh-fa#Jov@Cn
zsfXI8cTe*2Lh|zFOP<Jl&c1tH^6a~J?_MYMnK6a*$)B_rFg+8o8nA=ImX^?<VivVb
zG?%EEu5E*P+QB;55k@2+Mvg^pxDe!WciHfBu(3~yxK96^*Ol@|Znpc?c<e&t_U+s8
zSmaY=?UF$4qJx9-Yr?{W_2|Zp8`)=BVTf@krnktQYI^9<p-+$ZyH?TmF`YAQN9$n^
zSJq5`xw$!BgG>+*5dzrL%(6@EkBuxB_isTFrq<=>MTRpCfY+S67sl<{KciL`H8rdq
zEjp7u8oKOOQ*Y(v6&;!SebW*K)fNyosH~zA<&0Fqt$h`j+LtovjJ&+fVeUCaT%K+w
zgCG)2m6n!zzt`IcBQ5Fr1BZ-_7k-9R$3qRHGdc5lc^+N7@TDK%8#ivW+G6S-+oWq*
z+iy@%VBF8kGqb@&AT3Qv`|-~WMl;F-O{47`92(U{5>bYfu?K#l%S}@tp`AMmYxl<N
zDy97+&6tTN(Q<lui`~|Z);Id|@*JBhq8sCtNzqHZGZb?4qG-TI(EPDcK|ukF6!cVw
zj7kuxv=kR`=i`95Huhth_OnzmFMfCkRHG)++nY1FVf*1eQ^`X2S%)xlcb3@zYZ|;f
zr5vv{v!*VpY%8%|a-4B4&y><;a`4inLHF9onTzf$w54D+h$6mWO*r2ZecGXOrezb&
zytRT+#C?{Q;M7(%^5VsddM__8E@|Q2Pg^`9rf+Iw9XZRI=>`1ntRBr0Ux%Cl1?iN)
z-tdN5UhmQlru^&--w?39%>vUv&yokK(pTVI?%S{w{cmcTL%w5w{-=xy)%oWa`pj<S
z2BOLwrEFx}q@u^dsl{z7?}Ma@6rCfQ`=qFmpXZg9klb1|`S$Hwc@iPNmzR|Z?R6}8
zf<o%^^5_P*=9(~WxLS2yUY<tV$6{F!buOmMxZX%dN9V}l!?R2%$H8G{?DlqjI=(7o
zqv4190gVj}K~`mB9FiTqZKNd~omRtOachh6^470kA1XZDHaR&t1$!JgaA2}-Ty3}h
zSWQuWchLr^NP9j229FOovZ9xQULqjv2+dpTiavb!^4k(>UY@=>^VO?Y$+jjl9lO#*
zjs?JWDLXi1-r>|tw;J)c<6Xg+kg=L59&R*KIK<?cD_!|L8u~aBS$mJv+Xd8MAJc2?
z+O@vKom>BU8(26{K$C}}nN>hBB={qregpgUFn4>BHY%zMa6A_5#E-=+@#&m)CK$|_
z=#8vp>y{<wV8BAz8JyW(G(%!?hEnWG3339)+6ysLRZ%e=eKHi^TLn~yrBUt0Dso>s
zKxS-`y%tsR{%iK7wWNLNAN_<A{Y%50Uu_2~B=%itG|A7)+X9JPkbXESk;5~>P7o%6
zD$p4@KRFhANJ^k$)va42Ce)JMbWDlSlwL_wP0><CUY->NSO^nlDKHVC(<3dEq`Bj=
zt(tA1UI2f5*dEYTkjle53~HU1%+EYePY}mOm2MTbUGN{;wCv|k46cz9cX9vKEz-~X
z*{eOIpD{__MaX7yvHUzh^X~yt#$5cH56c4cf8|5&gKHtmiJVe69v1u?O0d8LBfTVo
z22&q#yT=(o7Rm-|qzsiKUV3m9yMqY=gD}L8H~eIXM7#KskFpfBLMZDuu7i5AQJ=C!
zYuWWsCd9WBq@StcTbAiD^+^l6ndgvIB;y0jArgqA${;&*RFH*018%h`!)LK%ncb`8
zkxdcgL{59^&s8wy5>k#Vgdp2V9e4wZC1wD!vk|<3@)5$}O%H^^^lI75Qm%^#%pND5
zsg4P}&bK%+!Z`z1`S7pkw=6EbIME@=^v%L*R7WGxsg0G()D?<Wv#}5e(m>=lwUWj7
zVH2NUi$`>~9FeKjQB-Q%-ffEoa}E$i!@%p?5g8AkG^MqDR>OsD|M;wL{?YPmF(kST
zbw+u}o*mT6@g7vR&_OuPy5m$QWG}LCh<O_>f-xuRD`h%5Z_BL1{|jk@aGdqxRy#q8
zCPVceM*uagD^ie>_sZ$UH?3hVx<MILlHZ_44G|8>fvVLu{V>Obv(mkZ^#O8nF5WF^
zdNcN@RpSa=R<ZW{FFL}Fl3zd7u9Cx^R3i@(U3A$80>!y4b2p#?TLB5u_OHtH2qsFW
zkD=S^OsaIp89feE4QgmPQ{~Qzv)lc|RC@0yN>edHW9egahk-m|tjC;ySa#i}>^R~6
zFYXs#YIyeeGX4Zo520+|q?ZwO)(*IQF}KI7xPY9?cn94m|DIBL{UL`nEa3fsdE2YD
zieKYPutJ)U%Pqer=guI6@SajeXWW*}ZmQA|_lpzP)6A>|Pe-O-LXC&Xc6NeWmm0G=
z;?h@1Z`aY*Z_eUxf)A_V2<q>reBhw8cTdRbI2)vCUGXyO^eyGtP_+T`r*fs~sgviH
zjbw2m;p8*b;Z{_M<2)!MfuF~_6(h`2IOr`?fBnC=;;{7AQSN>RZ?JzU-j{E7$>_dV
zoBa(fA*cbp|Mk3dgj}J1MNEj9A#cH_W$q?4<I!bg$h^|R-~N5cXC}H@3=*TFHizY)
z$+!6G1_3L$GWtJN{<NCS8f`6(h!{=eQH8;_nVb>GNxj+_Mt?bkkv1z%O^x_f$Xk~p
z^F=WRE*68C6Dg2Gy+}sE+gr$0;n=khnLHi1b+ubT_SYcmG*crY$Mr5}BIMlA{7IU$
zTqYMoEqHSkdw@BO&<-+BhZr~F@v_k)52S*MIgM;u*ZQ?|V4X@LL7qQ7o&N;?BeILs
zx1YHJPT~#Am=J^8En3aDN}IyLrq01Krz7WkO7Jzi@$it@=8R=BO#oZb7aKCyjLOPh
zo(h(0*$i~Uvv2;Ke?1KtgmS2O&k+sj8|Cn)1Xf)^R)U`)lnurT!D@6n1V_p4NVjO<
zuJ-|{+3-OqkcfZ;kUq^~vNWRpq@H*Z@i`NPN_wOlJN?-a;=p5V&A)%v({Xtqau@4A
zn~CRj_9)rha?`v1XBkOtpq@&|8s+VdkcBvM#E0I4wfmW6xD2PwpV>%%w|V*Aeh08_
zd7jU}Q3S7n^F<gIf(H=a^3C2al!Ve!wsZsRt;hn&M;3N6FV@e<$ey=k))_fbAVpYE
z387%j)OT(&*#KhKX%wLy(2%VChm=vyUd7TB?L2;6e$oY<yOMvtC8GDc!ev>HwE8we
z!9sv-8C{p;xvbjMe<P@}S=zY4RwsC)E#@+!2XFMRVpI+;3obF+dr)!?@d+^w_e1VN
zPFucEJ5-7fy%ZX-65NBfs8v54?2&PbC^cqU%T`gY{M;|v&N6E6%l#2vSkiwzUqe+h
z&|$uV^B^-A6NsR+-lr`;>~(M@pGcyaF)!iYE>%O?R%L<TcFAP&!}e^{>tJhyTR!?E
z9cqIGL(txJ#WME3aspDb8D2C1{=v)|V~J4p1J*+ZhnO%-PY3ew9NxI+3r4e`YCmKz
z$#@@g4IE@+uWv?C6J0>e)kHt4U*k{zX;Q(Q1L%fPjqcLqp>Q?av1!q#YoE399?&JY
zWsV<mWES5t_wO$*V@nkAAxEl$8hpMqVO_q2Nu_vVT}6m&m2b^~jQOQcp?y;+mIC&r
zQQ6g!1n|ZLjqX&=BPbcdeX0=ULkd=yW5}rQ`=&0_KkcQ?+2A4+WT3{}LIU>IAvHpk
zU)d57I-x{Zaz)brBHy$=9qbFv8`+!F9385j)tBkUb%j+(H#KlFxk|}F>?UpJg-}Av
z?^|+B_AsyOOhw*MRTqULs@hLK?}s1Qk4M|nWVTlB`(q!S4lUD4{_1lT$pLtVJDfmx
zk);&x-m8T7{6cEQRM;b#lVqSIo*sE?`{fOq@4_3UcnL-~>q{4t^(s!k<=xx#^r^Wo
zSSl>>?K6q>!nZ^Wd~p@)s>=@C*r<hWs5a)wvhcd~fPc}K8Z=tpy=~MVa{7xCYuV-L
zQx8R*n`(3o?i;AXreD4^N|Px>jCV6P!&wIc7q3E%YTap^50IH<S@j<};Pcy{8<k|N
z%cLMcXc&ik8Oe`YPjqBflpBOQ{k1HO2ogsgR>|O8)9j6ZnK?AE1p#r3-cPJ$`P~b=
zY-Z>g<P6fM!F)(kN_}K0kU&m{#j@QvcmMfUG*XbIV0dKNdOo#e{3^Q>Bl#nQ&jm5a
z3>;NH8D_kL>Gw^W%mfD~A12p{2S^J?m<>>)@B4KfBU}n8BbhS~?OK;2o9jNmS!2@F
zn8{H82!X}P!ycKO*KkrBW<mfZSc71SP6ok5K8di4d3yTE_tG7Xq3az=2m#k&ScS}y
zz*#Ar%oaPOFYTr$-W}n=ju6fQ(SQs>d)x*r1dQ}=We*V#kn4;-rl)>Yn*a+MVUG+>
zjP;2=zwu$3JG^MhbcT!E>OR|*z#pz+-`q?6il*K4JE=yY<#%dWf-UA0f>IX+fe6SA
zh$$Ys<qvm!5AY@uE((k~yd^{Cp<@)AN3FVKvN)%=<G+|@Ba%<5$xi+`4hVg~&l4+`
zIE@Gp=e5I8sos7_sq?bWNQnhX;0P}sWTssUkwudwQU|c~qWJwpBa)E`m=&&;yRK;2
z;=WwVb~-eqswT3i|ExpDnAl0vGQ~EnKfROA<RV^o;xBKdGZ|H($VUa+%IKKAA37*G
zjbPPtziItTYD=L;2zLr;Q^#5STjnTZPU@bGpu3VXGC28Qcf7IcS!4aR$szGtgwM|A
zLFgbc5(z<+dKOGm{!r=N1;ASKpe=gjVk-eK&9)eGq-M}i_U1>I2bBXG$`P9Jud+g5
z-LHE@Y5|P)IKLBpvS9UGC|I~nON48^v^XRu_x3ula-^0B;g^iT>rDORAMN(%YB<RS
z{TZAGFg=Yk3q_N>X@d*0w^^UonWXPxdLqRdI23|at3JT2gd_0=e1|p5wTAm>5V#dv
zlwE$wmxn$!zcXHT51uMOiFI|SR?9#fIq@;z`&@!$1k3pzEa{MKzFghHALi!|aOW)8
z2QBg_&gdJD<*gc!1{PLjmsLjpdSk@N<$G552-k$i@ZI`%IBYm8m7|Rib~CBSr*Hm<
zrU4Cs<>QPWnD<~)gp&s`EOh=fzaSKD?c5Llh!FNOoe($LYL;xluOh8xHnN-2riQI5
zOr=+(QQ@F6x`C8@aoN*$x#rd3&*`$OPAq$o3by@f_9|q-eOh!)?VQZQ4r)CB3^mcl
zZ)|Wtu9K9~y={=|X*~3ov9wu|cehePf2udc?EKQlZ1GJ+?*pccpvF82PwalO9GVV*
z=sb2^t89bo_9ealtTVfheop#xCCU+e5=S37E5^A?$&R6BAku8j0`Y4p;d{VE7J_z2
zU%gtL`R$s|hRg`a$=9$ciE~$X+{ldS^>`ubEYd{Ur~QSGwF_vt-ChNflsU-EPVgI~
zU;z>{5c&lDWiIQ)6~46dJ+o)hr#8U!OpXTZ(Q(<R_79^A0uPfiMXYtNX<uA=&Sh$m
zBKg!8DTAYKv%wusE*!2=9rbc{%9n=zVrIOJuF7Gb-d)IgbrA+(C-O)K=eoXd9>QzB
zZbm-6`Ds&U+T%^0C>H-mc%V1rint|t2dLAn3zIqJ?eUUAU)pb9&?PyKY&lGJQa7j)
z5h4+kwbAS&-##TN_vA)=ew!qaEy^^RL%QE4Xj@_0*EdQIF#`}ckBHB?)M>*>b!H2!
zl`))>p#8K)_88HJUVJeEr_O%px|t?3z;d(ogjIYh;+B5XY5xKlN^q`olMQWtMtC@G
zMOtu0o*@g)z*P1naaO^$7G(u5I%RrWVm`fr7M;v_3khv8E{HZ%?W+EipA9!41$Kf0
zNPho<j%-{?(1T)!CkD(L3wt<Oi5yZ3y`4lbW%6Kp%FTZeiq{m$e8<s;9DE4>#fcY0
z50Hl|kvT+QKsk!AX(q=Hu2L6~VU<(Cfl|t1=;h~7aaDT(*s%ZuEW_B4c@f#JJLMl6
z1{c{0u0y~ky*gR5Q8t!Z@cbLv8IaEHWu}vqvN)h_oy8f06f6Z*Bf7G6ian!OvvbEl
zKqpv)6zl}sVUIgEwS8q8?5R4=uQTemBLM@gX+=<$Jy2NASj8uTls>4E%=(;u1L>DX
z;vihtdhx_P&M(%+rFbtS08u*q2gzr2CbRBvbYT^PXY^^3vF}U1xmvX`Q*g~wcX8Jr
zy#QZCeHFfaM*_vC;dncMP$XCiVyp?dJ+&lb$`|7Cze=U^6-SgT1RW5S<Ywk0j?iU)
zIdkn}MTPV--nZ~t62SmQzno=mB?Npa_s4XLGRt@lEpxe-8ddt@oW9JGxoK1TVJ)jN
zy-ldcT~>nbOO1<;x*li0(5b}(%ysE)dr3;Y1XB~H_G4wdy>5_fzvQNMfIqT>x)V85
zsXsR+t5sjRF%*@?(Sx<LY0KtrV7D@ee@37!t(fM<E$5ew3Ef_>c|&xfQFOh_$Dw7Z
z`xUq-PlVzA`(OVLZ0}j#LCV>?VLO1+C{=S=&hBTD_Dq`AC^9Plw3)9~wn{!!OegCN
zX&@bVXZ{DMQICO6;a=FaYu9y9Za0^vSECvRhhZteGWg-cMKpReK0bcq<^GzP3P`(<
z8|j0PA<?J7!9i-PiLlT0v1;^KZjQ=n)(oCoW2(tiKxo;$yeQwt)kc}ELkPiy`3zuA
zVIdN{jL7f(f4C|Xi5xmI=|nAhI;WsotIDe}9t+`y-4e)$K=u5IPgBxuILpLm`ZV17
zwiYj3ng}&)45jv~+Dsf7PCPTw&7CRA<<7T(^73r^Mgz{Bhlm?Ja%Sv0fUD+S=Y|x!
zQ^UKf0((uJSTn9|W=n)CGPn1(_<O-GH&`0DmPE2iAbEaz4AO1|c`HzkrfMxm)=Ue`
zKEs;Ho0^)Uw$0FZy%Ywt>5A}wKzE#1AS=jKPf%4-BEU%fP`uGZf3h2f83BsUwg|Q>
zOhSMC`n8cnSTohI9Yuy>RioxoLuyr|i;uDYUa*>vvJp#~t&LxcXV)}WoA(C^3(rLJ
zLk-KbbPWv+OTglFN#3CRov-y|@Y;5g_fW%aiW=@DD4<_JG{OxtholHzdV5w9=gp;J
z;#@L>roI1V??fucuVr2Vuw_3e+DG-*gi*hr7s^CtJRhW};ns-xKQqwvT7P15>O&L%
z3_~b-WTDF~P+cc^69_a*^`}=H97fMN%)A4@g@@x@nQVrUM<6vfWMaQ*b$a@Y^e8Xy
zmT!-A6E2&Y`n(*mB4r$y30oiLu1e;_-`nfN&VXF2F5|AavlNQB^P6vlbjvor-1WuY
zZBaZ2hp-dVQ&e3ZU?45GhzsF9aH_zD5ief6_`yc{pfgeq(8}v1T2&F31w?Pw)>Om+
z6~9x(EF+)x$2QnR8vB+uea;!qI)|~9n+uzLrw|>HtG#0Omh^I)RqH7G0V<J>{Z6PL
zc_vvkO#uacHvmrY@*MIPub|poo9n1Mnf;?B4KugGRya5m{>C*7b4QaeEeVF38Uc8L
z++-br;KOJ(p&ei{yAT4viWkBVE&x27!8XT2ircms8Zp<T9}cWJ0P6w3?pj{n;vtZb
zAJ%rQ8s9vc5z<<P^WgK<r_LUbt8=$@a6o&E1;iZ2@aq<;qdgasCKo|<MNNTr9nMPT
z+=k=Rh726u1<cuz&Qkz5sLLBD^=jk{M2kCkqX3h`l8i&<rvSvB3FU`R^x+tgGXT-{
z0(J8W4xYS1>BBV5#42H4d;m~Rk&2?C0Rdp{6g#tJBk)KTkeq~V_Z=C%2G~HHUd_L&
z%PZ#2my8T`mbd8X=wPHU*E|#*0iZTE;r&gGjh$hVS%5u;O;nFA?tlSCBpwOut;#GV
zif7F%1Xqc*Y^i=+dj891v=r~V|9oVo*yH|B6Q~H&<+>;d;KQmDsws@;PojM*fre*J
z8x(uXs&v>k3qh`R-`o&9c8%NIJvTRpT`Mn7;(UbJhQw%qU>@QR)`nJ9H#9Vurf^(=
zhKxhbr5Bz{koArZGc8On&_+PXZ~q9;|4x!_LY;Wr&rI9}Q6{)JxmS(=(QzKLijiFC
z%yZCB%4n-S&e%=lLjd@T70ZvH4^EJuA37CWlzeDt8p%lIh~X-Y?t3?1P;<y9LtY(|
zcTeY3LHTLg=2VhLv9JbG+`9F{L}1ptcke*m-P6-^P^g0kT@E!K6)ansTwCu^j8N^3
zTRI!ux~9A2{IiL?Ja!5Mz;u-1Ovjl73j`qBoAaHDqs$!~G@$AU+H|Y9A!}qS*d_qs
z{y&1{K<Ve*4ai(0l||j@+x{^WgdKZZikB20OHr=!_K%;CN1%J6)of|O>TwnYAj{=d
z=mAqVhfZC9%q19R+5(QMV+XpplBL-R)7GN$`<v$Y)w-9y^WTH^d2o3Jg{GO<`S|(f
zlh<g0FZyme#BQ$IM+j~y8{D~9`x>F+!Md;Ze3htw9bNKOSAA8g5xjfH>eYK*c_;7O
z_e$VcqWQh%q0&@!U17J`$lCVvNUHUOH%4cGKN?~-+H5e2tL70Ad=zfPpucf+YdJ<k
zLj#rNy|gG1FMW71R6D#i%81EdzhQ&>6(EEY{pEgbR!CpehgT1<S%bZ@cqGNi%x`k>
zBpTo5<z8{F)<?l{#$%Ov#<nBx=M=VX-=0pNe7cT&))E&9LKJI_>VwRHjVi`-F+fnG
zhc5<hR8TO3m{u-@UE1S4m0bBG(uYt(hEsr~(Eep3md_^cVA^%KsM114ZMh$0sjO8I
zSvNx4N)|eC)QAQ8u+G7QS8R}zg@HzfhCMcdB6dK1Lt`W3h+TTvg$wU_Fq@vTTr17^
zIchCUj4WJ4>>!YkvwTNZEGN?Yu7Vp3a_aDRAOaL3ZfPJOq-Q+B$6>TTcrBNS9OJ-{
zwzf=Gz5BgKwK15LpHtcoes$u!KG8crnb_@n07g9|7Ra*T7+ggXRTkRG?Te3*dAs`$
zRxg*BHgvY>ML21f^(hm_Wkc$IdEFL+MIjEic00|#U9OgFe)ipRVhlzr(u8IW)s6@r
z!u<<zf^e>gne`i<mss}Ny_ROH*tOFcON|gqP1FvNZ3MN@L*iu_>7FM)`qfcS4cGuI
zCJj=x5j^nuAbaJ}q5Z6FG}dfYo9p9QnM>J*u3pK`Wrq*xR4&(6Bbqi(vdby`6l<RB
zkwa|EYMLJ;BC!|aeJ;wTAFO!Q7F!@0RWdoC1xM{>+CeyfL>J$P>TX<~)G-n8Bs$Bx
zb7iqf3P|v_h~LVBtGZv_w8aP|(;fP$oEuWQ>LH4-bUEtYoI_s0Kc&i*?YO5()6|`1
zJZ=8y=YGdi61QK^S5YAYZKeTiiKHV0114kN`SfLWG}y1E2`w?sNRHC*!aXTk#Q&~B
z#>$clpYS(tOp^$@<_cI6=RGuFQATeIlr64cw`6?;OIoUSfazJ@XAsqi%GZ_{M+EdS
znKF+e3Tg)B3Cnzt4JD^>g5V-6!5zp5<SUP6J;VrM2lE+1*rR)uul6PP#pK9dpf^|I
zxEs}Qk?5u5|6qaHr>Z6N3KsiSuW8wr^UFu}POK~{tmWmIR>`)b<kwHSgZ=Hjrk-Fl
z)=^Qm*!VMG?}f8f=*%28!r4&%Qg$^sV_s2OtEh+UtjqQ?pZRs75MziL4>9hf#uiGP
zReNP3cJB0A@i~?dxoj!83MJSIx|<6c#l!5`W7Q#p;%S)P{?{J14%#3J>Y`GJrF?qn
zx{hv$FxZt05R2w!j?4~t-kmqNE-YC}<_!3a9P<Z7-lgV!ZId$)*-mg<?V=1B@qF(5
zON_82dc0Of<hhkrij!{+*b&Fha52IaNj-S>bk-Zb;caEdU$VEFfiW2=QN{JkB_&uY
zEpbq>tsr3ClW;43kKTW9Se3Nki}1AudtC9Xs{drKuqv7w6JXZYyNLO7*?>Eyd6Zh8
z6|ATniX`_-2G{zbjcQ{8sI|P>`7b0Di$>#<*SJw@DDE#Vqh$1tyaFV7)$;7XYi~vN
z+I=cr&W1N<1L>tk_L?;@T3=;0Db@xc9OeI@mI4P^V3r^<EO_BOLqXh9yzLNqHDDZD
z|D<I7Y-!O)$j(Ag4Jp6^CVOf-($W0ynpov=cl>XEiAe@1h@-3oF_2OGk=eRFgFa$3
zB32XGlY3<a+H<F6yW$FV7!!`fY8Yt4Fn~5#Gp#a_55AVPbI@cw2~wR{-jWaHQQEI%
zi>rfoelDl%ULmVn9X<Z_GT6J#@((5aYy@w7w_K|{l=wfboZ8=0{Auz3=EYy$?{8lG
z%?mlXzwL!w%-^<C?$xRPPyL90aJ5mePF#F^)$DlpbRyK2TUlg>IfFFGuU)%1nU`NQ
zKQYeL2W5w#!kG<~T^j13;n50I(NJv;pfjAlgE^pqv&d-|jlHQ|4*QLacz%=dadGR%
z*|pR&0YtmPnb+_H!nq2295ykL%!E^kj9tY8pzXLHRE?6subz08M-4zg_gFkgY5iJ?
z@SuG-A8~UIAZ7>45{Dz*w&ECJ-PKiXwgL`>`VdYX4KvJi)Eq8=(AFSCRZ!QAm33Q&
zBUjgH(<UoH7jD{Es3FQiAYa*~4+=Kyj|^6e{=fnP8GuTNM_3tK7E{)p&k3uOCQf^@
zX?x(uIXS9SP)#U?a)*Y7#Gv8>LS-H1^&Z~rPv-x`TO$bx@$umaM`lYQzo-#ia&_;!
zRlIPqxqo4Dl6}+_ZdYNf?bBXaF7braZa7#w+Y`{%)V!}hVd8D+(1>$R0OBN^?QP7<
zNwg<<`}^lX0}AoQyR&IT0%4)Y6uVs-dUzhQ<;LhUkCPrH3hJwE!>Rq;NUB-L@9z>c
za2<D>4tBUi?_u&lzt@s20$40|uI!TUa3M1e^~*w@lz~AE)>HGIeJhXWVa5_&y?WKv
zzYrD-K5N{lZ1|}P^h*!)g92s<iI&?RWVVBOE1*zYMOm2_)c0g+971!$sp%j?kHVXY
z#SavS9|n@`RXW-X6B<E*BRT)lkxQlv?-Z8o8QeqR=QQI~&(We1RJ~&FJsmJ9iD*#R
z$F`3n|Fk=s5CEz@(OBBBtuO`~Sg2RowJIbN1qwtR9FDh6dE9Ztd$cu^0#a+qMTAn(
z%VTMXd^LxwL04}ktaIeZToGg>r6LT`0nPBA>|dhQ3^bUnpkdl%>6g7lZ>cX9O?7WN
z*Oyc@fF$#kCwOz#)T&!v<8|*9%}OGgbfWwV`Br695lxw0EBNm`DT!s_MblW&wfj!U
zSb8zEY11Z9bbUI`jlFTB2UfvP4%Xu}n4qwR?r=dJK5!to9?KB-ql-YB@TXS~j-!-7
zWyJ6D<;zRsf}9%An!gXb<^J6`<#WbE^4~YUz4}}?%5y4C^@tvL8(MQXy;hMn+7d?Z
zifm&~Y#GfD03GSE>KYmYB#u)NeSVA&TBOgHbj_#Pzg$;TGnQAZ#!TAp>Olt?f3otC
z;6X*0O`u5vD_zj3X=!6hXalKS-0ucu0_^u^CMq^Y<UD>%2R+$$pad(y%kEJt%~4_3
zpk$5^W@BqhS7VL>$)zIRyq-#-t<EV+(WiF-RYCoczx4-2S0xJColn$>QyPs&=9PSE
zJ*88}bh|Y5U(F63N%e~|VA7!ko{19A7HB3~9nSAG94Y92@ZbU2Qt-(!f5r+PGj8c}
z32^pKmI?az<A5GepL}ek$;gDRU%wtr1*bkjNZ#Ow2N=o+@^g=q-&++Z?sz>WtuQ@I
z#GEXgc`m)u@NQ@{3qL_YKqY-*{#he9B~rGBxeF|mrh82`y=})OWq_@Xb><J84FDj+
zG#GfB7f!XcwPkw*5KkjSP>P!}Xej7oGMUL5-Hhm^MKeRe9FSH8H4zVD=pH&0IsmHQ
zUCY`}Q8f(xkZCoSBwe>00<DrRd{)#JO*iFTvJ*F@qv@g>kPyb$B4~A7Z~b+7b7^(6
z{q@V`S`DhwrEr52ia66!-WL7Q3N`=sy6VE9tCePSDcC)yWc%4p7cvc$7SY<>+hzH?
z>_z{7{O1V#5`&dXyAT>$QR&x~l>^~y3&V(@7~y6Ljtj&(xrcRP6DV{bI`V*N?_9X2
z5LLT3mSs7K0)`W4kS_v9Os<zVtyfS$7gm~1Z|z*X4E%_Ah_|lbZ0!|Et_yb;@~Hh|
zMrj}MS~f>lq0tXi<LPjglcl09$oMGgqH~2a`++7?lgiO0eK%rbSz*+Ejb!_%gUmL7
zkRn`C9Lr_E@ZiCzgphl#&BT99UXME#4-Njb28f|zN=Al~oq%ItJCMX#u+0e>`WH8V
z(z^xfCRR*mC!%fb?946z$3#)V5x{%xtTfvUU8-HU57pJxi<~c-d1h}{QDNH$5CeSa
z9e~U^{N47nHsgJL;0S<9zkj7^WSmtMfJT<W-x#jWNeQRc>N9H-t}sbJH^qMUoI5^~
z)8R6)@#rQcp!MCUUt<}sc;Ri_P(x7vbB5U|oB~L-`O3m|Skq?@)2Gmpo5Lz_4q?r3
zSqjoqBqiVk2$@sk!euCflLl?CUlu!;j}&Fv<}^orjXnQPi$3=r-{~^o5<r~52-n!I
zCF9OL0jl7b-`SG{WhXiTt)pTMGrKxEI)Eu^Q%kij^8qIkMzzs)=S7M3rjC}~#(H{T
z#j^LI7*U@7{$n0+6a@?W0R$-D@1DT}@s2^1;&mlv!G@GZ0L%pdkca|$T_YpsO^w3G
zpNM{L&+ZRGkvJM8<FfXTg!vP_)VcS!_y_g$W_S>=ZZ72^qpe08$wtQ4DT$aaP<N+q
z3S4LVgOm<4Ll7^-LHw+gl(}Ym{Ug8=Ty@N^oM#K^!lBuTe*A%h>37@HrA+^Jpm7#3
zTdbaIo`t6LYFkVFzMqY?^@6A@^gG-0z)QCScs?sD>P5V=5xwG$PGgP2wKFHF{ZT%M
zw)Cva_vL4Ym>1R;8T(E?H7)|4ds;`=O8Bt+X}|YK(_%he%!{hP4&QTH3_)sP9TO9i
zw+V*t&Q?lo9q=I@9YGZ7ksb2hY}%7f|4xxrRtkt^-r2t#l9!iXYA>X)d)&unJ*!2{
z&!!ZWl=_k%_2OH6Ks9CZJNFVK(S9QY7lzd{MtG(_-#DVrG|I`zK`U?DHh+}buhIi5
z_-Vrv9hVyhUk*iYZRS6K?JY%vZW_QAh7~&QO?B>A`1kF&xP@074u^>JO)df-hZ?DQ
z+%=yJ7JeFbR1<|3&SXj<AiT^Gs1jqcKRwbg`8WYUfJKoD_ZBer$E`GB!ITSk=^Tij
z#GouctL-DMJvci7F`}Mtb{dzSTRivj^BTZOHlw#elxS^c6NrwifJ0(efSv3GXJZaT
zbUFJ;ybEVG!*oom!KhNeZI5$T4ufNRU3GPJ?f!)cTG=$L)cx<P82{JAb<b)L%lQE_
zfC7T*qB$&r4M8Mj1-1r@P$qgQd_I=fqgUxB(Ov7LwL-fY_>Um8aK5?~obBwh53oVx
zg%4|i^=GNWlwOmuXC0W)?xzWRvyCT4($P`F4Y#*zAIUPSd*^+L(u<WIYhc{v_;I0Q
z^>T9S`la6sfcF|H&rbAXUujg%%rB<T8%-xNjG50HyT|R?DlfkTOm#&FUZB+;ZQ442
z|1M$V{1u?YK+JdnBw7$ayYBI3n_0y!+y@YY%yC$}qKt%7C4R6LvzbGJK<_0=dcN$+
zepFeVc*nV*Fc9qBfRfckD1e@gYT$k>oau(Mz`oQl!DC&($Bs&UV{Eg_`_Ymp_Q2Nb
z&`OUKCrzLK5=TAz8o+*$LCAjCqY?zJjC3Ns)1OH1<(WnS?@w0RzFnLQupS3#<nzqG
z5S?FXiUAlD-wtyuUXc7%#$`Z)+35h{CFCyq0qE=U&>Ths^Z;!bpSO0p7<KOZK%j>S
zK(Ot!vyDSR02I0S-K)!{(<KqLeCJV?3*U&_kq2>I3%g8^KKgq1>t|ZV`IoTT*#!f@
z6QeY$1Ao}uj8$Eo^P8cmX_Py%B`z+G7J<PCes$p%C)sDQz&S!cYX4%m=5QDgMf&CV
z0sHYLNz3UmowZIo6uJP2(90Wza3Pw*(vUn_PrLwP3SjX9Q*mE48fd&~1^w(lFjavB
zZQ&gGR}Toy3VOhi&kgWIGRGXGDO<kpRrrtt_^J;_{lI(xC5IYjmhNjeI|>NNIA~xX
znt}KLbI~?<KUW7f`t19wXEHp%4o^iYbW2Y7syq@SAYwMF7QEccD8Dk)pbJpJrQ?WD
zz1lUubN(*CABkQ#{JNYQ(hwud1m9{aXsQIq$V82~xw+FL_*3mS`{e6$e@p&$fe)qQ
zW9J1&XGL4D?@d>v9Z0Z$sgKu<XiMi33TLuv`?a;IZUdX~+J(yp#~A>)ekp!hp#$KG
zD%-#&pV1r^1{5mT?Hn9VJQYB+(e!cfP1;;dh=`cW25(e0jtCYLnsh$(f@NgGWxg$G
z0|u$Qr<keK*yp_fxmp9<%Tf(EZaFeJ#kLWc9}evKv<|qGhaiB;hO^4uB>z6c1SVV`
zw8YXL^aqhsIrpT;iES~%uU}?-ApRrcUoXK4K~**{oP0V!!Us8BYIiw$cXmdiIK(U1
zXA`iSkN-<tx=Av(Gfx{}kLZHRMHgg%7iyTSRYfjz%y$j2$?w$Je(Oxb@4x?^g8&Pl
zPz=%!J6pk1h;KX;r>g@{Y%Cz}HXw5hv_KF+=42OeR^o!_0;Eo;>S2Bm24S_Uc}2(p
z7~pE4?M?aUv+pLsim5wbOU=w`TlTOPwz`ONg0;!kH|M^lrKak~{&crH)z7l&{qprX
zdjMSu)`ANAm&G8PrK+U@(?kT$%TBB2!>>#2>)fM{nz05m%e@)~f7vzo6D|Wcl8Va8
z$mmPvly27#GRDsyZ<fZ%N%n92D@SN6j-;`RU(k)1XDl_fl~h%uc^AY_&Of_R@4^j1
z9(B!&o_DRX%HA^Qk(KI9TLSj^R%O?gB>M*cLSUQt`IR13;k+J&gxj}UD%oLXtltbB
zO(g!SY}HJsgERaKm->1^FALR|OwPzKW>E0PT;?-NV9!2=Q}tU-U$+6{vJu$(sO-eM
zz=~Ugb5few(j>ZFdV-o^v>?|CU`N&bG+<MjpKo<L-~8$Nnyo4-i!<zPhshw@NC*8X
zv$uh{sFa$ECXg3e3OYe#<l!qZeD<wf`9igA&UERsM)MzkYz6U3-7N5KBx7(IeqoA0
zd`rVkiz>T-19Y3J<rC3-fcroG@y8$627!(kd5TM6laB7Qxpg4HYxEVavULD`3iDX7
zuL#s;hoMN<Fp35z(^YL9Ri!@U2{4DpZ_ZdaD*|PVlCc^XT5_bBDHa4F?%=Q`El^q|
z2&(z`c%7A|lUlVCveff2bTL$s`c2cXAm+Hk0XF7$0Oc|&16W15QvIxCdEm+RFV#S^
zGzfq23E*TcH@Nz_R6)Hz$9Q9s2%Ey0Fz^I%KuO`Wt9`?c^3?iu>q<Z`k3PuFJbz0R
z%mq>KUetbDK1+ajqbR*Z!pKGkc~Su|lh~T8KspE)_zE4sXo333_--3B(5cav0`&)B
zZ4egA1;qeMQ<^xQKO-5asKo(mY=%vYs?)p#f=v{R`3Sy~V6W*+uARxRo&FL_kbVt$
z&%PTkZ!0p7^E<)=@xy{NA!!0;Fgm|-@d~0Z{l<do{vba-yH;~pjN+Qof$lTT(LJ6D
z09XC0TnZ`o>@jm}X#<aWd;)YyRdJu%Eh-3qV9nt!7jBaM%X?s3dYd{GyzX?nOruuD
z-M(#u0(}>Dd$Ns9n0oMIQeO&3PhLTVa8&z|OTpFV1^)&G8mN{qmU2RJ+N%#UcahwZ
z0E`)n^to78*CYx)fiHLIlV$4om68}A&_&eK586kTGC?S(nrZ9Th7rok0Y^gt)Wxdd
z0+3?pqk6R-a~3jVL`mk!$^bg|Xf=I`ss8ZmEmMyG2~%3=A^J|(1m#yo?PT6UGWtJ;
zs*mLQ9upo5-t!Sy@Qjp{2r=dc_2$8^{&(5dzFa@ZAcZ3so&XcIw1fB@9E$oz|Bkl<
zQonp{*D|j2UtV+0*&~VLiI|N^2rS^@0dN`!SE*HdI;M6?zjC1mT6)v#fjm-rnCbVa
zr+Q&4YrPcE#;S`fkySALE~l8ZKq&67?$c(16N6+(j4uY6%s%Er{ERG@%m|OtfNcZ~
z(5Sb%Qeg-!>n#;H)neu0w5E_xcZANh0gEHMNBLxqm3gjkl3N*)h#xcd@#k-oliOBM
z*{2Lr{WE>w3k3Fo?$tgM-GK*fmV$Y<NH`nC&f<{Z)|Texw}8-Y>WC(+QvP!l|1Kv4
zR!QVEG&jGOJn&F;^ihwXLri_Yd!Sy270#mAc!le=m;eNvO%TC!%Fm{<q&-8hv2Cyr
zj`E71R?}_L>wuQ-q^fNnNy`>D8qBWm_i7pJ?=r#?eXn)@6iF!V^NWAD9(=`<_gVn7
zDfNc>b`@k6s*B3mi_dg6tz;P#;qB19rmf-pEbk|jr8tPxczSkHxZqVNStnO_Nd}2G
z&R&IL<1rH<wSoG&3TEj|tjJymxb;YUc$?XUCGdb29Ikr0(Fid%R{3qMT(FB2va2gq
z-v7h~SyjtpH=E$~fdeUxHvWOj;0B_boA<pRnref!TCFD+Pza%lm5>18&ii(KS|c}j
zXb0H-7%wBoCoTy?Y0q6Qi}De0O0F2fMHgq0Z3OZ#Ju%g}UU;OQzCx~~S=#at>_6|1
z#xL!Q!u`~%Q*Jv+(kVCoU4mR@Hm@Hr_x1g;I4p97+%{_|u$c3>O{J7USC|Qtp55pr
zErjylK12HQ6~b1@l~{kCY!otz6!R_0Y7)Yt`9BjCP^;uF9+2Xc88um{-a4RoxNzLJ
z>y?Sprz)ND8xz{<xYhzpGK7Oc241GMRI~n>CnB$2O%qN#=}yNS0n#BY0|R+h0&Var
zSsb+PrsgQlZdPqjSDVm6fUjk}?n~p88cfC3fhEN)Qz6FqKJC8^FIpjF7!&UMbRYm(
za6noMbkLT9n;xl+Y6fOZ1r1RSL~d<f&jt~b*GH<X==xvu=Ku9eb(hALR^5LWYM7)p
R`Ga(J_@wQLqJMc_`#<$tZ;Ai_

diff --git a/reference/plot-24.png b/reference/plot-24.png
deleted file mode 100644
index 8c67ffa28f6163d933041f4ae892d70cc4de752d..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 29791
zcmeFZ30PD2wm!_MRobF*T8|?NZQVdA3d%gEt##OpEe^<R1w@(WFvp?lsRPs<1exmq
zR3RW>j1ZDQtzb|<M1l}VPy&cR0zrr&K)$s%p5dN*{`Y^+^S|Ho-S7J@&(l~!viJT?
zYrX4T?^-{7=e&ExM;ks;QBhfe?)lbLMP+HciprA1A1s4+_?cHH;GboO9d>^U|H6-a
z2a^K-d>FCEJz7O&(^>q#Kco*%2B@ffs)By|^}hJLvA&dyKDPv|S&7~72ky^}PJea&
z!TxRAD<d^Je!q?kYU_VuSo~Y`F7u50tieXE#}mgd!uG!RATr{_5~p}wBqLs)lacFM
z`+?)<4ptjhypBEW?YkU(^2O2(FW1&DyHMn%+gSOd>C?w~DatFx+cbAoXu7Apk(W(-
za)e{7LRO(I#V)ua<rCsz|3CY3@Egm%4>vlLbR2Vba+*JiW$DP}2<Z~CW5<r+7-ru>
z`(b8Zm|t6gZH2vb)?@TLOo1ie>JCXg^eA(yx`A77u~ROO=RqmqCGNzMu)6B5(UioS
zHT?0JL3F`{tIhI}i2anR#PntIEZ|m;uI@N1$P!gx$rfA}ZTIc*X}3B?Omua1ljuAb
zxRMTc71hR?8(UpnEj-HLlj=4`y7?pwQWf(H`SYpDDTPd;R7HKVMVt#4F4P^F?7e>d
zx<1ysFL>VRfj54i{`SP)wei6;bwvIhE%yH7$B(IL=HmW_Qxy(UDZ-E`=Fcaxe6D9i
zr4D`cU|6@t(jo7VUHiQw)$&@TC3cjT646!eZ43`SgGegqsrUHfb}yC8a3MRG>fe%M
zDzUmsc@Mu&{U@3-H<K;AFh(%8e8mdW^q8GH82nI2XXjL1t{fhJ?b!25KjBI7(=Z09
zP*g&l+gH)0`0e^n8td1~9ci%v6<z$>pEl`ZV<B`87uHB|?n2y?nKH!JuxL(nH<TY|
z*rX(GWWlr*yG#^Z=wp(7Gio~SG7P?as)IU73mV8tmM_=n7d=C=v$OTMUN7!tRMlii
zzh9h(Lp6DqmI{zI6mxBEafqE!%p0z-^S!q`lA)cj@Z0FRAX=M)bmGaACreNxOG|l<
z$LMJ~&phwdE_`q*X*XhPS6QsK*9%rHBya5$hM2)*=x`sTn}sDN&Yn3?Iz7YXa!aYn
zH~PYSuZZB=rhQ=yPUJviiFVCZp~TnH&Te7idUw}hMsL)k@C?@iMqV^AoOG7Q=vuRo
z=Mm#SHRj)3D%elwJ@ZgE=;TBVHfP_wdDCBPb!PW!^X=Osl@k*aMQC|AgU`tq30#{y
zVNnLqr$KbPOa4u7-@a`{0^;M}-g0gB#L7^k%EJ%NnTH7K5+rEYbPb)iHs5~>K9Y&I
zk#GG^wmy7Ml6>NBMpe+wg&k*Cwn!B9QNM=@+4JRyp|?$HeEhw9|NAfQpDipvHgDb>
zUPkxn3NbUJOf;xUyjh8{WEgasb4m4)kdU$Zmv09q#$B89bw4IHwoTeI4^E0vWT5{E
zMn;sR9}(|#nC|i@w0c|TXZbSo=}1g#ZcfguKh`Dgt(iJ8J=(<xqDi(E)g6N~Bty8b
zUcIuO5%(qaHJnnVZ0AZehiBM<_s^_Q&VK8M58$|Q(|Y;lU}IxrhPs_^-9C!mU{%9#
z=%bGr88BU=)H&Emif^$p5Blp7G^Z%T@Ns(0WTQq2H5oR;Nxv52A#Xol@7`}K>y74Q
z`ctqKr?f1SeoW;kgVCo)O%mZ!&ta~oFt5&|CO+9WG&lEo1aECxw}hC#CHUMufBrl~
zZ;*!1{z805u~RVhM#o6mhV|<ceDwz9s8~=NJ6dSCyZ4)X=iv{FGFs@o;Jaacvo!e(
zJ-BNj7E9#_wQ|BWHf{30B&xZ%Yv79R$v@y%Uz5~ZxG|K$S8n0bPLVXwoKuo+WFEOm
z_o+-#%y_E}SmirU!Qm~`E2{JV>6i?TuNM3$XJ4jKFM9qgVN&a&lne7&;vRbNk?)0`
z*jx~8x~Yri$smK#>gx2U$j{EMIdu5&%d4APKE&^MJELj-LqWszQvGEv(O<hof2~ko
zr=gMS%?RszWUuF@El(c$=r^IGRB->Sx^MiA&&Roq+T4c^A8MoJp(YzD4tn)EVLlZy
zvB#*Vy&l8K*;&(3x{A&_znV-82*1Vu{QRe9pnuHOnfzXfE3+^40OsptlqWigP%evP
zGKr8fQ0~nvPWw~Wc$`Mk%vi#HtgFl8EY_PP+S_tuYKiCFkc5cIw@$fcG#Y)unKpe6
zks%3za2DQt993(>E!3&G(KXsP{{`a%{AuzYv_5|?9B#88bJ=uNAwE)_Aa{Mx<6P2_
zmy^?zDH?S6-8Ms11uNSkKcin!C*{H}WqXYBS8lYJxY}9jMzZGmX1JCMz%)0lS(cub
zR{oqxmbeAewD{~4g|%}knHV%W5ThIM<y>#ud@t#->^zeb(P}wdCZ0~b2*)i0w^cU5
zadC0nml8PczF9Rd#pB@CuK84SeNmKGD7uffx<o~<`BP(V%g`uh0vlP@;LVI{d6UTE
zQ|maq*%x69#h;ZDxMjH$W*W?^3(jUNCkA-B_2H_z@i+w?O*P~uC9(t+hV7(FqC)fB
zmcY>Ns=)&JOc;ZIB`<Hz8Jh|W3^WAy)Wlx;Ls4HC#g6NuE58YYKaH&aJNI;G;XiOs
z%5r?eAFpm%pK#Kgo4#D2lzlv#gc=zePdz<-5$1;dMRZ9_xc5FdFgYUIQ<9nXHHJmc
z8s@aK^*3g@nq|4#-MXiSKl+v|-iUMmzF+C7r5P9k7LEJKj)erwWW&X^I`UU1XND0}
zZ*Zt+Xk4c74Kmq)N`G70!qAqhgU4xy0iT*K%o4r0diAO*ICdRfU0<r`H^eahPyE?G
z;LfKynC9ZqYDqk7lTUSb@~c3~OrXT7{ho1EJzpZz&1pG;)y4Yziyo=lzWFX3-2Yzi
z0j0q-<(H>c-GkNXRvRE3p~J#vSeAJ@52=Y~hyDwG&?tPGVXwE_)`?gtn}un`d>hgo
ze=Ralw0ebg+}PV)wCNl$dZ~%iuRB%w<8d^)U51E(%`p+ZqB5U#Zuah`f402$tWMvp
zJa2t+0ZC{E{*gkPe(Ev0huK$vR+VN%P0x(;D_k&e%|9X5jZIBt9liVa?{l7nF~Y%x
zs?+)7F07ZUEAAaiAgdD_W-Gp7&?>OAG6sL#8x`b<in8^aonhXCi%Uuhw^_VU<e%O?
zb-GXfjpXpzm78*uluZZk@Vrl%YX0yqdQ3mPqPY@}QQoSHgj3P<4Wh{k9DCP$_7m3h
zG*aQwYiDCK=Wwdc&>+OHsMj;ti8g($XsDEK7dw&ARLK*^SPeTlIi(nJgURAvdHwA4
zs3)B#%XAgo33!y5g!#wByxEi_6SHuv$GIXRr&9u>c@@5t8(No$d$zuZ&vuimz1;ki
zEba98JBOP5<Gn}A3K^S%FaB@6X;m4CjR!L-d7D~m-;;pLJWsVT=aRJ&UtRT9Q?!DA
zt<df+^uOQV)UF92Na=y^VZnA@ybo5o;bNmc^_ky|eZjQBw~Mx7+wg_06L<OJU12?U
zM&ugiVP2MR6+gj_224ZIS}rewlkc?k!g_E|02fSA4;oJ^yywobP%OqOe?RO|iGGm{
ziQDJN2%DeUOyiHl6ntmj?G?;AGRmU~bHaU@apS(aBj7Wy%I~(fx35ME_(x6B5<Xvt
z58_az@!fh)dwCP=@Vf^}O*?TUE76h1WB6_aTfi-Pd7#`&e#v9B1vMCzDTE;mer!<C
z>lYrQxzsU<Okv3V-gyw>2G(Zoi6yEllaKzO=(jq<&`IFK)EL)7;2>nDbm=6-!%-Q4
zbBN!yKjf)R--mf$wj~c4hX>O=6qwCs?p~}nSH#J5O=WkDhLT)+-PqF0v#;Gavz&20
zLsE(w?5R5P$0Qt|S+82PO14Qa4o)kEP@Wt$;YoUHqJ=LPtR;0*)oXo|C&~7D91d9`
zkY%{ebb0u){hHD^2^svQ_|5<4=RZ3G{bT!Zzx)(mNNmntF96}5M%h$RHJvxgfM95J
zPb661%D}b)H|>NJTI?6x6x6OW#Ef(s4x{SYbm5(29X%eSl4l+jd!k%0@=PMj#n6<@
z>|_7UGrx*k^vo^bnI)IqeB{R9a(UO6*Hm_m&P!i5{=ILudFPv#j@s0;AMJGuPT|9r
zv%3QZGf-E49)bh@-B88z-}>=w4pv|uV4xqL5l(4B!&cyf{uF%*;b-1QxX(Cha#sgD
ziuvsP!d`G0en*ZB<+(PW#~iiS_rY@qocB}Lw0T?mcr`g3cKO=rMAk07!8t_BIVu&)
zt^Q@3AjHf!>qbpjU#&WhJBtBk6o|krsKfYn&rxLKd~zsO=3YJIU2<d?zj=MP=)#tX
z)XDR{$x)cDZZ|WOf0uDFJ)NS9=3lvDaR+|&#NdD*2@LGodSNt<dC|3*xhzEktT?CO
zMi>l@Qqh&P10y*|uaCzi{m|Q{Q$vFwd+nVRGj7c(8w>%X;ZEm~GhG)RTr?M0q<nOj
z7<FC%fcNmy&Yr>>r1(rD>v&4^4+G*preSlhUY{o!BGmBj9f?p<adb4xH3)-nIYX3m
zdZ48!BOSl5EHeCkd<6Ypyo&-}AQETkg*fK?DA+3-5(*Yph=&1RW5PYsX#gZ!*w<y2
zsH=TV!p#H-+Ow7w8hSlR_QZw_>(Z|BX6q1j^%W~@e@3WOY6ZajgTX2HjY0^A1WROv
zn+6E;XfR4LCJdr`aJs|VHr3MT&bkS{7<nuE%c~hxaT5t$uF+<>Es9_o8a0{BfAi+e
z3fh6uak!?Nt{glfdmGmjLJ%-Pl!3;JE)G#UI6_WTa`N1Z!Co9z3cP|h{S6c74W1wE
zs-VaqAgx>~fH3dnldC(jIz0+2e+}M=UEOtkatEsUeZ<zHvRm|GUSh;eOSf__OJh^h
z%a6SWenzsdT)9VqNc1Vv1%6cM92KrN_!{izByx=IQMjkdFF83m5Hqr|N#3tFxE19g
z*#{w5)i4Ve)XDs34hGL>Zycy4jza%r1#kAo<39U{Z}4?nq!R==KpuGUgXpP};@q&r
zL->G>(mfzz(t@C#X&&11rVQep)L^?X_H9plclxuszxe9s0261JSY6EI+ZQih#9=GW
zs%xbtVmNbB$xHOE?yj3_P}Qm<1M27-85yy*%5Tl7tgKveejvSwm-tJDH2*LDn(Tbf
z?n>U(r2^Ze8Lt9BttCk;LDDc^S4U^1wwzL4CvA2EU*`!cUZI;XpA12&py5No`OF*!
zZ}wTk@TR8bs~sbG=BB38T%@9}CVF@YOnG>xhtShLp(Asvrx?ZH=#t0!n6lbas_F}e
zF`;dH3j5Vg!^G*I8u2Zt5}0Nz-Oa<i3Xq}<for}?^2=4jlr7^QmxKi#ITEIo%#Oc#
z&$T(=R2x6-a{aSs&!*9ZD&V6&M-R|Err#hfuFY$QnwE=I7NYb9mp0`Pdk&mnRlR5M
zJikE^;4tmt<kYe@NaFcCRa`Tix|O?tXe=Y8&3phJvJ1P>esqhwMS(e>lsd%Cbi+kA
zotsV0j7+}^MBw-K4?qBr!IbbChTC8tbcOfco18<|L&&@R3-&_$FUkS6fyTzhBMy4u
zPB=<y$~6o1$VKv<13MwSRM_jaxkSl$G*9K+o?x<7zHWxAS(itJdpt)1Hb!ZMa)c7~
z7RE&gJoZR?cU_NsysCkimh<_*^ZfMFPYMS;j?LRCvL}rv1h7{NKJb1~Mw_<$EE;A#
zoEM!+i5zRbtWR<cW?6CrU`V;+aZK>uwqC(0(bzn^p>5fGvlmbh8L^D$uRo!_HAKs+
zX@ZA@YiMew?t<Mo2Hv;_QFn21p(Nj|A?}`5oB%}8e(}A5G_HdoE2fOsvq!oEEyxfx
zAQViL1F+nevoJfVI6nYRz2X@-^-d)aFDbrGQD0-GsMsm$uYP@NTDG~7Q51@ij+$&3
z1(a749;wVH3*_DB4cTw4i!y{l2rH88_2Aqa3L$zCZfPTe5aLPx=jA^=1IIb)&3MEG
z1Z$gpR8!>g`7SWJB*U;Va6NFs>%QcU2Ge$+Vh8C>AVyDY^}K6yFMjTykiD}X)OZ(&
zT&P}8N2kSh3*N+0uyG}V;5UkS%>O}U0R+A^4=VGWODrudyG!Y*RlHeawBU-U8!gYx
z&!67~H1mto0}ygI>jG4@1`{$2;lENHt25$KARJ9-d0Df2_ii_&<|c18_Z$-+X8rX}
z+*f=wFficT_OY`r9GI!%;^G-y`D<VXeYc89@e}={_IidcQAzJizxdLnOH)BK&Lxo>
z)|&_L(G<;rAUlXgu)ep!ol<)RVGN;f@O%V=KR^GqC+1J!S8k4Juz=p1L_q&myFk&0
zg6o!W`QnCQ>$S32W*?vfZ`8=zn)$<bUzCH><O@fq1wV`0u(_kmJ|D)XV}aD(ySIPN
z6;-3H^DozO-iqXBz=M`~j9NSBDPJ~fgy;E)Q;4YxEUYdEP(Gr=wS*(v@sdrjzM?sV
zA;#I{YvRR7dOPn|JkZ{YNmi$sLm*uX2%;}NY>{Y>Zhfe~Aq|LGB^7BiC`wMoNy|@m
z6i~r#=4PD(Nb`?T^mfWQFw^f`^OMnOf)Ga5>LfHs$8~C^d7g!Fuby1p|Hx7opZkgD
z5De@!OzM8Lz*%+Cy7Z$uAdWL#@xcFT(X7KiQpQuE9UUF!cRXke7iVXme9>mWD!(CJ
zfDA`q5-w_(+im5pIHgF|@@bua+}gS@e4sS;S;z0hDht&L(W=RBPqpT}Ct2<06&a>C
zXo|8~fawhlOp7zBZt;M5ZYY}luky!acXNV~$Xkuy5td-8xnYE_yi%ixq0}e;z{sc)
z(N!0{tL@9gvk;vKFN=!KH>s<uGpHdJ$!{k*sQwU}Tm+a5n`=<r?rlG-)U877?BXJ9
zABo{@e9`=8KM1kf^1t0k%%z2E@cb<I_x!C(1+Kv<yD^j87MD$t-|5L#I#2lA*#|NH
z%~NgGY4k%S9hTt`d{I#!X7q5YDvrBu(b<d)X)&D6UMd245DCsm#j2`n&OJalN!N0F
zZvSv!tKcX>Iz$W|%@e<`sRmJFmUhAstOe)KCiV;*`HSmZc;`ClL;ow+*&9D_u>)m$
z*}L<!d>2FNm4Q1?z|N7VO4K)o_SOs+DCmfS8nSNAqGvN4e_sb|*Y$Sm9vdJc+8m<x
zUvG~ONQ>C?`{s_)!F7K4_*GKt*ZcNwTEE^sUvvwhFacAOtb3jV2<Zl{jC!LpXnqnm
zJ344dhr6kfk(BM)JVotgL8694db;snWt*)B)L^qJx1#oX5;5#Pzl**#QG>gKXy&>#
zWkLz1x-YB?9N^htTDWr+I9v|0+w!tCfR;2wl^_C7kDnS{2Z$^bAJHqYmN-kDD|!Lw
zDV=m~E0N&{ylKpWGprZxf(>lFmUi)COSvaQ98i=|iY41|L$Gk21PO$)h8wlCNG~h8
zFJ8Fdg^79qWuC*CzAa+ObfTo3&MwM`;)_J!MoHfXv*K|Eo7zq68Dif4-Amq2%|@-U
zVHhPxRE3Z(i=HCG+Lk|yW)Mn&=n=ZG8<Q9e)Dqvy`*9bRg~%PT#P`}-P#?2D6K10y
z$s7Ana0H9Plo=w5`xCo-e;3RUksrZsBEyLM1wzUeNfG@u7nE@rCClFd>!v;?g6iV~
zZ`!cr%|g;eMK_{M!qlS1PDY~AbliLwZn-ibKnyK*%88lT7GV)x5=znM_UyLoGRBe*
zZSzgZh$akQ#B$Ch|7oMP1RqT3HAJv=|89sDOjE6a>4MA}MUqzPg)ZXm{a<WAtx%u4
z-wO9)$=lIQHe_vg#q%P<R@b-JZHs=u_w~LeiI%5Ek{9H;h`v_!7;#~XYwCgJ5lQXW
zW&+iFwDw5Tsb#3N`)fWYdCMvy<T-R~%57U-^{?|zr)GS8p3Gc7b(faz6hwS^W6O!h
zt3w}MoAF#WPOAHbW_pybcz)yEO_R-|ck+i$-ra>|H*_2{Ve4e4gt!vt#bRoUL9N{e
z@=Ae=_mj9|Lm#mj$cW+5Leik(iBUNLNFxrox4tJyYe<UunxDH&LON(NQ`ew)^yr=)
zabw+)>}+P-Rh@2F9X4gfed?Nv^ChR)4BN~Pi5ZUYcN<jf6v#`ef4^t0KAS47bRYu3
z{>QFqwxS<xwWCk-cMUm91)Sfn_&Q4OAe{UftwY4aEl#)dGfwkn%AwGjCgRPAa~oTG
zS+ylWHpHLROj@7PiS=3#SbOi}?ZRMtblf4HtzMW^Qd+O@J0<fiD4u@zp<vr{&(pV_
z^W7e=?Aj;~>pOsHIaP&C$ev$6sD)qgIMqbC`IxSvH(S)(WJ54S$Bn;D_)}&0<86&*
zlD}vHHdn;QEBjMIvF^>kKSLn3>xyyqJf-|;lz7P7jzcbQh~ZDHi#H)p7{s}2_Wg8-
zc;dgk9B;b0H=*daTZl_@WyOhmvOf(Gf2)rS(vhD=$N^&2uC!s|vHMMeks9qL-{<@L
zi9Hsd_x5x$=cpnRAz?Ma8xHu3bIxqGCBB_^r0|hfq^z^V`eWjE=f`AYw|;NH#Ke^>
zG)yuh%fr<LXR{)mi4UY5vQ(?%km5<P>xUO3>+QF#xI!eMZN!=Q{rD=`9fYi1W5zw#
zkw-YA`t`Ie16~q`r;dP_DcSy3+(0aOE2{h=GTfta(IT{Ncm(OaJgPOFnP<89V5T0A
zEV`IR<dbWO-~Z`L|BcH5dBKBd<0iOJDo&}kSptt%{$<fzGIkZA<obui#m$h$pH~iO
zI6MG7Kx2VzlFwz)G@W-9k!j2KRd>(TL-3~QWs{001%qjvy}_){xwfdFdvw~4^{f|s
zy%YhH1SvDJtSW?1Q&{M44Ll)>z*d)7p51*<+N@83(5g36G~zG~)PN}n4k$l(E-Dg2
zA!o1sji!B+JLGkSPMQ|<9yJk{(%IwnCzfb=uo9=MsdGCa+6MTO67s_Yq*BkZl4PlL
z-fSiKX3oI(;eW&jZc%EAvsulJ7d6_#O*SYuM?dIm!Bg}r;eLQMIfr4GkSE`rb<P^c
zjYe+0EEXiOpN&p$D9V_kf>uF}hP9r&76a)6Z)^yXop$$vyUMd+R7seu%Yy_1)tXfH
z0X0TPtB$nfkn_*k5_>Z+YtY6vsp0$C4QT2YTnFmhE=&VU{uCu~1p$v=B=wBXoTbk3
z5tdH{Ll3n*;P}6n5PXBj@5#lyS_p1K58j?xjmF*epp^(>I-DrhN0Sa<6<MySQ2;}k
zS)teQ(bu~NTe?h$ZJVdY5np?0!{gN_eX~VbQRmJvasQdEFFpdp7Y<pRW1>~Sot-x#
z+sk-6AM`CBif`*@rrb%e*8^_q$Ps%ZHw%NJ5P+2XB8gpf(4b`C+!<AB3|42$J@5SB
zJ%C1ut^HHEyJtV6Z;p4yGF-!aZdHS0JvJdHk~aU5^x!K4{ApEocUn1(wiX_qzl>0#
zJ{sagOlS)4MGiEt)<592H+bF+UHBUOQcdL1qp#H0D(gM;0>=BN%8PVD&2nKSsG^th
zyWovAfGp>OtfazOdMS>X@&gteMC-y8EwBd)Geu`oIg~+&6z)H&;d+ME3}#X1o=3G3
z31XX>AR1-*+FI7DSwOemO9j~tNdtO0$~nlzYy{$^u!6qu`$_g%-GoVC(nyTd`Psrg
z$Z{teK(6K;L4>{oq|+=p)gN5<0?;oGR4-u|gJR8P0TKZAqzOkiDW7xGCMPHJ(Pm&;
z_ZMX_5O?^(gLA9HpVuW!aeekj2BbxL6NX{pt(!eKZI1a;LCH#i!sUTzYAH~xxe!8x
zLx>aKR~zfgrw6+Y`~=KMx@m4pB|s&&3ZMCKYVy%YFTH?H*C|b_n4=5I*h4=kAXy}7
z?p1$i``g4sgf;gJWV=bW=&oJ61P6mz<=r7Z3;n0m1x(;q#lWs@0+yl$Gyo?aOfIfb
zwADv-E_<X&rzD$74xjKCO)1JixWF#({rvnW3;;#Qh^|}<z~+5A4-%K3seM#el#!&W
z8*ga!Tiw*@5$m+eJN|V<tEf{hC@A8;>LQ}n#4FP`%6Ieau_<isTQH8I-mewm1_QG~
zV{2<g=Yi78h)`2QL&1D754pW3T?Z<B@)45FLz)g3n9lK*BgWax#uutl`bA!5_fE;~
z`VDR~HH}TdIAha^g?q65r3Y+n(drE8Y58_@LHq!|e{k~1@~8JCr9RWQx>sE!%zW!c
zFu(I0tjwad@+>tbYNPHCE{(tc{`)+bKQYqQ1KU>)i3|0uV<a%I1*UK3b}f6*hgjlz
z3^P+xI*EG*kAp(-opL%7XPfNxj+ty=U8j-$$~I~FSSUY!<`Hb!N+a%>+;+*ix8Q=Z
zz<>Dr0vc*}#eXkOE#JKmtUT`8tbyI=4q0e_<hvjZ)X~GK`ePl%6fgU3qjgg!PH%@*
zs)o&`gO+yA4&W(tCk$xMAc$OGAwd+C1PLTc>Wx_I{MJ?Wdbz>0u8_XkY0s%~FnaT&
z$b2e?WFj=Pva0=@8?AA%al6a=gt6Nar&G9bE3CRQkT#u__Jr~X522m{n&0ge{bVF2
z2aio0^#-*ae#i4iBs&lZO>f@X15tOr=rTyVI5D{R>inz#6G3L3#D;Ky1-bv4a$e#%
zEZEZXF6)UKZG03}(4SIw6Re!Z)*`P<B4e}|8*A%D2b8g37qrixXL5V>1P~>#$}*z%
zV{N+GPSO@cVha)4>D~I`w7+5Q2e1$${{(R#Tr_dScJJR3^<~A;_`}$FtZaQ*v@cen
zEf?!~t-}9!eD@129r->i@yEbuOwHLdj`+rwkt*bDwCJUH&Y0@|%4CqBck4ed#cY1A
zc~5MM`V@%f9(%Wahm|=?pCaUyjdR{uw9D>m#FuTJ1{=BEn?H<Q-o2jjIk#{{<AT)h
z=S4r>blVuv<)bveu?MFr8F^1H5Z|ycpk&ot3mvs+G+XA!NwPO4(2&I`Q$Qa>{e7gg
z%Ge2?$_tBdDQ|yfbv-k1f<0l;Xy^Am5gw|Pv{(3x{l%>tSrJ~?T>GIh4rRHNxJv&z
zAoIQ(`iQ~MkyqMOJ@Lf6F#p5bh9a;Mn?_6GBN6W#a=j+w7P6h90x`a-P)*^JniI7M
zZ$2<^l0LAhy26~07orB%q20vp&HJobFHf-^T|lwGl|1<_tOAkq-pwd6qQ8G}KCO!+
zxL}m%jefgT{n{*x))Uu=@Y;nvLr9d3e%7Zd^Y`W~BCw{xz1jLbJ`o%J3a!>Dc9)oJ
zCN7osh+*V*Cv;QO^TtDCDSvwxykY30$g~Rdh_km9@yWoujYh3>@<i&($P3<<@j}ku
zPn)8<K~a+KH|DZZQ0UpXh;l7ZH-B-@=_Kj=s#=GHZ!l|BzunF$9>m2qIv8`kv-Nug
z1)>t^!&SdsL`Jc>wuUR=i2LNa;l*9?vDXz5Gh4sEl|&v?=*A5f?m!LfZ1omr7IASf
zD%iayX04qyb?$mc<|h2I_3N+Y{Z<2RtO@fnIXFCuvNJ>rd++HGx5Qs2l+|R}_i;Yc
zDX%z+$A7LoEK;FIUvyo%_tXj}$5Yy${_)<W-}tL@j6H5FX9T@`y3?ccB0uBI?~@Of
zHs1N;r?*v9|Ge3w$JPAH#}|K&d2{mDy`{bDK8Vj)YI5`9l63>yj#zGAxu#%q%d0U(
zQ{GQa_g~FlII}**E;xVMrl(DyklR-CrrO++#uw?R#fP%c;8W+uHQVF^s9pQOVtjXH
z;3$^kPsWCa#_XQWbf9su!=-_~8;CFHO34r{r=P9@Db^^O`X#p>ZCr`SndriPtfDPv
zAf4*P{%V?-qb&}B)w44RDCjQLhVD=dxoaUD<$jA@Mhgl=a?~KFL#Bo<1jrU=+j1rk
z;pB?;XnI~sx+ZK#__YJX4J~#AR-Yu7?<#hC?Q>~_6<e4MDiWa!BrgrM<w3=6Gn%6>
z7CE}b0sX`4@qimyTJE`R=Nu`?<KL!ji6Fu*mAq?fb%E}Y=-w=uAS~CROP3K&;a;`a
zrFFy=0&jB8M$g|zJZp!Q+qpF0!;RK|O#q>Kb-LjZtB3Q04-6IqgAU{BVR8N8wTm0W
z?&zC{a%G>b^Gw9M$l`j|A0)e(O(lPMqxFOumTb(G=<1#KA_57Ol96i&8IhktT+)9b
z%*mmKE#ImneUG*2$hk-p*4t&P*=D@<7b1{Vaqa-P8jB@vAQ4a6`V~cecy<I~#I^y4
zO45;6Bk8+-CNBIWnzcNBcWslpj=XbA)UVbiCU<DO*>smDgGvy$78Xi=d-!nA03_8a
zoTU169vx-piw06sQY77w2)ZAwfJ|5d$UT<Oc|h`-bNPVXbcb@8fI{*`Q)oF{LukP&
zA;ul=cpc^|Sr+e?`oV#*IIDa%qMs%!rMEJ1xoCv&`SXLNyjm7-7K30-v>pw!i5q(k
z5~v}U$6^lnK%6PAfUmXLN*``^v6to`$^`Hj)6v7Ygz+3wSVm`iFUpegXg=lrDciWW
zkn}eZt<`Z2OV`>Q1C>dI3tb7$onDmTrd?CoF%r@Pw43>vGiM4Qmq*U_$%bsFe#eSR
z#q4Z3jXndpIWY@-Vm?9!P$n#=3v7VLDy$7fF;&g@{xgU>#4*;1a>A0=%Wr1s$5S{-
zbM2a~3B0t8mQO7cDvrU|Ath{$iGT#w0I{fI8i|#O`_h3`HAmzFNLz?m?&y&fiW-Fc
zcx4oI?t4s)^QOqqHlg@xcZeO##$DHDp%9p9k-P1m@YVb(QDcH_*rBygF>ISRl~+5}
zY7JC5Yy>>TbHXw!7$e8E<?7~?WVllOgMxsXTeVU0MNzUp+?70YF9g3vzfI?(ts5XX
z+X?%}GEYRNI!;F{7z3eM0Rp|9*@7h5c{(qh=EF)<L>_3j(+Ad#(PHja(jkw!-o6K0
z+ZM#$%;f;Xf9!GDOBx-Lw$}k)r2N_$6abX<G-}`=t+q|NjR08Hh<{2gc=qULP5j0O
z-&ZHiMGWt+NbJ1Sv8n6PaakWai*su&sDLW^3cRyryV-}~V?V5z%4YS<6cuG;*-g|h
z3CF{TPCyxyh`1)#<{a^e?Ud8At}fXjnrZgRDk`3mX1tNfII(c|T0{%xG)udN%}bnz
z-5UsL*DR!_h*>DP&Pu+0S>G*9(=tl*W9d)0FFLN33ev2(eb~hk>vA}T8b&PpxW-jc
z7;Sm&8l)eDqZHmqjL%)rIFmFtZJJ+zy?So*?zB07+H|JJ=#*R8!O0QCtwd3I2UmQR
zL7LfWyD)Fbk|le`<E*m<jh~@a6y<^{{1oPk?!)xXqhiA%uPzWnXCm|S^9-GC1|<BJ
zpw0dL{ia13=GyN0{>pQ!^#!+qET266Rs)|nl}Db2O9W*9#HAAT(~w!<kY_lMSxMKC
zQ(>2~;~_!b2C%Cy#H<3&9J%F9_J?r}diuk;Qx7j~Hl|eEujb3fx}#mNthRBIg~fBn
zsu!=?uV?9SK}KHD=@C9QI?9}VaaM~>>AbanGfXELi17*-u&oVwPVf1C`)o3OJN_KE
zZ}i4W4M@MjRP9k8_SgnEE_dU`jdO?e2K(r|!sarPm)_ul_#?riVGYk<*1e=e8CPr!
zSorriCd-y<sAUn+e-)I}5bz22+b0aP=37ggFc^ct#Lg~t08Zy9mi#KFFaXz4zldqV
z@2T>rDM5as90=(V%|~N1K=WqbmK9B11nl2w%Np>PK{iawSy}=@0v_THv~UqRYp>^<
zH_|cxgzsjrXFLCva7_`Z7LxY_V*4In6!RD@Y+=W}IK38<#hX#H7)TadKum2;1xDkJ
zb4+?eL&I8~n&S4Obvo5EZ*9{+R;@ci_<@NfnF008gsGuYFdPe=LoK;oWxgCX2|aW6
ztilX(`eP7$q<qO0qET+8orzxc)>c-DWf{^TWJDrS-fzh@uNX=5wlpz$BgUmG$$kKw
zeJup6vuDp1f(Q&K<5YM@m~+gx`bY~I7mT&`WZ#7ug6we~WRIV&ZXhNldOAv7aKd;a
z`NmmO1cgs6MFqaWDau>>OP_|Ab%ZfGj?Ot1%-6$KO0}CRE8nzk-P?YcUs$OC7%4lu
z(mNe4p4DO=28lej9v7HxbxwGk0*s9gnB81zT|h)cL=za-M)gfrF*~r`5S5eSW?U8p
zD&dJWruFT>E7zRiWLI~GlsiN@VHeSYOi_g1pc)#6-|u(eb=n73Xo8{GXchBHD~=Hr
zp9%zE)jaGzi5=vscQk33o0!A|(57L1C-Z<uPYGbxE>l%ig3hNAaoJilHXdW8i~>zK
zCDKfta9VkVnZfnM|J@#&jyiv)B){KdGgoL2RU%3Q?p81*H#+Vsh{d{H4-nWB-L<A>
zB{6BO^=<yMI|4|Y=ckF^BNAtgH%*AW#=aqbwMa4LUV30a_^U%VgSI?PF%)ZRx!+j4
z*p`IzA5YkQ#^p!~B%VLMdrz$hyxQS3t#w(5i=(s`0g(MXlyU%LCLJW6KqavMXtAu*
z_OcLabI$K2kY2s?iST}nT?!`eK^JU*6a)^fT~n}l(uoK8yO*KY)=e@d!oBsJLU{iE
zv-X?05tv(R^tc|t#~X1J8R8x-wkPh^|E*8n$i$3qTl9E!!j|5A`FsF&BKiP<Jyn`w
zql-`R9}V9jUFR=+hjDWKV=>a}yP=j@ONkM0<797LXK<xsp)pKB%I%za9L6t?-un*K
zMOk!Q4bARIO=8UJ9-&W@`HgK($B4(wdnhyB)HBl(6pUoQUf2*l+l4f>zWdvww%#Yx
zA<6Wly7&Y9YU53B1qC;ZOH+<qmdFz0@hUQv<5Tuv*;4ujc<!O$W132S9@}M1lk@c5
z0!ExE$Q%h>8-!U?H#Yn6x6QrEV^b=U#<vbveMQ_oO_F1?M%p@eW_U!V(HGoiO)es0
zj4g(PB5!B@UZYLDW=1Sb){rn;XTa4)1%)H(w~0>>UsutUmZA&JYmO}4M*Ob!3Rde0
zU<ZzqG|w@dOT1aHUIU5<7wOydn<*sjS*-30?zsp0#4$L&cQOOHkFFF={vfGDfG{!Q
zo<9F>zDq>Au{s+rr6s1`+InJMg8Jl+d_^cYk82&)$L1U{b;L5S=`8CkHrIiW*hf2w
z=Zdfg#?grDNxd)(7iFQT#@yBC_b<}LCB;aB2hNt1+t`f}6!)PI<&_90LzKRTNie31
zqq5MQZJnGqK&!swULStTCf>~Z>hJeW(+r=b%gu>)`wMD+>3Ei*Oe_d4Y#AW^dH{-3
z{=<z^OuBV|8uI_^ol`PeD4NtdI2udrW0f@z+Z`pP%<txjMZ8iJS+IDZ4{<%qex_f#
zVTdMO5)+xC+oG)p>J@wY1_}+HE`B<*ZlQXeqTO=9Ehb&5-6Zs6khsRwveAX4H(SIM
z4<Pq806{zZ4Q<*IFXSnRz}Wu>fLx(V*os!S!7YuNEq+(U+4pq0(bQz+mLZD(s2}u%
zY(TE~)*wiWCDnifWFSiFLw!u!1|`}xw>pFR%sX!5fZTk;#c*)rQEw++$Y@Z+1f;-?
zjHo`GTeV7>06~=7h4ge~0&V&_&IH{IR42m9ysXwE=g(+hbB$%f$(}%Esx5cam~~R-
zZn^t~xYT$4wN@#8A&fR%OPvOZ<vp{G7h-SlSJDHeonI7rWr&85wc0gXt4~g{*-+W{
zh5+fnPUV5*kgP2q2hB5J?sSC$2ae_l_jPBfu$@`*Zp1{Sn7x|C%B$AJolMg8_I3B(
zJazE30`y^$O<5`d#OMbol8G8vLHw~7B%!kUvPVTwIcdOE!)!Rf{BB3V^V{tY-UOVo
z#t|!!GrA#`PlQ6sq%KL$<-ERYYs>9&&}-vEMk8NMb_YLw_3~>FWir3yRwr5bx<J|C
zip#D(JfToYp~#rR@w{IpAeTx=D0Ul3J>q;X*HNNB*D6k!`gSS<w4IJlPM7_yfWaRF
zAph-RjPnLJn@nnN=Y4r=4}j8DuxN!MU&#3Q&Hd?&%_3wcS9Xjf*CfrpbZ;Re3GHL!
zP&{bpD18OGG%b4!*uAbU4@1bxyl^ZaCg_me{CV&CZ-NLsMN>1l*T19WDJP?<I4Q9$
zxeLM~%`&JZTx}fsp8?$O;;ZX{Q#R3ksfuysFTBJD+$6%)kRAg-<Ach9V%ry~wtk}A
zQF^^3B3-fe<xsA9c<T2h9W})n(lQVh`jr@gT%-z@i~OG4rMhfcanE?15$oa(%pc0f
z)K{)tX=lwn3wT9A?e(><DTnphg3c-w;Z9ZI5&=sU%K#`B3^{%KDl6^+q$28G{K;)?
z+pt!qtMk1@e*QL+{3Z9AYF!GNx}BR#wHf$R)0?^LAs`fR0c1(}S>%n^#3=SPwY<+P
z%D6n`O^Omdd@LND7@GmfRWDUE7_EQ@F#{b*;+t1Rs1abhU^-6&4<l3GzBnuKr3ZCB
z1Vum}T(T5kyL#KE#gy^q5PfT)VGdR7v(W{R?U=ud)byWJ6GT}--HMFtHsNC@oc0)f
z(APmd(}+J02>F4Hk%EF+(}R<m0wNP{2pRD5L;=zbNX-dkRp!o)j^OI_LN{;sq)?Qj
zpsrfxOeXxt#Qi~t2X?pld%@=MZT5R1zKAY(Vw1>TG)P?%{pG?=Obz5zFOE!AZdwi*
zE@AQqwC)g1C+zy}be}{}7M{f*rhHuvM-NFjhB4am?*K_lRu@=1ANY!_6p}*e9t<5k
zonYwn6p|h<9Sb4B=L9FDBEvP+EbxyLY>~F1Ts)FS7|G1^MlLh2Dwx>>8SOO0*Tv|P
zzfs3$lF3xHV-NCtg*IIaji=Z8yrm=QNcWR@i=F`htU^aznqt-N1R)h(zzVQ2C(ow0
zoX#r<-%DI}3x(y?ppexzkzT#X3xr&aZ6tdJk2xwl=s{KdVbyOs@r?zRE*#Vv91cgR
zZVp7oF&uVp(A4xv5Iww<3kStsdJ)H?FXCllv66Wm!buPyvrRkLkS3IXX#+x{wFfCA
z!k6-OT3Qt6>+NqJcH0@_2R>q9Wbh*a=apvL(TYDuzrY*}EJqG}3~`}W8C{Smy5(FF
z?OWYl26={dkI`@>0je_Vw&)(b%!jHUd)%bnm*n=bVfMJdONqChI(3R<XJqs=zW;=3
z*dFOC*iB8XuN&|?9iMWA)l@#S^+~?wOd8t-VIgi5FBfG5W61{GO33#nbhTgKZODgh
z+?8$~2dqgmVh*BTUkk{Ys)vx;L84sBn=N*2-URuDRcHZ3ZinG?2|a&Z%&S|hF2w*4
zkPYHK!|$k)$9BMA-=BFbKfa!LgEhTGYlF3##QABK=q7p>$##_9M0_2dTA~+45r5GF
zG~6?vxlJ%s$ES9|BVV_y-p^!}5o^|8*A9?aaA{=qBD+Rh_20b&mH~D4F6JTrm}j22
zenQi!>hPst;&+R1lQo3QP(`KH5m>Lpo)61Su{j!I=pvP1y_m=rwX&^Sq{Y<6P?UYy
z-~ZAm@zwt20LXGg!)xE+C}$T+@e)D)d2fP!R4J*@v8;N&llbVC{TFxEN*?qLY(5{e
z2x7Ivx4$RQMdY%+sg&~(i|++og@^ThJh@~56a9o1<bv`>jX4goO|tsp!}c%V*z$&5
zTOzf_{EtR96I|gz&DkW2t`Zj`Uu$lmE@Tb)m`%_9B*Hoq_x`(=dar;#3@=<lvUJ`1
zSZxey@>Z^F?L-0biM-QV4$?_v+JdtQEvS8mEVosZVjEi~@AqX3_hIz&`!*9FjV^PP
zepkGtH;D@BQ6PT`t|IQ2MT022XMCy)HSoId4Y5M;bzOm0_vT%(6RX~MUJ|7<Q<UA7
zzY<~CajM17BTr^-?R}NBlOY*b^H?PGOGFiTU{Lsta2G=&TYKko{C39!%G&SO%+@xw
z>b;@$4V=>yXyI@7*F6O`=sQ__x89riNlQQcJ(Yvsa&ks4*QiFl?7s)(XPD5Jq#8>+
zOvERo9MK+#X4(+>O&hycThVgIDh_}6C{~g;uz~o9MOs#XRb9S(lY_JqA;-1_`X+UL
z-nmF6N7`6(v!rjgk!K_n?V6zWTg3bx)M<XNb$$Q`-yO9w&`*&Zq_m}Vy2P{m<I9#G
zA+-^RQ6fJ|VZ<qXZvV1+@xkoRKx%^%^HIuKX3E)jV^(qIHo;8TP5)`ur*~YT`nL)4
z%w8F)aJd?OoiF46^<DqZA=I*buwcr4_S<FR$!^z~CR%jG%ieA-c|!grDf#wU8G)rN
zD&83N&tDf$E-iT?OXwR2Oj-=8R9c_t8?nZJV<m)~7AF_{EGq3|?@;qyLkvQ<BTH1^
zsLCK~C!}g}zPWeF-y2hQj!(59`mK={iH~e-3P9O~BP!$}vWuo?rS%jLnCBko5zBgf
zpa$UzZ0w%o&s+a2QXahXn)piQPgTbuza}cdyacSaE^4wR_mo<^%Rfxp@ujza_B7dY
z3&iF70n|a>96-whv><OpojqPoUl4;#W#MJK-|j`kn;44u-lUg1NyiZ_$>?ZO8_-2s
z1$@fXPz$I6w0@<tKj*Sg0T0Nu?ZRFl*IpxqE>QuH8*S{iBi1cq!Q<{9!mGPCx<rk9
zSX);G@E|d40QhEKW}hHmw14V?97_CvwoWl#0kZx?;5XmK(WbZJ^%g>)QVK4KT<rC_
z#sw_J^=j2{NYy_8@j)R7M{1#jDz|lPF+wahUBf(n8Y&d8yEYF2L8ax%j0+V4%u;~&
zP2IxvWO{o*+0Di))?tt;?r}WJB*$9iKk}p<fPNO=W4TlSws#+Y)QCd{r>9WSUMthI
ziI2d+_GJ+xm}Z`S;ew4AdLJxN@BJA8U21C6H#UoYCe{6}e!L4LjRCyW0%ho3165uX
zt#{xXCQ!A&&J|sQcWq-q5v~4s!vPSt4r6tna;Ko$U@g@je8FBi@5F&p`VYLhiak-k
zfH!>Uk}9QAIzy8ZaYaf21C%A!Z4G_|aC_8S<=v3<c~Qnsc-1R8PVsMGD`LvEW)8Pd
zANwwH&m9esdI7uj1~>mkboltit+n(x8Ti<>D^{-b@eED@jz`(Yp(sy9*S<rn=Is9*
zV!axuPMnN9+3Z@Fc^(w7<%qJdvR7uncu=giZJW><o`Jfbkph9S&<ipa>O4@GoXfro
z%`ZFw9ty3@%;LN;b$}+1FfU&q-TeJ;R6`dGaBR@{!E7ALS=tM14`Tdq&itxur5|`P
zK=DtdLi3R46R$fq-5FaVa3l?!b62K9k=;0uXHc@8RBfH<dVuG#H#q-F__Y%!PTY)R
z<~xQ%sZlpRO89AWmc9UC{sJP)axI4{N-YO$VPV0OI)*#v69_PF6FIKUxsYA1q7Y&d
zD3*A;-j2T%HL$r~Nj!skx>~FMlPdLiSV5A#H2qz4Sc`B;a3BEAaB5&2k3bzo@h_G*
z;*~{~La*Iv8?)s}n+p2auy3K^OLCYJfbkJ*#qM-XmI)rwq|$hae!jlKPT=S#=1ULI
zHe~1J)g{Hp$4>)|_o8ONr8pxhsiY&}ZA=qk>quSz;_VVDwD7rDPb5l&sPqIB=>LE*
zWYg2EFS)vbvl*^pNqIjjnQa+FTT6D4*mchXQ(E{ujKR>A4?^{+%^eWpEO?AE0i;7o
z=r9Ydg7kkbl+cCeUAYnui<B=mra>)AN{Kar9uC^ImZq1rd{H#lhJe_@tu3rDC#SYF
zY4+*B1?H`!#3rt|9a^{KiCkcYXlNC1_YTxImtENw9f81CQ&Bb?)b5T{!WZ<{k$slc
zjf9ne%0Zml7yrMagiYkY&#O_rI-t=8^+v&sOkjjdzPH9#<SQ(ocIOVL$+F@?nL^S^
zExWmVsDyF`eXxnRZmP8e>SLh*f`>mWGcX-`Y1o=Uh%hA9WKr$`k?aK3zOy`t;hG__
z01a_czQ@AHdg!BSO-vac*#Qtw^BA2bKIN!@ng+J9ot<DL#wwVaY=%O&j@rY(|64*0
z=4LhNS^SAqGU);9Cm>m-1{#m8MH$dy<V%pu0Jzf|j8qHAg}t`W+FOaEQJCU&TZ-z+
zI!1b-aSG7Jfd(Wv06v+k%9^3Qhq<;TC@)?#ji=+Kk6hTgLrD1gjT?I_;kJ{|kfrZu
zQ9GQe^!#yS;y?};2Gs1lKUE;BvjzygCn_4c3pwh@eQ49wXzFIJHtj%gTe2UXyZDbl
zDp?2~LZ8|Tb;7AHpe%fEbOpnMhU?~UW<)iCp6Od`nCDUP98Q(2btaqGj>N$K`@I|=
zEZP!>Kj7_ycSE{e*D{wK+ynKkl7yK7)(<tFf0iBj44lW}bGKeNzL?$p>5~b<+-~{)
zJs<o}$F=R>G2Y*P7*C8g^NT+i;(w|CM0-)|dhpu&i_qC2<G<fUUi*%{w+7E}8N-Hz
z?oQjdE`e5$?e^Kr4Pg^9($dm2Z3WMv&ZiN&O!?%B+=6JZxvjZP;1C7XP&RRPct!}i
zI$!}*gJ|HV!Z8NjgXZ$soJ^d%VX9Ym5S<;#Q^&iQspK&-A@~AA6stE_3cUsp_d*eh
z8rWB{d9zxOnw8Q)I%S){u7sF61x^LcV|2!2)B&5@g9RcrlX$@n<Pb^QnMx>ODCixD
zVfT#0*vTGlPPftH24i!>usgfj@^P`1K9&q+QE@Mb%zwmP$PIbgXWL7y;WP#s-s?Z9
zOY>&lm7G*FcqNIka?_Sa(Wb+}Hij04kKw6k6hyh@5DO-hb_Qj*8rVlg7w>q<$1<`P
zz-M*oj6jEgz9cAWw&)C^i|wU+IxiQnk)}+RLIi>s^f%%7eO5rE4lFnsRI<B|&-`mQ
zB@I9b#}R{9wHe$I4E6j|V(aSv0kNgLfzR|~P!YdA+@+#+43fRJZK-wUTgYmYf-N@!
zYEH_&tvxXsk8OlT8IehU6ebMkX9Xr+76%Uf$V6SKb=64%n|51Vq6#3%-e?o_7KzJn
zwfnr6Ahr_+>(t`$q}UaRC>Q2B3K!nUhP|NK3`kRR!$a^~wCv4;>Kw@)>9ubD_GcbI
z2DF2k(=G#n6|JbC`y2>P*^#bESfx+rbtP4GclhHnydFh<BK?=kuDlo*Np^S$c)U}1
zW01qep;1xNOz_4GsQmEG5}g7(S%m;YR-v_i{YVHDg<=Q;2nZ*@Ln(K%+N7j~CS^Qc
z8slMzE5a=b=LC{@!Qm_wwbx*>lyzOviQxyuYhJP}SJQdnq0evs&^D50UGbg(9PcO=
z;v`K2f^dNn4It|;m|?EIzJ4FG&$*Zpxd$5)a&RBr0Cn;aAYZ0nP$=33yFOhTrBIYK
zbOr}0UtvZT7V%DcgHGrH-EI!vOABvbHs@GSk`Wb3oBokc#H_;=9;0V*+Y_{Ak|p2w
z$sZ(>P_X3dBt3yh0EQ@!u<eW+!t??Lle^M4S{hBstZbMmkaUC2Jo4TJ04^<v2At8v
zV|dH7c@NrjBXB^aT0n0p+-M%AP<NJq`>NMKtlW=YtU$<nnDFrg=(I0B^AP5UydgWQ
z<i1huNd4+B31X(tUlPQY|7wDGx#qYVSdxo@+aRxZ{1~8+wFhgVQAhNeLmrY9k9yu?
znU}o0u;#dd>^Nij%9T(xKED<f^aRlbao1Y+?%hkCo}ShQQFq<Es*e8PEwN@DId7~H
z$e8=+!W{tJIS8l%>P%4Rk#*iU1u~*$-63Wi95&>AUC;dC{+T2fGA_SM@2(1XM6CF5
zs(6_Fbkvgd*Zr?Cv2u|+7rID$T$?L^ja0C7HKK8ZLF)@6VgC0)(ric|!|DnklZ0D}
zi0rdT+h3tZ>gP@FTWuU@^h|0efqk~Q`BkbN`~c?yuUxtEb|2QIQ!}&gw?4dF+4g{a
zLNo?C2;~brD;+cJ`HxxVc9w%4xaBS=Odj<#9uMY~$(=zz+Tm{9aThQbv>5?-<6Qf$
z%}6;MVMbpV=URreN=za>3o)Y%TX+`&TLeI98$j#^C|+)GZGHqiSPH-kDE%LjZicWf
zz_Q_1FKD;-VvthSN)?XHh~t)rJ4-WRGj9T$sX>Z)^<)LAnGY?Eo@d)E25PCEnfmd=
zuN<llv+)Yy)I-5P;w&9}JSjkLFb64gh}r_ZDg?1_UbR0Wlr#f<NB=FA^{jUkb{RBj
z3;(jHce|OH*(DM5lhX$SsAT!re1Ut0h}24%KKwgOgW|iqF{a+_8&K2C8tzV)of!=K
zOptpz#;)w0L(Tku%Ax+c@BrLNfBfI$1w&R|oAv#uu2J$bmCPmj|8E2V|BW%tEef2f
zFK{#fHXs`p`DS4@<|wb+JE;L2=kT}x1&oBd4MBiK6nH|w4n|fgMwHf|d@3mP-zg)1
z^Is1|LZLrG$`@Qk5KRMTA>Qn3a&mO^$`$QL#jj(mvH(y+{qr?P)dzS&4b2`Ac6CAf
z6E?2Bmf@Hi&QdwT@eXDsfV@1CBK$kiTkGr}a1?rcN>J5z?P_NbD`kcQIT*AFDqL{Z
zL2wCaO54Q04OE0GhD*=Vkpzg4lE22*g4>COE1&QCfs`dmv*g=>=FOZk;^3PYg>haB
zlp}N=<1(-gc(_rWVuv5v;{nW*GRNM(;)f%@^|=S(IyaXqYw+GD|0X~7Z-s8!AT|%w
z2A<)r+yLSsL$ZsnO{qK$G0b6Y+CiH>jcNkLKfVJsGBcZ*^%xC+g6J)d-S|!G<wv{2
z+9PzJfmm&c>@k^DQ6XvY2*$j&cbI{<a{w3Y+0@ur?jUUdpy`f5-F+8uE-_|dz40jK
z2{y49+`C=RqeNg3W+;WCB39limDDRC5Q0`Z9A6r%7-z`d+-C~bXEJ>_#y^4mX-isC
z5)L+K?;8@%GNikZe?AVKVVavgfbCi}<*l`D-4)2qao?D&M6r2=43u0JZ^sa=Ftc2V
zOB8hOsD<tnmAKA`W6EX2-8J{PLYA`vS}^KyUn2@!tafITb=-x*AGoAUQI+vVHI@38
z%#cd8<Jv=2U~@8mc?6-z!IK7_Hxz<com21ERX_f(%Xv{gTB+?Gxokk?(9O3T{62>Y
zUr#3gv++mn5{f^WXoX%h99Vg&yd2V$;sQZRRn5Ol9fYwo%tF>#_HBp$ks;kxMFw#z
zZvPBjOmA0aO(g^#6erHNA_9#}l4M05`*eKx8Vr~PBU4k;>g9=olu!pfw^$d6MYq>~
z1P!u9`b8Oqwxz!<9tjNAOwF)vA{f;~tF9^{?=(TE&dEkD?z$d(Edkc!FIg)@SK&(o
z8X@qY5x(IfG*?Zu8vQ%57>Yp&(q%|iNeqfWcw!zm-ZKqtezpT-`q8!dABX|Gxn~To
z^-d@dl{>Lv$nj9ii8jtESAx664SaCnI^4io<y`KkNs#3}THQT{ptR}Lkf67;Na_Xj
zsS1r@mO1Ddj>cHYK=lfcIr|Zz6B~TLGzxNqPAKpj$!}At`EWvUR3yw3;n^u`$oHG4
zLbCx4%*G1Z(pFX{d*P$%x49GIHWA0LJ=i`!-BSC^ymVgm0)%^bV51Ls7Lj{let5r%
zay-X)7o?CCT9N`7uA5HH)%K%vzd^Ufk|CdPXehKlIAyQiU>c;Ye?r=5&}bBdm6H6C
z#qPAj6<{Rr)}eqCb|vwlP9z5F?U~Q70=1+t1xw;Vsx}$n?ryz7L~6&oGsiz#46O&9
z|1q@wHGrvVs1^)eUSlhRkDtB+RZ+%0_Z8p1i*4|=I(*>&wg7!jIo_9aCEv~4a6-~!
zJmUE4$cw|4p%y#u{5k#?n}Xs8n?G3gh4jbAg36Z>X*P!vCe~T*{CM#5jcG>vzWVO#
zS4+3WFZuY3C85UQ2Je60?DT$4;ypY1r=LF4d06!1*%9;X?$NpJ)_-i|2Ix>7<7VYH
zZK*R%;dJ@*LY%Zbo0VEUlL)(b-FKfXatTZ97pqVYeT6QLclj3|F3#E?v>S-r&$eHX
zMJ%!`<NZZK;yAX5pZ@0}Du0`k4L)KO=Uq@rM$-tdo8+PfG<rE26^xah1rh11+R<{w
zOz!{~{RrrX8Vf>NC+Q%PW$HB5Du)^?{;;n6EJEU$v}C0T7pWrW*dr6I+o25W4v-_o
zyjh?xu4BEMxl5>Z@zBc17dXRtbvqzdr)96F6U}9btaNdA^>6{2`WZJDio=YD?y`U-
zUX6m#o$VDxuuc{SKB<ew`#5FeO<szjL^iC$gy!139yM@@+J(uq^;FZeHf%^df?dU>
zA)Q^ml=+wr6LW3(eUH&W7@JTgVM1`cRtHzz_(on6EUFw{WBs-dfB*g4(hTW2yyJDk
zW3UyRD>tBp(fF~PAxqRQQ?wgQ%Tm?-iflv+x^lhRbVqU@bykXWNeLB8tYI+f-~pg#
zBFK)M6fX6Af&z1n{G?8`=EWI})|Rm?ed<Klv%Hr%qRaCjVd&Ar+iMpT7;wp}nz}Mv
z1tw4*c>j~LglezX37tsgsOkbPNSRuVx44@&r_BD+r-e%Yd@9TDSNg&}%(|=0eTs(Q
zg$~9+H1%KUcEfLfT>SQ?6^3LLU8-xd<$Oz9o0}bwxzm6QY+lu~!7Fnjp{1C%HCdc|
z0MmzIEy3!3;mw|eSP3swXmyk}LdkF{#5H7tfI5ndOp!fwZSHpH9%_dUksVQaYn-ph
zILdj>?PsB{&GTNr(KvXoNNhg4yu^wVeCP%K>p**W8bG&Tyr%ETH>2o;=8bm}8G*bR
z=%He$#5Q7H0ErB6GQL_+MmdR8KvS{`rzi)#L3gK)I&EC7EX?ue!%C15Js?b6^o!;S
zky2`iFSIy|F0UEN@jib+#)$rL@blw4bG^hWhg3$)Z%53o*mA*ciCT%U{<p0ToSSsE
zF1PkHPrvE2%oEUt($FaiI;E{GDJ~XlXVgKT>05L`fxlaaNoo2s!Sj=iJN~ckt~IKu
zYu!TC<M5!;*5MO{RyH6*#VRTSLbSF6HwbC*wn9KaMV`u2hzJHNM>)M7P2C9~&(sGX
zYCr`Fk*80LywqqSZ_z-C0V2VG2ubeT^!z-(?zq1$KQqYA-g~X}eczhjoO7Xrv7m!K
zBjA-+6<<A!z?xf*?o=e2VZd|!w}J|+tEkfs`M~*W(|LItyiu-$b*%CVsa7THCkt2p
z6?6(uj&JQ>&-IQvDVy7!0zhb>^|cebR?SixkJ;%5b{>XBpMG0K$MHnUzZ;v$3Ka5=
zaLdBYiDY}(T^YPUbfMPK;@>kUP7_J$`l4kj=oIy^>x{psp%d2$o*Vjz(*5u7dIRBu
zSpdwBNbm)Ne(k#a9GQR0xVcN%^^T!hXK`?LK|(ln!&jufLF~xqGna1(qUR~bzR=B@
z+qCk_>cxAHJS#kxG_Bx>x)!#Uo<0&LLo477lIjb(=4Z7`=;Vh8_|7%=w!=4cvWQ>;
zm#O>hifM0IxiqY3TI?5ampEz+ti^t}>qIWG)YMehG0$Xk4vzrsXEj!}8Q;_Nyxa{G
zD&5D6?VzMEg!~xjb7s+>y!%B9XQCDv1xw=YtQnIiR4Z4C@1ytmMetuL&?bw!jt#Ih
zO9NzTVP7JHI2g7n5fzCLg-(!T_}-bA3*Mrxi1B=q_6>8?eSED&*R)+AXH!$dfaSH8
zmX<7Z7dZPUM>L@H>2E)#PruCFFP_HCj=DY3sc_ySQ1c6?G@uFA{U#s8lmji}EUgql
zhjLPwdJS%<=$Ja<zS>c%4(9oAMw{)DKW;ZZ_Vh6N#hn;56b^}DLS`p`8M|{GCeu(k
z{WJeB|Gci;?li0~3faH^Qrz3KcFy+V{Br3gGQvGZ1NDX(rgL16-%cN_VFmy$hjQsH
zoC9$sP5#L9i9cFU)p0Q~F+E8ddPBleN{l5RE@u}^q@H^48QVpxn;Ky$?^9Povn1(`
zmRn4Hl~G5qeH+e@x8?IasM&;Qtr(s<B|U(-G>uD5)?`3!d`+fwy2V#tAsD%GKhjwc
zcuo50G~kpeWs+u*qtviAc)ga{<EYvB{@#}+@O*z4U$vXe+rk_fB3r_<`d7P1ey-|r
zW{6vMq-A(U&<Wuc!cb(k4-lKN;Cz1Y@@wFKegU7Oi9eq`Fk}Zq!+t=hnU|@3$}0$+
z8kY0rtW<NKI*#@-Cdq8$*vAD0p&ZRWX3ezv6)Ka}V((OY`5Z(!wny&w+_)1e?-0FY
z+{0biH{ZZx6OG;m1@Wrv;0_v_T@ae)*|IYxfwHBl*SSnIq<KC$LaoMS&C#M#iNN%_
z2o$OpxzLYdzGUee@ya`lWslGib;>TVjy0pIq5^66+T(B@LXHR*9r)#~xO~0~e4YW6
zv9ftrfS%#&s)A4!?6fZD^BXJ&EZZu9A&#cod2ShTwKjt_!-c%OS`}J*r=>T%Z8thO
z9q%F54VXQ@figMJBlm!*(!fZS{HeQ0+%(tgTe^(Mb!3A^_Ak7>QhKXA`THCCxQZ;f
z6>n>GXclvMqsXTA<D1D{aQrhTVG8X6hYODEZr|*n_vxN37hs(pyGC)6_gx!EnB3N;
zCO8=#rr)@H#JLv>xd#HZ^QA2g%4lql+cPe_d|X67Rs7T7lD+Pd&UK_yWi>F&bz(Cq
z+%{nyTj!PjRWI=Z_i$c*5_e^WZEi=i%WlQa11r2duzl(yBbF^&7Pjcl%(uV(@l;o+
znrCBsbl;q=sQdF2LGkIsvlRrJ*S5WWHR9$0Ct117Z?c)Fq5@%G@zJXzh4=FGJd+kg
z#zYR(D&>exf8daWpRfPK^&Nc$RNezn1~y9i=#Ik`nseh50isH2@1JvxjaB4xC8OU~
zlI(mfzw-6)CH`$YGEC^Ajf*_6!XrjOBs3!KAI9oQM)R|QP84@Kw>Jl0#1uC1)s+T>
zI&Gm%r?N|K;x|FauqZVWnDJo!U<MZ{K=bz!dnUV+884#CrFAIPxI1sJ6k7p;q{V?9
z@c1m#dmT9Owr>ve=)D@`W=^s6(-b^LUwgVzNmrzwlmMEOT0!Sq(J(rt)oK19>Y|%B
z$p)8U;W*?eQu^MoQZL)t+lxwI3&)`r!=Q4pJCo@zLhkGUW6?3`(b8aAjRwX}B}^ag
z3!fRWO;BAp9soFMT`qM=^Xx9iKbM#t?z+4}SsxRU)ep=!)?28thGhT|id366O+Af#
zRT|vN@J-fgB#leo?w=IY+Eo%5BCL>sNDik~jSN^q8+B}NI93cd`**bbKDGKhtLJX%
zDYHGU37v@P{<QRebg}wtqjMF2fVl*hsrDgFyHu+MHu+n%#Z|*$;8l+Z&0o|xh#wkl
z9Wqz=*yK;A(LGplC9HtNZ+`Gq-u%X0GUb7h{#qQ4QpLYJqLtHht$diN#u+lb+&?>W
zK}u&Z)i0VnY)C4Sm`FqOaMoI%+b3?#Uz6$p<F8XhhQ&k;9@LL=!;vYw)O_f&;*gVR
zgAsq6H!@+HS$d__(9NDRP{D4R9_}xzZ~nv8dUd$4sC+oLBngIacRGeFOB=VWX~&u<
zr4H><|2Nv!oRHcBxQGII^m|%8_zJ^U$K`TMDwV}mrf`EpwCY{`^NpjC)M)HK{WnKl
z!UWZ$9rYtMqAN}&)q`*eMv)Y;ciZm!6Wytq-_ulRb6ig1YHggClapf~GjGYR;yABL
zouUYwg7NI#Wh=#5@YfR%Y?~b+QMWHlCJc7x^Ea_OqZ=G~ot%yHmN$FN_E<^Vd?VO8
z!k^(J68Ao#1_HQNoMHew-io>(T@#v--xrQ18AX$i)}@9<Zl|C7P(+@g+f|$glO>0N
z`8wJd{6Itk+5$g?*Ep5hDZFYP$AU4Zq^6=J8(W({TTn(@p5H?G2|Jf$$}BRgzu@5m
zLpM|lb&ldzt2TkW$X&84C{y?T8}RB`FlEdjA|tZ;Pt6{pvw`GS6J~T+l<@vNO=Q93
zcxKJMVIy?<^dH*8OSDoH$sX`6)v3ttG`Sz3!c?$xr(X7*U6d}vE<y>{#jUntJ><Td
z#dWaf6C_?S*}OuDr(dMKrT4dJd)gD%WXEE#&I}-8zP3$Soc@D#U%GB?cdTMHEMNkX
zM+J){oI*(^=rV-(dR9TgkBdfju<4SG2K3hCPi+@!E|SDi@1LP-PV@XZFy8+Bb(B|U
z(0Xn|9F>}{ZM^y7PS@~35_rw0Dl2t$)FqL0UvEV9$fHr)?9SuCl8q$}H|rPyI@+5l
zxfZlDgpZn4;{3*iRylJ~B^@Jc)8%o9iyw$}^rmV*6>j`fu=OK{n@vv;=}g;|8k?eW
zor)NLJlNDGOY`)2RhdW!CC5#4Ud=+j2n`8YMVe-Se%|P2t<cC8Y9D`=ZG%XbA!8-4
z6}x2>hz7mhKo45O3TFw^+q~CQAxkH1MF(`?26lzaUFvEg(tfB+0>XzgER$@2vN!@N
z9$Q|VYr(qzPDbT)55=pt+$`Qt!9hj3UD1|7YkD1(-&y^>Zw~#$ElkSO)z-h)X*in8
z)MUIlHudY=zk9^z^Nl|=m@0yGGbPlyxYsjR<F1Uc1U;WbqKZtk$OI%AihU{*IN81A
zK`UQ<O1@tTyZiP09{q=!-T$1qZq3E%`nalVIFI0Z11!pnjodF)YB*JVc<^pg)kCXL
z(pbAz%!ASb$MhW2+<H)nR0WUwd=g6L&Rb=(+=>MCH4devNA<Bn4OuQ#q<LC~3qvb9
zRP-@5paqS?*s`UfMdMk#FRn5^JtlXJS445#7FV=u#{`1k<iglS+(3f1`P2<FWA<z=
zlRiKuO*G5RN&3qI%=Vr-{szBrZ8I_M1{11s^qN|oCkKsJRQC3xqA`{=d6wLD6eJ*$
zvfr&wG*0(I@A<6G!A=lS&Qea#YUGhL6mT-PaA7M`onH{z(wtM`wERdzW5Z_m(cZxL
zvk!pdM!+hCFV93s6hC`_P;nYDfs4;Zb5%*fGL6ii;RGn)h$5Hm3=Q_)^KDBX*#SK8
zS3EMb$_DdYzY|K-FiJs)_R!V;C^$o(G8Zho4}xMU+PCG5TQ=5(n`I8m+vy$ZaxA_I
zU1KoPOkEs?y8?#D(GP_c-N&{AVkXAj)pc0~6Qu{|F7&vH5YM5E!i7HcexPJ8ojd62
z#3Z7!7eQ_}^l(U?!SwMDL)&T|o{I|*ojrS2>uue)RGVQ2=pm@@Nt!BCVHwuYwf_@d
zQ`iP9*H`o&1a;a9D10$3Yf;%q^mqKILFAjo2X$B8!Ga97w^cLVAq$mi&F+vfKOt4T
z`MmIWfMjEj_Ka_d%*0*Y&QgZs>C$GbFJtv+`-g_1{@6P9d2~H_5c$!6S4<li7>&-l
z5ia8f>?&ifzjPS-?om{KX0SBn)B2p-D27A8)X^WqsJ(<>m=J~0$5w&Thn_(`U&Tq9
ze2hBK-^QOG^#;|X1avbKH*5QE#$Ew5xT{@Bw$HX=^B!{X)M=g(*UKZ;<okAPY_?0c
zO@H`-aa{G{g@gL*`fqXaSc`Ae-C4|4`CHeDU3g8zge$cNHfSnCZQ#rOa}SOwP4vC%
z;nI4XQIlOD`dWOS2j?6ka?DC3<lnx`SC^qiHM`TF6&}BK!+ftr7=aqpgxWf~C_e^-
z#)iGspi_%B$zhbF#-#c*;lJy-TZ|7SBsmc{8;win9PgGTz28;6zh?xaX{{Kn@`<FN
z2Ga@|OYbZm<>j<(Y)&`d82bX{T!g!1Yb4OoNY=1C;x>VxG)=Pj-|xQ8NiE1R3(+yr
z3`2w0@|Ub(T<S!^JPYshfk*%P?B8hPfB1=QRP`2dEs@-re7u!T`AJj<?~q4~F2})~
z%*&LYCiLCOoG>cp!l>_aM(fK|`=C|gx4dc#@nYVfJE-BRUFulvXp2He`lZXwc%9FZ
zlQxJKlBO>s|B~DCN2|y%s*!i2<xosn%jm~8-N<z+Y2xz+-`pKm34&*g{;t$j`iq2}
z*`=--+&!(kRg1Yl6pnEXftZw~(*(gKKhO*gYLA789mMbvoR^nt)ohJ_q`UeSp$Lo*
z8r(@0j&l=YD1r9_kqSpOkerR;JsTX)jL-hG{>I}&ZzhV!h|JxpJTRSmIH7u{x#P;S
zmiBhySSl&_i}ll3Sk8y}O|LPU;dyGRt#4Tv)fyNtK0OgMpd;rqc8o{a9<>{9bQZKd
z7<_i;s=tb7bajS7J=ruDV6InAW>c}Z;~salLOYZixOtAR!X9TEDaJl)O;i!Nx-kp#
zj?vY`*Q5Ggscx1*z=@rJg6rQ*xl`KIRg;JGqKvj|L9?P!t5`^S_|zw~I)+RhFuIyS
zI2wn&Z6-~Z+5OiHgC!v}qI`2ps%xjExLY?`+{8$<Bga;h)%->%xgP{|RGaq$qmgD}
zaq#AmDL2x&i~Dv4nU~SF3Dzx3zWVo<-HN5NmX~^+5hm}Y5-leB2-Q%<=$}OL1$lo{
z%in8T*FXB3LH(R*kSL|kO{*|_@+M<XrBkVLCgbbU#vPpu@uQ}e_s5S01%F@_F2(>&
zKT5#FhwUBV>A%<bQ6-iMyD68kaVJ82JmdGnI8M)&#=DoS>DhH#Sf}kL%-vp%vFEqP
zk8UzJVfYn|Mx`hQ$UCHIj@&kL?~d_pmnzcQHXRZ0<<IRkd0FyPBxj}gT2kcF)4hc8
zu9!HzCSyZl$`%6ykC^+7v@H8(fIn0H6poD;n%VQCJHz{{YnfD!iZQ?M4}vSc)iZOx
z!HEUiXc+!)5=QW5QI(AtzBU6yaM0$7$!COW=kunb*n?D^N(OKWW@?3mK3%>bGxzRf
rF~MG;e?L7v{=>TYKmTY+l-}%LNq1Df?~Y{C7bm~>_^$ZdfRq0N4AG%l

diff --git a/reference/plot-3.png b/reference/plot-3.png
deleted file mode 100644
index edd8cbf9f9bf8223d990cecd0780944dd140af69..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 26115
zcmeEuXH-*Z+in~k9Y>L|j39N?9T4dtAYDbpO^bj^2a(<dgn$r0R4i0?LN8IIOOY;J
zM8VLd3W0zUKx!bA7y{(%gqioe=X~Efzs_0f{c*CEChWbRem{4)uKU@~uIp)X{Lb?`
z0)gNFuU$4kAb#~jAa>vRZ7+O<824%k{@Qy-Tk|sf4S(nwHNo)L{=3&qyb%cgC#-+F
zq9>Q_5Qx7J;N^=q{8Hve0}^a+Vk9>`x_;fe_m9w#%e(&kQ(sa4yG6`Xt{b&~gkC)@
zou0lca+lnfgoG4F61P>f_c=|>pRe^#|L%P1tp1}M{mVDf@BEeS_aNtwJ(H)NMjXU!
z8~1xqoSZ3ng`}JPHH;Piw5|5MIthl+Xib29Q2Dy)es=Hv|9}5)X5c-Q_6$g7@Hyzp
zo(Z%<g!EgFmhDC$LZZy#Z87+jtxjphdP&6lytnqlyVyOD;@NM>`VkTZjsME}@#D-t
z7fJ{`2!vS2;od{=?SAc0E`)LJfRO)nNK@GV8WbQj^ai~h1{fjf?vRIY)fqzN{UQ%z
zvd_A<mo3GVQ_E`_kgZ#8NFs*<4(rW$iw_JFM<K#1&@zU+1n^?b6Uco41@yAf#cK-r
z2SH;xCE5@LNgo7)@#JLfrw$YVr9&Mj6{A`QH=YC2Li9^ev=DvhNa2?-aFokU@jRgv
zt0FXMStCOTH(bn>NCv4cB38`Z&`2e1Iu$#gXgdv5W{pU4WsW4{^{AOZL&Bw6?72>S
zX5lCk9_ahCYpXbN%lvOG)*6-*7s2|{r=CSx)So9?wq2nPOjPOpHQoTX*+L(0NaJ`!
zl^ImT@pM@Otl*bvR-|cD*LF?ZitfdZSicx_Y<ssDj>@1j<h1boP>RLTwzemjrtr5$
zx+=+rM;<i!UD0K_7#w+ev&F8A(&P#a0z>Xk{nT#uGG#e1_RgfHDIi#J{C;--D&zYf
z#X=KIlcbN7E38F92Hd=g;^V-6HjXD-Xmd0359;Zqn)T9TLHUgKKK?0F&iTknGsxx1
z{0jYGC8{Y2@Zec?xoZeL;q;7@k6fweM!xvk=TOt$^>Yy&ak?=XciFaKuFm0Uy{yQy
zk)SPREsBu|_&H!V&rH_4%Q%NtE+!S7xLX(9x5}V~1Yv)g@X&U8a*Hn}6DhqHDKn7K
zUTZ!pS087a95&a7^y||0KNtLU&KyhS-lTDz7&GL(qpfvhzU+AYAx?q2sTC27dPmvD
z0L#5&Ho-Eh%2p}DedZq8t@{q>9&&z)rl>8%Cyl4-dVxvm5kD@4qK~L+Pt-Gag^2Ll
z`|d?tFsjX%+7tm7lb{<GSeXOpL4bLh@JuB73g?8sh>MLx|Ca1|lfEtuL+EbPu$<ji
zdyNED$%E%QGao>`$z_;B48R0WHQwG9aS2;rnE{BBG8Ou)F7;~2{uD?)(OcdHni$-E
zB?k(~y`*?McLpmGqIoyqwrW-{g1%jwv^#E4s=+{Fp>_;OlpwqJV}F*o4YxPkO^a$K
zyoEw;E^1b7wU>&@m|t=&!fB2Pni@7}YlUXJhY>ZgBdW;5r9yTC-$ghU9Izt=Q>H$$
ze$!Fj0fVO468lB%y=8dap|`WSb?Yt$(D^}yva#1f6{+`a*5<tJH&A~W$8W3!EA*RZ
z>rT{4LcR`5x1tr0LkcUjJ_(VrCI^ngv>ej)quRaqKg>ZR2ECNeNow$3at*UOl;N)W
zU7L5KT=gNXfA@Fu`GX{IODvMHy$_Zu$14_hrl7cH^;h<rLjVPZDIt6CZq&WXLK%kA
zdT=K&2P{Uq9pnk88LR&HH5eK$ca-a2p?Z^YIP+&@+b5XX(u<5!Cs3iZO0x(L>2;5%
zx$YW88qUKWdMgG%xaVyRb05L0WvSzs{t#}i<nh#xT|K}e%NB(U^mUvH^o021&ad0a
z*>q*)dz(3EXr;fgYB1{wj<bZDUzQv6x6r^B^{)<<zUnHL!_*t~ZX3!iOy~1|Q!Tek
zp<U?-3Z-d?e=IxZ!&}?m!KYT9wZ(Ln&T?}plujlemctC|E!BHnhzQqR(wlW~8lkBp
zr0fP@)piS#97qZFhVa?I4N^Qg1J}|->%7Yp7?z*Zrw4UvQ%l3@8+0%q+>>!~l3Wuj
za?`=x-O~l|*KoJPwAyrq?pDbWK^JaOk<tUqH(86{c5g`Z!~5RzgdJu|e(PojU47Bp
zy{50|A%j_dR1YE>arlH=*xbKSS$=*12#%}vWx|RW@#d?np?Yb-;yWZk8B%#^J%Zug
z;KW7*T-0FhTX(Wjcvva&Vw1=*c>Y~hS5lbTmb|_5hLS6om^*T>Z%wu48>Zaaoc1ZD
z!eOb<tUa0c;ew$ck#t~KEjMelGOTtW31~<w+0)9vrr^nWfP%wZK(>3q@DLDeD(g#>
z^|MkNuM@N1OM&~-jDh=OIE6Jq+(s&@b|g!zmFhy%!gLz$V^_19+w6dVZtLEw56y?J
z50=$7UDBnvd25XJ%T04JJQ}l#nv!tk50ayld}L1$WeOR0_#{xEy<V>*R9lm^=ybr6
z#`x>6T>P&p4!W)T1l4t^Yk&s`URFC#SQODtzClSX=df{87&mSE!kU?Ta8FLhu-C{8
zt8!*78sMZRgPou6*0f1ByK0QyDO?K=tIld^@yJzdFy9t`x0TRhRY<#{-V}ISYQIOJ
zBD1GGnBP~=%TKpPmZ#YgOMWe)eSX#KciSpwEg9jCXM^Ajt+X$bA&B5^^rCC3A}*K}
zY7gC;c>wBxCcDwnhfrP(zfQY5IyNl5SPNQr&T%pIwSJX#c^NQQpPZBYbRJ%2$E0Vf
zAeRu`6TcE5T;))-NlvP#XstXsESx2_&@D$inA5;G)tt22Wncsg`|UkPY7y_Xd@e#0
zhW2F(iO`6jJH$*SL%T51=1Ut5yu$+Wm&KwTb7}&gZns_O9sE{lVE;4H>m2euabOBl
z40Q<6gPmuC8fepP&a-p&Orc4pV39{o@D*swz`m)Fu&FAu;PVV?8y{A^1|Bm1XxzJD
z;>DIQ(Pnp%nid-I_K#Jyg<#`eU-^#cM33Z>%a<C0l6#rEY1O&|N%BvWN>0|RI&Zl4
zc4xc0^|v_>H)NiykT99YyeO+7m>S|iB67>7R@&~kM?<w>CS8BKrf-$E?AcIrSdb4d
zf7FItA@SopmIEWu`-fG7?f^7NFFn_~#fi{KJOPi~*~e#AA#}KV+PpWvNkP7E%Vbt?
zAxJxEu+t)Hw*F98*LQ_G!wKYa?E~39;dPR?N?fpkwO#Z`P))ygTF%K*%XToVx|DZT
zfOLJZBHX1@U4Cou-Dph7eB&MQ7CaK()%RldNFv@so7Sy5)z%8i!wplL=#3QRa_f|7
zfPSHM{$rJ9ChZy(EwZ1a<iUHYd2*_66{S@~>E{@JdKUD(=6{Si3<#HTnHG4GxI{;p
zX5QNZa?>bYDN-bd59b@j1{;XR!Ck34-JfEGg&w7up#$6?hA{mqgwcH0nGqB>!g3-m
zIMj--XbaY=5+<<ou+eZ^EM88;wZvG(Y-r{Q8p&>l^K6K%f9+dRi(5Iu0h*MS7LHXK
zK%lbywoAPll8(8om0ma;B%RzI={!RFq-BwQQUq+ClO0jrLTN-)>O$gjTMF4DEyT59
z_hj-Hu#oTWXhpo^&yJI0l((9rM>q(!K|5O+fhe<=cu|a-XzlAtF~gGcj~9CCYf<w$
zy--V?8NkZ$wIE%YMrC-fB8C01WR&d4V2=CV#Y@Ryd`0&+eKahH`$^?sp|Ncs{@NBL
zENc%1_`7C9`)7L*A^ypR&<!)^Sr2V6Bk1=2!JmQ9zHHapft#gZ6i5xu;{W{UV726}
zYTCgsQ^UYCry@zzP~L`jU>~$Od1_d)KMp`yU1p$!70-6Gq55V%boDqlFn8S^!ovIh
zOs9^C=E}a^!&_$wvWD^)p1z9NY%mQXzXcQ`{}}TsMH0fXBf<out3Q|>xR%Db0WS-Q
zFqYT|3e$T+<b$!_6b^3ML5$46uhO5hfEIfy;ge*BYu<X>t_3}i`VlvF;I(O|x;DHs
ze}wq39NL#xB(?l;j1ZD%1Ax{Bug}MoOp%9inu(ISCTG8;0L(aY-WAKaTGChe?ueR{
zo&92}p+HUjS)_^g#nuTTS7!TwnbWrHB!-e!a&=k*%a=kXiC9srvVbS`Sm^17LF*tn
z`)GtG1h4AaFwWL0^^<$3uLY!;H6q%j<>d$MHngt@7A3S`YZX?WeR9&#QB<DX_A`aX
zv=Tp6Yrw9s;%P!n&tkL9#6LS5mbd@@gO!zCfbYD#Q}L<z7PY|#lZrrGgYpx7cOH+p
z3j6M?`>z4N9j|}q`v3p`KVSwD3KO07T@Y6Ji$Fn|^nNf_28I*}o}h$>wF<j;JSlOu
z6_XRVXZ0-i0b%@LKvx-|^vw=b!`r$Yu;Gt|H_4cjwHW(hAxFd3gCuMg|8J~6?^*D=
z<o&|>CG{(8j#wXlWa9zWcenmO7s}m-cimN%1=mqt6rcvEk{}>K30*JZu?cp`xD)$H
zlau>F(hFOPn$IAhC{36-8p*ESxI*lURFDIbu0wI8HUkb>!CngBVgk7g$j)e|vV+On
z?efBj3;aH5`cAUMxs^qr=o+;Us1(al_l;PNB4*)ecYDfqvt^Ro_IfB<14=&Y)~Q$G
zc`<nbO3`=nwt?JQ4$tqow#4ep1#P%N%oE|EdZTR+I&FBQ{-%Pkq7q@UZjcwtC-dS5
zYh)pI+M(!;s6w1oTnV3sdZm!R)8v}IO*p&DKlbb=%>>-d4D1Z4z`wz9x~m^!m3gTR
z_+@)EXwhePxfK=QI9>nkGG$83zb&MqaxIFy&Rfjx>YIo?l+{!tks|(XU!j$;Qr_mH
zuttF=_0PF1{`Y6Sue+;j)1Jgq^}wKk?qE#j=EI6m_GDIrq399`zUvLaP|>+QO3Bk=
zR<V#jJCAU*wY}Iq%b9~~km|g^zBt)+khH9PvPrlnSncKZQ!V<)mk;dLZ~eZXv>f0N
zZetNcuBxM=&G^`x`^R7VN$&K|xWWKeNL^1^`~Lh8!@Kf$Gppf{o4TR#)+(n67&8Iy
z!MGNdB!%46gs*L0gml4y3aJ5nM*DB|8IRc=x_uKnZlN}SOa#5mAaTPeyH~^>%S-Fo
z;*o3-e3Q0^a&P_xxf?)T^STS|hVqO0tl~>v4J(4%iiAwiUYPGRYY2!x8N!OEktQ_>
z6A-|Ozf?Q%R%B8&_1wY**3Fl513JnfH_1P7iSJMXuPAj;N;0`ldJe`MU=ZTPcxptp
zkbl7B+9%L5nJfp^e^IVP61)xLJkFmdJOk@%Z>QK8+9SFmWQ|z`5YHE|BT4kh<54rc
zQ_Ha%V<y)tJ)mgYIVA7)UGql05G1?akWN3#siUPuE{?S9SYPyvAu~^FD-QS3lM7@^
z7$SA~%Q<29>3Mx8ngV9l0S@M!yj^Ro^&((r1OAkK<f)yY!He-c%vU`Zhabugk|&6Y
z7DXFtU1#0DKm(Ve#+tS06=A-mu~aoVH;Fu)pqv*QlY9UCuSVsdu_BzpK!uk4jn=_|
zGP+%6E%EbV_PUCO?xFllwGvTJueeE#(V}b1x!Iqc<h>fTv)K*<el3*Uu~Z_~bu7E1
z0wn4CZ#5<r=gEdJZJv3uYv31#>&I(Tt?H<1V0|ii8C<_s5&*Si<QD`)TPAma+g0}Y
zH8)!w%K6W*dfP3|me;#3P__`=l(LqJ-PG1#OaoQLS&LvSRXARNJ!#!`M{2RlYXB#T
znhPk)mnlmSayizx0Dse>Kx^e!?53q6BEd8~IR#KaRv`&o*i8@`&Eg&e+_Yqp$r(=j
z*y3;Du$N*xiX-Dl>@sfwc@wrk{Tt+Td3V$B>4m_*-WETWW_1z4-NS1rf7NBhZ|?FK
zyHx%K`z4tOZS3ZG!tn9tVc0>XM2bhSwaxoIZd!?r0e@?hkit?HfCVp#=HH8Z54W?5
z|9Kbss#!K*M4DjJ+qv(TeXHB6-`F$ueox?E1lP6}bsZ`PYbfuvZ1+kw`DsHP>`mJ!
zQtH<*_+e)U=yYe{PSgtPD*wTH@`uvgOLi6ixr6@KY7`{<BJO?Az^X{XE5G!Tgy%Km
zW?QL{U1|v|<0mLYXMx;sq$=|;-ntebGyo>V55`^Bs0jd9xzusiJ;c#65xcz<u=mg&
z)?d|~LN2WD|Hk6kVaq2Sgb3T<VybOCFo>)IV5Q}re8-&i_<NbtjWx9@C*kOPel&YA
z2@n=i#H8=+(6%@vPctPSz|ZjOk`l=6tEfdG0zWn-j*P>Z)U!$>B8H}b(z@53vS;1w
z12Cji@}D3nmK^u%T1%8P$&-kq3g{*I{>6Uu;X$n>_iJ8;nH@u*yo6(^1q2sz*C|;z
zADW8T!C)PqA$hVWO8}A9st$VDYJI4jodPt4t7>WYZ6FPyBmAt2-nxQIdShQ=sci;v
zHb15Z0Lg+{HD}kIE~mmO2g6y2>pIcB?j1{QTzIl06whD_wqPV;rTTKAJeK<4-!G4v
zi+I3KSCC$e1f7;=wQxHYR>$=#XD7>d4|m#b2lzJm?U<tKJ21LzY<H-2!C_|mtpQ1d
z(8aS+RD=uezO|zc72vK7Jl;J^Ls4VHHY|=yvzO}a2a|Mbe?_kgTf1VZT+Jug3hwP@
zBCNfHZGT?g=-SsOs4s$7Ha*X3{k_&n_y)z*GyedjQfaj99=_w5@B@uRalyZR-1pRz
zjcrm<C~96=sr^c4ojT2R$0aiE2mDSnW1#t7YJzfwiU6y|d?nf5Lm_$zV2Ke_E*`PQ
zB{KILKE>AXzywRfyw8Kh=x1sQSr}A{BZrB!$m9(HMrG#)y3BZ3@EPJ(o&3O!%%M4`
zR+6v=s>P5WEc>u;nEv(Uyh#I=ED9ECQ6T`G;?5qxg?M-tTKKV>6u?NBegkUi>D;Jb
zRrLPssYYcJSgn5hby^^vTnZNIQfGkxloDGzBQ}UBCE$?j<VYu|cg@Q#v+zC)>O2E0
z5*;(J#jqt!``eNt_RhqQr^;eWW`<{G@IOXV$Q<B0{#pk1<c_a*@2X*FrSmi(sk4<&
z3OAH@A_lp>&<oqqh21Bbg#4!40eU1Lc#V0P;tCl;e0}%-XW`~MM{94QuXzPRqCM(g
zh*x&}#%@Rsp2Dkx!v3029US|}ykR>be?8mdVbl&A6Uf0n=oTe9o98fVpi(|ZYDc{_
zT6t>Gi96E&{vNpGYh*My&Q{!}r~dIucMOsv1pCB0+Fs>I?F4$8#bLmoTSD6B*gyUo
zr6vb2vkp{~oyxh#*U9=I`?0Xq3ZPV6&;$QNne}(bpPvs}hAmG5PmnD2QG*;(By^en
zt!nlU<76Jv{->$J%CU?5FSN+5nnXA-wDpD>Qryp!-G=#vj6ii}3(f}aE4T7~%Mf>%
zgm?=Sp<ujAL8waJ&imt_FuemH9{SPX0JR|LD}bObH5C{Xt`Z@90!dgh0hAXzRak?B
zr%@C;5<|w;xwIb~3f3s1OaalRAAK~Je*#a|TRpf}%5R0Pfs0tOqR7~Uq;QoIGWY&X
zs0;A7gLZQ%F6x1Yw5T;82}k~bqcBH`X)zP#@nnzto<UOE1+2*{fZ3d~X*Ezno1W;b
zq}D6dEUmGMS9aF|>p6PPP34}D__m&9$&d|mtU2alc!W?2TFOqj6dbeaT%xu6gX`RP
z+OrQy52SJ(o+!Gx!C&<>2(EOMFM%v+0R3*1-F~J9_SrGmk=N*h%CEy0vEz?@Rb-ol
zhlaJ5I{AZ#rAQWM7cAp#YjD8qi`W%O?j_P9(WO!msjr}?Y4m;AyM549<Tw^)0uuG>
z=R%~3dP_nCM!=RtxCtw8(ICnZOmdlBCO732moxS^e}={9Ll1vaL2xCE_I(9vGJz6%
z$Q4hv1YsRe5mpRi2dh8}(*J@-gxlGPc^nj&oII(79?7<^ZJ_ky@x)+1v{%vM6foz!
zQKakTR5VJF`#2gV2NwRYn6t5|AJC${0)kV?(%?D>e(xHUBxVKPWf%=@1kpc-NaJzj
z71%BHLXk=p79XEbZB(XRY;=WAA0L8u!R?#Fs&;XmA54^e_LF=>z-LuMa`wxGhmO>~
zJx8>i8!O-fhCF{696d>xjGe2`MkULMl)5`$Rq9Y%0<qh)f~rpQBR}7qV2&Mla^t)I
z=Y*ni9#wfWChtop@gGW7T|f4dGV+C0(dw|wb>9{;Q!k<C0s(8Y^4Y<m3ZSJyZotyd
zCfgG+<Bv1_J*R>_NGaq)M{Cz#m#ppcJFsA$OQ~!e(8=7+#mv$Q`zXXqwGRWuR<tkn
zb)3=DGQhaME(*>|3BJ%uDbqHr=Nbd4Q@F(5WBr_`7tdW{W(xVeQJ2KN%Z8UkY3@}b
zle2HoK%h9a7dKw#kh$n%6ZK;9Vkp;D1#C$neT32xulIAKe~){^*i68!+4bBL6U~-e
zrCcd^(96AO@uu4z<@e3!x~>)uXZ5Vxh{(n=)y#s$j|&v>Y4?p9)>)x>UNH@h)}r66
zs{23MhMRoTOq5{?<~$M4ae+;zQLZvNWmk#owWSX~7Ly9FPanz`I-AKo2=K&jIREmK
zgqcmYUs5YTZ&S~8Wx2Ik#~MeS=&nc_R~05W&EWei{8a1@dZGzivdr?I)TcILa*bvd
z@?^8q&2UR^&jyOFFn1B%FOG`GRtHVZy7QUc6B9mCD^;Cszm1;CG<+B2<K1pV%eBIi
z2Q0TdXt`qrJ<OM_bma^&PpM?``0)uhszR*sqS{M0xkhEQd2EUFS3Qwub3WjqYhIR_
z9d_a${k^8*pEypO&wredl&@=&KJ&Akxz(UJ+4M{w?o2Yy?)6mj*Yn@G;r@t{MuOWN
zycxQlVY*8srHUWMdVrtxy`uI_O1r{(X8{V{`<+_!m-9)0gRs9zYf^>VPz=3g!9$no
zBDX*yx~$OEd?a<4avsz6AM0Ni+<tOh_ueO}-9NbieD2)czSc1N(<l5ptei@Ed~?Sh
z#!s#jCNue-h)btE4pLYlOau)oE4&uA$NC<rEh>yyp3{B`Ywz1v-V_#BE`1q>5yOhu
zYD%Q+g*cA=mfWKviOf`g^VW#3ZM`Yo0XA{<<VGDR`c`w`Ee+z9=)$%NQc(#D8K)$V
zG9WF6M!<9-ssuB+nQ9PZWgbV)0`R&fTGky$#<WY;3=S)Bd<n7+MPG-O9>-QZiR7~@
z9ywauB5yudFtib%^j!W(ZOe^jzvd$>3vOII<$DURThbC*Jv&c&d?>%K&f_0J(d&R`
zj7zxcR1<Pd?FCWYqL@3_?8IE1M#Nfen6X+MJ@R~I$L79A(M7B3VuZM)?P=SY{d4?h
zCoIlgW;ln^HVy_cIrGcZeOINFUx_p-C(D%5-XSB#u7=WLTDj#LU{#y5oZilyo>4q7
zn{=c$`K(WaY$w7*p0!+*&Vgs8Rhr|$a8L{nJwipwtm|E>?$AfqV?}QwmQjhQ&O=^f
zupRc&&?;G90i#+62ww(i&0DsUBNB=D8AcjT&V3N@cbsCX`7bznOfoAqEA+TI4Rn`+
z?4PdF3d_`kK62MM3fWbK2iVUs?BUq-Dqm^j9?Hd7X_Cs0Bw6}Qq_`N`Yj#wLY#pub
zAF45f=Fe|Sne1itus^bK>PVrk_VL=>p7jzj_u~|uitRrZHBP8iDMwTRKMFwKf)r*r
zi7vQSWx1x$;Fqbc6P*aW^K`v`52Yo{G5EqQhfwsZhb3z|^ZEm)-A^}~ZPA)VPDjDZ
z^47JyIL;bY7h7|d{?mY!HoYrAea^ssE~wONkqoGNh#U`zon;)b#3zs|!q)puuneK9
zrkFt!EZG9w&Y^9r0w@$XP;Tv}<)y^RBFK)do~CVY)~F4>5wa0{_h`;YVU4e>k>@qT
zev0W9g9~uv$<w|+qLRtlLCskOdpsZ*BVM^b8353`03*9Oj$54|wGG$eJ8f@|ksvUg
zqz}P8=+UAZ!8m*O_7pIdQ(NEr#~@6;H-Rt}vw;!eHIzd_Tyu|bCWT|yOnlStFI1VN
zSuONZrrH~(t2qxp(H}K->q17z?ibD$Yxy$Z`StSavHSGMOkgrTeWS7YCH@5Lhjco@
zjJQi{Fdwtf%boVC+o7(3=#hFkIQk>yONGSF^FB--(Q8=insZGi_lmeo3K5jCD#{B&
zXj<#0QjVr1cK}hVYB4wJ&HdoTl1JP@`XXaK`zLfhBmr|~Dnf2<y8ciexY?F<3>L@t
zU&VZcsgHwV$()7*Zh@AN`sW4a(d}@683VzH&*Yq}K!C8K3><mgy-w+>pr{&i6lzl!
z0Gbv!u<(NZ$U&VK8fl<zJP)fN3ezib97Kp}`$^E=bN}~?_E5^OD6?xMi14f<)@zif
z20g#!zjfsMJ8`7R_M8YMK%=!6Sw_i?r4D8vYuXiC5{!nIBI342>wKc}r>o8uYmKgT
z#fH2s0YV#xZcqAYO$A5kp;nrmii*nW%3h7a3bUxHqDW9`F(;~ZbFZ^cDOfaL?r(yx
z=7{Cyw7Me(ZyTS=U2bkA@<~tZ=f)WOqNT3#Cg)>HWSR}+r{U2X#|cS?Q{%zyfL{i!
z14%LDq_P#ig(d_0<)E6!#kb{@WC@cZ$yxkZD()jeWdE>L=4J{>j^JV-UzW1cleFxs
zJH+jcC(R8C(N_juO9(D)*%WP>i4?D|&_K&rGVH@86vW>>pKh?-dSrrr4^p3r6^tQv
z31z|b(TF6$L`0CsJf2#YqD#Min*lJg^GnTj-I)Qc7<nhzM5feu51^oTk`>z;>&s^X
zzp#SCZfi8kOFIlN&D;(vwk&<QP;erssqm9Z?D970Cq<y!eGum}<bbU`jkG=w6H|4$
z=xq-!&rv01+ig|O=Y)eK|CVf~YDMPMrYrOYyEz=<>CnRz;JxRecIW*&#?6g}z^@9d
zB&<G`k2Zp|Ys$Gu7&XfD-<p5CpY(|j4hIAWT!L-F0xoZRatv+;4im-Z+erzP8T9?X
z?kD-fPEGYhorreGL7CFep3_kpdm{oaWiAA1)M>zy9p&WmEbNKx>Kc!OVVvhXB|;Z8
zGGd^E5FLW0FA8(7)n-K#7!sX)K`#=5`Hd%raYl(*Zct!`QP5RZi}^U@8Os+X;2=NN
zJccX|M^^>$ZpP0Wbg7RlBN|?i+r%SlpgLK?w86lj>Nz_?YK?G}Mr*_{j0b5#!JwBf
zgx5i~`b#cu(Iau3EZ8llcuBf7laDK_{oRmzIwb^@e#m7kF<={HSSg#ucVM9g4J(V~
zk0LE_VElHzJaIf#*&R$Y`P2%>Bx1r1sq_Z)dV-f&*_^7#*q~sRf8r5%n0hYoFl`Lt
z9Q_<2^}-aUc`MHoW`6)C9@p|z(88%yWU}}x2I5Pkes(YrNK}<2yoLfRGSBaNIg7_`
zq9IjmN&>mRm*Yv?d<vO%fAVNAse~1owsq#rO|6%rpl}2EQ8*K-ON);3R<FsNH{JdS
zG+gs)0LCc&qauVS2J(|&YvBx9$0R#`k%%KlLn$MoviuC>(D{YGZYQ2UM=*ooQ0m3Z
z^6?tVW6%;;vbczafsG?@P|(tQ2p;OK5)KhwIxAF5LZ8N$3K6R6)COT>-Jnvp^>nOv
zbcaB%{ac+_Bv0HE@1E_FvxlBEeY+bmcOD7jX;y9pAZ-<>S#OoRR$}*5trAIpXz8Dg
zJJH}RpY2*EKb@1gcPFOrDT^K7D$t@QhyS3#Um^!)>#vIlmi%}M4;JpjyNV;V{d?x8
zuB_IS8w8O<tLH=~mBp1#z3dz|4(1ARuhJMERc3WI0)BAr?BpjP3dT+6xcWM7waObm
z^yNz^0XsBUY=_tfX&CXerLmF5HeIv*B*~ktPsG#5nnWihC31*a2KBBlmJwTZDg*)6
z_^;lL1|(B_VW{lw2;4{20)y=*xRxJ#hfWgIZ&?Q4y|`5vOKpfge*EqGU=$oY8vha;
zy$kXF7dNEe7@j<A=|2d-zVSQPx_eD2<)A=pN#k%@owowuKNjjsE5dcTi|83Iz@dpa
z=$-;cyl;O~*QE}EBt;hXeY<KVK7{|1I4JOfoGh|`WE2~CmJkO_3s;@9(;eNys^kx5
z&2|F{lILmjTJ)5-?~nQyx!^S<hX7AdipLS$1M0%`Qg(pcrVo#lWzA?>R02+}2ZBCh
z{3<M6DrfhBX$dmP=vNc>HKCC{%LMj>i)xr@pKGUEYC*IwA}+J;Fjigx2<mxVZjB%X
z5k@lrpJ_IcfjI8yy_b?Q1Tcq{(-LiMAbu>25csza<4bMcNl4YU8QPETzQAQJ&0-yI
zSMg}V_<}t0m5YY3v1OVv>mod3)i6)Z{&o9Ov+Be6*l+sa_FsRS8UB9(e0sVj?yoy0
zMvr3H5X#$MB3<@U!T`=|UKgMmp`oD9L6ZE2Dq<b`K|NGS#pB{2b;1i!!g&YWd3i+{
zNw0}edxMSjFl?o5{407NoH|IVgUnNG<Bzwn;lP)$S+MVjT*+-0Moa2ZKLa(wLk3Qj
z0Z;&vj*s;Yj<X%y0j1wR;)m#V$LsO7M6J=ORjbf%4`NqP2^S*7IOEvE35hF{dq^{h
zrNWG~%w>F$kiT2e=5j)#G7KrILMn=^tVE>UfVbJ0C~Nc~1?D*V&^u+xoHw20sW2;n
z%3MU+L=_)Miy>>yH0&Df2h-&1x=n5NvIzz;Sp_?_!&P^)X2dysTtt2FJl+w#UbmpG
z)hHnk9p=Zy4->c%HGL9t2ZnjuZjRqDsqR0K%gb0zeN$tdvP(~Yl;`8kuFnJDNC)^g
z09k;A!KEmEo7xGX9iXMHy;24rXx{L0F-2?C;&@*{QQxT26nC~Ai-W63+lBFU$Nx4a
zK{ydBuV6nHIbVHwhe0;&8ECc`%fX?IuhWLK=nBXC_^*pt^%#g*c>6&M0h4$@Q3m<V
z3{IJRjWs6))JYRg9(~3ZDdP$l2?#1SAbDIIO^l^7TLWdhF_41MtQv_``d;Q{32u^m
zqe_mT4bv$+W;={Yo)@!SQt&P{v^o*sx;g6sOF5w4HZa#JVP0>Y$z682=~L`R-QI|D
zm&^sa+pp)X|5>Aa7pIhPD5YO?40CZ5j?5w0Jfs@hF9{UC>^J~zzRfhQw*N1DU&!`i
z40$aPUsZ#&6QXQopXl3jFsU&>Hxu@*mn&?y=LWCX?+}qK1zmRu?XkD7hth0fH!jNN
zIh=K4v4K@PNwlzxlwWSTv*oNcm|F4Bi;-B_pDE(2mH5rv;osT7+=4w6##tLxn@~*6
z)Q4SbaCV871`K)ZAX2x_Jd8WZZ8Rp4Q{Qm0#fpP_?nP`xOQZ<Nv-C3T*}!C!t*Q38
zd?8`LMAnvw=gd)0H&8>D-DOK%>>WkGZ+alM&Mi+rqD$F~Z!Q$*T8M2Ret*Zhe_@}l
za%0UagKcX!(WyY&zjJ}xTFxD<l?Br*{71{GxO>+$LCe%{@kx#{@qJ^kx5O%V`z9~4
zkulUR`9>;xV<X&&B;_VVSO*tNUyi9q{c!+WtO;3|3V1h-3$dFa%4Y+2i7K5a<x-6k
z_*7_q<iqj+jzR~q9D=L|ZlyU%d|ATchUzK0)aCpU-GSpSJIK6@%X&2s4^*A&-QpST
z?m1ssc8)Da`S;-@28OC>gCb$LH9=SV9Izsb9vVIh)t=rO%&rrHxj{X=<Bs%{$Z3UE
zj*0aUW$@#pGWXAm_LwAEx%|}USRf@#*&Se0)zP3=An+LuPY@w-^Tvu_O?^7PE5W<R
z<%owKb%(FU*>U9kjo%-0*`W#frO!0hoFwSR$~y=Ty?fc=3wGv|RY?;nCVun%Whcqv
zCDJ8}AEBs1Sg)g^U{?TIDyUW@d;ujWbuf1j^XxKP--(&M&TmwXsDnmM6H<GF4J(u=
zJgD<+!~YKF>r%gw=3YZ7vd!$Z3%NO^4@z_l%zcuWGJdvK{L5X>%%GxOHVdXs0YPTF
z^b&Tc&X-6xcxtHtGs%{Dyv3Yj{SB^2pnH8-i~fFHgEd}6bw&Y_XEBky7CWd8QilO9
zSjFF3oF9=NNukU6RUUK<Frye_i5?%>+)KU;Z2M68xP}yRH!gEO_k1mYnvEUI$}cx_
z5LziM^x$@2WB%`{gSSFyp0QNV7b6A#2P&K;{PnR^v?-4_vik0Yt!=Qr=XyiiDK=rv
zat-87YSIk^%A&;9qWG={U&!2i?75dsUWf&|b&PKrL+Z1!=xam9k>4L3#9tF!^-=zJ
z`A=u<Em{1y>N!Fp>@yyzE&Wt1ag91+Nv{!)Vv|exUKBl;oNNV%uu+XG(iNc*{zV=Y
z7RbRlQvTq!2q6uuk0S?(NA2)$zqrBkZR(L9SvLrq+ztSa3$!3Og=+do+;{e*8ZX0o
zP9T<*Y+nQTn`;lUW*YGl?vA7=u^NX|7&h*)p&e&|ObA4eWhlBOuG!lEegHIKDzNhY
zA8I}u?!CBpWD|9D2Ill!%(&+|VXCj_Z?lJh(>$Od;9pc<HOxn_arE22|DE8Ylm`!h
zOG%0#4GJWd_|Ey(ttj6Zo=>(-hQ}`wy};HVQW+ITYiooQ;kfgOBV!4mP-5{Z<Oy+>
zCk)4pF}^4-w^p5zDw}U}c|Ks_OlpG|p*`alyFAwtwnT_Kxp0sYaVOp>(=73^WIfOq
zq`pZ~gEKC4i)uZfedWj4^ZY)q8y1L@iKy-%Rl-rMx#Hyp)U8%c`0l3F`l5<?MPB{U
zZQ97Qw!dUJ)6-A2zC8c^(UBuZ&7Nfyf8o|Gn{ZW@ceIq&a8bT&!~1M1QqB2Pfq&!?
zGr>&$I~n@=H*m+VL|#r0efhNe6qZY3B!Hyun_B4eJkPLZcAjTl5#O~kKb>zFP*5Hm
zFsSC^vE|%v${OBViF2{zn5+6ha5}Oa(ZUiZICOm{hi2}Q;eOK5=hVlljQi6haW~Oc
zn#UP7Zx!P6Dt8ZM^f|S|X1%!l!Fg6ah=ItgBee?0CnHan5i`$N4g7m8MaL#F0qYCq
zuSTt}(AzjZu_u<@^44^E;P>Aa(#&w`|Fj!%n|UKVR~)7d2m+t#lb>TUUq5d_1657r
zCfh?PX|syzb^{MsEnXnX$%WEp0goRzSOy~j%|kggB5?LlRgT|$AE0n<XW9{`hjHZE
zdzzCHLRERwv{LNIz{=NYnc)wvWyy!x$w1vPSyz{YZhl{X1O!{C_+CJzYC8DK(D*?C
ztn+jB3L~y|+rpr-yDqf{fO9FQbEhU2*52HBHhLOco72mFBt)@Hj&KoNJa!=0dZ&<M
zdzJ{_#1x>O{K1peUBp1F6X5665CAADcIeD3k9T&7bF8xE_5B|XSxL8E5f(6ps<23g
z%$Q!IJ_d-`TKkGu-7&CIgk}}%#v*KX{cK3$X{?bZgey;4Wd%_eY1sKe5vDFg<az$Y
zzD|w6+1vgbY3!QZbK3X48}KYAl`0bIAQgbnWQcsb_`<g924$r_8Fqry(#Td<=L0mL
z39O1V!83KjRnO}RbO4*uKSK-S!Hk|JIDu&~wlrWGNq++{Q=Fv;QfBb!2tqw5kW5ZF
z+-Y+xTHB`+mhrYnQN_m>63Oo1V)Cq|%E|sy3)+imWMP|CxSvHxe`iQDqdMAM%ZC|w
zilBZk$ci9|t(7Lc!3Lgzt$=UT;REl@mE=cXBnHb|sxb`rZPEQpVk-K|0BmE`Qt=yO
zpBY}zZkP3ddr+IpYSn3iwn1bdKbas}#(dJpi=$TPZAA}81oOcMX@(R#=Y;%H%BM08
zHlP+&t~SuLy@q0fJhZ79JeUQ;GXG8PVnlnvU>4&S+WW)9{1#=5@ckb@ZN7lXw-EX{
ztx&Wyp*;6$V<GwNjb}vt4a{Un{h~pizaeB{h%^wiQrX@-xjd3>=n;^2c{C%%{Nugu
zJ<+8DI;R>LF+7+!;sKXc`UAaD?cR0;i(OM6s%GlB7Zoqnj6V0ZO1fa)F394t6HtPB
z)G(kSmRy*7B!|}^GG`P?_-3N^&Ow@>8MbO76j_P%w}*B=4}#_FGcYzggJ!B>wk#aR
zjJz58_1a^HM&lX{54-Dlf!QS4u#a^-ywGNucS<R1;6L`~QV+nS8^us}ve+eVOb~Vv
z(Ar)WU)G<)^nH~^UVx)bc7uW~eG6T5a*LFm!}xmm`>q5RqfEy43eQKr2d>mu>9A&>
zMTwRXZAw&5RJ4WM4_NP0wtuXqAWXlgPeTW+dt8&7^_3<VZj`CNbqv7oYBAROC0JDy
zwY%6f`s^M5ukfU#nI|kCsqJc|k{1_6`@}UlgG;WCjiG2kzZf5N2WgM*_647bA6{@&
zIx_{a8yhR1sgJMe4N-b>+()A5%k+Wg8^L|09z}P?^fAM651;7GmNH}RFki{tF^Meg
zN!@$r>xU^brL_uSy5sWC-G+7OYM?)dOcC~nBN+vLSM7-8@vZFi(W6&YIvX`+>w(^y
zY{ad;K67bx$<~KzbqfyO`t|j%jNx6GkWYM0jodwJZiGV*{O~S5!z^<c8#viwrsHLp
z9oUkFIRJLf<Cy3A@0ejr0;->V<g};z@Ys4g5OK4r8)HcN7`W{X1!O7F;5~K>j;ZqO
zH7t7tOmUJq`b?WE^uaKA{PZ+)<x?U19nUs0V$%bE{<g4aiE9{Ut4N*B-4MKqgT){2
zo7<GRnJ$w%wW{g`Pq!Vta#M|1%A86cp}4Fc_%t!NHXq;xO+cIHu~s9Duj!l54igvS
z&Wdd2pIqBk=~p2fRA@a05vZJ}!Xkgxog);aQoS`*4!t+O?_O=C>&N_UfpoKMC3b7A
z@fd8xyVZCNF?L)2GkRYtRonJ0ta=;g>F0;%9%)4Bm7uE)8E(4NallHCR`;!qv0D1-
z$25ppr&yIqPH)uN<cS<Qj#Y8Imapq)bGXG?Zq`F&R?^KVQ%hnoaaIWirZ2}?z1HS^
zy4H6u`-$Tn<yQ|9_MSjL50gB$Fe?_ezLvgCbMBMO4l#BLvI@ENRk)3LQK|T9^4Je2
z@9vN3HfgLp$M>6zt$Rn&b^AYm0<5&QE)!qn))x;I;k{}k2uF>6OmC&$pI#rTJQwUi
zsjYPI6$nOm7#qFOx$MqwpiqSgr9lVvj=Af3+3C|_ZZ>Z`KNzfcbx6<HPDS*DLL-Bk
zKFV22^`@Gc1gZIC4gQq8vc-u0gX{gseYdy<?s{oLJ!W<ROO{jh8Iwu<BXKt?%#>oL
zr8PdQMQ0@G-9A<$M4$U}V^c2-MMHO`Z{styq9ii3#3UjYF(&%VpZAWo*wv!7*6PVU
zH&)hzB>F#p@7J?7mRR-ejj7hYf3&s<dNg_Io^ZsUc7yk?EiEQHU!=)A2)dNVkj_mj
z-3_J0xf@ifWOtYhVZ*CY{3j(ss>&4!^s24!O%ygzsecdIXmu#~VpxM?iSN#2nTa&k
zoHOwKYF@0J1}&c3-&8@fpL|?{FTF&BH;lJWJYa+t2sZz+{YVV;GnP1HqPBs)5pWEH
z$+l0HX$-KaHZ(fu<*Lt|nRq;;H6aAMeVhgql9`SVD!zf!?DDq$6^rn(T`tXWH$;O5
zFw1=F(F!g*ilXgOUQxxEC?UCIQ+3oyW4OL7r$)SgLwsKbk9bX))bKJ`f3vUVFHvDr
z3f5V=WX=%j(xod`u93)x@y;$naYZkrF8<a*%S@W^)qtL4Cx--W#jKbR9(+<vu_)C|
z99PZ0x^+5yql@=+bA_pHs6wuqt#Oj_{qFUBxyL!h1bO3@)U$FJ>(bwUNpOLWeVIYL
z-VCk|US&p9g2}!19sNR-Hc|WCx7+;fJ}#dM;i|UZG449C<K}tH(3>ACaeL~-+p;+V
zeWsU1d-!K+GB)$J1Z)IoG3n8$K378_zpgxe=F{~0Jru|FYq1}<E<-L#7>l%7jc^TR
zpT+pes&MP5v$G8j?e7p%&F;A{C#7&=b1`?@THWt`&;S=p?>jblal{$!>l5L6w=uzk
zA=TvsPbPS}=}_zR1iKbtDYhFIWt0)i&#xCn`_K1BjmM7j=fg=P{9U3^D^)yek>kHk
zeQ*|jVMT4bfp=eHJete8!8~u3xN)=tb{M1@T5~z0_^(gDOE(ub@paNyJM&tS90zB*
z#O~R$hh(3(C}b-4{G7uKyJ;)r29mj3bmJPvR2|#ZvS#{joPLu$te4=tie27!rAiOl
zj=JJM!ihxOa+5pyIo|i<B_~)*&V6d$HEBTS==>xdB|XdNf7r`^#or4`{Z#L@5YM}}
ztRb(rg>+f~O(=R!9c`JnNs|A__}OD&Af8k~el0hQd$y^~8@c!P9^J_Cy0SBzW@|@b
zH^)~w_l(u~mz@pACyzR>`VXd!&9y(MwuPa`d#y}40w-*0U~N71b+l+@rTTD`*>IvW
z`L02v8cE4sy}jA&Ikf><_3mo2m)k3)wOv!gX1%HCsVI%vlKRAlaQ?yY!Z(xpnDvKy
z`vUIhPsiPOZL@DEhAew!%lm4C(~Pl^?NUyksIrgc7g*hY{~9f@{54bI&#jTyS25m8
z1F{4~KQ4F|PSRGF<%mJmj<Sp20}<X6)-&}vspm2z=8{=vs~s<#ZF+v;9k4G$OGN^s
z?0vHT-U0YbY@rzHkwr}7*!PhnXLn*|HJl!FKTqV3<Ij~ib-jX64X?%S`98U^GA1G9
z|M~Yd%duQ<6L+TD0QR}fnDrwY+D2qLUbsI;veB%|`ABU;BDu?8v?0KjUPpT`#=LDU
zJ^Et2Ni2isM12C8ak_brRAr~u+6P}<NY#PwzzXYpGGeN0r9_2LUbeu!^z#nPe$2~e
zjJ4!<2X)_4XsjUbPW5#-m|N-LsYcP0i~MjY4;V8D{V91>#kWYcM>X1d88}i~7YAk4
znfV8g3^*`P8~rHNA#oD7Mb@T+>dn3^MmtyRN(e~C2Fi~XecHU5WcD$#Tw|x*mwPBL
z7q)WM-%eLTo9Xrq9-qgr5I6iUXOBka(*C3eSAHn&(v=O5!pMv|iG-zw@1b}dII*?+
z!6z{*z5=7vJ(MUQ+rCWhv88cY1M#;71kd1&U)J08thC?Y1R!#7=KUA7X9&^Clh%6f
zX9SXb?m+QbiC5^0qx34H-55CX?D$Cp@F=;4#bx+7j+b$s+%0&!1B6N#Tke0KhL7&1
zIbit`;8kJ9IBSIkWNKw?1N|Rt9guXtJda1HeLV01&7JoG43r}Yno5(|S)_DVI8(zh
zPtgdaRsMGPS8!X1u&)<`s^JNKF{|1a2XN}iZS5)bz6I@rBi+2ilQiwMZy)<XwfNbz
zpMtqpJ)^%PQ}JW%)W_4iOyAH%k50I^XE&f^yF9Au(0LB<?vmo??Xl=yn2?QnOgJwy
zN~>s4oiIjR-PmZ<s>n_(DmnKjXUDL3zcBrE;7#Wkt_;&}>ysd9_yX%N>B<xcs*Q*U
z;>dX?nlG)uM><dBW@W3K>xkIgn^0xJGi=OgXMvshKD@pBpBTd@S}v8YYoS*cDedQv
zhX5TNjRD#kTX01SzD%H|OX;_S-Eg+V6aXIqe+(#e$@B40pQPf|^}O7m345;JIp|>f
zYn?W^SXRc`HAvO~>HnZL3vhr7^1!E(r#s>AWw>;T&mV7jc<tGlnFz#$bSQdT45)I*
zyQxUiqqYHX>H#3+&W3(|$FhI|I0Gz5oPfd35Bmp3r+2;rD_|WE=Nw91Tl@%6^t`U&
zF)-t=4+W<<!=TpnF#&0&1Y*JKcHZN|=ys()61W&bbe(EZt#$Xz>_4N`t)ENl5#hA;
z-l1F6frIt#lj~_WUI!9qS8q&vw&w5Ng?Oni3=GBHXJ(fVxymQIrBw)Kv^q6bqc7zB
z3ACxlkjAu=4I;HHbjnSwld*B4LuuR7R`;Fbtg;t!k7u73Cj?+edP<oyM<dWZEUaN5
z3uU|R(d{Om++oj)4gg<fCYjp0og^@IN{psyT|kv#o}#S>e>coC6wTo{b*V97cbt+7
z!0n2kTk|&t7ZO=1YbLpcb0$|((LzIZrn`PYT;PcBRDubF$ATS72yP=d8h?_R{IX7p
z0Ka9R<Z2CyYf@PLm^M*GiQt|}?ZT?45p-33Dv*kb1iRERG>V#&%fi;yYR?j^rq-G9
zzlQkxnm}Ax8$!RA8Nih}9Gm3I-u~DZ&o!yZswc%9XZ6BaDJ;lcn@asoTTA+59~^>>
z#>_HH<KY>GkBlbf){Afy^xp`jJxUL>?yCRv=kN|H5Le;vsIL}qZM2JPa-W<a>(v~H
zZXsQyALSO#L!dOefaBwrt7y?yBhK6aH+{7{IvXJikss>2uC9=k{BktGS?PgN>*Eg+
zyv8nW#JEMi+NP|EuR}xf8A(rA>Z}7+Ce8yF5f^ml<ODzQ^uS#Z1=G4kD9Pl!J7b-H
z4)fo_HH=&rb6c6bS2X&T{PR-v>%t<uE!>V&mU@a0+@Cqn(D2-*x4s%2=yZv<;$6?g
z`TOcXuLCN)UR9|+DH}7#1^2nKsg3vd%Y*Yi^&L0tI9j7wOv$=qHIwswnIBw~qDwj4
z38<b{bPf0%Q-4ToHck4KpYX!!jb*&d+y&OCUoOA_dVmk4aIY0+RIWm&_w~Vv{xrz^
zhKjE<#LM$MvqM1LmU;LJJsf+G&B!DK>vl9I14STpjzxqm0ZfQIJtOf~9Gt<Vc(UKT
zR9g*CX1Ym5`LBHFLuGJkd#P8P33jSw<J+v<Sn@)yI<j%hmR?(uKTig_W?gk7>(k^7
z*%_kmQ)&Y8M0kT<5QTj_v|_)i-ns2EAoF)t^@{>HThc2#fAn$RK(!IFO0NHo-<wh2
zEUl$9wSHOFa6;sAauxj)GLmcSPToA?LFize<nc_1?8u$cwRC*jCB$GuIn7HyfNAo2
zuT%=`$4Y^<<u*1NM5|A%Fas8Zr=FTi?poAF0LAl`sSm29k_|yUt+lnFvvA4{mfYMq
z(AAh*xKNx6rH|=QhdR4VhQ3@^59e58qjM26;#FHYqZf%a_Qxh_hlB}<CP8kU+Fp*R
zI@pAltEdw!4WFY8#CEVa*?YI+g!v`4;Mh4JTT4w%h48DY&!P<W+H}h{SO_vR7R13r
zGkpQgAxe|JH;3ztQ+LvT+q+kKbmfDD`+qU7lh5LDtG+?8GYofE!-%wD%iv5fc$*#o
zeYH#K{N_1wk4y^#{$?nKY!(lubO&j?9j2D%p1%C^&sl!6KF0j9?@30Q_WoY$AMbJD
zmE}h>7K!0XoIiTIUf<ss3QpHYMg7xk=8Vt@wAm|?#FaW1A;EzWCGpX@r42O(9zXev
zs8G~B81Vc)%0<3CmxWTm3v!|wISNdCh^bb7UOm@8G-LTNU~MS)Zk(_mrS6gy=b2f|
z%OR=kzYjX{lvqc|+ZmvZzx%duZxzXDa!m@-hL|g9W0GdCiYQ!@j0k>qV)9#|2<*7s
zImWR`vyFL*ZYb?dDcR^s)tKXL)(YO{hx6WR=fbHkMa133TSX(ETk4dQ8j?zqUb1e7
ztP{&x*J(Nm-@bM=AD$H=a2lQHQmH;i2uh_&fE5MlMUU@cse=XSY^vinhp`(RoQ2ID
zGAk>HtWP92rc#4oHo$Ni&LBD{XzQ)N-FS8<v)$Cae7|KFLkpU|*Gki4!a%n3x=%}b
z;1}neV`Re~DSIfD&90S}n1VC-81nqhRUbe7?Vht6YzTTF#Pdjbi@d<6;q{{2ET?M_
zBfq4m<=EuPEmnFeESO$bzw40-K!NV-R$RjVp4poo9Bb^?@|5BBc+><s4zR01tk^W=
zvqcec8_vRJl|+dDb9~TXYCJOvt^%@G`{SMW^z<dR69Q|Uj|{AseC^-%9f5F(Ib7Sf
ztPb%(wl=#l{Bh35D1ZI~{vAtAbKXM{Dcw1+G2D5|<$r7Azcl)vav>SE?*(^nKx5jn
zg1@PxuXx9^QAb!}p31U^@4<F&{uzEYSPzrvyD_Q~{bxwQPA}O^FCy#5>W$NWs6?N?
zSi|o2J-%}o%28xTONGS!{X(AqybHtU<`9mLvDlO(c=gYht$5>ih9BYsA2U>fiS5H2
zH2=FzTP!hpUwyh!o4-q!8U^t3{+P34o$9XnTLR{i_DK4SB#~V>*4AqNR(o@4EC>|-
zm}>ofXc>M#4O<>fTls2N98MBcv3!ZP6re%J%M;Ru=2NWAS&zR@f%mGUu#i6*(%6`<
z5h8>`n-e!qv(@xeoCuE)MHcLwaTWD<fl%y&s_<)y>>*kP&*TnC@yHET@^xshzq7tW
zaqm<Fkd5^BfXwc%t7Af!^@eA1JHO#fc22QLwZGj%82}Xi^ifN;rKKssh=`y5bTd$+
zbLy)I>n#?rW(Z42w_X=WwXM{nPlE#Zv){~y<H%$n-X~~GmGC-i=irW1CH&?c3ORlA
z`Hf!bRc|vWo~uFK^Ult}odWpaP5`HPs{>~)gPy+ma%zXO8@?QhUSGTgQ+Woo;q<^8
zE3a6CN4UWUNdjOzFY2TZ+P8wdexR5=WpnUB)p<<S84o@K^vTib`3=G&_J%I|lx4E-
zZ6_a=7~NA>dxnxo<`Ug;1a7cBlncu6iIu_1uzC2@LQ%onV9H?@s176f=MB%b$G30}
z1z*<R{tm$IB>b7eAFjpHST=f0%9d#ZXWKO&U^}?RqJW<;#e;XA#XT?Lm<5eNtO6m&
zp>enP==P<rH9{tiF9m6QtXke5tCJwyX(AG{@3VP@w|h_Mg8YL>m{e`vyD^+HTVGTD
zpq}+4A|bukFmhb=Uu?*UMxq+0Hd<n&s9oj{)&sXI06*gN$c<h;phIk@y0X476LzN%
z@BRLfr+*1pd!w@dK7|o{h6<~63_Qg=6542ewj&YJ%+0am*BM-e=j0RUTUZkPI_WpI
z596r5pG2rRPrxWo?h_iSXx*-8!Ev(cxS;bziUTj%icY-A?$P_<({LNpT$86Aj<cDM
zw^P@a^hlNA)QA7eh0H7TpC-Sur!`;oxJh#D9Cp(^JJ4+RmYds4cFngrV#ZUWI2FDQ
zcJAGALQLCt{EdH8&i_SE9yfU%1@9X?OIvIHySCm3!P+`i`X8_7^KX{3KPfDOU;Lmx
zN)P5!dj8RD&-G9+>-5hiaC3OEb3W!jS4;h)K67<T4~5+Fy8t<4By#SFqHK=*BS_re
z+ktQn4@&YZoRkZTAdfAqk2duoNui;y7p&m7kGX*Ly%(%l_YgW?W^%7CrTxp@{oC*A
zHux`d{eSa9xzoU|n8$z_c0P`kg$2S{y#J%UE00RLTh}yvTBp+0XgTE95!4)W%2_Ip
z70#lTS)rMq6Pk&sA)=n_Xr(!TVGc-XYB*+^=73sid}yNPlp~cWsVSubDI#}k=iYVK
zy?@-h?w@y^-=DA+?DyS!zx&<K`@GL@8k(q=v>CG=hwf|at4(FUL4f8lHk4lKC>TXp
zWU>1YO}XBjw*1Yl{-i<$T4y%!=a^G1MH!1M-X4-o&Yg|2|Bqv58Z4)OKz*Uc>I|}E
zf88-gw<%l*#MKf&p(@vK!}M4x_Btu$9Zi=(MI_t5#NB|aMLb>l>4ZR$0+w$Kua`OB
zkJL@t415sx$y5yoW|PT)S3v&9RJN<RNPfgKsSo;#cC_b*tC)cccZ&8U09pe59{!uH
z!C!tZf2rh@>uPQM6(s&=Kz;rVZgoc!Uwd`+otiM~pwedee{u{Qx-BL*DS2yEuKX}w
zY_!U}G+_<y@6#qYO{NG9dp2{MQZ(pQt~DRH<Bu>9EHGV{U6tK}QC9GL6AK1jZXq=5
z$A(v0vZ_zaK?eVx0$UM~7q=0jLJ#5<$ZS*cVi94DueK`N9uE~Tzb3Yp^|P$@@HThs
z2^~<@!G_i*4R=pNIWl<j=3jaPki&jxLePw|kG3tQ?H+zioiUVXQKK?k_U7=fXs2t{
zdMhWD9crE;Qe{XIqBIA-puypVF2<$_tX~}-wj`JtFwAiy7AG0L$}+#)P_UmkMp{rE
z3#m~+fHuFD&4_TGZx2`+wcgcj3w&s>?Mv3O<z}dyeso;c=Xz){HcT;-{S`E~)!7uU
zL<0H>af3i(#K<s#*LFjJUgW=?|C<NYc2xoHk$Zq8;heJZaYid(^LV2PAT%2>1YkZH
zzil+z7Y^Uu3G@N5fM@q2Scfq{U5^Q|CsYxlBCJEe^pzUZVRe5U$S$iNE!Vq$0OFUp
zq&FhgWZJys+Wr+RFu%(EdL!A|AoqKc?6kizk!o?19|O7=W3%|yFO|j(T`$70k%Ty6
zpuGXFH{K<<S_%-C2ypAgf!j;JK}t()zfUyyvd$D!pMW00xJ$Wg5%KuD9U!j^&_)Ql
z`eBH0N{|r<aY*Ic(uzodxz+fb@`i9kOV$s@Etai$_&dPwLU*d}oUp8lI#3}^-ZEIZ
zXVYx9meopvUgB1W8@GIB1wUt-2l3PbwU<+PxK_ux_ZokJc+y5II08A}eZN6MLHieE
zR_FrcqM(8t*WW?@cR{O^{|Z|BUJaat0QyiG*pZ20HY5@nPcXI-zZp050Opn4NlETB
zb25f9C6>YdtlkPvSGQdN$8;t&SE|*P5OdOim_c~9QxfEqFn#BQ0-{D{G<jJZLl5=Y
zcN`qIo%3mI9<@ozShcrddHUM`_3RjM`7_(^Qlm^;%%6d8H-6CcWI<Ne=Yt4~v^Pq)
zX1RRqme@xR37-H-v9f9>z~w!IMhzKU=n96^yjR$w5w{gGjmrE)i|B5$LrB>z)@X>u
z?P-0xff=4OV^}Ba?!4jXO-QWq$-uN%Fm3JGt66F*ATYOV*MP40*9X5nYQFh3fSwyt
z*^PM%yHpfqAxnW9m#SEUKpn=#M3(fXi?S*ovoqVdF`3?TNnTq#rGRr=*cwo*;}4;>
z&%qs#7ll!9#wGINAhbJkllY1r?5m2Bny~bdlY*R}&)WCCV56)=*@SPpbv<B-55`A8
zL7q!Qmbud+2wl{(vxQ5c0;sg@;<|>zLiDwfH(McPK<v>FVWANE17vpn;TK3ML31~T
zV8am-;DdP``#XX^^SU1DJdp9csw=4tu^oiNj5QeIjbo!FYl#LGj~(acvZZ}xKyuWw
zHkuRA)KCP)Iu}=1p$GwY7m@+We>I)`9#GgZoXwrlW26LXYhUxmd59W);D~=c|C0wS
z$J4!Wgy1fd)GgEIO@)Dou-s8%RP|GfBK-VdXXtOD*FZnZKJYO>`A(q&Z81Uhh&pc9
z$sCOXl{toI8oJ0qeb74rEsgY{XNBEiBSYnkNpi-5D5_gxwdDF(^RX<0InK4RCeqfH
z`i)NUe%^B^u0UA0eh(_h_5a(x-K7fIA7~Vr!rND3ESIhgcDI(MNJC2R00hTz$RB?c
z_ohD|nD9xwxKk72NN+-{UoBdj0P+M_KY|Pl`$usFzowmZN%BNnBi*q3is|E{F~N`%
zDjSKy>}_|a=*xtmh>mQcna}FyO;0$|&ww6i4duGLB_XI<h0745>U`ifLI31B5OZ?b
z1n4LMYy}`>7|j|t#g7dduYahab%*;<_^lVB6%TL#FG-0}KsEwR_^t^4w>~<Vu$BR3
zoyS}!bP^j)2O8U3lj?Ov6Np?gThWvbq)J>7OuNN;s5Ii$DOAO2&Poq7CX?MjWrpI4
z(I`Mi`sBh`4-WHcQf3qp8=(XDxYuf)3CGB;%eXUy2v}Wd-$r`{nH?*OAFK<{CXT@e
zzAi}@x@&x|w;$!+O2$Xc?z!u>D);lo!}``7$^y?Q>J%BfBBwT=J?kEAHh{0yWVBPu
zCqpi+m?Yz*`RLj#oUcL5F2>Ix&sW{80o*A>Gl_ITnAO`)+sOatK5Xm5PkJ!xgo9sx
zSOzTjW%rDxvbitKDtj7e#SZWWX@BP%iZgxDZpd;&rX4w2LSeE{!$H|B=Dcm|v}Sp>
z{#d#i)|F$@m^Y1$4m>3^e-JJ#u%4ARoj!gETlr#Q666gIma9>#sc0l;gzT!sZ->eh
zSoc@(Z-))oeP_0^9FgPN71ZAKJ5b3%GfKU{vroW$9a74dE#+g~=4oa#LmyyvjXwHe
zT##zD)Gs&PVx8c+nnhQ9`2jEQby$%Az`xSlS?HnYk|K&fC^Fj{s0DuviB9uI7ekTc
z*9*08!-|Re!-UA4#DXBVJAxb+r5TA=&6Y)7TM*?6pI>v2V5FvgB*4CsYJAIke2JiI
ziCH%h@`zr7?%&SU5XYY^kUfT~O_?KngR`S%C*njSW&DXFD#^;_Z>sk-#w|<%K3aBU
zm*4CJxDv*O71=_7_TYMApB-vBTjGgJ(`k<pAmDHp+#fZbS!c<oO))xcB$sQSK40Qx
zafnHzPny?%@}Lc2pv=h*p%!|$6<4L-{Fufrpc0JbkXJKWam<+<*R~8O_uGB$Hc*dy
ziB@UaLoJ`#hJ>*w&P!%7Y3%#deWz(kGlwFXpKGAqHZL7W@CX{Bh%RN65KU*10-87l
z!fQ4h`KA%);GTOuCPNEB2-L((1lJ`Wo>4xGLfG3b#(i<)UU^z^N%^+#is5pGRRh}D
zZZF1>^Kln-p*10FRfC8LWBL{Vy{qt&x`=o8^Qq|6B(a%(n6|_0t7^tjQXv-))|8uJ
zG<@8!BSrS5IL&f(79SUlSnc-Bbh-aPJ+WR?80)6Wr@+}C3qV@JvU}MXTTkck5efMK
z;m?}d`S~=*LZb=8UiK>9PB)6|Ug{_{bLZ`~8Xwg)+F6%6)faIkb3HVce>CQ@gafZq
zhoO>K&y{topx&J$xh&n)QV6cysi=KK`;BR0^Ha;jlkoFm<L4QHe=c(7EzE`IXp)kz
zF3;2S0_5#bao%e~FHL60I*E~*L5l|YGq??4b@1UerCXhQn-t|fT;$3v3#F;zJ9!H`
z@q7pMH#pY8Rrm84ypv#$WwNL+Zz1jayNk*4QJtxiWzPg=ih9r1af$WXN0AZE#l2zb
zFd8mJ{OuTi!MUGS7Mg^zQ@)a-o*c0jYf@$Zo%tpB>np45XhsK0@(|C1RTbQiyXn|4
zFkzbiu9t0aUOD@dT|Dc*jO`z-efYS>5AqFh(_iOZn@%@tscW}9g3=H1gqo@P54Q)C
zjk>jFr7zcs5Jz%v9X`Ygvr-gI8<drpbt_QM(#pIqJH-_2O|(H>=;vmP&DeeX)>~v>
zfvIfBU;P%W`?)?iq?9A|T(5IGw{$(8M7TNBUrGUdmyw*&rJv6Qp3UKQWB)!p$|r^J
z|2*$*&a3#VQjp&fxgP1P+B3A;gHx=sVS(75_YMaGeAcUI0B@Uwtb+?z-?6X?^^VK*
zd%i=r_QQsrS<4!HZb00)dQE&|=ADq}l21xFzW)+^Hx*|^g0teH>Qxop^VfghMLkq6
zoCJN`7%bs35>`s8DS5D(Q1-=*s0-$LgCp_C1<{&qGFrn7x2b$IglV?VesoAiK3rPr
z!!k`zBN>5;3wb{WTDFFgYSx5XAkKZ<k{}4=GE(yZ3F?91N#Ij=5!W;ZwpzxJ0uC-a
zO+-oSmZtbW-_K#@MUAG`mzWn}4vZOI9KJZScudH3OmDte)lHF^Y5*e*>9UBqk}h37
zTsU-$O2Dhz(R$H6@{LvcU}zIq|8q1gb=XM!cwC%5p1V8oI$_AuWhqMA0W3RBaWtf*
ze}jA^TJx#*aE`uR?BuC_sTjv`2}MTZ&V=Y3F+KadA@uemXy<;7)+T%N`xe4sy#ZRC
z;Be3pZq|@Nhq+t8>co*lbZEsJ)sb7;4s%>N$e`v%<w~@M=;mf81QuLNoH^07h-S7H
zn)@M*y~<f*wIX@P^n`~G%<_CF5t<$FlHtPTj$8Y^G77Txuar}#Z;n5n6~Eg>4r1x7
zddzhQKPAqfO6o^RJKOZjaK4PjD{<K73>8t<7RUjBfW@LDRj_SpBXE^I%F<2R8pY>w
zJz`>;F0Jj9#R*UFZh@GMcLa3V%_M(m3MIveB2&*Dno5lCr=bbsfbaP@Z!KSzhcho)
zTAFrI>Vgy0(r!5yJH=1!czZG_DG*6+ews}f>p8BNv6`4?_n~@$3?nv4gn4u-hiMz~
z+)ZYNc0-=LKuzr40^y11xtQ@Z|Dv#w!1Ifv&iATpJxhzMZmrj-QcT(>qo1q7|2RMX
zaArapvA7~D5+@uJ3akyRb*TH?GWx$(85S1oz2njG9cxB8=_L-*XR>;ZkSTJ=7^ikb
z_A;%nWCnd=cByBXW=0+QfN1ssAp?RBJdX(~He3U&Kc%qo0>6lB6w&1eBHw|p+<STJ
zOA`Hcc~#<?e(|SBTv1nq<TphFrt~3R0_<FL72uBQOg3EXrF)T5Y~|aZb5@?`t=G8Z
zFG^rRn3-LUxzRDHU1*I|FnZFw2VvpfZ+`t0?NRwSV7PhfFv2cj*;Cshy>P<Abh&|*
z_8(G;j=o3N<E6A$8#Q@{evUj=dlQk8OFw5$iJs4T9dY(yJ_G&2JDqcFpI{M+vdNC>
zdfWyQ0dzc|aG#p_Jfj@e4QK^>M^L#75BLHHj9*;H9t?^D;Arvjx$97_fY{jZQ=|u^
zX4IZm>Nj)F5XO;Rc&s(SSO4aiHf+VcsL^0R_0OUL;lSJ;D&eB9&x$;=+m>cU3U^qH
z0}A{Cb%C5Sxs_v*qP^1B+uH3#a0u_f71lSWAPkLbbOsU*fC}zIoz>tHCvm(MV)3F?
zb|MnCC&l`ujZ^TD%l)2=!mK5}W#)VNMh2TpMiF02<eNzoUuUTN__YckkS8gaXQM`a
z8fGPa;`&@X@$lbyKQ$D|HPu1fKwrLpAAfJ7jZU<=sGG*(KiGm#GKmCPStw-4e1TPD
zeNx&ap&*iCgR$?j@Tfi$)mlETqtv#pVujUA`J})J_H!<N<~5G4tLACl8*JM#p}as2
z`QkS#t8s*f^;<>dwxRt$jBUE}#a4hc#lj-qFQ(lPqmwKwE~GoyJ=&i6%mKE7q0kK@
zY7mMs7ix;!cW}F$EfyY82`MuD&vEX>#yK+Reo!ZVD@?SUHHR};{a=Gq6A5MwXKAcU
zQz0YXIn7T;k#>6)KlID_uS`N!z)KEvu3uq*O@XrWh;yPO6oNF5MzWtZ*alJE2rM=o
z?RVCjT}rA_W@Qk}`E)>LUJHnm_3|yo9*t1}(A*UgZ)srSGO<w?aPQYci??@aJCnt)
zs7)9V3wf1NYjAt4WRw0*Xw)01mN!MhMws@Ci!QrPiLB1zv}!<w9W+X<<BBV)x#t(P
z5$k4*W0x!p+Fws(2Wa&eK9j}1J|Of(;#DhS2;xj0(OXs!?B`2pw&%U+aYjCDT)S30
z2QGWwpR=9qgct!-v722!vuO_q*|`bIz@DaG_JG~y_cWv^9Hbe<e`19H`-0ZSUj2Xf
td=dvsOKp~0w~J!tB{%s7|IJ9VEnENi=!7F)=-T)SVP*ff@-v=y{tbhb;^zPW

diff --git a/reference/plot-4.png b/reference/plot-4.png
deleted file mode 100644
index d9bb453c3c988da9f02357d2cd75e3246b96e34b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 40312
zcmeFZcTkgE`z;*%V*yk|R0J$kX%eMZD+G|<OHd&oT`8d#QA9-vgl4EJ0YrN5pdwO2
zkrqnm)eu4pBtU?i9pC4j^ZRDb%s1b^-<*4fBqVpp-q*gab*;7b9p30_tFaura0~*0
zu)x&s>O&xhd?1hmkN-Igo*~44S^;;5A8V@J1^3|Cx>FMj?v6fHH}il%xWk!0`%ts1
z2ngg71a|k9p-=MSsCU2_uf66~n!x3A8rMxB8S1C*g@s%>tE(q_s3`TAP}b)i+2xZn
zeU0J^pF*=9OSEMF$xe1ht`uI<FuJRudrMdM*4r!5r>{J>4Gs>lA3<O+qbYgM2Pmt9
z%>oJ(u^S4M(Uh*Bd3xO65&hqr{~G~j1in_hL0`MbvuszaKX_=F9Rm4Mz6Q102VQw7
z6EB$ks*$8!aM{Br4FXXui;E#L6aMe9bNSoxe;@Xr*$4mk;g`&RR`&nrv8q%PpA&ef
zy0G5jc?dix^uNk{%hwo7SPGw4f@dxkkG^15;D#>M?vX*)$FGNBoA8XJDbe{$Fnd#k
z7rM_>|63d52F!#1LGc=!K%tWMezbP1XBgHM?Wk!kMC6Bc!G@>Mb9}G^{3)W@>OB?c
zOm#nVec1jDIbD4TShFcA6H50l!=6WHYWm&jd3CT&z08@A71S3aw}eXY#A2`V^+y|C
zBeKJmVSWZZi8lvC`uC$3aLJ_bHfP!==)CP6n7`)mWsjY}%ar<BL5Ff|-uW+X@)>I*
zS|%3%nLhPWpzVH+TfC@z$@p$qPmk=Hm0-60MZ$F>{jkPqD<@wUC+>>1HvW{8(epv*
zxqI332VqjvU$Xeh;%=&F)K8*JY(t}n@npK^<Jj$cy?5uDUz(1G57-8MRFAvwK9+h6
zqxw+P^(m|YE)a5V$$d*<{ssEJ`nA_Z+A&sZo4*meAvkG6^=c&zGl?2((s>&zgFx5J
z^gL$3ArSu$NwX>~?3%wX>-5|>I>X3+e4MZ2DU9(6TW9MR_e+1Q+c=2cx<2uazvE_;
zG~;$8HauV{912UZ5xR;g4ov&SC0lxkZ}g0f(iL6~y*2GVtpxY9h|4aj0#T&cJ>&2V
z_M<C@`SZ}eI_azI6J8=7cVmm;UND+NtS3=c#$XixHy1*bet*&soj57>C+cc14%!W!
zgSDZG#azF)n7u7XnDfIB@CnUk%^sioBrf&VT<;EirWNas7ADZL$05rfWJ#y?)V)Gz
zo(b+Eh|mT2a21xN?S$M(;ckeVZqI3cm6K^p5I6bG4#SGFzmtB*w(&Cgz)_vnk+<+W
z0-Mny>>r&39iou+P4ziyd%?)o@viAqA;Q$zck7fDkK-FOy#?<gwc%_uYYdT`d^I~m
zi_JL8OGe`>M@Ft~D>ZOI@S8t8ln}iDajVC=h$GT?_{j&viOSE+##vYPM0?Y>o?v*Y
zsJ>}ev-8`pbEQ?zaL4vZXw;$4Pt<{B3EE%#|A<B3HS&e0ref?*US~#QAE6^X%;0>z
zq~Bss*IsEA@|g%cb^nMh;vpU3FI)MdQ~Ge`L{9BR%&JxReKQrI8{hln@Y{0ew}?<#
zSCY)rY>Q8E2?mKU_oaOpL0XX`@5&%sYitMI@U6;IQ?;e<2J~koSi56~MjSQdP3KG{
z3>FU5qCL|zM{9qb?Tn$9r%a;fPQbFRQ77+>dAA0e#g?|7SYbu!8`tg|^Gr#~c6MR{
z0puG`o7n4OKG+tUd#j2N@m|#t>7`Y5?=&@E`1ckBbYG3SwIBAVhJ5Sg2fU~a9nb6}
z6CJw(Ni7r-#*QOzw?_SEM1Q3CzTd}u%TF*z-@k7rc;39VF4dLYFJ+=QDC{$v+v%J}
z#f7+b)moDf+MJNrJDG1uksdetUBv#b>fnp2jZLmAy#&cq<5eKSRfmQ;;LuybS)PQM
zF~V<=b3?uS9YUy^6w98Q>*^jwnu%^!__W31da;3KX?_FgWdq^8Z<A1zd$n(D%+ZKh
z%Py6wTm~AqO%p8Ama=O5M`K|sS{$=)j4QoQatz{D4HBvxB-Ev_$5s7D!F`Nfb?sgu
zwgdVd6|7lpK4-jifP>(joanyHBByDmXe`qJOPnYcweC@tjOBGm#F0c4dlZ+}wk3ER
zs7Rl@%#rYTAKX=m*7d_X-w|f>LJkIEG4wuE04B9a8-H#4r!$etmM~D@;ghG@6DCel
z!6YxJqs#q8ol<|g*BbIm7I>y6WjlfeQGL7m`~(D2JlgPC)a`Pw&C}BM&9of8IGolA
zPpn$?D<0hAqu8oLZ2UJa>nUm=erbD6ahN7AulZ%DUC3GvhSR4`VaD3vs7!GO)M|@N
zKeSF~_sLArA_rzHc4+27!T_P|AmfLZx0&~w&nT1#CU699XBx+e5HPe5^vGA^LAfUW
z)YXRa-T0Q&>2bq|xjyQW`)fgn2z8$a_p45>@w3FPS7{ZNmp970PQTO;czUYi0tx|d
zRFGXARxSwX?-DO)rgL{WS<ke*vmtFfotB!k{EF~P+n(G|7S+D9wK{x_XqJGeySm4_
zP@HNZ@z5d?D`FeEM&VZyAYQ>hX7`|t883Mws(5@k?h1==Wl%=6jF@<={|+JnfwUkM
z_`J|`a$|S0Xm9WJc+AA~WPcKLthn4yl(N85*)hA}ycKC8@&d7(Prpe}TwL2OJC<_9
z9~Zduumb)KJv8sua$+(n_tEv9?;<UtzDMA_O~jFQeEC@j<nDJ)BM3y7d*|AXTRq+A
zUv2N6bD><J74O3dGeto&<XP=n*DTtFh5=ZnTiy7F$u)z^z4Sh|0@`F$Y5x`F<NUjS
zR<@s`%`Jk{X3^C1zDK4CVm&h%UlCeCv9n^+Eh3}ycGrf@A&{uY2YtX=dqGWl(m-gW
zVvpWE5w-rY-82>6`Fb-ruesXF=8GmL3NhXsrJUqLj!<b^Z=q0Dj)mHoz?T~jwRNUO
z--3D9h?3lt7^8?Pi)HIp%dDMVn=KZojnzOQz6@2Rh`LGCpDD_Hg>9T_5iyvzgA)G7
zC!dMMJaov^lK7UxUlx5n#-mCzk$Q^35TF$yb)slulT!Q}&&<lLmZmH1a$MZmG-@71
zhRqNJXwx)Rm|$j+`0Bakwe6D_c&)aQd~mGcRP_9F=rl>bO1no1+Puh09<d;ZJt#he
z==2f{_rsDd*vny*Wn#KP&j>m5mnk3xRd9hl0$JZsCT!okd2p`LbeBO?tN*(J=h<?J
z!`JrS#5ihJ-zZvGEV}P2>TlH@&X#?dxIJSeD3F03L;BQ}W8a87+;8ze_dZZDzHYfF
zx1_n)mfkbFVKA%siwuW0^86Ix>gw?MJL}tj<6T~((ZhR&{_}30CER*pWmuDE%Tl2C
zM$wzw6;d)$>#wy!20GX!`y{#Q_~yTPvN)NQzDeiHj%?4PWS|f87dZOICh!|oi;AF$
z9lr0e2hMDZrwOue)@k*yLBIEU)(p&Uys&}?%V`8zD8_SxnzVv3BGGQp4tHNAHGzLz
zj!b*-dqtJ1>fhj<#gay}5dUQv4|Q7B*K?w7-q+WGJwW(<9<o6n<|7Ldk2zUoH9LrJ
zu~F68m#3c#5ulCyqv1Rbohw2G&ovVRP&e-=wOyEqb}iCc$;!V?uY08hsSVe9C>F5m
zBbZ1LA<&4nyc;Bvp0OzE7^>EQ-^A{Aq3Ht6{KVEQ?qKz3`SJ<aA$+wI;+2RgdxmAf
zw~90xapqiw48A#w2az#7qUH79A?hN3Uyg3RP(^dcMX$S_Zri*q@Oy@z@FIF2X2Jp*
zYIM+tMdf*82zFkvErQx|H^c<(`nD*@Rq5$frIR*6lc*ZScz%U%Ng@tj$s0*RH0m#r
z|J?x8PrZ5q+CWXsF-!=W;IeE4-KSFV=JR982R)Kd_A5sAw_u#=em8qLh8hncURkvK
z)II*jNBotS27lmbn;@CCz$<qJr};;Dnz!G4jiuX0XphBs=GTU&JvHF&?0E7Q!cH+E
z?4h?b%&r~Bp@%1k1Ee{c_&JYji4$h8xo|DJU2V5mF9UlS&6;qtS#A3?hEOs33x3b}
zD<4Zj-|e1Q*`7KP;_KKxp=w_Kj>zIp^(N(`lWyqYww;jE9S6|j#@jC<dJzIfbg?!+
z8~zQngVRpioX3s7u!2>EXNrIco@{!EA>^}_@wgyD;T^WWLUYADG*=G8aJH{BHdT9?
z(fB4ayr{_XS?F^@I>K*GI=<?jUy9a4M1Ff7-nwNym%ts!UqYti-evnHW7GM{qKD=i
zS`T`}w;hbM!!0IQL-mG62ajP^5hUEf)2IYr-pDA7Y>wd^evi%8b&{ccqkm^K1^QiC
zW>q_snuCq8E#J~XSHtABGAni-!`w9e;Ge!d7LYHjmFPXY5x=g6p?!(Oy7AWu6VLN|
zPZb%aBoY554S&1#|Le_S{(ek8^DOE=-}&a6yazW6mjUNM+Q|LqY357cc$rJ{_t1TC
zYT`fC_TMM|Ki!2W!JS|n_~|z%2Ke=KTAkGf&dkpnMUEqeax<|M{<<qfe;9!;FM7Qm
z+n_0Lsej3)j52>2Lz^l+5$Erih&%N4MG?A<zdyof*Dg#s>$6hzeXNyp-I&O$HBNI<
zDmS9;A}obpNz|bpD~6U<PZT9ygVBHK#?LS)v~y0o{7|}69S|e;+nj?RIMrD;)sJ7{
z)oId8yMhU%zQmq}CD~G&YtZ5>IjDp>K9>0P?6L<qzW3<T;I}vy&Zzb0Bd849t*p3U
z5fUTQeSF@vdCxB={{>(FGvQ`!`Uf?uHICsqH~xNaGYK7i40aawCCQ+AWtoGfxge1J
zEq$!qhdRpzZC1-Wx5kX&*OXw5W9BH0rr=w#4^rcYBJ+hiyK*iz;EMZs`CJVDoO|Yf
zkr;(6Z4Wa>mG<3K#@Q~6ir$%(hIv$}mMd7!jhtI5bgukdaf_Pnbl04otZ?;}e1Joq
ze`K?Kc-z4rpZlk!F&{r9_^1R^5TPyIZK;p)(zqI{T~S%-eNDzE@4&gv9Q!6WWZiQu
zp)?L1siRjF-5IGqE)`k2S~=dvct~Q{fk_r!q|c(?D)m(F`sf<;h&rW!wn_a$i5}kA
zS+}S#?V9IC=f}}}&vey0KN72HzN!NGf^=sJSXJYen43Sh{ol9GH}XPFxNR{!=&=|>
zq28E<OE4+h$d~}wdxRfVJDLbPo1Ij)*K@N^ibsm0`j2(pk=WygM^L6+D8Jkr#+eBj
z1Jmh!MtXv+>;Jf3CW>I--|s=6_g%z5hE~z=7|$B)5|8iJ)=r?-T>gtZ^mf~Nd3za;
zmu=|G3l2vQk=gpRIdjrZD`G1|$m=LA9~;3}7JIaIIPH{LYk>3_SmM?B)R-!?orb)w
zz~ErJhPBR|+K-E{UOp*x_p0axb-!+NM%FpJPDP82iZ!4d_0hy#MXS|aRqIP7<EWS+
zlO}Y~hABglw6(#97lT@*wo3+nZkLM{Au7X4ta_TAs<7GU{21@+Twu+=AhqL(9fi7|
zQbl^9=IINMyh#}(Y`oSo;Uo4&AeL=eF5ac?hXt+=CoGf_>8|pCxz;8A%gXb`ism|Z
zC0ZB0TUckRJ<^?{E{_QLa^Oy6ObCp1jn{EF8YtWsRydnx?rZhoWo>O=TvZv63a7(o
z?XUF_@Wl9?qj>j%M~8P}xmx!YT`%7_qgTE$;iW+3y5?!Q{!Y?`8*^D~ak>B*yM5U9
zm6gyyCwJ2hUJOsEifBijxfK=!c?*^}7Hg9?do?|l1=4?53D6_7XW7Qi$95kFv}Y)G
z7iz$58{LcW8xN8oy6_UoW5!OlzwkH#a_+qF`D?V@n+eBlYm)Nj!=dKK-DtRjUJ0=6
z2(5`$+-+aR2lnN8Hz!h+e$RaOwWNE}B4?aQ3xzv=q%Bqq<T0PWdPqad`iTd|Rg;lu
z&#Ly6Yq<2>Bp1BK_V9L1;0Y3E+xnQRNUz0~DMqz2;RPT0D$y5aB%Esj%6;1+m@S?-
zoAN~tY9nPW*jsZB@>K~ST8Q7VjrIVGx#RWREu`sa*UwsAok%F>u||ZrEzVXT>AWcW
zxl*tAD)O^BtUe3~hp0FF3_oZ^ReX!(W}NzkYc!ru-2#$Md_(~ZLA6n^i6}TPu}w@Y
z6$?mmYSproHu?0%?taVqeT^(_JY`C@mn>K!CX&bCau{ee{L$R&Byi&t?k?uh@iv63
zM7{MGr$>dk?@$FNEo%L1yU%wgTff6q4?25?{8C*q$k#mHxkjy*#FYzOc~Dol?f+m$
zAJ>-Eei^mmo^ny(hV(=Nls;*J3ArPxQn@Hb)PON!<Oa%HMuHU7YqU+{ve%|yCHL*D
zYx*-(KfUoyt{*5bFK<ePFgwr4cOKvS)AObeY!#MD;x}R)((<F$i`t9IU^Uv#J|EkC
zC2{8VDO#$TdCk&@#vt)1;yrU=iW&O-u`|bRvgife5*G|o4kmXQ`K!NLnBgSJxr_E+
zI93+H@PO6aof~Y?6KY9$QZ>0|tE`}bsCP3|yG4>4w$1X_G`~bViFw4X9qXa1gjTvW
zxp@ZP(GhX%&_{oI`&G+0tqB!1yK};zYCaCv7|un|>>m|ul)^3eX%f@agAyjK2OW|H
zer>AG?e~0%Tl=W>v{i(WgAlJw?g<i{$E4Q1Qad-Lm+-OO`A^YY!Xga6=)%KCdt-;k
zDLsupMx9ZCf9Dl!Zt+(M<|<!`*2A;3DalH8m<Lt2n+9q-;ARwE(*wDXGBCGltW~;|
ze?(rSK0qMV6K(6#9wHTHKS@<+9`9^5-MR7=RvrYa;UE07tY+MKafz{;VR$DXidnfI
z<=yCgG^E5cdvMlG>xoLU<7dJnUWwrNqxe?|Qo_!lTY2BhCNSh?(!}Un2c34`@Q`f%
zUcZ_?Gn{Yw%j5l=zDR+Oa&V)YcLN@g;)8X1eYkZ8@-X;UnA7J$YL3z=)7pK?74D5z
zZh0Q&gHA;!x^=BT@$M!zn-$oyb{D4Ire9w>$EH(*YG*6Bg{e_@6hc~?60ZdRL$j#x
z>9VI&+K5XDKH*QI+D!*;VJ0L&q;<(+(OaSPtipjG_8F09Ls-m%dHL8Mw#1w__{}=i
zj;>lEig-+(jjH<Q({HY<@@glcZQZfVUz9ul_n}Su;f?(RVj;tVqFtw;bs8SIYJr$l
zwgP;95BCI1+j@{w-fyf@xX?OWX(G!0>)`aAVbnZ-$`$rAZ<{D@gPuDt)9xT5>q6R*
zyx8B*ruxz))IL4iGHyMX#+o2^m}2Y4>sA=A7D+U_>YQZUcCc-6H${ZKWynrUMd}Lq
zsykae?2{xi!C`<0IXU}^CV-2>2nh+X?<`&MIOUQiVUvr<aIaU|;yreVm>)|lbLJ(`
z+ukYp-a3q<#AK#_KC&ZTg5Og6Nuv6sj<+6UB?)<2k_CHtSST$6@6&9gqRIcLz(di>
zv3>VDl}@D<<1vE0SG!!@NG)2bh%^zF7}EyZrFo~^T34#6VDG*$9EZFF`up<y^3Osq
zeCuKYJEC*cCqv|$f9-9r$XxsILIivtiId})vtYWVg^knJky8cD7beOmv}zGyJ=f--
z{C1QERnj8UYFWIfH%Nezo)K8@zwso})-B6`&8fjFf)>E@Ft0&)&fMS0b+$qow#BrH
z&iCH_bUdadwUGK!!1eoGg>zdO$!_d*N}|gx5RC1s3xx>^9gm&ot;OdbKIRS+(3X>a
zj4^;nS$_%{O*@n3dBlekxAlW`J&X$bTvz^>Bg!)i$qF4poyLTO59l6Ju^=sxN_GOm
z_8UYiy}uCFKg%M=8=hr>Bb&{*Og8UbbstMZF3+~WE0MY+j_{lNFc1sf&N^$d`~1CH
zR~k=}#QZKMUAJH<{}fKU2tIzLmtFiUQM^khvBA{gm0vE!%_*e?7M5vVz<b)ptaSpJ
zD^Scm*-Z3sS6S`?5f`NqsBXphsw-0)&O=?(d9|Gzrp>JX94aG4NkpSscB9@w+<ZQY
z>b(irJFD&|A{w|kZI9u-Zc{gXg?r!F?XVzcuBpNuzObO3)wRO1u@u-f6X<;338$0;
zSK<*iAu>CIs9LVa;mLyQT=`b07@y!1vf~=_eg4C4qf;y^!*zrg8lAc^Wk1#^+dKy#
z{rhcJ_J2zyLrK2n?e}ymCGnI?zT-b=Y;rnKO6rECz&wE{ja!GKkz~EIOK>?Fe(>|}
zdvMVTCmsp*kS)ETV1+R<l=DvX;%-?2+Qa{;Xein>y6QFeHDXR70tk?P6_K0IOe=3=
zzAYJt{Zf55fDDT)Xqmu&!c*L!(}D`z;Wu*UFOmcH*$;T<X`$Lb-Wj&en{S!!V(m;_
z(#=waJ|CT*9Z?u{2rZL`R^T=5-rdP6I+<e|s@&5#t%Hvu$~k6PW&i21N=>_oK6*^k
z4th}rS-2GOdLYTHD=64>rIE~R3X79G8i-dFPvx}%MZeUGPZgoT8WtxxuDmE3Do!J}
zCKxwY2E7+T85WKY+-)9<J5ApaOE|hv5iOTs@=}aolGy<lVUNX?CPJ0EB5Z$g2HkvK
zyEv^9&ns&qMhn2K2D|AmHp44fj05*;jQ@0|N~VfXwrJVyA9m?&{<j<Cpv_^n)CY>U
zI`;4$r7+koUX1;R3cY8`Y#&7URtP7AYFTxRMQ!5d(xN9OVlV(RCgmDmRqJ}BX)#q-
zd+|6+!TF*rKhY#RCEn$t$d084TW{3@i^B-YdHO;=?4-Q7<+dts!q78(w&9$toTc4~
zi6v;=A^z<0_NK=(v||cUwrqzcGUMCxMy3(3riIvY=8>3)rVcz~a&b)6yDgBqOnwv`
zex!qJV~0(CLDYW|TGc|cDQgdNav=xknW_m)SY4)#Jw{ed+F(+}J(MF8yh9%4Z0uMJ
zI$0t7zICB7J|cgM;6kg`2f?tQ!Plh~x9b>JD;inn+M_Vo+&VP-I>m!o3e=wlS4&B&
zZVF~|k*|lL6&DgLP%ik!{PA%7w#Ahmdt6YtVAMN&gaP{3wKIku{*{NSZH}^5TV36K
zqQtvctZ3Z2y<>HSz1`)qhx1w7yEnB~c3cALj&TpI*rNqKG#PqmWwd8%vdVF3i>|G;
zpy0*S&ggy58fNivH?{2sL+PJqFk)9wyZbRUAC^r1sKI0?wvo}+wn5Ix0<>1g#{Sr?
zXRdE*gL5BXU5g6+P2G&$gKUqg2<UI5pM+ppz$Vq82e%0O$*AOJLxKQN1wRg-9<4gm
zyeQF0u!;{am~L{)z2iT$CplwtWXiay&^5D)3|!-g3l(E7Z%(Wh!ZJB^bTAg8{2n#-
zy{|VGA-N`~v=UqKonN?voviW1PZ!6It3z*1sS8?kxhaU248`Q+^>pro`d>*wS@s;V
z6?hnpOyO(eY0w?^y`ATe2o(*Er=NQcRz~S^jYU^M8(|Xfv3bx7IUIZs#wBhr+Z2<j
zF0fe)TXG2*eL?piUugGN?P-FlT(J`AO{%^vFwqa&+mthAA7!PV^K?vi$$GhKh5OnI
zhkh^KDcl|!r*jGx1+|WJA)_SXLt-pe@DYJ~%{@*<ksB9>&JyVPLcan=_(&8P%c}N5
zIhmHyCaV3ikuvGS7fVd}mMtbVU1?@7aALl&;i}-uF?+V>6JDIEinC*49-Zr}rh<e`
zIcnoBAEjy|r+t&a+eTMI3a^{-xCn*aC7tbv#(Na|?GF!T`zpSy0;#+Lw*LI{S0d`I
zXGU<8D^9*_a;TUjR`@KwBsKW$W!{>B*5_}GvWDgEb=DEa)b;l!r+?o+Y`6F=mC!UO
zi5`pHQg(met_$%(i2o-04J$pmD%fyT=*+~%v0=i#{(&EGoWYYXFWoGA`V*>FB;rcD
zA8je%DR1zUM^GC~knjz7cFc-<<AF2OdnB4Taj}%TnPvT|Zmh{Kves)%vl{f^Up!|Y
z7dOizS8l8gMrobgJA~n}e!#ce-9`!sb3Gp!KF;jHEe1~pLpj{-m&Y49>P(ze1yobs
zZNpkeyZG@QGqPV=AJcxi%~i$wqPQ%A3*MNPUs_!5T$AWeYjEOMvfCy1ACNU#9BWjv
z9UI7BTf)?dFS=DoRv$n|mg8n*fV=#kYXSC<Bo`jHDk{p<V0MRRP{mxkXpbQcec|Z%
zzTLjt4pXBs7U#`|ug8}R+<liFf0d|Dax37iojSF5m4!WT-xn&_gZ-A89f{<rkT&UV
zzOE>tWJG9s!-HG=c3-qkZgEw&%Vb+Xh`|2xh|P}!dHcdbbE1_Vy3owK?Az84S_*j3
zD@QfuOzfc79R^_UsYZX!c_kTsebs*Q1bW?ekc3>BI}-Nl@@v~J_%iac9Klcd&Ky0K
z`_GaX4UR`vSv$^;-A3vl3sRltTi1*IbQH!H9BDV#HMoRFXSZNBN8ol<qFo6`Oa1b$
zVqC2FiC&nH4KqBcqFIl#l~I(~f*e%2A9!{7k`Xrw(QVrswD;3bx!psIvG_=@R^V9-
z_Qj9A$P>8<Fu&Gq`l+p&c#VPig55^@YJY*bmUkY(`Pc%){@2>m0pU5ie>_GU^o?8B
zKUC0g1F&lYXUR$}MSCc(C)kCD8SPJ5ED-xA-EhA2#cz(;wZTs`%9M|7_<1Kz^s&)d
z)2z?Xi}pjzBEXiyBBrx6wF0;PaAkq)l~lIG;*g(C_6w`;3(w&v8k3ffvf{3GpohjX
z+bj$HybTOKNy-X7nSb`wN#x9L53c-)gQP8yaKt_s!%#@*f*h;s=3PCj(wva?%HVrY
zCC5fLO_d*hL4r`dbM$z_3dxV;R7*A31p#&;fBjCvX*1WEr|i@j<E&OQ1I)84{6x1?
zxJ90@BCVSrve$O7LtTEzWAMYakqR{?)1PJ6%9^7#7U~v%F8>W>WAI&4p55|zM0)N#
zUlNsKN!mKsRkq_l>Zh!2=MI&j&bL|0p*FT<FyHn*p?M6~N!ilJHWcG}C|gN_M0e6M
zTb;31aY(!DOdulkdCH8kvLDtzcb+h#sMSoyQ%uR*FT%&&N0*)!@lsLoqj`+<jlFzV
z<Cb<E-OCi;l|<z3_jO&&vV~1t)z<My<3F7r1Y>JIm7AM)X|wM;02pKckJ}iD1zL0A
z(ms6kEwZ8+&X!b$)P7^TAASqd>OB(hCptVO`1e;gr!etq)_9EQPx#rPt275Evw_zG
z5=U_!AKEL6ki)kz1o<O!eNAF26$88C5#>7k9+%G=KJj^LD0im^?9cYmg*>}CqlQef
zG`Bmy;{3>d$DcjFO|O2D?EiqkzEXEFUfCq`$T-6`ROA-fFR9d4K<HT8$gst=8<~hm
zBW-q(u8&$DvyJZ#N8^i+k4nQoy_p>L&%I&r$wky)JvA0t6|Hi;lnpn3BFrzcKe|sx
z;KQldn|+tNVt(|n8vm*Mr)9CXmv0p82kl^qE_n@lb3XsH)HaAAT-o=OMY~o~k<GY^
z70oRtK0W01E<=RgRIzZ_zNQ?6GHM`PhIyM6s2zXD%NyU^0lj76Uqi9&9Ct;h_T<k$
zk|x{NbPQgm!M7Uoygt7_v|v{o7ZjuQvEP+zKQzrtN2s*UB?Vfz{*Q%Qz1GJ<*TFS;
zsLI|B!8}WCu=uV@Fk6hTgn!HW8Y{0#S!VlubMqtOmFD}X_3Y0T;YI%ll`@R4?LFdO
zJ0m}qso`i_2iN@Marui?xS2+&r35mt+Ar6@`qO3ZM8Tp(^*0yk-90>+zr7=l4sWSN
zB;^jEkm?CfMn^#F>&((drw(+EAF?^0u3${F*h=Maxd&GN5|@<TADnxH5NUWZqN(Nw
zToY{m`e7ngKQ`KwB@3ETj^&U2K_~1!P3#WPIkBRAb$QoT#ZI{I`1PKeQzT<hV@uV3
zGxSm{4uHaXW?62!mYaE&V*EAF<93x@xwGN2qA~4F8FguOq2-M~+7_mZ;I>uTCBx(3
z4+TWuxj<m&A%?oA+5%{5O3Ey6KlzU~sb2WTo`Y^pB*{NUGjXNa>aDd2YNs}mxe{TK
z33yiAm3#gwFWh;&6$sQ4?hx`s@d)cbL*zFD(|+OWQi%O)@Qtxge;Q@gGgIK_`3pW`
z^^P5Cj#=Fzh`)l8$xlfcyy9m2#_pe*^{<nOzOyvQ8nxBk>5^2P`NdME=&;Xppw6y3
zBE$sRD79FK_t*P$X_<y6!EM$BE=;flk0YN9(CYLvJ66Bmsdcvvy2$D{!229XuMo%Y
z_1CS>sU<EMCql#gjDgf6XlEW74g}lt+)(!16ESqrIrw*@))6~7eY?}~+)(Q^m8hzW
zeh1mtN|N6!H0@58)Y$qN<N-$!0kuJl@q66*lc73?B=AEypCEq=jdVi^bq@qK>7Q@k
z5#MRlpa_Q7;&KsEw$vu@8H{7FUfcby+S5npu6cI<!Lxt)R2Gneb0=W%ejXfwmh&s_
zqjsXpS3DvX?_+i)4d#v<&AQ};wL$pF=b?uNauA`2tkyai+iF~d%!8{Ko+vIvXjFla
zXU@7Al<w|%H!%@fC(Ii!*MQ4)u@E9^^Pde__0QcNiZqSvRVUjD;LWd<1k1EK)25Am
zUAI8jC|8k5eP+hPbMh_q^&$sPevH=jHrJ%`mV^4hoL+n!C(?T-@ST>#@-dMDHiUBe
zh5NZ_@4Br0zl3P=DG9O%j6Zg1*1$+p-N_^AgPLFW*TPX-;=by0CG!`YQn;b}esv$)
z`18Q1N}Gy>6xa^S(o41-Fk?Np!hWS_BS(`v=QS|Y4I2z~)=(~3Leo$%r`FW3hlzLU
zs-oYu7cp;bBj2|?^X(P>_&X1;t>nIM(}uBP5=(pw@t37{zD&jBs~-!vRN_Lnt-b=@
z^=&B<A~7<k<d)j@kXt;#qw7B8joj_Ik!rbKj(yapR9Q+9^>BeW>*+dgW-gS1i?d{2
zfRDgIb&;vJEAk5B>DMnqcKnU!s5^#i_YOSlqQtis3S^(#W~z48gZpcrQi|J)o*7=i
zDc-G#<QNoqKg%3#NP6<09B^+NXOK5P>mW+GPTG=zvD*r~Fi-xq#(eMpACKa@mwGuM
zQHtLm76xXL{?72*lF-S2L|Z($A%ebOXDS3MFAGj#9uH;ig>ae)3;z2ncxLksP=)_F
zz6$@Jwyr`P-$Eop95bwHkR~&;vvrlWt!Q&?+;RhB?}dWrpX-J2IzP5K@PcWHT}Kv|
zl;h3M9`l`?qtBYT!wak%>R%aDpA|Ddkz-aeSCV6#_Z9Pmdl$cNkAsubIyxyyVJ}i)
zNmkK)=++T#&pWQJOPfVD&C;57^{VAhoj7Z}m-m`SMn+cEczkC<T5gv5_K&CbEg@`D
zqDDWLMMR?#rJR%rU-Z%y{5BT6e*FkLPV$*ec}Kl*enXwJ6YQ?=0CxFB`qz7|0o(4+
z_eQl9SAPz$DXss`o5X2oYElxVUHmBHj6GV-Sy9uK@CQlK+`49FX6c_le_k3cb7I3t
zIpO6h#)IfCE`)OKrC8dZ?)0>RFJHclf{{^n@e<wg--aHAb4&Bd$awG6(Z>7>3Gzlp
zMsnb%N*w1t8h>v(P9ln>cK9BBljZP>nt<{24xDSLzXI-;{AK^99`XCi*REZ2EU>C6
z?@(B-^?-%5a;Nf0KEBIUu@loBB#_QSM7L8U;>*PhimO~9Cy#>_`uaY}t$xcL<~Q;D
zh=7{70udW?IM=eOSQd<dz}OvUD4FW(yAPGS5VoBO<#}DS@!;Uq%4gI0a`VX^sbltO
zfyN56ft_;mxe5~@2kMiR{Ha6Mfg46HDjS-mtWS;2%p8tngN@GPQ7&NI5~aRF?~!aP
z67k3)kOBsw-ygcsi=IvYI-cgYkp1Z1y?f(FUVQ&X22mohsq9P)-jO*^wudh0aU(nk
z@f}_VLQGFi9MTng$0d{By3n7)=GL3p(cIM3Wbq>*aHHR(j8n|K?*aNgiYJghSY%^g
z1|lu0qN4I4EsfWCxRh9~DN*OyPbz5GTjI11Sbb@6KA27zhoiC24;)G5cH2c89oZ~-
z(*0Qw3~ZU;T0_QZ4IEPEEI>-01=&IX*+Pa9-v0bKEIv0gV;IJ;HZ?VsdphxLIZF$(
z#s>TP^=n%VV~<X#&d|Z`YS}6EJg2ZhzJ}LonLt#Lgkpr|5{t~p(;10|9rTLq!>Xz(
zqu^bVHK*s&h_6kpgWCHG8-V@ZeDYf%8YiHWI--u=|6H{{&IDdH;<o#Wm3zCRLuTAk
z*`K;tGLUB>^Xb-$!vvcLNxAZnmLMm5ZfgU%oDex&>q~*ZnTtzHYncAU<+h`^COu{9
zQx;M5vEx9LsAjxqwv_u!OFGDu)AjYa78Ofw-XqVN=igtD`ubYN<umy)$Qo(RvuE?i
zN*#LlG1rnE=8=Ne{5v6VEi2fDo^Ti}e8im=u-57n6E~if7PL()M&OoEf*qiQ)*AM<
zNuAGCKRSL3**C5xK3?u(k<WMN#rNd3wY6k<uLZ;HY>S|vpu#)hth}9=8(9NNw2@xq
z`k!vf;|EDkxT(WV1x17kw*m?=D{ZgJra3a{YC?foi9iF$lP;M4y?agfK;&-un;{U0
ziOeKdM;drXN--$fJ3(}E#gEX?(4GUVvJ*+EO8(lsl8=Ax{P=OZd0Q(<I`7HP*QcyB
zH8diJ3m@GnNls4YX1Q>4Gt;8Nwf8;<{+r%II56qAt_{0X*NC%X3#R7geh<M|ePaY3
z)GrPcjME##*%)ov$uC&t|GbJ5F*&QUH4`I~{Pf3B6~?SYFwqDUOMB*nqzfD%fW)2*
zTzHovrJYV+s##v1RodH}?g&3|-n~L;i2xGe`V@10>asLn9A=5nh1YGfwrWAvnmDaJ
zxs?MpQ{YNk@uFt>BJsu3llH+M{kUYGPVfpmNc40lUb85+ZN=)S?5<b&^xSpgw25w&
z)@jqKna}Dd>22~zr8`jghLvPo2^Z?qK(<sNC%!$;>P%6n@NjT&h%qm7-0i6kq+`02
z>5CR(7UiO};dr0OH!Dybc_KFLFdNT^&G*nlYHovtpIX5wl-+7Ck}J&9e|o6F#PV6w
zGsgBnL;hEz9K6JM0Hvc}yScEAVP5VmXAfq~ChMZ9*qj{`AyK~>TIDfM)$k;^I1S_n
zF6C8>C7#hqv@MO8ba!_Tb;X@_P>C^OQ<?viR^ChAi4ZLC87^n$2*77-KbKqWv-JV%
z;8EA!%m*wU?(Scy<H3wsL6CK&YP^A-uLv}A6_?wJ027?=hj4~uZIA{yHT1vlV=?l1
z;xxWx7`ykiXN&)uh)52X+pvAAZ)OIcl*_Q^EpQmktchDy=aw**db7HSZgm^@pDU`W
zieQ$~!JWx+bz{EsDdMXEW;&&;-ydF=7(RGb8z&+nav;y5q8_=mHk<OpT0%sG;5Jl@
zpo8!&RyIX&qJo#}XmFI4KfuCu5EU7T2%oCCwm6Z9=J#ACmiqcV3zROEVCjHI%VTvl
zIv=i2c1;$<B_s@3R(m?k1FXrTcyuYzh)i5kouS+6@T|=xx$!)kjWbDY2&7ZTn{!3P
z#9X$i1UH(>pN}v6I+Eome)RY@06Hn<a_3GQ2AH+U@O`wUmIC2REw-8U2025hd}i~B
zkE0^HVCqgpz&wS_mLx@?q?P(o=CBnJMQw&jf*B2?l-@bmUZ;nOOKMysvg<jae3<UL
zHq#ouM|KOYc}BwLZIBU3GH$N2f!pg8R)xhv15R;E*`L`7wP`k2cIvnleO9(OI5^50
z<M{(zIF|RgZ`idj5__|F%Y0U+*lRqAP#+K~wwGCTYi%M5aj~(UB4C4PcXJxW3t;^&
zDyF!&_%=-5b^Nwe!8Bv;l|+!3L2CixnwVIJdW~%G-cEIrw2QndS{tM%Kb*Afq7tLW
z%$xD@UgA!K3^Q+(!O(m__0!c{DIhb4_vqscS2_TOjno_VonGPGNK>;J`JmyKJ`0B-
zL+>Pg_Wi_;opSM9nRs5$`@T!5(R&Cm4o6{Fo_U%1Lg~&+cJ0kFvl2TBA2G#ytf~@8
z9@a^5ZxfzJos<0Bv9m>)u2z+u`(T6D=y}T_@^%`Q&@TUTI*qYoPnyLgcz@MPzrO{j
zgsWAR`)rc}a`8_I2>EgQ_79RP#W-<)AI2_i{3ON|)MYb)vrG^mNDgoC&MF6WM*a9%
z(aM09sLM2I$G$&VD>ng7l4T$Lc&Rb~{-87S1#x(7B_onsVcU76a`i_ukN@H}AQ`;z
z64pUl1@O8~ubgEpEJl3Psi00;X`|1evD|fZ*c+^4U|Jw`$O$luWg`Hdbg$*H-MO~5
zwh%rR9z~yf*Uv2mu6#ez6_2f1s&?|(yvqlgN|jV<AcxLSk@~Ma)eU4PZ{k9$0*bQx
zCm1d=j!*;8V>CLf*lNv$o|w}rUDGb!PiQ;7<g$UoeK{s>9h<Au;jW|AWy?$kQM)-6
z5iBAqYH6jXr$?It5OE3me%$7Vv_lV8^Z4=O1Cd-ZONkvxGS(oS{TVxyE*ZXJ?jYK7
zJLX1Ou0d93HQ*|9l|j4PDbApU*DNnlezNOWUWG9MY3V7~+9gbLE5F!<Khf!I)Lq&1
zRpok`^%e4CUtix&P+S9*{<NhEb$#-1*?_fIzY(ssXVXQ6JuiHG`scFueZkc^4jn={
z07d8eiO}PziL%e8;<%)pD_h3O=hK4s769{;5CwI={O(3)Wlv7fYU$BdSMPn73^t{X
zsp9FNT+KClE{9(%ZgWW)DzQf`jZ{(kMT~Q09SP+w8=2ZE!~P3>*%y=pDjh*ws|>QV
zmq)$F0~w^g>|TL_;z;?0FYnQYAT-{fk8Z8LW5nmm|Foel4vFy*7fZT=CbvlijNza<
z?G*XA6b0|GK>)AkqaB#sGu3wv^BwKk%AcUPHsaCsfXy0iX<0iv8P2;Wx>E+I=&Ukm
zK6m(ufLiU*9kNoW5p}tRg@x|FKi(4Ql%0N;WKsHLwH*L!pZ{pJmz&(EKw?r-O(2+p
zOsVVWHqByso3`ArhK<p#CQiIy<Ea6Q5TCAwa+Pu(EF71x14Jc`Z+3Q;zmKb3XUG*@
zrUGer?7(TG2eKn)q;gmZfJ<yLWyG!KP7P-o=+(QGp2Qx7#_(fRe?CVTyN%cH*6;1o
zQK>4yd)aYuab1ME_FW$KNoJ&;txh#H?y`EE5{}-rFf=TkW-_95lL9!ZZoNk{Sc%ep
zBOw`Z5EZMZ-w!pu`2J8kO*!zFrDylOoY6Mzxo2qDYH0+hs!gUO5bjx4k`7+p>;n}N
z79*$=bba{ujW?xfL8R!GXH)YE>0AxIO+DpG^u?0itJWY{bN^<oBFILyCBREjxu!**
zrvU(vkbizZN|+49ZW3Oal9-^vQwu2`15Jb#jRhhlwmPpD6zp372ji3<$0aOBGc*$h
zKmt(w3z7gY?N0%i<;yJI!G@h#iDthzj=kAhES4`H{-pKm9mNaM&YL-WVpi3q_i`P0
zf;K1b0Hj}Y?EhLeiLp<8L&WW*qm=zt01P;S^7gaaY`R+sn1`#v`tLjT2LX$Id6fNv
zLS8ROl8PS^ms~(uG03q7+P)9aOmO7kC6)iYia`iE1w(>>cKG$Vy3AL9hn$)wt)tQD
zo7yTZF*IxMQRDsn5U1ADW529?Cc{)T^Z{(V2S6n`9yGcvc`Dz&E0yuU_^F9T00ZQy
z$L_c12Y6$#`wko)L9TbJv5nRGdZdF`wEM0$ajL9eQ|N7WDE>(hAFiN;A2Fg(C^+WC
zg5^f{c!i^>y>3dU<+OJK>^)=JYp3*db#?nc-3dRBzWHqG?trwwnDjmf^2qwcp{R{=
zw{iF-Seax0TffWwKo|1TH%K&I7y|(IgT0?`*ty;Vfq(;C!TmlEPfH&pZSsj@b#x<U
z8JlOVPd3aXBqY!sud?UeQFN`EjrX1Y^<G3oSeR(s5VUIy^jL5ONH5J!c~Wi}5G4x9
zJpt4q`y;TPQ)(9_74VWb=G!s7mgsmf3uFKs4p%c%IeGZSEqZ)M$e9Zg9!y%GZhi`*
zGmvl9@Kj`aCPrs(0pR@r=oC9_)WysHje4)RU$(1D>Y3ob=!y17wJ+|_L~ad@ECE4V
z@XSvMNSAZ}9Q+D8dnZ26BAC~p7X*E&ayn9Tp`71_5qbWG9Ti|BA$V&xK~MsuiU(jf
z#CPX!@I;os$0%#Z@TdMqXq^T9`aX2JwKGknq4W2rJ6*|4F**-g+W|mZ#wGwn`?8uu
z0kJ#g$LF5Bxz+hc0OKp?OB?m3Re^q?V_y11FWfqKw`>(8ToCWz&{#Q#h;g5{Ca+r(
zxH|cO-^%_LKPksPp#&^oN%+@-R_XWOUpvftadQ_8(0b=jI%+R5CdNP*c2zBMkr*!!
z@U99F^_LAFx3ljk0_j{$F6mO{>d7|9dIJQ6Q_N9zft{*g#$GyUpn#zddb8^Vpk$L;
zZc-2G0Xkh^)5Rv(u(O66Lsx}-$<56LO4}?Q@CfFDr*5>cq<#G1l3V|Zg{zwSS`J9l
zHoolI+FE$-##=QgP8+`m*;ylq-~al54u5~HS?@;tg6gG0&8N(e=L1`Uu#@S9&fWFj
zCT4Y&Gnm*uWBQ7vW&K&AD1a9^Ik^Vj8$g$Q2k~)DmGhX>bQvziz>SRJyu0slT5*%l
zY2T-x|DPFt1u)j-86G7+MM*?kY&{_Jy_cCRicEexC_Mmisdd>8S!Dh5{rDvTl5%qX
zM#<c!Ht)!!+&fGbNG3)&$Z?*^fGzj&-u^ToKpKFn5gjw%nL==Cddm@Bl~-ZckvQ%)
z;5d+<GF=JJ&b<T1GTro(N+wuIK;K40R0J@m0!Zv^(;IfdB^5A;Io`naZcy8FauE<Z
zZVaGLwDv$7PAeLa5EUH}g@MKJCU2e01}zAIm#&oub4;*#zzLLE&{uM~GiV)UeAypA
zK1?6q=`}(YQ;&SD@c$*+b7#7KEv5eR561Q2)+d}9?G?WI3xt1Lf?|vdycjp0GOX0l
ze77<=d4?JT`?Z&%5*#$zovvo>#9(|s!fUzNCQ=a4Q{(N{Joc{KmPc?PU@y<ra-zY*
z!xv#kXoAHEym#sU+Q0fYir%AMCP|Zx;Wg50+0H7X3(<d`onH3-<vEK;0+>4{0osN&
zKM@tYGu;cM*LH4YrH_Z6UN>t^^SW?Wvin|Nwt;_h4H;5clYT@3qQk^!Z?Lq47YfKH
z&GFE2RNHt^44;(SnBo%VJrIv|hUj`AdSCvch+Mli%m)MK=6BC^sW56&`DW+m-KaOr
zUX9BS)@cXrZOtcY6X!aTD!fZ;LEkqRs&E^Z0vb1^%z3Ez2(w9xfL8hx0H@MiX9}G@
z_>@}#X?YoD9kexDR75Q-?v0I&<+~X_*UTHdIGk<+`boG3vpnjbne!m&kA`uPvAMg?
z7qTlWH<MH^>249qHFm4Bf6#15ZVG(DgP=oH3W{qO*L=tZIRBTn9oI={@lAaK8gs&h
zAk-T}4NpdVLc)A9aAagmjE(V&-RaH8nY!-@ta`B+_o~36qO(NJ2^tx5hgHrt8~e}~
z5f_v<I7u(cE7kzIzZekK@n1D;=x%Xox3Qgy7(m;NV19o-g^O>P%fykhER@h*JioS6
zB#nHl%3LIBS8j8`9039FFR0&Mr~x@dC|s_myRz}JPS-5gJMrl3PQ^W`IS#?xW`jdg
z^XsU?2u>@{ZUk6sgCFdt`o8F0J%__`L<94$auz7{p0!yV8`Xje;Mtw2&tv$rjdG0h
zZ8|!D+wh!^+w=D={vh0%K{zO*U{EU})ZOJ^hW`^lor5B^{0S}<Dco+pOio^K=+Gf9
z&|Ry$fD*L=j#6_6`lfA5REGDMuRX%dY^dG@?mH`~qr5dE*u8c^-t!L{jM%>!Fh#c7
zg|9i0k<vAlNe;c4DZ7nqDz&4)Tu5<su_|*k>R?meoKS?x(?&e(!zBoSskiYvfT`yz
zz0)@;PqHPm8S0>vxBu7=Sb1A9dU|?ZzEX>74||4-7xTu92r*T({@k#laNvI%_EN>@
zr%_EYP(#=FtAlmqW@97p1=izJAqQAD-KT!Oahso;JIvG{8#svu#VXAN@fQhV7Ss@n
zVcUy>z!2N60Z*kE!{Mnbm6@RD5w4O>N7$A{Fbz%yurCMv8(*<>r1?y}d9erBcF=q&
z{2D>0!VkNk=t=MVuvq)K%mwhjEp#l%9ofh=K(DA4MUR`hgGh7yW@K@Q?J;oO^4)$W
zC7PO=5__Y(bI_ZgxQ3Hp7^CX}qO*c`r$JP#{(CJzO{oRY!<N(FWRZLzNT>#xTG`g-
z=G%0X34oyjN#x?79q2VY^*~v@Lf;J7TxoShFa-=ZU$z>G=f-rl&1bJaLnZUg=ykxq
zcE*9<YGGt(c#zMdW>&KKq6Z8d-dXPc!0o*c7-Mza>Mt%95HyB=p+hU0zs!%i=!*w^
zmv;2vej`8%7=cK7An1txfFz{GVkK>US@xFLca_X_rCARwHI)KF8%oNr$=?O`fw#F&
zBA`KvS70c`wU)xA(Hh3HxpvIP;!QA3&>h-^Yk|?FQ0hFiw(LsJ293*(P9-nzbOrDA
zekm@70_XU=577P>025#UeBeIBG)F-L-mnhOasWt2L$3pL_GAExf6;%t>BoOaw{9TQ
z<FJ4Z-&6T+CG1yoE=LQny6wO8J4L*k$-hf#%KVRWis!}vzg(H=Gy?;;s$rZAI}38>
zX2qDFlZC$md0H-2ervRW`742$-nlzFJH?<m@AWQ`nKnR?7YT1g2X^Ge0l>dvM%$<!
z$+9j^Kx5)!y2ZDTE@`wnaRS+6fd@ttji1OAYxCu)<J23`|0Ni|+~D2A<p$bvh$_B7
z0E}7VKGW~Pg*Z4sp&d_fN#{0J&8O14Olj7&=>Dbss+qrEaH`^+K)K%oA7c+3oKgzl
z3<PJOVH8-FjP>-QEr89Ehs9zEfYXo?buOL!0$B853orunz0KizdZ#`BdC`i#2{g;S
z)jr8?`4OPMJgj|Z<J`bB#@B#fP{%Y4SajGI2^GsRj4iMpr!nIo|7pMgfp>v&Zz^lg
z$**1*JIr!nv5$RfYd($P0~F;3aHn>U3+klqWVOZ!GK8U{(P}(f4J!<$vptlNk>SN;
z-K@z_gvPkALFT&GwXRgd*4FH-)a|Ko-k{Cq;{XUhft#mU2!J{n%e9owl<1N`YmW}8
zG-j?GFwQmYE&>df!Fq82cL44EA8)<j&oL>eZ}e)6(JfD~U1a-K@PzzW6go5qBf8Lb
zN`3XG=Bvh$_TJ94#n$NJS*m?Q^UH*<U$2Gf?B|eBp?d?fD-7_l+FR%)umcnPSLyM-
zfBW_=vKCM)#-IM20h_<wlIddd0~cx(p$J!LhJ7X^9s`Zs1#mpSJWy-FrU24bfo0b{
zk?hfF##yfzO})|Wx2eRo8)WY}B~$Rk#e%W7YZ7@;&V8;YLsJ2DL26+RcqHUldjKn(
zGR}if`$G>M-#eKXxeA)ozgXY@<Gsw-W+ETcI90lmU<G=hidM4hY8%sBERuE^4yXm(
zw4fb_Bb@!`U?|W7+Zg~1JU}JUe*#BZyk<Vt|7>;KDqx@GBlUs)WWZY<0Mq1|Wu4mT
zygPs__XA1ij{}<-6i>ha*pStL9yQ1SKSBgdOQ|t?go)GR6(jDr*H=-B?rXq0sk<g9
zXs-sOBz1d(6zo0qMzDSxJiWUCc*@R*Fd3Xd+Xf<}KoJakys#~oX$B2FKf+V_1;|{T
zTEG+}?qy=GKl(vu+A$%$z{lMJL{+>K%p)FfWV0G1InAQXQHqY%6!*0SR5KV!&|zwq
zP=FjKmVjMAV3r>{kOs$&A2$ITNz-9&4ZFyeqGp~N9G^JT9FPs#+580P*;b`-o_Pb%
zgQ|cSyIw!{j-;lcf>4Y3vj;gn*)RU<N(>O@d&n-GBtd>;D&55!S%Gu|O|gt$7=QyE
zY`j%m9)lV*ElWmgtS*i-{aH{>WBN?G4MM~K#?gxSz7A%I@dAVT1^VX0Zy|&qf`>~o
zres@V?-K42n0A9EWv)Xd0h{DnO$=?B1$w*nOL!XPdi3(U=$$zZO+q=?bqPOm;@ryb
zJ^C`kOIuqT0fZIPcRx#A9%n=|Rdf*`rgF$Wb;IIeD@aRh?&cwGRd+BmtNM^7ZC1sV
zZv^84c%Tmq)`|f7dP8dDMFMEV<J;3HmGTrBx3LgbZh1a0=@}&}U@5!7tK1Q*20BVo
z&cK8lx35<Iwsa*L87(d0^@M^%5;GD~Hv#iyQwHurGjND4mTjw*uOkPhZ}yLK@(=4g
z)B!0cU|Y=^!Va|07wP-&&aJ-VvjV$N&)&b`j9{i!0}#qZE?L?s0&DlURlO!Ftgsz-
z<bk)7W23Fz843DFKoqh1AuXY8IpQ=~5OPTY?5*g!Sb{Vc0lQx-DCM1%#xdrOO<a2|
z*p>0~g~P|nfiS8~ZV%XZDQ>4cqSRMC0hH@Ez<3m>f4Xe~n5qkE#k;ttzZIzWdG{d1
zR`6HMKHa<kAiz<Yeu_g|;@0&$skZY+A;gI+=76ea=I4(9cUz*(Q>oA%)=#_lhRP0!
zO8IBS4zAh*M?1pW;tYiya<JokO^P|puPT5eM2RgXH^W+0Uk#?jpCXu@(p(+ES0_LW
zMk~h{#;py1GAE1*nlbm$A(sq3*s_H1r3gD@XmySF@aFKRNQAC2Ujavy@jh5rn?gtp
ztVf<S<W!f?Gz0#b(BC6U`f8a+6!JaRiH}=6yBtZ5<0^!!X!Yn~#%d;sYFQche@`q7
zSI|{BV(%7%(Gug4L%Y6f$uc`l`sd$QpI*28^l=efoFOGITA;w*4!$e@_c+-kc<c~@
zP5pYo?vJ&^gopqAKQ19c<w9p-AN-nmf`oIxTK+v<RVw~n$kp1gBaf%K{blqw=^GZI
zzW-hX@MZ3wmCUiy!Ui^->jlz}f^~GRgRkQMJyg#mIL$xvp!_fH-aD+Rtm_-b24g|S
z;kF<MIwA~$ARr($Dr1QlP&!f-0ck;mgc1TcD9j89njkHZ*r=f+Eg=L&iJ?V-00Id}
zCxL_(A|>HHi8J>-?|WU}fA8~M@AKp@JZGPM)?V$m*4pcAk&`y&UHP{nT%V>6{M|a>
zGIyfT<wetM6gZ(J^Rhn}=R36!=6|9h`LAb-UYyHQFHe|0>Ff92UtO8>i*NGlYE<;I
zH?G#$*Q67AsXgGXN2r?@U(X$zVf5Jjz3=`bClqY_FV-XW)D5+njZg=Lzot97AH0_E
zI`B`ea_3{duU{j(-M#+0^X1QDhjm`Xx7QI|MV!ukt)er=6RONy2Dd4ndU~cgvh?E>
z$lp}|^~@8O<&ldTrB6dKud|iMftHfE)t|s)=qE}N68%+Yo@O@GoX=AqZ9W8!p@p6h
zJu@4@vA&INHZ)ZPvw?QcSjLBlt3p01)U?s+tq{21q^POYz5iVGy{gy5XW5Dkuau=x
zZ%lRqJdQQB6}&Q@enOUR4b{6{87R=oGFXnk<{(n8O&T+Y;uP5nV{V2LjE%uZGaf>Y
zR-X7l<)z8_AJ6~mN5yl$Za;YF$KdPxj!hp;X&Z^Hdl2wv^YfA9*xlQ7G=AIm@^tKW
zWy=dcR3<!pIP1(6Yy?}nA@33rZ5Uzn86ry%P3pu)^O2?NMeex-BiG@O9Jbr5?G0lg
z_K2F^Z%rqonQW;nO2?=~Y+TK)u@-$r*!auAjJB5JDUP9l<`m!B%3N5@S3&frnWl<3
z=;10qdWsMHAt9j=`@L1d<MsCMpZFg4k^!1q@eihCTg7(sEH+Fps?^lrqWE8jlIYLH
z&mH&L!DX%eTFpK>br$*iCDFCm1=;!11}j^BM(^-VM);GZz_v6Q(T#I8cO9j<nT0{M
z1uEi!)t8%yi)LqSa5d$c;tP7B3!OQI<iJ}%;0^rzlVigsknp_5&6io?3%!a4PR<d1
zwxBa5VS7#wKi$?YR)yj%TbcBQmMv2QdLu5Pfla@Ec`3g0p9P)x{=xDgwJl<MXu7f!
z_;3;UUAp@H0HxE`Tg6rv`}C*pM<*qv{+~Sge`|_DNFp@$M%_i;^Z1DKBbwbQZsMiK
z#wm|@DUuxo^GzX4&yiKp8a){M=5<mWARwrMN2zXB@u*-~B1E+MCV~nun44o3KaWz`
z!mZxcAtf5mPD5V*@6DY5pQNlfkX)QxqOU>NCt};kr{g28uB!GarC2-DQo@8;OF%L9
zCbEK1L8uwHK_U@0S#N+NY$PO--sW92C{NCIC50<Z#gs;_zA<Zvbcb<CEeiIgch<gD
zvB31w{WCv_#vYPZ4|kwg)pGCEvb^ZW^Lbo;^;ZrPdtESa4k;GWap2RE2#S4KhM~}f
zm&EeHTR~_7FwBo9+dam&3xig=Hn*@aMPan|g*HeHLNuvFul5#hE5a5bG{!kIa=|0(
z2>J|N3NKq^(2=ytQTfU(of+$f)cAb+*xrx=oN^-};n=smL7jU;m)DIC)TIn!ipNU`
zl;ttqtRdULAY-2vj;YwL%J=RJcVK4f@-FiiHtG>aV#5EZ_bvm4kY9-=iG|(jekt!L
z1vR#DuT7b%orOg;{}j@ZOzPMY_9jZkzp^s?_k-IxMuPBn5+NmTi1uWsBURIv(tHu<
z#^KuOONZKD72*7%v|3$nm{t+)e0~hf*>v*+u>L_jz?yhJz8fv5pR&O8T+xG1y{*-+
zRf1MZ%|C$}%txtz-AU(b5&qZ~irmVq9<QfwR!yty{Y1n&Xc0tGam0%jh^CYBcz@8n
z5T-F3Zj|-IRb>^pv)_6rS8#Yxss7&6eceTu9X8FHT+{<S=~Q69tnq8K&nZHBy$e#P
zy8cdn*fzSRFYB>gzVN-Q(OPfq%}}k7)h_4ZFZ-2N9apPFfq;6e7E_(OZEj~cxooYQ
zgpK*+MoB3a-UWr2M&$`ViKv#(Dt;&bjt>EGf}SlNn<{Y5>DuWtG{%-O**-f|3HyAi
z{I_IcTaM9+WxgiiaedT8+$LcfKr+Cm9~z8K_FOj5<tf>Q;S3uscv-%J;d~``W+T7P
zG$K*}5yA^3MS2q({1f1)V~RAhdM_J%$r|aVKcerT2Se<~u&#la<ke7KHfu0H#m50x
z0@cr%Dq^|fyG()SB<_8zyx812XlGiyb{DagXWVH*u8MSldGcPfwD8UNVVB#T6odK*
z(Rc0=<;hNOiAH8_IrhjqN>f%rl}5EJ7@el2O{-B^pKtUW5y7oQ?7qt1!gx`YYc;Hg
ztOIm9ePaC(f0R{Lq-x8@Cl_{>F>Oe>)T{2Iu@YOjTV3>!8WS(mnBRST(*bGm4bT6#
z{PX_>7gYFSE0bKzF&&8#FEHwe`Ho7EBeI-OxAui~U|WYc9&C!<LmFLq&6=Q3&@n~p
z#Z6P<!Oy?G>oazsl|eHM`2{QiJ(o_iQ{B5<FTQ!VE|wocFSDmO$P_X4mWKYITj0xV
zW84`*O*HYO<xuG(Z#V_QiZ!Onpk6jZxt~OB#`7UGQ{r0u0}x0wOB102rS7YRceBoM
zVGtu3?I8wbZasI_<7*v;?E+oieh*@Mv+4aOZ#nS3vMv{m6{*_cuo1#zbVQMQ(Ufg|
z@dYa}J+Ep95MSM#0k4&>a-Lf}Ry16!St_h2#8ltX!+Z}p(Dc3G2DLsPI7(XDoJKkV
zAA&b74R;Vfe*BH^Gc#Ht-O8R;w0D?YA5$1z0)CFjS21v3UziCAE7L57_TBN)O00N6
zfHXH8y;rqmQsQjS5r>>|zy&lN!~~SQVFnaH2fr~5Lm@EwON__$%V+9iKJ>fX40H;W
zl_<Zk*MBkSLq&ZNm3+`i%3f#0v?%fpPaX2xfRlkO*@i?Q+@0Krpxb0^ZwxBrd>Nng
zC-*8J_11Q?7t8`U>;>TP-u+>PW=!P*IZUmcjB}{$?)j`nIJ-c*|MMLSybJ&#k`?u9
zzrqrB@Pm(P`c^(4Yq@B!RYH9v2!Ht$;I+!D*Lqc)<o$;}ZQ;hV6d~`Ll(_y5QH|ua
zAUYAVVO<Hr1^u))1|GRFy}cnZ6pH44XxTDp^n!YliI>XQJ#sy?mi2qHeh_qNfg<u(
zv&?FIpo5DOZ1N?->4`30A9pV*A@F92fx(gNxjwhS(vm@ay@-9TO+<YKcg728d85(y
z&KkgjfuxN;2T1FwD(Wd9;s{*A<3zR-Z1HJA?M!)a!&kPtb8Rhbs`PkKH{f1&n+pzv
zT;SEQ($@#P*v>@<?KWSnz0;Z8*tmvli(@6n@mMpSKWG(VosbdF0TzK(vllvYghN#a
z87(d~t`T-hk*%MCdJf@&gwNckGmM$hp99+`-Z=<wxkvXQ4l*RwJ-RC>%X6D!xrtCu
zsFILI^J9CtJB7+il#enOo+IGISCj=6(&n<zYMFn~wQ*YC@uK;ThIIW1=P$)F`d@B-
zb)~Ot%Lw7W)!!jfrSddsf!X^&h&{WpsLbsF;<=(}lY&^s{JXAi;U?e@+H+@{?AaU|
zMlxG$DIO5HlOrVVxh3(a)A`K@7~;Pqq`V9EH^}4jt=RY?Q05XrsRU|R5<?!)zAm;N
z34bOqVXeJZPA|D3iEUZxW>T_Qy_Wt!+zKw?xD-F|^6Nfb-e6iV!`Pd3d2i2mYbTtx
zZCa1bT(fDb<LK&E%U6qI0F7o~GFKlKJ3W~3SeG}mjVC_xv0QV{{(0<a8}7{(7}19{
zCdCr35<m%n=FmK|+j4z@*upZo0&&6ypmRYif155Z2bkrE4{KBRx){?X*zbVZLG3-{
z@+3Ca)P@}L{4d*nByv!hg}Cdc0m=T2#5HQ>SfBJi2t{1vZDfws+LLeO)TLC^7GNDn
zh{>TuyoKc7g!G>oaiNtdS~C!l8;RYN4|K8}u_&kzTGa<rD#RLl0684+0|-zL$JaOv
zJCVH0mTHR>2ks2UlbRWQ_WT^T?|jkOBR6j!h{syzxRDwN(Mq1BIJ2;uyfJ$3gYO}a
z<H7i$m$!OX=`@HE#g+o0xpz03e}|Mb-`uTCr#{slF?S9PeR4h{er@^Hztf?=p{4NI
zKpfwmo`WB>zpOb%=qtWvEFS4k`?Q4{=1^9|361UB{XMgoJTrbkntYu^+$JOX8UTne
z;$PU$GjrGGGRG|N2W>ZRx;oD(PXDDidVpAUb$L?wb%#O8EHQY-A`kf`#wWD*fUx((
z0`mAEy?>q=ohKYW7xG=c4*>yab@UiZWdTRV0zi5KkauH5o*DeP;rCaHB<J4V!&ka%
zDH<S=CQ_d=x{?(eRGh4M_q!ZdJPTrLS*k_k63fEQLXP&xka+B@SNFN4hG?Tc?d~z6
zy;50ev6-zaXv!o-t^ij(@{Kj_XAT3yc((_|uiC2z7*<NDK6h-R1D~*G*|0ZGnl58Y
zcFiI5g&ef?aoF^)#)zrj3o(a&ZfZPogM?fQ|M1h8zI}{?0Ol^w>AhpO?2HFd`C>r7
zTerW#K>jYadfIHx>BB~tMbi3X)(u3CaoK<?135V)i?_N)5rYG}|GN75+P%w;iHiC*
z<S>vY)JE{GeL)>F>l4LiVeqf;>76E<?^y4*AhwHEcIl(Dz=0Xo(SsYNWl(56L50zI
zgj<tCkLguK2A$dKZ_g9N^Ioz55>k$i&2Sw~TPCLvM||%>K4TNJxg|_IP4Q0ixwpjH
zZahWbi4>W_DwtXI^l2GtG>W<zQ$%&E^8Y-U8EL*QG@@k}YPsAlwW9S5!%ss_bu)@6
zdhNTn@-w4NZpYT2J)K|&`h3U#NkznEp6~=V<}N~*^q8tz)OQL9Og)?|3$ZRo>P1no
zKaWCBKQKi3*@R%^`5B@Rn^L_jaFKAFir*V<RV=KGoLBJVItxC$iN1p_+1dMYvw&k*
zVpXg;NXhsT^Dn6T?a~}wUWZI&*~GVMR)QpU>CJ;F&*<SBBtKxX=c@K$L$nYK-awCe
zP=@J_gnoudNQ5BZyro3ws6>vfX~<_taV5qfhH-ZI0|8I)f=#@X-!k&)%$+8LHS9d>
zU3)!##R$ubXH7sZ&>U=Ih6h|etjhp#u9CH2)vP7z`5M+28NN>bIyCDyk|;F;H6!25
zaSjA>(gb@SyD90@2c;(ai;i0v7NQ&3(TN-{O=1mnE-@(5pWcynsHV{9cJr+S7?G0f
zBZW>JUXckTPY$fH!p=%aBs6;Bp}}Rk-veSG80XK>;2#obg}IxMYWpL0N8q!cF>Cdz
zxW;L>$(71JY#+>E_G~eD6Ug~jd<$a3%+dSn3?9{U?=He~6Y+sa*y#xL5&9B7UH?nz
zS0-IbiZyk8F%06yD2oK6_s93m=}lS<$j&oKMn0er1}ran>bi92*WXw>wspWRpku=Z
z&n*SVCL2wRH!cLHxP{7xe2MO;HrUbN-WUH^{_+yc4mlxEYET1e5|^eY`9`mtO_$FD
z2&O55wV++$eI=eLUxu^<%cf*ullglk(${yT#9Ne?gAyY@8uSd>bo%}D$sYM8GJL7z
zG;;8oDfU<I2tZ=dH-{@D!`#>ZTxH|6EZOO2cDD0tUPOLwUyzrY6{o0&UH}e5wX?^H
zi|7C8+IrBN72MzuwF-Ag$uMjTL(i{I4OERrAkdgRS8E2Dy0TxD=CULWSy7Mvxcn`#
zgtJTRFh+hn^Bsg<Axvv~8VLrSJ)RAinFceaL&6B<%$KauR@Mfq7W1%H>+Vl0LJxE~
zzlF8N>S872_d$d1#}Qodp8=@>sl@bQKN+TRdyn(pd&$S8rpK<Y_(1HLjfF<g$qhzc
z3i8W3nR*Z}RqfLzE=+hFk@wrdorN^(1J3ujH;X3S)AnLE#zEE2$01#7j9_KpL`P6q
zX!TEGe|h?XdC7?29AYKYv|??sJ;WmL#5_cow=Tt+xL#XLeca<VFwO?OGAbe1z<QQf
z6EP172lc-!j{f;5RD8g;LRb8*@Zm4mSpF@#|5O8~P?zT{#q!dCJK(a+MD$Cf1s0q-
z>;|&ggu-Uq7liinSU1xU@Ak;QAYgAR;C{X6Wmat2qL8cr<8m!b5ye3XXBY;D&7C8O
zrzqYH_)R;SKh2B~R@&)bd64^A*OO@%Sm@S=#Z-{;*GOB|#RCBV+Y=sUD^uX69m07{
zYB4D=eG8?l<E0#orDnzc3cz)pk&*#7(^p2-k+)z#0b=Nra=&TGTIP(NcnXc~y!}~x
zUAh)T>9YQpbbh(Y`s{%JcWjYR-#_0uVC`DD3-_JZu5yuyqGR+WB<%nAyA&o0O4|LJ
zC6=;v|B8#mv4Ir6V_UoKuUEvCf^yCNzl(42H{Y&5oZOw5Qp|BO%_V$U9^%?69Vw6z
zO&SXWP>0WJQ`mGb8ZH6&R=-xlT6_@s9k&i1sJ_^|!jhvK<DKo5{OZc(M9+A*I*uzg
z5b-`%Y#Y3ii^uoy<W<!)rr7ng4G8{M-7x9X%Wcd#T^?VfSR5S0?z<V<L<C^b`=0}<
zyjdxl{~iE1u>_X1E+vNe{JUr``2y%?QZjIsQYTFh2`#MnyBFY8k(9kmh9S8mlKxK-
zY5^wI?V6lJk1JCDcX;Cnga8h900z1UmbQKu<Dpz23cxb&tUrnfz?T-6mcV6}SH+9&
z%>j(lO8ShLJrW@|foDaFmQdLK|5>v%XWW@9_G>;DGDnlL504g&j=cc{q=J0ECDFyJ
zQXfUvBwpSAF3Q-ahXDif#1xm4ta$F;o#e7e3*M?<81FH?xk#!rX_YY}HcW{$Q79N=
zUyB!pMs8@)$BGhHv{x?DONz*Me_A#xSrf3j{}W>ZqS{BljK|nb>F<Q{_j8!zqbvkI
z#rCp;5<dAK!R7vj6uf12uT@D6Ex4m|ravTWdo#$M<o7)KP;YmUyI;n^?a8=k6R%hx
zOO3t7sG;g3R8J9}Zkv~)?>@lTE>T<qFi0t*5LA=I7S)MCTevS+P>Aq_!E21P2@opY
zn{TWVBja;bAY$`4au^76H?NTb<@)9_3PoY-A!b$4uwTah1{yy4CE=d<Mf>OHrI<D8
zf>x3!>o7xnh9POP%=&4436EQV{rgNA<`JNs?Bwi11DT%P&4Y|P!{f)=*L?t!bA0kD
zWuT|T$}Ccpm$Z}^A7pHm(CE2t;Dp$`CV1}U!gmGS&A+t)h`I%BnQJL{=l0-^1-g5?
z!sh>`LNRu(MkQ7SW_G48kwL`r!nMYk4B7mt(qduMe<JRc5r#dZ)~F0zfLLO6&`Viy
z^c|3uV3I$~-u;LMx0vG)_z#xH)~cxA=em%p3G2&4x9CW~r`DH8_Adhm3<Ezv%BK!P
z7wcJy^lg~gHe%UeCa0Kr9T|k)PkJY=!N+bpJ-&;HnJs1NA~zgi*I-pZ&=&|}go;gF
zLh9BDAaW*RwiEhF=BK`!Zdm+Zg_&5pN^A9St1?ArqyySCA|(q8f_t~m;hXb$`IUiU
zkJVJU=g2&^e|*wU2J1PC#FI06Q;L~7_r4pp|G*RHGub_NPH+P15cHqMhOfJtmAHQ%
zR|ybP-~Kmpt6obPzQG_6BC3&~zS#mSs3gdpf%p&0wlb>*WIsg%cnMoj=pzN@MKAx;
za;>rpac*<jO9U+-P?rB^F0@5d$BcyXf9hiZ`xM&$Ooxtmvrb8!r99oXEM9dv6o|at
z3~Yj2T40ykdvO2TG(bxA2Y{6Rf3O>U^P)HZ`EG7@p%S3YSR)%U?4Jbp(R#?vM~p)n
z$om`e9cbBd$G`7i06D}~W?PXvu2<C-D4@*w%;vEc=oFvM99LK4f6y01UmB=N;pE_#
zinJUq(~=?_s&gY<5zv+uEQ>Gs54d35Fnu2Rp}iixa(L<a%66h%S+SzaoI?3uKKj<j
z*jhgXP*(0?J)%EbKNMn)_Tax^`Q-J9MNobXV}F-Pvn8KFDtx))&TP(_t@1DXt~px(
zjwreYX7u)Pc|&O|)C05cj?D(ZI|n-qP_E6D6z2n{lU}Thz|i8HE7ndQ2ra&34t$R^
zV(movN=@ck-GfPS%<ZugcseMfQ0eTaY%v8IB1g4p8<g1_lVR5&zN8Knr8C-eVv#Jc
zd~T_}^x5wzw{H9{4|h<Gf8X*qeL4}JL#N<rR_wAV?QJ}<HkMBU`IQ-Apdn>R9<Q4(
zlg0(<tp+*rvM%b22~~&;cyaG80*3f4<hw_nw*wqRkp3r=|L7z8+*0vHRi*dURR(W~
zc8Ha;jns6R+fdN}kMX_ABZAK>+f!onJ+%Dk2M74V_iemfWcJ|cCNb$vSB_k1S|3>1
zp0>U_{85v5R`q~fXr;c@4-L$KGEKKhz2UPqF%@Nu?-1)qu|GNDmC}X>Do;$H4pq4t
zHxM|tvo**d%0I+7kKpc3Qk{Q5XkLCbC))Y!$2DU%6eHiily}NIMF$x;jqfNwDJpyS
zus+bO?8*wFE7ydq?!*AUOgH+S!N<oK-*=E4jic+=o?SEuj*YI46u!hr^vVGH2gHiC
z8I~R$leHbQ`FR}ZX)AzhlvdXZB48J)gJbJ~bZ$R)N-_h0;MnRJeY-GY&NMV+`Rsv(
z=!N(a*Lxk(dUHoSs(%P~pposviiOU+#O@TUQboHQVpHSf!pdKjAy?B+Q2^=1H$bp-
zXTE2cNrcGPR&NG_K{1^%8U$&;&EK}No2(^8LScN?JCz4q3MSy|>Ev`}k!8l3$H#_g
zcj7JmD^azqSymMtd4T?-h_k27wlY6_aT3_0?HZKumW51rQjL-KgTFRag3A%}%))_c
zKozDARcYbOK1R1pSJ6pD`SxYSSX;q4UjKH{|1Koa^wVDb5z@Wq+6G_lJ%5}uRLZiW
z14)!Uzl2%wNxXj^YmVLjG9fJy&n=FALzVs_dP-u2ea?PfIJm`)5(RgNaRPRS_`e~<
zu4L~msh@hH*GF8-wH(ffHl0PIsei=+$N#QG{9h<9|IeEu@Mh??q<7o!I8i9k=(V=s
zy{&zRMAHlIKDsPkIuUv&QxlJMcCDViG{XKSWL=gYruFC2we0&u<=3S)x|}G}jVfQu
zxoEIO%Dkz~1bfpIK4NHfBpiqd_d>J^IQ61EBamGgP{5ssHIy3#iZbVzVqxTG$m!pm
z9cb>MvQpF9bR~NSO~Ssj*SES&E_Bo7@sGE)1KT5C>Gj{~2@0Vf!EL-c65J*(sk&_R
z#!R!8%x-iuzy4TT&oZEyRHIW&qCS(_CuGbbZYU8p1{QT|(gI2eAHHHD7f75pqCB=r
zsj-|5pVe4BbbLzY^io~X8u7s-fxA3b`e1UKgvWMhbc?Cc8oU-LhlbRTnonsEb~x5Z
z<=nrJuZZXowFWphetU-5>eid_${=v|X{~86U|DlhwAB>g;vqeQ?=3U`b9T)COZ6QC
z)P39F?bnckcLb5eN*w6i#iYY-@-EZ8=p5E?(UiTahG()mo}DL#L60Lqp9I#u0O;vx
zAWSR|nF6Kdbo&(d>*A86s6D}fBI7_Ny<fuM`|Qf7d@=lGpic)+0EM&2$!{-N#~}6#
z%)H3ku+`;yRfm{oTye5<TI&30E8-pkngsVnBZ2f)xkq5l^P$fb)ml<aYN&p0!{X|q
z2N)o^$!H!3PP&Wy`P!2$tbO#vqAI*n5!}|Q0A3Or7h+U;>%!D=$&76h88mH3&2NX%
zN9r>*B>V@Tc2hEDOH18CWt$pB8@{!)p3S=VbqsnHJ_iE*xN@A{%Tnu}Ixu5^pW0(G
zWgv>q=jqX}*Sh`2MBu}qcjHACBH_4Y*jiwLV9pq;l$A!`2TD%v;-cGG8+TpGw<+7p
zdaEBMsgMgy9JJyX_XK^v^wRFpr@+FYavquz9-Ey-`ieCXAqjzsl%JWd#_%`(gEi80
zB+$cAu8*{UwdVmF7m?j(;od{_#M0r4Di%;z+R<+Spx0o1`vQ<j+ENU@y!`45dH1Qx
z+2!MU*LT$UGVUEB(X@-$EUPqJq9h!-SgW0GF}$vYT=d#Reh!pq`g8q3&_=v)gc&BP
z?$nQD$*(1cMTG*lEDllHE{?&e*G##KnG8C*$iONE=R~^KgcWH;tll|MlMALW#T_Wg
z7g21vnTngYdG1T5;s_+~;UjhKab4bATEtv3>jfRceH89CM!X>|N0eU>L3Oh(PfqHd
zpqFKg-<{wqPSqXNMK^}MSqTaw3Vja4Zpej^??u_(KY=b0k0W)j*Pi;4=8jI*6O5|X
zR^vG0hBQZ6;IW`E9^t{{+Ep~BM2s|l=kCS-XX*RD)O!A#_1^rO7vBNrNYP~jQ9U;+
zddQ^7MI`3((w*U%lps6MnfdH6B@uc6$zs^{hW4>X^r<2d8pN`k@m70Ii)$~%{vw<R
z>ZbhY)}=hc&=Chb66Mf(>S57?|NVQF{>`udil!@(=M5|X8gWYX?TzvBMW^VT9V_CU
z9EH*C4fr5)ULc)smr}^7f8QU1Czmkok#z&rqYSaSzd8oA?4%U9IWgd!3;YZ-p+77b
z$aaD%E5(xesMvt_GUpAjI_5w_q#8o#TCh9Rj#RORj~W(<CU*NH{KO+V+B+H8#B@c5
zzyi?R=3qLtJ0$@E9LFE&jiwU~R$s6#sx!sSs`oyMy5k0I$!9}08dUJ%F|})CIuALQ
zwwstgMZ>%4AKBdOFxf0-t*Cd4o8nF%*vfpyPt9DDDRRfdHA9kJ=|Q$ynTGtHxLL8g
zKV&HC<mtaA_$@A*ZLp#%&Ya8liWHOGWI*dlR~UCmpYvfpTY1kilwSI+3EZ7Be7ea6
zKl-w^p+t*3g%edh;LPx`A<r_i&EL!xt7uMdq48CSzKuj?Sm3Z(RvF%}?SwlPF}q4$
z`g5=G#{B+S@3Vtt4Sv{LK6YgHBcw{k9;h>Kh!t~6Z<-ZClkdQNU>JwB8!yx80{il3
z&lb7#v8;yU-sH~BlJuHeOId4jc-HP10f_i*j_W3R*;2gxq#}Y&5h70sld8uEY*CFo
zkT}{}Pr>rv9jv|8Jt=-^Y4#?2Z9`sIszu`EsP}i=gS*8DhCGLb%tWK5P2|!pxNdgf
zVEP7IfOmm^Nc$Y+sy7qnmLkYs8i?wT7fo?+xGs=WK^AgxT)t-A-5o@^PP@UW&2?d*
zJE5V9*+@ndE8VP_8$dWv1;lSHfjPXD5c_h>E!X<M^&w9s0W&73J3e?L3F#{b@+_7H
zq*GgB<b5+}mc4O*71l0UytIiEjvz0?)|x^(H|2*$v)!u&S^k639cP6U<FSns*mc`#
z&%WM3476d|0#B!1Pg^6@*y9R0XWx~v2mA_Ss<H0K!QQsU%BypGyu)B`f1=^+k?fN^
zoi~W;%~-w-Jvqm@rnDMY@tAI2G+1J-Wp0#fR74f7tly93opRQ64FP$gH&a}ZkioJ>
zZ<x{HPV1NhTE%eAP^H({;dMD;-kLQrY&i0vd*6O~u7ycVG;9Dkt)b@q6Z5W87TUrD
zv?@nrv1;65JXYwL?4`7D4w_!-Up5@wzXO*E`gZnL4W2IiB5bBJhiF9@(S5aUywIg0
z;L=usugDPsKjEKef3J^j(PUJXSX}1kb}Z=DAg6BiL2)q;I*;8h#8tEvwS{q*$k9iA
z$s9JJ?j}oXaS0!^(GGjH9#>aIC1V_=S{AyR>qv!>J#Wma*80!4!}yzIl5tAf)GyOJ
zge0<i^fBdQ+qZ(RA1ywn@!Qc~dk_8V>@ODQoVNe^e#gBnxaU?6^MBswq^KGD`wueP
zlaE&Z@{94-pPxU4Jh`j<^RIiff0V7kJx~I3)?ayd_T;p8h2;B0dUK{iEOCP&`FTZ=
z1OkDWM>v^#(AxU&0gmovm#Y5saTPTKeqg`leM^hrQv*ERyU*PZ4qsi&bctAO9H_3m
zL@H3<OshnWDP;1pW9N5gMxC$FqC%>on^{qY6SsX)=t`_<_@*g3rNTd7!H4ZVpbfXZ
zN~(i)TKSsD`^2<r*FgI&rr%lXK5uZvcgRgupfsy;#FL_G2sQlTORop3z_><6?A7C~
zF?kz~hMfdO&x5L&D+L|WXoZT04R2=T)hPYSs>=0~KP`6U%`HdY$@y{8Pfgab4C71b
zzoo|a$Al*_excf>JxSLOT-1g`RcLB&N-6zQ=qTj7^@QAPN^eJu&!S9u)1GLP+@FUm
z7b}#B$d#pYsH&nm#lfo_k-L*P#Qt45-TTw&^#$ep_CpjZ*v#ca@GtDXvF|&!UkIk}
zDoQP_7MU+Lk@{4e@@hgVMV*|z=s0@|oD7lEn(tATv4|MTgL9R+jxKPq3oa0xYYo3X
z7^0D`8u6)8cdzI@=k7mN1`c^Pxsw*pFb87+o|b=)bbd5H6FPRTwT=)5x!<O2V411C
z`g^LJ0nI8iXE9q@X<A^~1GW@TE1Kqi&d<KpzE3VzI`~&q!U@CUR98e9&2q6<S<7N^
zRqraX<XbZMew^~~)A1hVeAW60o=_T{4f15_GgMNKuOq8rb}a)}`R7n*o7UbO)Q}4G
zv2`mW9TD89PQ{=$mQw3#P}!dn$9j~9pN$_NZ5~M*4*{E@#-+i`%yD_?r7OfVJMV*b
zii*$1Pw0-1#eq3y=HE!(-JRNqYDL0T6iubxo<~NbTs}eIa4&aMH_GHR)z!KaB8A%U
zL8YihgtXRA1IL`-leq^eHyS9y93&|z)Jj|ny@|3w2Us~`b3SZ5?}85me&ENd1qZ0s
zLz_+J_G7Lm&+_TLc5?o@jH<l%Hd;zo>yI6pdaL^$cA9!TD^T;)P47%s+lX%qa-$5(
z#Rms!yletT4ZX{H&1tk}wOpf|$-DZs*H(jl2h4~j)Uj7K^p|xij?#g>9f?CS$<7ig
z1cxl#L$_yetA7RNGrXfP;j%SXFxH~C(t)=cGn}B4HE0*qc%61|$9>t)?%1xV?#%AW
zw%M1hagxlH0N`a>z2?K2Z)+5adi!!8@LSZR&sEJ<xpS2BF*Lome$zz{!qNmIbvry>
zdR8sHTb2bw(-$)+jj6loUln)BIDURJn9Pj^GjZbvHi~#tL#iEOp_C2je;HtHSUK!q
z<KPvrZ>CX`s*4)ZrSfcA`x;YQNO<LXUsS%#ht+uP;GC!9H0#47(Dd;E@ZPdf<Ggg-
z?)XunSGQKIEBj@;W|2vcbVF^9ZVG~-g7sXvX&QQz8lg5gzXa~=Sks&Def(#-N<=7X
zOt;_Xa{eAwoAkE1RHOybMYcFzpFMeTPKt7>>CbvFN*6s;!A@G<@RNMDscNICd?~_j
zKWDd^D$33R;WW6o&>sDAI22=>+)^=21bD};43Oaq71H6dmwi4Rt;iYQA2OBfOVBXx
zZojF>l4Lq->H8yk-Of@1uA24)1sEGsg~jgMxY}PLpqZi9l&&mho}4v5fHiIjhJ1Q_
zvA3gYR!qoQGb-)%o&sjsT`TTs5Hi7QsxB5Ctv8gea}NQU4J&gTXd6l+-ILW%gq*yL
z?&mUQijrVXHS3y%qZ>7bV{6Qlr52M(iB$U+v)6PjB0e<{!x0k-RUF<Ys@(YIICHbN
z<N^Qnp2fjCzfmZ5Ut@e9tll>2+ul|?O{kxVzBg3qqwy+#TRW>j8I-DXNcafsmSL^U
zFL_s-YXHfl8iR$$b5-A+bx@A5<gJ;hr@_+)EBgYi4R7ngs&Ni4qFn7uVvwD6X7(jY
z#y&3BYIb?~9G-ZfSsYS7h||$UqBbs}tim$VBcgI|l3$+xHs>eUeQN5_mpYYx69_+U
z?0^S#uf?LNnkOe9#nCKn{42PWX(ySmaYr%&;T<=hZ~Il1Q4D{&qEqWtwp{7!hi?6>
zyn+ZLob+@}n|qxE8%W_69~d>(o|{ZrAZQvx+N~9rM_e07_miEbBB#M>t7cjg`=%RV
z?6(+YCF)U{0q#dmx$>0ltrYjD-b2lL=M4_H8qg{M%bndBIp39AqFj4*(HOi}CBnV6
zJrg-~VzoZCbsY|N%gG@(JzjjD`iIlNO-m(o=cyu|%E4Wm+RI-eyw{<|bD98wgRk$u
z_hb2i!k%)3sgl1@lvmf8n~l%L6T<fOm2*pa-3Kp3CDrbQbwk^l$E+i@-BRx}Lg1)$
z)%u{;hi6Vx3BBfhR*OT-H90ljq7mXQstv*R$sHoP9FHSR{jOYVl^M$NCT$9CqqH=&
zQveq3_WlCceC&>T=+#FNw@Qa>_@k**JLsFXx|rZ!vzIjTQ@~RYy9^aiVW9K}#CWlM
zP1Bq7`csJ9bchY0*eZZxciYK1Um~P@9HG?R;PIupEF+A6rlv;*yQPh1oqA6VckNAt
zi1b^r0i6~Hj#iE~8GKB-m+aIRdDmUibP5X^6~JcSVp3m>3C+4w*k#HQ9+xAA`~hRb
z0y<6<)0gIAN<D7l-K<#pD>-vK%tp~Qo8RK?B;COCR5Xy6_3r;b_bD~)wm+pm=LKzZ
ztJ{hBpz`b$c$UuzwQEUx$~lgoZ*Eh1ofF?!q8xMM!}@W6usZ-@-LrWnX97p&w6yM;
z_F;m$Y1(WT8D4_Rs<ou2e_wC)0|PY@$Nk9yWij8n<zGIAVxs6BOJc*zjz05B{gcCq
zdW%*^1NNkiT)^dSW()7y@RG~>njDeBbqBrkt_Dxs{b`+K?E0+2>$UNWH`F-yAT5QK
z%2r3f_YbST3Y*Xgb<(n|!Ys~@C<BY@QKsa&ECuj{*o-dr>~FIgM1GL*m^@TO+fA9A
zCyQ5>;V9zcomHjM4Kk4XzFC61OO?tFGb$LwpZ>qms|pAawd-$#cYpfrt6Mb}EVtyH
zgVczd^BriOe#z2ZT0ZS(EaFk#@WYOO*g~6Sn0jtlKNXC%`%slLPStIpJ5ZKdQWrKH
z8#LJ;zz3WZ_gXF1oD}Z9jZJb2STyu@Wxps<=5{>a;++i(7(rh6l&$av*p#0J|9ak5
zGNSPe)hhSe`!`x%#FVS|Z=$ShznY<Vv!5zUMNaf$PY019(z~=C-P?Wp*-HTUvLj@d
zCim1#^hkb9p!{Is_o&NlvL;{=(hmi^e%AYFuBC%laa(0h>amzLQ%VgwYJCrmaaB&g
zVDn4ghsQ%YFjx4c59Fl9dv{u&*j4F6t1_oDU}Xu4C=|C$#W3<h9n61(WGT67nmQgS
z`;KEZ)6*rhS@&ovIoL%u^i>BT0xLqKyI+FUq)EI-E;04W)H2n2uunaP!M*cjQ}^&g
zvL76}WRkM{Ml4&|pm&oM|B<a`8+^DM{|(zeue*+FXjmKaufl>ZlvdgOVvKCll+h~e
z(s^qUf`KPH@gt}E_e2+e6RO?n0~{+@Zk(?s{cgB?SIpb3OCLf@cczt<5F(fWHD0vq
z=PNWebH;Pk;?jnHp_ZLkz0W0iqD@Xs{LEbfN1Ig%DsLigOX?QXRAif}<ir&AVwO)6
zqN{~vYN|;LK+ym6DP<Q{Bb}?G6C}UNSS+q7gL)ByA|UL%e{pxji+X@eu))MvcRWnD
ziVMM9mtmNB^*>c!ZM)XkS!x^>)ws54rKw(}OR%UGG9|x)97=E*aFw;&VSqUE^;hD?
z(oD^_X}Nw2K<U~<@}eD3plcBUDr#hoz^(P68t&?1K|35>rA2*kAU)X3wgmL@d?UxH
zgxq-k8+gEy*cSMOtzW26kwqAw;jcHYq&O)}oCKn*^JUcZyzH2vkEe!DrskBknwy63
z9S&q4!28IuUkH4ob6gW#>4=`lA2Jvl8Fx%*93YSZ1&ZMrbp(4Kz2Z>6_~YI|Ad;}I
zw~}`b3cpYrk={*t9dgu|(rWB=0lRX-uxuA%N^S5oWYAYD#94n%yEOv9rvK<0gFke#
z(>?P?&PQdX(szwVy<0feeO~$a#e2WPJ2}Y26R+#l^uH=pfq9=ybvAi?&onIC8;d)y
zh`H7avxY<aiu7H$I{={uVF~peA{CqdqzzWY=2SWMTA(R@X<ys-EG}0haTnGt?CN%R
zcY<Hey-BY=g}Eo&TLa+r7=YI-F<x_RKh)<ezWMkIMcFWa7eBl1sq+rkX8>24Z>d#1
z;CGYGsw`xFAmBV#?wc-W7WXwGJS0~I-YO~b4O&#=qIs@UIOh-rz$>`3L9IB_R39C*
zX`z#j#{Idntf^~T?~Q8H`Pi3$9y%F^sr1j!e!|-QrxNzZ;W3><h}lOLmc+SlElzw)
z`9;U~VZ(1UzHkxd4lujH8MCuqs8ph|q1gmAMYv<rzExJS7Q8t6;w%XD*+7!w<-?`v
zs=n^H54+)g@P=aNmBr#zcABiUqQPAH2c+8IYOL}mHfM0L{6H7eqIXrHs*n8whPe}1
zmoIQYo9IyWbk5eKTv=p*J6k)));MjQv+*I2@-?`Pf!1x%S?Ct~cKg<`KWQDE54wo(
zOK|LVl~X|<$K91;YId!^Jdw|UM>)%xW?ni%@jA{V`^U{Oe+}feIkln$jRv814%L|)
z;3IF&;?Ce`wJ}OCDCC#ff=WYZ{XpuRcXXR(RpKu)DGKOeu-A&0Y>j8x;QNqvx3UBv
zBr*VVo=w7RVj^IwhKzvHo%WY8u%(xPDIAEs#CfWYdmo=Z<+JwcRh9mHO(uMBArSJ1
z=g|Gy_|>6b137I@ZGXT28K(|YAKa#HqqrpR^G)S;npfFy`>8gaw%MOso9iT*H`@Aa
zR8khs&WB^dlfSTepN#b9RlJx_61xLHrpM!5oVNFE$BP`;Q^4)Fwdfu{+RkMFY31T^
zFb6mOCY55+X}Oq|DhC49s`!Qh*v(LRcHG{^VZFX2)IDn)spdJ8lQ!N9hGvhm)E>u9
zsjTKJ2SFA~mE8k^WT!t8`XDi=ajs3I8sk))4&qwt*h?SP)w?K=Rpzx74C9V$#%H2+
zyz6SCGALv5E}yD4_6Ua?S3)-!0!frh&kaR`9RID}6LsR!<Nut@^S@vv1Hq@MFO+fp
z^Jkn)6Y^8z%)!k5+YRVehl9Rb5j3|F*&5Uaq_qZAGFTk&+7lh6JF!lJP$ITC`mYl@
zB_!mzadKzQ9l3fx7$wk+5^9>`K&j(NJHN^-r!~XTUtC!~L-O!#Oh{Vl<j<||-kc>~
z2~m&ZuEd?nBRT-#RWrKbG?jvmIzi0^o28FKz}T$uy_qGYcUF5B&h`gl`K$=eIy$@8
zIR=_8H}k16)rfT+<z6ZrZZTYG18Nc0HNrCksz5u3SI)CLh`aSmdvGy~-)SI^3*XF1
zr+v68Rm7f)S=SuWNf^yzq-OjwbLg+dspLK_z;T=aI%bmFri}uJ?%zP+z+Y&^SeVIP
zIA)3zjvd5}|CTyuRyEcTI@2qcnu&4P3}P!FI!<EOJNL;nVJ8pjND$_iq?=t&QR5C|
z?u$JU;epz%>zo63xvDj+s-b+jNm=H~{o6r;N3PzlJq1Q=f3Og?Wx+jroKE5mr|9pB
zb-fX>wftHcSAe{NdL{pm98)bM#QQ2>yUb_ObjSVOTwha}FRqo56LOQDy2#YeS1N~H
z^?h%uF+tck$K?g0H@K<Hxbj^qH*?ChsJkQCF`46JqjrmQKxm;sf?RN^<8>q2tiV)_
zN;;4VG?1Q7Slib~v}c2cMyh=H?;20!_64Dj9ey5U6MxsS`V5*Eb%vzg*VcLKpxNhj
zciDARHrAVWqOCTW^44gwes@^g2!h~AiJ0k<*8Qevc_gxIE@U<z5sa*RR0kB`)93~g
zeU^*ZJ%<fd>d*7)xfV;3OkZ83687!uqd~>O7@{&6Gd`3`O*YFJUj;+;PvDkbNAxMj
zJkhD>0u<ocxC2T-_vhpwi$#AhkX5n4X1MVbRJP8EAcNwr=#8FJ;OK%~F$wzz)0`m0
z)c~SKl7B!>e!vNITVKJMTI<6*0M<4zr$E^8KC}YJCkyD|k8y(Q-`s2UF#H#)Kjl5?
z+4u?=+n2cM(dt+gWTEzoqW5vQ4mi6B9Bn*7d0~OzDm@zyDw^*L_%R0MQoDAoTlI%a
zs~(cW-G<=gjN_#@4y+<Oo#E@HPsdeJ8&0hY!j9ChCz&{%nwdgh{TD;Q;AnlXp9#AW
z5k{)H#lt7Mah3u8#$gDB_M1?DDO8kC(7HQ%+}&EG(I~@uxbdu)PI>U~+Xf2f&5}zX
zQ>T7<vHv|r_gl0IM5FfP)O);gge|I8b=41LVV#V=>~no;D8H5MD&I0lULk1Y?&%$X
zjv%A<4}(v=2TF2p3L3Z6*`p8+WvaZ}h?`pN1~$->S^D#qm><0B&Ih=oHq&zq8k*b6
zt_{8W(ruzal~;KkcG@_|QOWp3ObX^nt-RgWtR&>d$jJ==0ZMDU>>-E42Q_vl{^Cpx
zKo=jmQvVj0I(dBR%2I(^a9^AJrT2&uWhKA)FYZ@-l@I&a8g3+N2Or7F;dRo^^b22~
zsapnx<j3RW;N348hHj@XBvln~t4b;nj;%7gA+2p%mVZq7N{@cKf|5-u&mo%b^*&Ww
zZIH6lpdm^p_icHJ312VgPRV1P`hvxNTY)R-_SJs7v#Lyu+$~5Kg@gHLGMKe%!;Hi4
zF?%BEjrZRp0uAPS`_!!hjsZJ-&o4GhHzfFciknvRG^+yTO0~kTmgU{imTur$&J^~H
z$MdnL@>cUCXI!(p*ppACm%=dGzFL(J!oCdZ;GKX~{~}s77*JpSoyak0OvJKgcXs~M
z>2IlSUJ%eGYUipvlT!=%eacVmv8EKFjQoW9I0sx9*T~!*a{~?v`jBb98qoc+kyg{n
z`Rd#Isn5MdZ(cF2tGm|M#2p-Q^Ul-3FPF9jq3=8U4>S>TtTjuS*4Xu|GxIfeDhKV-
z^<k0wr{eM-&aGvKH@nn3tQ1VJ)y9Ta+zot3OkeDZN!iD(Wi%1B;_k0#eHZ=RIv{J5
zO<xYWe@QQUsDAC*=DrUX&&;o+??w-4O?;-7q)I-rsjLbf4LOy+TCW^tb^nf5$8l6t
zjO631gK|SXZrJgiYL5@BTOY<jjZ(?Mk1r<%jJU4UwELw=GWA>E{oHEb)$pK}?2<M9
z)h;U`|JneLZ~j$<7n*IK$G2NJBc~OtZd|)sshk3q(|jRxi<ht9n5s117(@o6r1H)L
zJoalsM($rR&L8J5pPbdYHr!BQX7@q2<i*_sZQsnMP7$>Z|H-vN4H=2bZgcDp$iY;l
z_2~Q8OvtVug2AoVWmH!dtLTZ`WVMNM=w0U)cxR--Y~Ij{sTLJ&-B{-TdbyVylR{Bh
zeV~jh97W!+h~QxCRo^&?xoKzckQsipUz7HSvs}B~CoK!U)r^4jOedguu4<`&S_@Ef
zw&e8j%MIk;gE?#^wS)vUploe)rsx6RxM=gS-><X7-?G!+N^@Tx4ySc|!~DKYt*W2j
z9F4H>#pc`}(`)suUwqtO=hT++u~WX~#$d&?BooMXi@gI!a%wd+UIv_~(wlcPI#w!B
zlRdD~+!&tjt5RjvW4n1b#HqzVGT_8&&dJ#wa)RTZ=Hc#7)cKAIaK_1vw)}@B2z!0G
zW_T*$SfKG-vMgbm8^WwaEMAY|V`|QOX4`&P$V^fszX@`Or*WMh|Ioei7h5_<=To(V
zRP|ct;HYX}*aeK&{Tnb}8>O;QjJ`KKed>KTo=;Bgng_KCmAQwm)`GWjr6O+)n@tJ;
zo#{0nyS9D|DE&LWydmA-td;@}nra;zJmUH&CV;njQE%Ro_#oV6|H^V&*GBTrnLw}8
zCLiD^pKH|Ns-kiOr3K+j`Ea0GQ<YWyAdFdA{(I%K_THv-S(~l~@BJ&4NjQ^&IZ0+Z
zP||~5?Qgb@WJCNU<d__U-bZttyXA-aJW#2UjOYxq-y!ZV4oa<K6(R5S0zMB<HF(u>
zVCzF_!4PXcL6PoZTpay+!{BnmFtF&srolHuouis|8~KmJV*siH)2J8exsaY=+FQx5
zL=)A<qR34%n^D5bb+ddx6eYWiIWuvhwe}sA@mI5qHN90k_Q#eclNuW+P-DJqqgXMA
zM?{jM->!c1%cK1u{OpJ9dJP;wvs$>58>G6Lz~UAB0r68R>{T_rPyPW@b1E?Hk!GvC
zbtKv9ez2r2iEB0ETfc|*c<s&odrqj~BDZ1W)o<#khT2<ZSho_Kwn}?5p6By;XZQPi
zW2-L=M-SJ0<W@bdbe5@|p9NOfwfU4w)~zp{A)fYXAHNei=3RpKd=r#*-#9U(h^LQu
zY*kgDQ$H)P-JD<gA_pSQhK!m=#*|%I>{JUzs*NRR%1t1Ocx3F?kfpUUKY^shwSyjW
zz9T!HtH<0d>GPa7t>*I?Y6iT-<<F_|rQk-l5_ecY0ld4oc73<ayXf4TxRqag2UiAa
z+N}=|esSJ3{MXdkAbhWt-!X=7a=S+i7T5KIPRBZ)=WMkT0Y?StwYMQq8&*LA2f&7&
zy&X?+x0I6#snwK^M1E5UrXLW9K(j3C?e$#;bI)kZwb)s!%w4@&#)>0NbTJn@{0wbR
z2~;oUL3;Z?Xy3%-4y^>{2K3pW8q}#%Aro`t@?D_7js6;q%4?bk5M8=qQ%}F=Zkvg!
z;<NKPi%-v}TLcCEm{$hcHwDVuI9;c~(|7mzV`giysdR%x=tLsetlv52J{PD;nG_k)
zJV_8e-#r+3Rn6~b$_jcD<o}4T&LKO^RtILDh;dCV3CWWJqc$#AHzvtX<L%fGpG%v!
ztIbL-GpBq$wt0cE-OivB4nQFVt|l9PVq#%Ny%OOLEOX=I^(yCOzG2Xp5(-s7kqmyU
zYob5*RPKzj;?u4=y8B*Ri{Ur%jhe}y2Q()1)jZpxf1WT#UP2*WC#jn@2-fkmq{DGx
z`wlp%q?2$gEPps-p-Qwu*OC~*c>=YQ5QD-iufAoj)lt_D<O(eZgg170yb_7jLO{V;
zL45B<+H!fdl^K^EZlU!8lZxe~Yg*!OCS4p~Sd?3u@A4B=m;*ciwlXO)W_WefMhGvm
zt3P|`_2*yH*Q08CPvD4p7WJkoRFIdpS3w}#O@%Gl;spX%-i7cx3h8Sar0Bux2IcOA
z@i$v~vn1Ul2+O5=pl74?sc0bR%z#tyukYSxDjBRB;JK;sUrXIfXlp+#LVHL9(9iR_
zu%JFCgpp-Pv}8&G$SjVMonpE8k-27Ex&C5@U(AgXQ0b#@DEhit0zGBQN`CKcRHX5y
ze(t_|X}KlGwC&$&sn%Z#5;Q2wTs2Hiv7D&@aBt46G;X6LGBf+4X~n=Fp60MJVJF7&
z9$YYU9<`_C7;yYk<uw>husbbVzv*Uo)9!QWjhW|z2>IUgce#l~YUiaQ$^;Q8)OUNU
z`@md5#*`F=DL49BG5UwcI&+bW9+IGb7XxFZDaEN$`kHDOx=^hXh@wp<Zt#^O&hdo0
z&R6rXzs@N7`dAa~Hq88iZP<^9DERP5`02P?njBP4vsUrTi%bWeYRUQ~l=F!Q^Q*6>
z-=->o+0633D<Ods5%jik4PpLG@-3hG3{H&BS@Q5P2W(GpD#A(zB>+bk*9xVoeEgxX
zS8G#7SCl*gMUwV!Hu!PMXUemqeG26#K;%C(eKMYZAmiWxu2g<QOu$tx`SqDSImEV5
zmSnEVm~t=rq*>_Iet}$4TaGf<KO^6%K`Ra$Dfr_^ObMTl2b+DeHo{ph%Fx*cc{$iz
z{0ed}KH+WnsKP=+13%0imC(MA_Ocy-*%slRNMiKSZb}65TPpwHvne|*G{p7k+y2kI
z@nF2SaqbI{8im5he@OPu(wk@M#03hM7+!&zJoGOR$}_2>hr_o_GgYfXDxdk}4pM9M
zJKpZ`o@)$gp%pbg(^>7&_GJ2)HvB@}>B$^SANP`T*(B+7c)KZ0Z7_#zbOc$I(0Ro9
zt6JpX=;K}&3YF`Bp?)E@GVY|Gh=2zPI@QKv_(~zs-rh=mVcB!D0kZ_J)^QX>wRZ68
zeafE06YKVK>k9UuM+>R5vwCKU!)3khPrW@qajaH5gWj~A@zHAwE|+(>b6wG@V;tLF
z3t3#24A%S^X#Y|1SxEL13>vk42<3$d|4iZ<GRoF5;rF;)2lNJ8b4f47*nv$HZ28H<
zstZuCz8aguuh+(;gu?NW_PfSj*Tx!UP8gnwyY$;Owa1n~X@i`G(K|3wpo{}AHtmV`
z@5kwAP9hMS+;vb$9`Ef*A+B?r8LRa1xhi{~ngK_K_Wb;9_p5P3{<_mQFv=;GOMg7@
zwlhfH2(A_`zF~)eqZ=ZT>&os|7B`e@Emm*jf~Y<AT;3_vhD&SX22t7I(4Qgu$+Urd
z^u`U|+7YlDs_S___ie+oF7w3$;OJx2l;q#1hLp*D&cnFHxRE6Uu8^{u?00pWK-RkD
znXp?UXb;mVIgZf}RB@PV+F)jmPE9a*?=WsW5{yk7_mm!U`4u=p;9BQs4t;G-#3U|c
z5%>>t*gzbXd0u9y**bnX+#RKR_<ilOw~k-6Iu>IBfEEoTi5H?vsx<<-@_q%jh?If-
z<z9t_ji~*!Z7L6qdv@!VUjFpe&Ljx>grjEnjiRB}(P#TLpAkO%3P9|IHh<fQ>^Cl{
z=Ve%PslfvKKxgk8>7*OzPj6fg9Jo<~qE6r~mq=hHdS*Ss*vq;ZocC--xt^O*C&NqY
z+xv)Yd9|!uWp&VUv1rd>Gu6N_G&eGvzc+W4V>p^Mc}vYy8v~6YZH)f@jj&;cNo6#n
zqry7_uU$P(C;I%JPfp4Ty|-`B5YU&w`v|tSZ_-Wd%@Pp#5s;@8$_|`FCl*;?j0k^n
zCv4uqk(MTg9js>-=nXh;t@QXP;Ra7wu(Vmerhtf$obVY~e51X<c>X6h{qSp<a~FZr
ztNWwUl~=KQRNl%>G^Aq2Q#4Op)_B)veRzuyezlIG5^-zy3`sBd^0)GBB(Lx;+-~c`
z1E8*>9M$|im~vAqN%^xO(>&b+8NL42hntu>r#vR`YM%Yck(IYZO7J3;Qi9X<F-gze
zvoQbg40~o(Ne#4BlfAI$loH4Ib_a#K`y)r_3lGpbDN=mQT4ZZjYO34Eb>sH7GdRmk
z1zwx%yQouM;pj-5yV1`>g2)&xtHp)?HObE#^_xE#=SqIB{pSE2xRAWee&e$V@S^2=
zAIdeI6KyP?KKuP?%CW`f%kCVwXR?%C|MlB|Ntc83ru?^fTKM&Ju$tbVwNnc9|7V;@
zPcL<yw0o(E)?U4*?oFAz1>(zkl&g+U`Pg*+<<s}sntM&=Pcd3}WwZ2=t8Kds|73gK
z+yCIpz8ROQKkb+@^Huz$OUj?>gZ8z5p1j94wda!c_d|d4Cgt~Z=3m`+J`tGePJ8<K
zNHRTJtH~P}zU0q`F9%irn=h5I&3Cd1U()ls>g$w`rSVgr+z;ONHNorf)#6EaG*9wQ
z-lKUs+v3hyVCkUmEqVQ$q-N@}{qyfn{G@zf>*PPFvvR*Hectmbb<@g!5_j8|^Q_)<
z>Yrxnng_ti)7MizA2>K=*ONj?p8EwsUuWt+JN3)IY>(B`_Nl4WHSgcsFXI7kxHGtV
z^5)MBYlk2!r`QQW=6hB?-m_(?+^Z)i7Jm0zac}qWNuQ^xn6JOI-=F;@XgzZFD_2Pq
zt(t1!*jw1?W}_AIelINaKU*0#=SrIPeNr~w_pw|u?xOea&mX6NmN#Cw@YN{eg2n2`
zdl*)y*R$=dn18P@?pdKi>W`UqwcmdTCfZ%7e*f0{O!)ncl26|LlU)K@3BP9M+9}8G
z_NQK0QBXBqk}2$3ZP2R@XV1P@7W#{0Oi#{V`sA~fW-^=l(s{e9PR+RgJT>6Dte3fM
ziTcm0Q})aihTOEt`EItOvisgSlP^gy8RTfE3hfVld$js$&${c!bicn=o1H8>+lcdi
z*`(X0b3p4l6OQ&~J=m*rvUm3xkpnWo;R-n)^U2FW%Z9cr%X_d_<D_>|l4-)<xig*!
zddAIs60F&~bNX_hh;Or0_<b`uAe*9riibZCd;XvKeP892b}5I4Am4hr`njxgN@xNA
De%L_u

diff --git a/reference/plot-5.png b/reference/plot-5.png
deleted file mode 100644
index 6a8677b3843b156455da3a1698bbae791944e159..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 40720
zcmeFZXH-*L+b)cK+i)u?7K();U8zzd8xcZPdJP*TAT1~z0>N!#BT`H#h7v3Rq<11U
zDk3Em2_<w82oOT(BtQt98K37p<NW!?`TvcRF$j??SLV9sJ+C$wMBO#fJAUNC5gs0%
z<52zEW;{Fxf_ZrM-T&ht_{mVx`z7%0;C(~A+u%ERZ0giRfNzH%>D&78@CZh7zxKpY
zm*G4-mw2GJwJn0v=0}3k_yTsCmV=Z@-^|XQ%)Iv0Lifa3X`5Tx<EOvFcz#(VOxBtw
zWQ6wXYD#|+XFg3kf@_t)ozal~_r-^QuS6et`t*p|Y3;ezZwORGn48fb*p@bz(gU#0
zrmJ!D9c_d*mzDM@%wibv)bA1fzo-9K0Z@Tkj|fo7$HAYmQ}NU7`Mu>Hy>3Ab%ZGS)
z673MW*4z>`qd(z2_K>M+_@ij+bCUMqfhq!^m+$6KE&LDsPYu8RIsV_*Pb6RQ$o==n
z7WeiC@A=Q`cdzcFmB7ot_39+6_ktHsthw*qf}c+8ftr~@^$a|&=9E-~J;V2+>))(Z
z;N?VQig(7w%hEWa=`M-AP%}R0NvT~^k@h!=1FevxuR9{q`f65WU392^uBbZ>KO+*8
zT;6io*cz%MA}F4N!~2LDC$2rme>gT*R)p6=#}*42!(-QvK)Xf5uhGsxJr%B9l|Bhg
z7frvKLud;rF(a#^ap<)--@*&FC5!y{rycsur_Tvrqn+sU<AcVCFfO~-w1u!=v>~Tt
zpk*T7IZw8dvc2lFRhxca;gdX+ymq9`6FrqUxIenQS+711{xk|^gt0(REN4~JCso9l
z+cu(Op{;&C(2UHS$=lFg&AzTa6Y{X)rkEBi-h$$G-q}Nh@49c^<e^uT#|flBgjfql
z$sb+d7xYReCCHIHMcTgg#sgv?5@S>*r>z($v8W#x;xHMHiZSB3(e>g;VEzMncE)0f
z2`Vn)yR)!?$VggN{)`~GeaVL6Ma(9B4CjIq507R4`Vpbx<x{ts6W3f0ZW43WwBOTs
z-NzCwbh5GwlS_FCD?|2VwK#XJc!KI#Uy%{vrE^|MG;RA)-r`oKSrc4dJMZ0|(Mx-J
zy>K-sI!2ClX{<sP4)=JegPqjG$29NvHP0m8Xk=pd2R}e3k!p$L!>7grBo!S0<`8*l
zS|*r6jnVD_M8@yR-8vKWPX4G6{^+Ht=C!uA^4Far)T)HZ)8pw93d7GRzJ1NZsQT}2
z9238sRF|iwoso|OV)e~U-X0MtVA&F{U@u=x>#JG5_>;t}qGxMu`YtlUqv$xdh1-$7
zhH%vQqwdkm-bSv?|F~ceg3+9n<iuS`)cj&Ev`F*Hm;~^<4;BiLn<pXCnU~aL;33Or
z(f)?^!rsa3y-~48Vbk`mw1OBsOHHz$!O&~jPpr+AQ$WD|GN{8DuAB^!^#B_d4*yKl
z5>q8upV(%To>MEDlhs*$ARx{to)Q;`KKQ~ZF`+bkA=FZ5dVOxTct-<zBGSRNG|1)U
zY5k<DImnc9{+rUfHsm;2T3_>;EJGycpNS@QNxxA6sTm&H9{EYt<Ycn+lr0!=t8eT+
z?gTzD@Tv-2Sg!ahE67eQM)Y{{;SX76reehtJ!SJ}pfd2imsI8J4m4GJ31fP`y_e1w
zs%1W?wzP_Edt22&20z4Jn^W{;3t-2i?Bn`g@&-s1uNWk{mYVHuWJt{jy=SQ<@MVPy
zwpI%7CSs*Kd1ou$9G1^g&q8c7``f0g=oMN@E@B7$2gl4cdPJpn{Xab$I`^XrU8#mw
zZ@JtKMaL$yoS~zE%@L1N%AJ|xf^V{<iJHMAx@J1VPk=jq9*f5&d)~abOpPef*|`Dj
zY*p2NoNZVgp@x5abV)Xe22GwfOjf)Bbu`D*W=c6FyaDWJ!wZBG{S!g@lhjJKk>kSM
zK%?o(HH#w0Sn{Eek&VN%aQ5)JS;n02iG?h>>cvVc*?`9>Kk4Tp8HQ6@BCmuGRrQc)
zdM71@c^22*=Vn!YH^9ox|191wvR`soqnT(Ax3{by-jF7jA5{TnI$~cEBuA_=7X)X{
zjx!i{V~1S+&QQg>PobwBA{EEfydaIvNPERES})-54F?d~ReiFwZ^xDl1~h}Zd$X&J
zmO3SDtZ@%S#qGxLW<ylHcGDV|tmih_bG-2HA_0$~{H{+_M)n7x_SI6?aUD!mULGF#
zFXw7`c;1!1{Z+iZ2YL~{_xx@Wtt+Iqt$+<Hgy$N1Ocx`v@D;M|5g5fTXGdI{4F(l9
zUL2NVn4%|;v^#-zjXO=sC56~TQ5;s#M(>smMFymI<J)f+ww=q#X-h626lL7668?A{
zP9HYTF2S?lKXArYR}{_`4NW-=teuFYT4&lM$;~`|Bz!bm4UPAfrEkXu2wK?w8<IGK
ztp3pP`1kDHeQ_)v8E2RcD~~f&`z|`>Dod*~N(y{qxYc0pwb=8FA`>5$WOzPO4d3z@
zit+eMgfMwmlJ8G{<ju_RAvjUK_e<vRAq{-SEP6C%Oel3g4HT@2RszjB<foe}F|)-;
zEoZdw4#_OpSjFOb<#kFlq-a;M;}Io+8j<Nga$}&?9iqF}kxRU$Wpb#{rKB08+R)5R
zz*IcM+`T+cS|<)<X3Sr+#o;fbwGt0k_864Jbid6&V{cwdPn4w(zYNl!uJjR=rK19E
z7FcQ>jGI%<uaD0&A5HW;yM7EoK*_DfkCK#~oBOIR_Pp6M$5$zX#`)S%RA{>v5PkiQ
zx+mVM@{ljvSKJHnk?_55^qRMC-qlgZHQY<CUFlA+JkPCFX*w5M_UwTHrnL1+uIe48
zS&jS0V%BnN+vgS{?R`ukaPP;qGl4@)t4L}hRjtrf*ZPZYHLlt@$7^QcTGb&1Qna}o
zUFysjMN?$`QU8Kva;=|8-4%uBI=Ppjl6n3`(e~pNwfIm|>q_pb68R+UbPP<72+bPX
z63d&cXytm$0TQb3kzDTejKpjqyvio`!_$X~kG8Z*_0PJ5A4nGdE|wg?4%R0+*_>Iv
zba4i8H(Z3HcP63$VoHvaNLrN2iCJe!@VaLtu;Z>IHy>l%?VZZB9V6?(s2p6O5x3;}
zcf0t*!Kdc5y;<YO;#^`}o*sE7IwJ1woKh|>$|1JGhY&(@<VN@iONeZJ7SZ(U7#0_x
zEri&!WK?zFXAM$+Oez1>#-GzwsU6l<AvtNH?zNJYO)f@c1~2`fZD&M~&7$%T-&lN&
zL!Z-C_XzB0vFfXgyee7uRzuC?fL;;7sjEe=LQRurxa$Wz0`DP7ZG$+E`x*RZUc@%%
zp|`_6i-ocZxBtrZ7d7BKsOlE+G~ttnip~DAAim^vUvVJcjB(#eZ(yFN=_&XileCIj
zprQ+37*_k&<#0@LsgYxTDvseEt^0e^n11<^%FXtj?_sGfq03oz1c!xy(ay_`)zCJN
zD!gCnqZ`ee<BCxOyAI^-u1r*pm@P&U^k(z#rG4uJw-wHCq-PGl^}yRq_YL`!ld3OH
zg&D3l_zJ5N<bBn6b*3Wr;_c9g_Ix$m>&Lc1goc%W@rTOpCSTtGpfYepA;%{7{huPV
zPV|Cx<AD$gG*q*3#SSf|ia#&Gx)c_k9N<%vPoFQ9^>R{^hQM-~J{Mf(d(Fy|ug`AK
z0#JAB!hbZ^W315$<s!_n3SXQ0YmKEd4l7l+$osLdf0#0@$KiXQt9-^IdvpY)QfCs>
zt~s{b&A{&$4$2e06k+@y2V45B)sm{U(YU1ZaTF9IrNyW|;C`Ry+2nRVO7uFKH}U*c
zr9?xx7%cCa$v+!?yrdDBUrV^x^hd!Y%A2)C#m$%QSq&=!*h}S_J3$s^de96~S`8|Z
z0`R5tzxZ-MBX{HHp$8)CNj(->*4PpD^(%htMP1+e#LfMY+mY4yW%g$Z93r{;=W+CT
zscwmpz1T}8f{k$@O4#urIV_Lu1QR8OC4CQ0Ry!2_c^PSls_b+uT6Y)A=@;Ei?Vb;+
zybXW)69!98)|g5b&*`}w9@WN{7#pmt#^cadSrRJqmoLPwekQ_%O>`9XsjL5mXKs7#
z3FNtqX);h{*^u6>x&8qUyUeHPYc{a2_9XU_-8t>-l_N_Fm%BZ?O-Igp%#2%3G#LNn
z@u}=^gsiLDkOPx_3r)%v2Yk~F(@zPTXg<P>&~FgGHvyde#=CS!v|!1W2=zsL9Nt??
zUH;#k@aJ(Lo2~C%;S-PyuyGRO(@&4rJ}=z-I;dLdnTDjD5qSVDlso%sV<``R6%KnK
zheDtC>2-G)Y!w@+qBkX@V;|c>3|yOEt|*T-DAPs4r*05E;bIR~J)X7_4*BKaQCgJ|
zhGd$rFzOrBGT1aS1c;0$BHT{-Quudi8-r%p?L{yq^I*rFs=fI*S7;W{5fN|RyTsNk
z3$f;)V*J1!GJnhPONQc059P4SIWO=<-CourWlltwx;wv?LKn7f7%n{c6LQJVJ|bl6
zUKR|o5YLJ_8iDCCnbu48Ur6*XY=wN@c;ANiAZ89A^6^^o4D-8*o0{n{zk#i}V3upM
zZC<tx{}ujj{h>JDPfsz0!Ur<gAtc4XPMU8CsYhm`t=Q6@S&2bMv0oU!<Td?Ln}-)_
z?Zouu^`<LnatS+?JGrw`SPtTng7Wk|c}ze>qUtEIWk39E%l;>>EdrQVM{qk2dkgaL
zelAbn#=sFAls&l&oz;Sjzb;J;SqNQ+VBckRf4{$rF4x7bKV;FZuzzRp5i##Ll76xR
zk~C9bMAY=NA~}1ZMI@tdZimiFeefd<$<N$!g1R-Y-7k=#*+YvALd|}eXe1S|{NWn^
z(H>mT`oBLc?!0+b$UQ9Ze951y6KYEA;5zQV_Nbn5{IA2}xpn>ja9dZt9R0ta_5bS#
zTg=s*GW;Qt_QY~1*$!$h;wa`W7wI2oDEX#GhHZJw(Sm#ct%V*KL!XW_mCO%3lbu_a
zrHX!)xV3JG@~W&X`h#U3aP$4sR?`_F?AAK31ks48GGA?Rhq<h(<sR}gTv%#V>r-m^
zc4!FF=VsjuzYTpW_)~+1=a9K$<>4(`Ouq47b;9d!?*P|U>2%GnwogArmd1<a6yb`!
zEk@%US0*3Ft>=jQ)e0qAh`2v%TMjlSZ$j4}o9U46LWS+w<u>GN(Bq{Jh|+)YK~pHL
z7r$hscCGq$jj3U63Yxn?))<#qDEn1mGvY(3L(pc3Pu-Pn*B<GNTu{p|(fdThr6iK3
zA7k*t_;OPldUc|?W5Zekd2x04NS<S3jz^KZPG*E4qpAEz?njrWvNxdZlEu7;>E@my
zxc(-*l;}*!lrN3zL_tZ<=GkAZQcg<r#l*8%dMe4GxT*I-PIg1M>0&KRXNpUZsFzdj
z+MX5->?Pmo(rj^jRA~PwZTWV}mcChiWLTf1kQX%bA3T-U-B468J=51*RY@Waq1BBH
zh)ay{r0=@=YE5HtyV@LkvFm{@|Ar)qv+K;x?5(xr*cc|xRFgW9c`qs!I#zNnJaN?3
zT`%ABWJKE<D`Y1l5-!?pn1^fQ9}<8fo5QrjpyYORsFl7<10fQx-8U8yKQN~MZSu>j
zoU?n2o${0hmz9jxZx&vWyv7$+`b<<x@)Iw@S47Z*!+G$>|6H=KnwB<;qfOtuob3HW
zlEa?V6ozk|nRYWS#2Xe6t?g*LZ?m=|;h`9xH$6^doYlZRXa)S~l#Zy3fxfxq%mbyS
z!D&-o`l0NcIEFEFpUd8@&A$4nmb61h;jvv_nG2%)iHbVpdbDfn{s+D;p^xJ#cXn<~
z>lI%o8E01F-3k;TAx7r7R`zyhqD6Zv|3G6&;W<MWlp@g}ivyhRtpkJR<Up*`xN!1j
z>1@{BX=b0Jwq6x$)~jHIaLC8fFSE^aee6}?ztPI_0S-1@XC%aHhIY2#!WnTo&7_4o
z^-<@#k!M|<*AGU0oBeu`>DK<(3buNBQEBTR5)*lC)Nr*}5sq9hNgPwXOieZr#5%+8
zqdN>sc3c{fL(|zdkS@F3N6pl%`>b1DmuX1LdZAkF&fRxnHdG<Y4(iKH$@TGqxhA{Z
z6wP4KZcjBah406Cg8u@tp*a@mQvXmcY0VltYJixQNPBV{ezzeMX8*q8*16aUoSWrC
znc)E^bhgz)nbyvQN+N2YX@>FVO7&=HJ>+?9spx!X3aMa*?eZfm5&uAh!6**DNUyRW
zXVWC@qr5CMipesKTX_XD_AWn0ja^R+EQ8Ofa!yR0zYFDWea)`<uBXbATR9*<8RV3?
z5un8LnF}tzNY-KhJ~b3gJkZwUegEI@P_?<%;Xr$lRo!xZ@^#g?)VFUdZ;>nHoi<cR
z!>>|iA7PjMiKfj+JpPWvissi=ijdp#F~-f7R_$11wE~2S^QIDSckq8@mV^O_u*h<2
zO-E~)hsmcA;Gx!*t9L0s(1jS)z$nax<3{&{i;CQhp%KYdmIJtcH@wke(MnH4D31Sn
z_(klO=bFE|)v&k`9jd@}v{BI)LlKUtYw*9u@&rSt{B-f^IWNslbX81wY8#`?Q<Nj!
zU6T(>30;t)-N8~Gm4W;Tm(<fC+73CBQJnV@A1zYnlR5ToL#BWkD6Td&>E@K###C{1
zOtGb!pwExKvze~kLb}A>6Wd<Mb{TxF4i7ClOL~1n-C$wHWMUbbfEdG2No$v6ESFXx
zL*KUYOL`M&Rg!=s(nv-2yDUHb2ZT4w_+yqWa(#X)Z@>4RhQ#g6#0S1_t9tV>I9|Qm
z7ej3w)+{lZRzO*wnp!RVn3zRe-}Pq%$NciA+<kz3vMqr13(Yj@>?stQOIo-9T?wi~
zw=J%xCTs>oCwnHXsP&;)lY&3iQ!tm-W1AgV^_LW^SXR)n76&&R+BsaL8wa`E{bNnP
zOa{<F>D0AGAKP}9gARnD)_MDF+2qZYWDhgUq6=ayz_3PHB1V(x96Ky-AeHDJYF1z8
zP#M=UydYh5qU%q*L(Bfj`||vQEhdD%f}b6PL0&`inZ+rsoq<euG9<pdE}61+U)6NJ
z>`uqm;$qR!(7K2xYeB7?eraEQmD|hq9p;+oXKTy?x~53oT)b<h8LO1^#@{|AbZ6i_
zF0d-%&EY@9v#@FRYg{kSduFleR>?XJbIy?Ay`zZj1vd-wRoNk*<ioe{6XkITHm>zX
z9~0wHnKOI&0=MevKGb#Sx6;wRQmTO_uiMAxY7VsG7$@qD+s|CIp3N((bfH!P9o+1O
z){f^4yI9)~rN<XbkNkNYDJk^abD<1owt2kk&#6^e$BT(KqcRwB?&NK2_1){gs289|
zP<7}QQ15|&Xv*(;w=^{{f)9->`s*m}&k4HiYUD@or-)@PtAo$_0f}qb{uglblOvDb
zw;dR$KDN4i>e%a+{jGyiqtE94ZhW>U<Kc|laFjdb4NWw87&B|<86IVHVjxT^qYnN<
zpuC%4e2*a8SDh@tB*}caSU#-L=oU}B<h+S`bts8f|K8{=Y&6@|O6VjO#_tUmYmI8z
z-(^NizMJtzUh|f3eiKf9(i(<K_(AtnGtRPps|h9Vk=OijAH5Lp$+Vx>_*0y{&j*Y7
zWho-&a6)gAYUMpFrId*OB|+vi7!WM82yXit<7hDXI{yxn{(8H=1e06#TQMeGS<C^C
z0*IDwNSr-g3v)j9LzZRdpgcw-(T<Km-WN@}h*>LaQ^n+pzpFD|Q6sWbXgk%ty00>d
z4oU9RN4M^0?RsV&C1ZLQQFqQnEQU*U<9dqc%$>;R8iy!TmnyrpvW8d|g~gS`QF!Vr
z-ldo8whzjN-a<sBN!Hu_sv|FMNI0kV$~RGMuoS6(SN>QMDhN);&<Acfr`vzovh_@r
z(veiT$LE<Egib;pUK-paEZr-*^zDzOgPXO;Q9FvL@%ij)J3BWm#caGHegq}ZWG-T=
zkPSHq6N<js5<StVa11|Ak<l)+Ee+2XU2IaHMb$V49ViHL^V|%@vdY65@lp!Q6FgYq
zHu>IRNjKv^LV%RG#KSi16sIX0H5gtmwwX?<bg4L#(M;8kEYC|+dt`0D9L*@?4M3VL
z%el79!{8=-eyRt86=VoG6ML$Z<@sl$^LNujovWi-_g^y)KiO%n5ziD4nV~8<zUy&9
z@S1Ha$;F*7*v;RdmpH1A;;`HO5DA8&>Y?6re*)PHeYg$Tb>)5R3CZ=xRV2Gz3ZBa3
zuaWSruh>@eOpSJE6<oZFm<Vi{35^domZdE2sP#5RYc!78qY&!HOEUtC8mkO6t1M$D
zG<T{+3(8o_3XxP4a|Gvb&GLjs>PX*^wh%!t*H5~e$Hyi^i5xp)<=d%^{Od-xaPo?1
zT~fT(@^YEc^%ifN7?%OwrL)&Y&6mih={1r&AD^cl%4Z;s=Q!Bk&WZ$9Ukgf?eDFl2
zc{8JBCOtWdnA=jWenXX@S&_eOSTXs?fL8bh-&SZXUxhMkqE(B|Dj8PBRZKC0vnK=V
zru|0}p>=ZRbDqzNdo8v#4A(>c;8Fkk7>wtS3=UkRd->10&-ShYr7o$bcgZw=_ADe}
z>uA0vbNoK`hOT)DQo$dyDSzsv5s9O{ygV_Qc)s#|1wNUII+}(tYxId$zv({GCGYeW
zWwAl%7?I|?Yh3v=wDA2SM3c$P+vmiO@=h<)gdk#>-maR9)RKIUSG$*f>SYDx@-v^?
z<++3kHJW*?=B=k~XtiM0UbiLBc2hNO^d6(Ds^0ZJVOebAy77+%RSdzh`d0dmEqP$f
zDZ4tj!g%??)Oyp4OKe+v*XCpD#<?v8tuA|G$sG?e)QpSIx{RsYbV_L;7MVNE{?d2E
zol$-LL0WC(4nCLI5}cgXT)0Hi=2v_{8=au&JOV=_;#J)`kG#iWb7faI@0z-X{fd-t
zCZTpiLQFR#Skl<e2Kp3dyR4vMj_=;4N$c6QZKCbKX07^PVmDdBU=~v0d*7-WjHr*$
z0MyN>Nnq$I;wBWCfm+wYlQxD6P3nRKx{^4y2bs7Y_KE1Rm9%kwc0KgS7sP9@5A?i~
zH?SgbL{cCpv~*Cb7#A04lQB$ccRi#7d)&g-!@^X}rZI9aY!_~0Z!9wt4>JoBstP(K
zc*L=|ixkrS)jp3Vm>xwn0Uo!z`BzHqGn@J1TBGGlsmViaV-|Bbi|o++CgFma)zc@=
z-PEiU&8&FF-whWd%w1y1ZH~D7=tz~Akt5~w&oBiI1V!TUCVhSrk2OtP_h#716SU6S
z#})K7Xjv`V4fz<fcJKH?%$8SgoI%!nkGfwi(O+-yl|{*sjq+%%brWk|3zM_jzUhSt
z;V8F;NALw;b~zgNrkkE0_C4X{U4n}+v_lAIoQIOS8AwUyepA?9-Uo?0niFeRpH9i#
zdOMe*nctc6X@cM?9(Pd|D%@HQ3GvPFfB4?$I^jjzro}advn*xomT5wdWJ9&<6>C>=
zxsGU^IL#;%K~%BpWO|RA%;P#*_AI}n8cf$-6ki=!?^N8hI*U!)bq*gDErWbyx9+E<
z4F<C#W5gs@9#!8#f2)GEOl`F<1{f+A*?zm(hh{HV9;kwm?$n8sw%@AVcAfn|Y~KIG
zJSIFN>N`c!9&&nb{|m5<_hd<?7)O_26D+D?>K+z(ju)e%8+_T!N=sh4w^YH*6t?i$
zk-XvWQGG&YbFe3sM_od9MkQBub8%)YHG`d}?kv0wE~`-c=79x~2+j4hP_vN3VpP)W
zPH||)*0t`yBD^0QMv{UfV&vW&Uh+362{)y0I%$zg#DzmUox^W0-wK$(BS!dk3MIM?
z^Ix_lmuWP{C5ES$-%>AK-F8aZ8W?IfwK?nWlsV3(TksD_$U9kyH}4OzCkdQj-|Qz?
z6KNj{wVB-}lZv&`?F;YMy~)1pjj%W6!`@@*{n}b`gkv=ad`9}e3d?2i1)liwop~8d
z%~d#RKHe;&+a4ZT1s!cll}S2JT)BY#<u@<<kL*#KCL>*VXorxvdu1E`T?@o$0A;eF
zz8Sc3oB8J-h`Mf0f)tcMDu;5&Ue?(Cn!A}xT8rp`$*h^?x)|$kfh}nO!5$S_KZOww
zR3TcXl?=}{+4(|70kc$i^&rdeN$3l%f)iZA;NA;xhZH6h^XB{=w>+bZpug4dnBY1q
z%fWp-Xa40bsm0ajM0It<8QyFxI&)*b8={&YG}q|Ny5sRS%L^IqyZGbXw`+v@z+=24
zVl$)OF)lxvF>O<(#xAl8ey0}${4qL%PWdy%p>l3IL759sYTsmYWyH5JZNzF!%dvlo
z<BZ}`6<mG{s2jS5wG+j?eN4yE0&o2=c^fjDV>i9ZvfKCx0rHslBFRC;WR^yB^L{Np
zQiac$Y3&(JOU^vw)V^EV%-#;USwJYd@`Rjvu$AR>Sz%sV5c?&eZkHMSzJ*V#>TqdB
zq)q2ojTc5Lhp^5{hLRYQWzz0d-SssRr8rE}t>c&xWC@mTHC<7Hj4Lp*5$ba@zW;Nb
zKXYLd^2Kh<a3MoICm)UVH}GBDlO_|m_zNaq<f&4{YJ;UO5xV=+PC_H?Bjo)1fB00H
zZ8t@`8*tPVEBZ_;c1(BP?0czc(O$69_al4D!JkgM@MGN|o=$l8e}WnY%uFw`pzQ=q
z$>rtaIk{C71U6lmcqqJ2Jn_5m!rO<(Z`pyk<x*IxsHPi-^R9HLuX_Dd@jhu`c3U@Z
zbJ|3y@zqA8vS=Lff<+ef4m|W+UxI&PTYew9s$I)cP8#`oKu=WjKGn_iulwpEs&5)`
z`84Uo&gX8_cHRV;-VVZ7_DKxw##*V--42wukI7)iLi%i<y)0I6X-m69Gh+V6DjF&3
zDvv1V9N4^Ab;xs`QQs4*>GO0xFJa(!<tJ|$kXz(6i}*?Vug;J9-NQz8ibP;S4h)#G
z>t>Rsjj*GZJ`y_z9A61%OojUo1*r(zcA8A;zWgd=@Xh;*ihGl=!>I_1KPJnvDgy;(
zy4`CTx7`k2YvQ#?43aAsz?P_*Ca+Z(bwUU=Y9f3$D<CRyLH$`36HQcle5-N7LO^rL
z&I$RjuZ7<jWt936i2bBi<m2?yQ(H2#0ft!==`$hunQeuXwC9$K4Edlo4Y}^51*z3d
zRSRrfs^{*lrRePnVV*#9ZjIG<OsDrQ-h#3u!jVH(%cHkY4h0(p6;U3NT3?&KpQeu(
z5pX9GRyTaE=FebXjkTub7?J99Twgc1%Rk_+zosDM!-qXgK@VtNO?&F!KBWO;7-4Ve
znjbhCXc5S48=<|tv_ulJR;r538sY2ed@dhRj2nA2F#QKzH6&lBrq(Y!!L6(bW)$R=
zCdNEcT4d(RrYpC+<uq>Qou*$wFPLcr&nd!OCAFD;mFr<o8@96KCou+DF~%;Svd$t;
zO=MM?LbbjH<h9j$X$p(3o*(~?Ak93SS%xWa@>K^yN5#6us@MN6LpD2NvR=DXp=HZ=
zP!#WmcwHB<{&_jRbBiK;r&$vw^i8o$p{z+)dPsiY&aoD}L$<oyd}Meuz|UR(I%1G6
z^SjHsLnT(0?-S(iXj3}t>T5dw62(Wm$o?s;%;Mz(892~OBKGUB%#8|@tUaSiz;S+m
z4_x0hw>p<l{_ifKrv8-J$>QFm!}>czgIU&QTjKfD5wj^x(aVb=@&wBvCmws!no%)D
zlvP{YU71`KVKa1^)aL5L63CTWZKTMmTMFsf($ju)c@=EjIGi^u)VETBK5{B)re%NT
z{a2<5&g%VkFv)I15>DXbOCz7|fWy?xBl~g5#i-A1OMfy;WaC3L<xC7FsY(GzM}vIt
zxoR)ImvZ^OTbo8f<!x{M^_tV@%;is_eDNCtt%L}(>cUsTr1!5Doo-Jn_wk81rY21D
zdb0f-=2o*Rj3e<>J@JH~I(!}9T#D4H9W#FIb0BJW*frCT|8qIPa7SLyg=^*k$D?Tj
z7cjJNmoa$(;Lf)oSeAv6I`8pu-FVJ?eEhUX<JfKmT+FZ>>2=)j9xgl#&Fb<V8=11F
zRZ3|cd|Z06PdRR@O}%TsRNQg-uYD<fEB}JaiIYSb4Q969!iM*?l_iZ#zAt2pBs?%p
z3XE@K<CBd#E51Wt=90?Y7RB+NxocWhTYj49|4uX^u-hl3-bk(LSQt$o;zJs(q`b;7
zzrhG2A+xAi4=6dQ4O^OLggf4cG{eP7Q`?<Ws|)u%H}ebxFIBAf#xlZm8i!Ut8P!C5
zz$fqRG5(J^wHN!}-o>X$9m5H%SvX?1DCHt%<N01Nf+ttX&x_H?OAMM5scC8@6=Z{*
zJ)Tx55-S%{q8N8?a8ubv?(3I4_}+avD-Q2_`22&b-o{F54V89Jn)13U99}vo5#)oF
zv$>I3uUiE6nV-IG<IlMr_(-&Oa$h6ylluBvt+KHr#tmO|WY4HWBi$*tXY!Vq)~TjT
zwI&NqSH<`Qpwuo!8d}_K>$@H5i4N*yW1MyF<@2YAPt-NdV$M8FP7RY)^-lIU_7FJ7
zw@&e#fILGGrfOVXZTOc%aMETfjL_vT>0Xeik1G&JQjck5{6Gk+spS*)Io!yM$Z%l9
z>jhwMRS#+TSZ_HCJPqrcPcE)~QCsq?g310|7b(vOe>8_pmwwITGl^drfnCggug81?
zwzu3fs^K{fDzLq^D5vgs@~XQEo3);w1mj9Ls!B4tLi`e(DqD2I56>qmvjJve6dyBJ
ziVUiV>k;|l^r|NX{fQ%mHBbB4drr8USh@P1KH)$y`kk}1LM^gXQZ-44sJ3&-i>=$_
zvF6L^l)wKD$6h{swk0eI_K-?f%&VqEySEq(ujU&q)|p{#hZZfilNc8AhnKcwR6MAB
z-5q>EqH05fu2q2=vfz#nu+LB{d(kDf^Olqkn-o7(u-)zwdg5%0z%JnHdyb#dk+l{t
zczhH>j`sIEF;Dp%E|Tp=F73fime^)ds2bQPBk}=(*o5;Y52_MbN3e1M;t!i(30=?C
z@YUgV@{4(I1JkZc*dnnl8E0*&-o<lU{~E<jPSw^Wq}Dp(8kAwIUe78MHa$}|l9U!<
z4;S+dv0A=yi2?I4fz0A3-5}nME}=#RggHf>>AG3P_iwtjLI;9W0*gw-1^dJgekN<0
zt+$%=K*esA+?~GZjXs>ki#a?P@{3jWV@KN+*R(CVTK{iLErAiWk<zlpD<Id+j=)r$
zAOZd9_(>3XIvW$iQ#91TDEpLU<WpBA&tS9UgCrCK`?d0Y(=GaVipEwU-JLbc@-y}m
zaTmJHy~MFYQ$_KI_aaJHI3{8@+2UBwYs3J<Db{b3Be6<93!zAx1p=?h`5otgTb|ZC
zFjb@(ZER&=o{`?OD)|g7g3=5ND4k<a7r2{L7S1T*6CR#zq;*z7z{O{$c%EFZySh4n
zk-P@Br=K|f0qzHAu7@<2S5U0|b$NJdUT}059l<*z2f;hngiTw9ThkaV$M^6A8@qn`
zgAm04=PZuC^xw^#z_;iA_Z@fOjsL$r9-YhqJAHqcN}r|Q?F#sPb$?r*j=HT?G^@sw
zGi3#CH+-HO@SnR9K`~G7fms;*-}k>P!Eyhs)Ta-BEbaYdHc0S^xQ9S2lpW=l&BuPO
zb?d~r-Tv5hxFv3`HxJVL`SGOv)*8)^TpP0LQ9~X1pd0<GH_tl4V{?Vt`>apyb!bBJ
zw{PE~KXlDLw?|Y_$6pm5eKJn2h>jHSZ62RebS@97L@rci2Q<YBM)nlgv+9S3hu!k5
zi*{~L{)o{FTo?}dB10JVT}n~%%!S(>44m$Jxk%jIp>W6Wpefe-tZV(5O!>aUiK*r_
z508=&T?ni28uOaNf3tL+@ZU<jff)YGneNOOo9urO7ryyr--%)L=feDa>1^<ovVBcL
zBB#QmH@DKKcKJt$pn71~{i)_R-svmCnxSjmqN1V|TTBMh>ZjVvdvPoGDN3(vdDUj$
z8GZy;Q?poK6L4<xMUFh>cymH>;El7<4F5X*J4gazq<T2ZI2E?bBJOf3J0ZK<aTk@n
z`m@W+%k7|0Xu4&dmEd*1>9%e}{pNJ)V1=j3m51kRq~sV$_7@Qm%>5GTZJjTREq<uI
zU-9+iS8yjhRFP5{H0G&M-&@O*`<h-v-1!Sy%^F2!-}LUT4`=6;*yI;+Y=|h1Y8+%|
zV<~ifE?hXe@rue=*lG-9dr7Ai$Df@ZI5-#19GD^5mbs>H^;$diw&Dx1Js%|{Wn?P$
zLBWoxVQH&_$|ZByV57yihl|yL47K2aQuvFmhyL`8-R&9nt5>gN`>O*yi3RoqCVPjy
z*eR%j7(T;2Gv7L$;&f7EcjP?d?4N1&HE*Fcx;3^{U5LH?BPYiX+!cL&Q3XLfBV{Ep
zT<JZ;I(qb|3ScM<#yn{wLR(v#7Jpf%K0_^F4kD5RriFz=F1XJm?CkC=?&2``%Q`R9
zQlQ}q1$rTGArU(t!PPDrtBkC{b^G?+x(*xC6~~qbIfbuV7HqE0u;TGuAN9sG0_Xd&
zjKPYbwB%%d5{VS8<UQz5-<-)HZc~$;P^-~$^}#(pJNiCB9t}a+;kvHFgN@e*Wd_@G
zV=DK5DU>FBkm7mgJ5=s&zLw%ys5%AyPfPG{|3gJPCV|LdtahZU7qL<Udp_NB(<e9b
zJ8j>IFI5km|6KC!-QSHbk6&bN&gB&`F^b~7fekN@v!}P{MRGVL&%PYc*|`2<=LT=>
zR#LeTk7QAzl=bw8f46ZYXw;HBSs_#^)tks1^n@7K$Mi=V4K*4K)<~FVDJJ3dMa%+Q
zM3bctn`u39Z2I=%sORY8Cc((0gansuPN_+{8gayzB$J^GPW0W!ju3d`bzpQLAGo6m
z%)q$E#?r4yO=kZC>z9dRO~Mh|%M*>!i^~{SYHMPMQv@eq2Ap0wcxil5p+LvP#H7vL
z5e9?hgB7;efzzz%Ob;3zK&;JnWx(xV@u6W$Uk|NwyH{v+rW4{dQXRmarc%!=62?$Y
zyaG3u7$>jZ2}`UEcXxN6s-_ou?vf4V0-Q?jH|`;FnD*Kz)J`?G7gI;;kP$b1r<$t{
z95|2}<$mPoQMs*&Xu0Z`b2ldXrf|n#tH`N0dMQq!v`oU#WUQlKyVBL7I_BUQQclII
z|Ix2kXQU7%yK=I!9U)*u_H`?*QjrmybqWK#`}3N(cy_+_vxA1Q>I<Jw2!+nRr?spn
zw0-9%CCNyI&6hM4(ne}nxuCrSAZ{IL9rV6}p(5Az6r%s^Pv;-zrkd+^;K*o(B8q-6
zSjnY19%6#i*xW8&vH$$&*A!T3!L=CxzoJ3+*%J^k24>P{NfKZvIgT|<lf@x_kDgCA
z0*5NSE5MqZBnkxfJ72_5iX5usi=Y^>NzoiU-sP6!1x{W2G4eDt%Ue*@*M;!m!v`Ho
zq2rLWxVRe<bPkc$^|9&D$2&1w%@Q_~1Y;AgvRKvWw;8)z6e3eXQc~?cI4I`mkt0@t
zG%|HWJR@v{0qWx!xw2hukMIuu@Zu<c-<n;$H7ZUoD~+kt?52|&zU*frk~C1^Ny&K0
zD>z7B?Sw$Wz*=GV7CG&*0eFwa({{F0>-io1yY6v6NH)*5Y+4S0Njo*C!<oHN;WgmP
z1VF!FTp@?*=(gz4WDS?iv|+o&yTb_ub`>7mo<()a1EwkdjQi&pXBlD)S4(%5J@Y6)
zD32;Yh4(@|Y{i}QdW!gNX7m!hyL&9@<sqxCV_1#NDU6J0j(c}j#>~u2a^{d$VwK-a
z-vC%JTIe1y;unY^=4{u4h53OpHbCr)j^I|H&>X$7`f#L4+RbITDAKFw=sY>Q@=vP(
zB6>o$H<pDXmK*t5gH0dG%N5B0mZg%VZOIZh+Nax7S+LsR<<V$<IW3Q`PY>{ohSCd?
z`6%kffBX_BmX^wkPM-uQT4E!;yep{XwgoVbzD{uxPH>oBp^}C;?6F-03pDn6x+aoo
zc}URz=QO_h4ym>EgL#&b>Jk^d3P(!eBpkCk=Aw%7)>6ZXPL0`j&tfxFd`2?^P4$TZ
z`)ciosPKP8&Qy+=zc|!)G_8{W@#=YP8bSUMb5RQKnDCSk<XG)*>%BWt<V56iLNY3#
zQUC0hyI*31sD`@Bdw)Dd$p{>DuLJPcdlX%oSMN1Y`t9-Lk27b_oT=Da>o(0WknQwI
z^`g`OT-^egwDSvXD68rT+L9PYsRB%u1=wWk9fvhW?0kJ*q0po@X4tfb3O5rbQjsU8
zf?ZnOot}j$iyZMa5vi}#iR9JH!{hN*->P3jtha^6ZanzW#=Wo;j>Ug}EdfJ4Umd(W
zf#mCk`L>9r1I7rQnwcp=?s8bdfrG%tVcJtw`1Q@cY9G2ryw~uBO$?#22FurwfKp`F
zxQJ!=VdON!HSr<0UCuq@e<JS4EX+G}!bfo9#{8G?eaV`DH<Qpkxt4jlsP41;{4=kG
zA@$HsoTg7a9&}zo)qE-GY)z)>#J^Xqpk#8U;n8P3Dwxn?7nSFWqxi&kGQmNqa&>Fu
z8h&Y;>krvX2B)vcDWap{1SBk=E_5AbQ|i)kGJI=Ib;QJv;uut6Tj~PM1`6R!+y$k%
zA8=AqdQCW+i2{74;Rb|{j_l|0r-v>mJ}mC@bmUB_0PM&F1my$3#0^UJCX6UwdF)#S
zKz{V?+aS)?x(W^eUz}+QFmRxqd5c*LT6RW#EkJ}-WuMVn7F0alKgxUr1`zLS0KLyq
z)%n|VXZuPo8WH>p)v3jv-#v5sbUs(N_$^e;;Jkh41jG}kY)(mzP|f<MFLBy#l}8)`
z#TfTLykm^qXcAcUEOTww!r)a#gC~c<#cKl0WtH8D6&L{8s_JXR@wH3(sVV^7J~A#f
za*rQljYaHULEit!6)xWlL9G@*GwvohSj8)=c1^WlI*aao{P7ZSNZ3Y_jQy6+KxyVK
zK+6EC5OTgGY<0RF$(kQn-6E68W8(Xd)QqgyqtviMabC--1AiY4HR%^K#f71?wZG<q
zK6NYsPkx}C$1);zHdBn4S7(>gK~p^d#es3!RmKFA9*uvQoTvKKZ@O~ygu~a4Kmj7{
z1VneoA9BWt$Ls{B*_{wPwQ+B^xnZ9$5g?AuZ^W4{{0YRz`cLHHTbV3yp7I3zKHi&u
zhg5rmt6GKJ`Pj8UUL=n1?&$QP*bEh))&vPwW{v2=Pn{Eva!?@r0;~!hu>3B^K3`e~
ze66$gMo;Lx7?itM#!PSidhRqIz~wA{r?6!*#zc*_bymhsV{;U_xb6tL$n*F`<zZ#d
zzWlRj|GAuOEx+|%>j>{`zzl)zGac!aARue2GpY}@$eoHxfSMr?g`S<yvT>nv4O4LL
zX=i`}igQa3=o3UP*0G)$Cd+Dif!PXB0}`XVAQH3!=W*8uU4ocU&7OS+^ZS9Qf*d-2
zK`|5@)Rdf*q?O)(^@r!klics7HM3$ECZaFj_M>{xLf9r)5hnm$dyW8%T~Al`PPy~?
z+<Tzpedc>|y6u6?gX~gr%(IVUdv54f)en9GA4SyysH`JOt6l9tJ_3NMUas@v#fwy-
z;PHP<(WJwJV&c;$W9|^zu?jOO%HD(9TDQNv(|(o|z!+Sw?klj5b1ZW;CIL}KxS$ob
z!3)`V(Tg}B14+Cdt5><f@%kgwL?jnhx3zWP=-JnQgw_O>8pZm65>^tIz8xj3(Yj}T
zajW6OrM7Q;rW(uHWu02c+kk6nl%lb>+{=F?sou7VzY^lk-|!G<Wp|aJ7lC|V3L}MT
zc_ixqpDoHkW$emrX5dKn*`n|7JYcZV=|f=nPA!C;Y;^E@n1#jZ&w4M$LN(TY>J$)s
z#$a{Esmcnc<aqdg%f^2W3Ek8@8?7s5lGcA4eL=}%El&iT@RRP(YOy2gkfN2<RT7w)
zn~rEP<CGqeUS+q8aOQW5JzIVqsxx{}Ag~LVlW~w~I3ep+<LZ#CS_{6;bNKV)7PP40
zLfACcO?Hu-Niqa2zf&hp0VSh{ZTyOK6AGDn^MK30LRhFEX4Cs4x8nO+TU&iWV^uL*
z#52~mpIQG@R=o#W%E|E9ss0odpB;PF!iW<Z&uz?yMFA^PZ7U5og$5>AWtV4F-+T&M
z9F=={W;jtEQV(DNrU2GGTf}{7{Hux}YU7s<(DVJDb)(OgfeRQ28fh(@T~(9@hJY)L
zjor#q)pOezVow}TXa?`G2n8HMO2z|Umc_Z!%e5h3>0i8xUH&oMk#3I!i$=UNO?9<g
zp;KdZlrS3V`7lmM(;-qyf1YFO;2`iR{g_$YK$)9LGQKgIUpPHA75O$__}BfCx8WCU
zZoi5vs02i^g-#lM)Nnil@F&9W-o1NqT%}<N(0**bJG=9nxrr*=lG`in0%ZTDC~~!3
zg+3O}8gZNfhHshT6yaS8u=)a14LO!1>salUTMimm%3P)s7y{iZZ%S>Xlgi+vvy_tF
zCyxJQ1a=vpV~l+HXJ3TC%}3)AA{;laHpYk<duRYjEeo)x3krwBy8t{nbd|Yvk~-Ar
z`Igw8-f2S2))dB+nT&^Q|2k2&{W|D;xv@#xN1zIwZc-zLA2)@sGlzUUY2-}r+P;>e
z>!cHt!7kBboiE?b-2U(}R!y=e{OIIH?0E&(KET0hVVl(CCT@fIafLFiI~x*@5SH4~
zI3Ar5cmo4OefN3BbgZNwFu4>RaLZdF3|QD#qV7Pg_L99*F_=}hw{6!$5au{XdZ&Tm
z`gspka*BZh2>SA5PjdD}6{*MJC=}}56W93A?G=(7Qh)kG5pbuWK(AK--1Y#MS=Dzn
z<>rm&>Gc38>>&UF4AleX6ltKSBR&(~qaw~LKKwZgren1V2s%d<AmS@gb3%crsDr!$
z!45^RI-zv60FT`a)E4at8jZGNqC^??t(Uyl=XwQwYNBsn`KKPg&+^TGs_ys!P)zc%
z^YT+{Py`J&z)_A2kna}gh0l|*qu6Qh)uB5Bm<dJ&1tx(sRpo*Xkk@Y}9j~(I%Q{`5
zlCrXe^lTXFxh?8k(^dOzl=6N}@l(A8_CcOM%X6LwgaP3AZp$1?R}aEXz&D<}(lTv0
zRqnVp1o8s4k0ZnP;0T~NLTA4o{en3>R?wG6qU<*z7Dxxr7XSwV$#){6#zYO1_Tvgs
zD^o2G>?=G1)=VKB|1W^y`vSlJd(uhOyT+H=I_5L!HCE3tnf&7Vtw0E@=uv5Gl(g2y
z__G<#l)a73XBq$kRA_gWai>X&V$Ro|9P?BnaPoR#Yz9Fu_S}sR4N^t5^3-lqNO<Xo
z6%76aWcyLvZvaI5|2Xfa@fe8N2s3b^5?Ed65M%W6NKFvLbFkb!oHKRQB0;^Wjl!7}
zNJW<$7(6|}ckWKjH8`P8efsSKl;%{tPF5RMA-j{y0ltIW5QNG<2wFoG)wSMz2gnMq
z1U&Gl_*#W8P%-aPMs@}&ms*YhGo~ae0f9t>$Jdf-wgGO#Pp+luCt&yx5bX8X#o;PH
z`fDN0Ed%y$&}hg^8b2)f*GGT{S@XZVj;?BKPM3&ROhrW?3sy{8E4OgW@gQKDVE=<C
zgNY#7NE8N$@>y-A`6@B`&c`8x*85u(F39nwX@={=S|HTfxMeORM~l!v_`$0oYYCtx
zf<9P5j}WDSAObSxJp^W>yikPeo{!0Ywfh;$^;?FbIiEgVTl(w?CPJ1mM%-;JdC=71
z07QQsH$&2Wqc6f7aLZUvQE-c0Tz<}(PJTohFu{dPV=K`;?{y*xJ3!;t4}8>%s{qQI
zQu*C}x1POFGmWCpCfbkiK!9H*M%})WN%WTKExKs17gD)h>4?4$O1iH3_})3jhlCNP
z2^z2~K^qMMFq6rTgLMql&gRTq4jBbL8Hp#28D4}F3Y<`z)HkS5y%$|#U|j6jAPu8b
zWyUS2M_>Mciph}6AOdcO`taY%H)k2e`8<C7xMLY;Fs{X32d$v#*QLuJlY~Di`}yzh
z;1eB@<u4!O1Rj0a6D`u%TAB2aoz?!2Fnp*|0w@j;jx5^~n=a)Cb`^}EAF7P}L@YeC
z_E6blfm#Du!Cb3Uk%PwHU!EKVi#PkVcww$pfn6I(#H}S(ZSIpOM%=UyFqfU1*$%J`
zyF?~qhy{d=_bQNQ{<|O@@fph;L@?(Ej*Gej;V;*U*C4K+0|D5#As=Em8~7+z9Z>1K
zrYg?}{7nLWZAF`#o1O%(0kd5&yRtXuiyP$U+$R#8VlF5FR0Hi7TPFl;-KvO$ii!%g
zKNt*P6bZmI(#?!$@gwD0%>UdAP}_5SY4Uxf(AHFLUg3pu;BH>O)QfN9X8@hLpole8
zbFPRU@LB{S<uqsZV_Yl-4=#+MnC1NJ;BhWN_!n_$^NUva)^=L$PF5;Ht=9*eGx!OJ
zVE?WH!l+$!$m(>RY}FL{Dp%{q)lQwcle+!&kZ{@-@DH`4<?h{rh#9B)u#k1`N-%<|
zan`!sM6-|PjU2H<Ne*nh&b_AH8RgZwwh_2E8vX@vryYPVEhm5lwDpII7gE7Tm&*~i
zS3ZN$9CPG0s)>8UO;Z@?4;6kHyESbv?>{ETDS6gD_P&m_e|&T_q9CB<sfdRvLFG~v
zsXUU9(51clc|cW*az&ba`2NES2ILp~u#Ex{z>GnH(b3WS2=?4)Y?#~OqBOuv!n5Tg
z>Wj7A_s}3ZRJ!`mR(r050z-OV)CgXBhF%+gc7k_tplk*}bK56RR@yZXa$q%)fnO7?
zVFM+=5=9WY*MU&OqkzE)rY2fnC<K84>q$2^qhc<MuoGnLD|_N@Ux5RoybTrF-%<o3
z&@n?YfI9MOq%cZ<AP<jZEf|W>0Pbln4|*2KI1mtyg(3y8y6VBDUk&Ow(1}4IIT=o{
zxq}<Mpw0yVL$sgPm<2QL|70S7DmCVYN7Zw=d1^+0cud(GX(*6MIm)2dUA+g(&vT8&
z9kk;8@Bd1)gsg!<g+oJu^xDw~U7NK6djDk0uq*2ONxd{{3yU5F%K_BTCvK231oVI~
z2LGG$Yl0Tdx!R@)VGrgaA)?pPI^gDNn}VsnwD`l+v`<_HjpP+Xa5^J)x<7-s!OYI?
zLddTd{1^Dop51u>yyhloX7&!(hO4PM+Ayqsu9&w>1-GTD6a^|?NdGhDj(9CF>BpZg
znA$wW@zI8ZCu11%obpaLNWhWMFm6U2$dw@yAt52D*~Y1X$>DBopnf+iLHd`+4U<1|
z**DF6K)VBD!5#!@$Heu)#!FuvCBka}7l?Kt49ZifAeMdN1Qz;Tni3E;c&^q3?xkll
zCmZJx2>9g?&XJpJ%7MYs<4DPJa$(###etk5kB^d^kZ>Cm0huOpTv#g%l@-9UkK=m5
zzUlzF2XSK}S|`gSeK2CKCx>I6Ne~~8$Ah_MaJVpw0;Lcj)qu6XfB^Dv&^GhgbWr0N
z>##MDq8x`+`_r7;Q<T_2BB!Fb>I}rFvJN0KltBZj@eUpYbku90O#_6KJbKnaBsO@t
z^G?hKnTI`}x>kON?65gnVsiB!OCaO!i!mHQ)L#DS(<eQyZUkZd(Bm3z#2L1+q#nvz
zsO!G8z;!z_wZLYKfqu;b*9Z$?gD31;**;>#Zqv>xh`p6m9pjbffJa9~^yXS3>p9G!
z86&Jb)f8kro94bLI=re=5iO*r<t<UMAm?-A%1;F#`_tH+;7PEL%8frJV^k2vN;=`c
zXFT1-S7Scvxg_b4s&?mL6(F??(yq4@U1|yfka>W6l}c#;PEFSKM=&wO-G>%HOEI|D
z2LW-z>XILLi-0lJ3US`u;RIcXQ)v!C)9PJys{fPs`stOY(%_*QJYS>$5PUjS2%0!q
zga+IzEfK0Ep7!XQ9Ef0cEzpHE3)N*93Itdl8G`qaRi1ga2_{9+{nXsC0T}(p<B9JC
zuIl?LjtidX21sldpPE{u1T<MWz)T3iqt~M^WG0<8o~G)zbKhsTLw#|Z{qiE>TuAff
zNTY!VahgdpQ&V1XfX$i1ew~Z4AbZ}eU+=Sz08_!%TDK!y+6(|LnDhKF-^&W`Atnca
z;m#?Lw{a!5jdW$hi71dQNBj-E2}|YS&vANMS{fPb4NZU<@EHXLQ^)}%H8Q{-2Q6W@
z1%eAM-tBc6lyxPKItKrG0WsnxEZkcC0^Ycu*d&CMCbhSZT!^}TA4r_-OaSD5Tdc+1
zSrCKEeO~QQpUG^!3V{&k0Y}ySD(^NyegLcr>33K6pIzz$sfXNoup_i)L3>&WegSH|
zj|6-{^%NixRHI~3J~vbX;5xR+VY4FExH}B8mLaI2soPu+KLfmi&>%3?6nalinLOZM
zZh|x0n@3R?1KC5So=}KOYgFhr55m)=MX}~(n`(b|8R#oeIWFB+ph#^=z$Ykgs@69)
zHmU;KR{tI#jU8x<9Ze87rK4uKId%%jv4_Q>&kmiamw5d2y1jA-=G8=>?8`2oaLI@y
z>zcznD9cw)KlePp3O1@v7Y`@SgPBk%uql}s2dVQ>unook@N=X;XaQfKQFr<m05M%0
z27>9X+ns|?is58g@+jOCDAyvVN-x`A;25p2=J?BF*8bs=gJ5SRVx9}$YrqHzu>gYC
z-RoNarz8)TJx~k7kF?<&*TcRSil2%y%IEG+{(Q#Wn**jnT$Bb32un!}XU)x4M!_0-
zPy^7+c;bWLEz3X49r1NKAt|bU)6S}@szXh$&##*3QV!JAdpXLH5qo(g6@G7t9hIwn
zGyr&bdz8D-gMNK`q14b30WgMtdU_g>e)I9fs`Jcoo`L_KPeMW>O4_z`su)bM@k8Lm
zp}Leq|HnCh2=Bj}83vFrPCr!SVu&0#w8LxaD9^xY{O|F!0RNbnl*IdKHuP^6I)<(b
zB<wBt9qunJ|9*V>@YbbZE+=s*$Vv?RS8qk0Y3c7jCUQ&H8T?%rbdMF84~<GXjZriu
zSD=|oZ}g%PmCuH9x37OcB#v8aTu<G;_NHg-!T)0K&7+#iwti8RDk)2lqaH;>psGYD
zR6wMYUW>Ac3aCILAdMm*V2CuK?-WZ)LD2+62oM|Egb)N0BqRh21?i;;O&|fKlRy%>
zB#;pDc0$!Tb?<m@jQhrW-??Mld-(&%-fOQl*IIM_X1iBx{{5Qb{=5UzW5Hj>L4}$s
z+5b8=CIiVrnCfa|c<9eWfp`bm2WGB<f_C;(e-Qj&@UL7hkP11h@x()~)Zrm?@u2;=
zK~A)}LiE;o>De;C&k6JBa|=mxGy21)v>3<=cGWrciG%-oCvtLP+Xwkdxb>~Xtcd+c
zGemX%-v;k^sBi<BOep<T%li3~sqTXCf76?L>j_m3U&hXNI2#H3#SPe~jh~ASdvrCP
zbSyp9KhuNV8o;f1w(DQ#;e;+e{IY3?5?|DjX{`67*d9Sl{r6z0YP*#`YV7<W9u7hu
zoOrtOXMYuHZ9pn~Zi@|QrOqyC+5MpUfq;<XG~ks2*-sSh%7R8sm;Vz-4^Z2$|JtKx
zBM-|Pg1s{fHuaAlRGk1`S~&ws?0S3p&!_(SQ*vDTE-B{TVJ$|%?WS9g&tbG@tU$Yd
z`KT;w(ExJpG0v`g6{&+g2uD_v@6O%{ZTN@M{HywcyHA#^q*ag7b|)g(2aKY{oFZbu
zfnSp%YfdDocwRM_iTJHIIWnUl^`PF?A2*HxN~&JumlfM-lCiI_m-OrFxjS&;8^GB@
z@{_;I!DwLxU+jX~4>gCHrDQm2@9Nl?;>w-}zy7lWf2o`Hv&z$()vvHgp{J8-SgR-e
z6x|;fW^6P{Vz}+C8Ic``0?>0v?q94SOY^TJ@;TO=p4fWe&5~otuwTCW$)VKA_`zVf
zk5dj1zM>(0u(vQmr+*f1w_eN8_}QiVvMHmG_Nc@@6L$`7hAG4pJTl!HqK(l0>Bs5p
z_V#rjoRy!d(RhA#5M#EU!;7=gn0-?}K&T&>UZ1FkAhHEQ=BV_Q+oY$J3fwhy%iGp3
zt)%O8acdN$QH37jWGwxZ^zMq1v`$3s=w<MHd3hKHDsmlv|KmQ{jG7!9EEA`lbsU>V
zT|5YAo~eHua1ayV3Ah`V-S72ps=4aCOZvG%cwj7#n%<t~c;$+U>;g%8!LdD!ORCV9
zUPywU0lGq?Rne6t+OiA0X8Zmrt{Sbecsxj<B=V!2^e`r%-HC=e3#NrP4l3U-lbqA-
zdz7Si<PIF%;NZQwn`HagaCYm)#-Bd7Ax1|pZ<c{W($k$ALYJ12B%{W=;F}G(_%G=K
z>J~1a_Q!ozk<!wgh8^;KGGcemR~{<{JF2~%Vk?x(*6jS5hPHFL9`U90<0o*lkQ7Lk
z(nZ-D(a&;{vNEGOmJRPYVT&Ouiq0~wfA$PjCfkGm%q0HDm(o;^w26LE{jxe3T1Esz
zz!1o=%AjnSE1Ce#8okddrzyKAmvX98YSAT3PfD@yCIQM}$R_jp$1x8N<#JCK#`WKp
z9P@1wC>GS1KpwGdPF6+)Flyk?MM`~)f1Prq0aicPhTHmv=8o-M43kOK&VQCaQvs*v
znb*IF@q;(Be%H25nRRKtfL4B<Z(AA@gQV3BL`U&@Y9plwkI0opZ)UF6Me|0o$EVF)
zKZq(>wwS}hemxbUV;9;*ydJ)eZzr#((n6L3WE0rCZ4-CRI>B#^#rl@kPbqtYi@5}6
zwb$y*`FlzGYxuY|YMOh;Xnd63<z_&)2kNANgAle#FFrOJ-^Q=%KUG&M+tVXI0zE<V
zRbeAtZdJB;te|R-wC&MQfFO}5djTYJoxPX6pz!zWvD57<aN0@HmFb2VULEG&{ZLBP
zhjm8^g=cwrv99AQEIS&bxY-?Jt5#qsBcK}Ck~;@)!+CbJQTuf6)jW^*@)#dpmwj-3
zL;!Lu!U^sS@72D;aRk>4mrf;P4^K0s<&<(2L}0&~XK2Yex3ISvkNSf>#B|U8A+YtU
z7_N8T#(w#x-lr`=XpgX?Wv1Lof;(ok^l|CBpdns%`+)DaTo$Py#uNEOu)<=}Kr|@U
z9h_o3)bp9l8e!$KG>f`o)xMYQ-E3}Bq#@B+Vi=$H?j_Ay&F_N%SJb7pr9`&#=j{RM
zAQOc%cYuP6u^1V7;op$U-Ura^=^@E)#UXJ>_RsakC-kJZt^{uO1b<=lHvhu`m+$Fe
zanV|K)on9#q4ta=A1|K&NlixJp7RZsHi=&9$JOgDsY=@%c_Y7vA^k^APIUUehmHS(
zlmGQNsfSfCYggDT@fFTdWVk2-eQY{Mu8vu0FyS7*y|IxN((OG+`<B+UDUeaua){};
z{cl*gGAPAL)Hh^^(tiO~_Fde>MPcfcugG9H3FcFBg8!0s2D60qDhWP*Q6}?p5k6mc
z*O-;AzaYJhIRo!9kK0gD`}E@|@jMhQ(`Y)OT6htDjaSY3dZ%6{`*KPu^9e6wJmF$i
z0!`%aX-SHS=FwOZXMx82tW47#!M#lG55h{^#V3&Lefiy5gOtu>jIu+Qbror$!An!h
zwB7;jVF%f#Xr(km8Ut(AMUF^j71|SFE3>Rv?4z&^81-hjKEmxJx@h7OFeFdjAgjs<
zZ~MZ6B6Q>BvWl>uHgxt^9N+V8mP?`=qa5_fayEv>b)JSuF6Lc^Ug?;UjwRQ0gCNh8
zAAWOP;K)D9^`t3a9Gz?OV<EDe2ddpRaiKY7ofL9QeqC`pnV(J*W0SE+%s>Y@;xarK
zKF6xwDs6gHV@WYe8K-ec9Lvkr7R_oH$Fo}jgw?~uLK2;z<((J}FLg+sd0t(Jg$#$2
zRP7YVY8XqG#J(E@PJli5v*)xOO&L6Z&-co!L0Nc}CES==X;$FKCgQzps^mOfqK$<o
z;#x2xt`LX;&eJ_Uz#uRo#4ql(m?b-h$y<Hjp)pEDi&CFod_hb9_gp-2&aQ(k8vdu0
z#D*AC=Z4t44twxyeG$y_P96SMhxEE!g__;BWKEps6$(Dy9am?`-CH5s%?8pj-~s#{
z)<&ye6poN&fFsB6xOh0q4D6Ww*Z{X4{@Z{xhSVLtmO>-Q^Qt`e_e}VQp9;2+G5*|*
zlHKG{7fPJ)?FY}Rk?rkp3G{_$f8}e`$Z?Je_DuDGk$l3Nm|u94EDWns`<=iGQe5VM
zk`CCgS8?~eO_bkSKIJWkVl=L)T9QRzsnQ_W8i)bb0jrP6w=kN10ceUKEUGKl_C?GI
z-W`_GkedvNZ@zsoZ(}s>$o&9Eqg3(JY6R<2A%#Kai`d`QNGeF8no{0Z*0=IJ88W90
z0s^7#*94jS0Ecv>IqnCBn);o!+o;*=W@Vs2g}&pyBmj<L`4=~9mIaUAsJs6N&;r!F
z(uE*<VytX=o=Lm7WHb(q4PneZYAm<CXb{wqP8=7lJz%Y9-{Brc-J@#;EQ({B+jB*x
zTN7*SHPO(@Q;p4jta5>#4DqFMm?Zckd*L0>2cUsrihXTvORtT_i=9#ak86xwTioG`
z%d4gK!5Cr{&L}Lhd_#Y}w3nx?-sV-ztnlr@FU#NK{F@}k(&0V)n*8U!@bah-O@<wA
zXqH<!?Aj>8(-L9NmA~Ow$J>M_xgBBJVYu$$yRdUT+OdINm0wTGfyX|vTxs@u7#|Ob
zvRICL7~fp9U*Epa&U$;wtg_X{i9&|0<6PvPr@q|Gqz0U65S6hE50g3BM&kk{fa0*!
z?1jm?C^0MVF@6LXw$^NP`Z1qbg-ubuy0m>X;rj#g{(Amb5<PjWQ)ZhgHroz{GIQav
zNH$*IBmBUL$WqVBuiASiV3<U&@Lh6ci2!rLu%4Fcwj>Lj5DN?Ih{vK)<_SwDC}k$@
z7`vizrTAH$w`vg~<T}S_|LI_cq#rz&O`EX~UI7l}c-8XIG4YeD=60MguWn?@o%wOL
z+mad=B8loA2r{>WdgpG(k=j6JbKSNP!>0t5hw>v?svZfG1T}gcv@vPbyqfjICwQbo
zbImT5<LeT)s?O!);61Odw}01gNcGT|*bTr!iLG@)B+rs|9-A7O6WgdQYF`(&*N$TK
z_evSb$)0$;#DYY+Sv(L~UprDhH5bS+7md;%#@BuZA7&Ck4RzS~`>LgA3ggZB9rFq6
z53%@dQJYRr@|!*II_rI*Ym4ovp5h1k)E-SEZ=CtpEoNT_71kSG4PV(US&h^r>jUqP
zx)}D^_XqBrUtelp2*cvRUZu=ltz~g2vLVuRM0|zUCeMG<9~b3E#-^8XYkC*b=U(<F
z`!z|(!>%%EyS(?<7~^uhBaf8VsX2T@Qlr4fiJI!`NdwDt9t&zT4uR-&?AIK9%pH(C
zw~?HzV|odm*GKPW52pnFL)ydM;veOim=4-Fb{FcKOvv2z&fUEO4kZaiB)>6@{v0YS
z)UF8Am@{WB8UoBFB#oC-))apl$sN7cJAig#K!s6$yj9l6fV}fq0sfpoMpTYTp9a~|
zvMz?c91?~6M-@wrZL#dpMKp1x#GSU>%yaJ|*ArrXg}VUwZ%;dE&p*n{J4398Sj#Zr
z@X_V(nI5&TEE0uRh_?k6{-&;P*4}vJ+P-OFjbO|m)an!q6l$(f#JnStJ#hM_=i#VT
zFdD;gQ+Huh2xE|Ho$NUQrWbPtA_%VBESzI4vpnp<LYXT2E=pyrXU4~|GpNg5=4Q&3
zwf^bGwZ9{S=QZ3J^`ukqkDgZrq}6KK02S+s>hsK+7<+hKJlS)@G^_Z<F1UNXj0GDO
z+}_X(l%sv$Gsm%6?fbp1h`~GqnlFuir(TBea!!CWIOeM+7FZP0tu8{Pt0Wh3twe91
z#!;j7?0a>x3t?seZjZFDILXt#kzVjtzW;p`^_UUU;-bB*0pA+{nmnWN16wdM?bV<G
zU~N61Nm@^s$RK_&8IbM7()Ak_>uB0Fd72C*z0H7bunEMNSa>gMoYo@WFAGL_>!7w^
zmYxOR8jG7FOFGN+OV?{j*MEa+9J^>M8*P>eKw~;Yf%9ce38SF6EOUb8o*Pd5xmsnf
zUIMFSeo<Z48hjI=E$3fKhj>$V_1_M$65&*1r$d<rI35NMg@0XM+hG*;7`r{^v8@Lb
z?X(g7nC<zriQ7#>VMKUu4h3W8&}@5|5;V3ER_iDHSab6&2$}h|>NBS>_PJ!5_Mvy-
zFh)t9FWc7&j?(V=;G3131tl{Ju~?LgSqX7&Dklg#n)^5%%dxq*F~<380G%4_?{0RL
zXgKFl@6IxfjMf+K#K_j~%J;xNrDv<!7Sc(7%+?zl)eBBgWD_~^8?dsX_O!0S83H@@
zEeDgmkf*@fSYKUW7ZRG=S6K07As^-9H`+e}(7l$dzKPwWZFbT9L9&07*WMD9`+`PV
z@4hMdJHw9XbozK1;qyJ#2ioV<;2W8g!Wi}|ly`ttt?<Dqvp22!_=Hl6$Zx~aaeyiJ
z`KqN`?0UHn2S6`iwQP}U{&Ec~bxjZPbbHV+|26~a-=w)uI>7HNkwSTN+nw2t#@F<P
zdQWW~?J-~8+UYhGPjGx$i>;_;X^izUjN_|Eg7NBTCaJ*yvw7px4vocoS>CNs0OP$r
z!yTWiQe$6u?qznlB#zvG^Vfx|tRHDq<%>Z~Q?NCwzj^tSJL6_IrJYfTs^Py?MmsgG
z<z6c^1&i^qd7nO`?$b?fvMtz4D!gX~SFaqF2yXe%<R4S6Ia;j`*769?tme2~EINA_
zYkGtPNKcaCUSRn1yoWO@zBG5#+V_F%gm1YIXubYZ-$f0SV==u77+z!Vv^66RW?$Q}
zBPG1mzNzB%`}I|piTfVL=JRuPusBM2;@CyWdAGXt8Eh827j9-b_o8rxdy&?pq*e|9
zGvJ%udHL+ThqIU;9_P$k&=zHJT8mguX+Ysx{w4PC=mrlw>w7{~7Oh3hy^a;J%Cut_
zG)#@A#WWfXHM}reS`5MOuh=lZ3y<Dth=sz@5w~a;;mEwH5j>0;Iqw2knmBpBjGZd}
zdB6U3$1_$b4T@0#X93unX@8fo3!~$Rm)-X^=$L(MipD0ug~tn8>d6YmL;eGv8(<)R
z<&pa{FAtB3@4aLunRlro<&+GK<gmZ~<sI5VyKiR&Hp;<mY)&Xi`7WvOsB?qpDXU=<
zOL50eS@R>`{t0tweSrC=$!PuKiydIUrsE&y=8G7tv^zo>!(Q1ZRoFvqt6k-uVC%w|
ztGq9)FaHvYRQjuH=Xrgsc-kXOV({AvnZxR>dn8nu$)*_<j}JR5dw6dM*z2mVsIvG_
zzqN|r;IdgB{UvpKR84ytHSyYphm!W`6=1vjH>DVq$t<Cy9d6+bvFX_hs@pca4ext<
zROb`U@TQHLVO|$y&V}6T>VOv(j%B9h-pu=~`!s=zAlsZ;8`JT3`Lujq`S`C|0Nbx9
zK`%qi9V>G2QY+a&=)#JRdi(Too|n~~#EqLpvU%?<1W+u0@P_EO%)zT3mp#Bx&9Ac?
z3zr9oyrZ(SZ26ORyZwcGa|(Y2`2b9lZ>&o|s(pf%=$F*j4@SuLfY9_kli+K|d%`lJ
z71DIDVMXI%7zO=u8Gw4@R0F@|TIbwDE-imHA3$d%W2w21yKo8D%$7{SFl<7&8GiI-
zOeF6<D}+Ye6f4sNr={S^#mXyS2$oCRpG;5Y)8vcBse_GV6&~(0$Cbus6=a8{2UO+=
zPT84!C#50{skd(m7UBMz@|<612iCu1KNVv^235p1vnp7VMtEw4MNET+bc2$1>;ouj
z%qgp&`19tnmo&?mF7a<#&h@#QY_+1o{5so;min>WJk`VyvoH;F;Ak!0M4l%VBmAI!
za=x)>TEmUv_xl!EPa%LKBW`0ov*)Magw3oSwCK9$q8`>D>yiCkrVcpqN_+(1M&yk9
zH$m8boywhQiP6f01cExCXEq|JNrk3;dz2I){?2E>=$ku7IEO|kovSouXU(8WaRb$K
z2;ps=!$3tvioP55^^4<Z1a{mXi)2g#+b#{zUIW~g^qbB36^<puvauXoZ>=V5r=lc$
zHtZD3A7g3Xd<LSzzTWwx(YSRym)MyGt3b!IAL*F7T3fwW*gSx?83jyhIl1<MM;L*#
znS1yF?3``RExkzg_fy+f{U_fHUbZfdX?vySs*^wxvhRyTt9x(ZJTHnDzMs-?7Q^8w
zwGq|)xLQ&ok9W#!b6LEdSb0JRu+~T_1M%1W$euO(S5Oh(kd;f;2lS`n-U`=eR$j1j
z7=1Y&N^$EyFmUaW;F`T}r=H3W$jf<0NpO>E9dA%BHKjI$yS3?oE4GKz-yPUFzkTwl
zJqJ>CD*==FEjjWliVGYD5e#PFVkEEBHucv7T<hf5gc5uq4LQE2`b!s_T9R+Bj&`HG
zCdCz-C(<7$2@15aIg!5nU2Os-+SvDN*Lh#=yJpU>vHXXZz<}FX_T!nnxa-Zf><6jA
zJ}$YdhpF8(l>|5*Z~1x1arWal%rfP1|1*62oR?@B>s9mBw7+mX@$<9=i#ih04Q*$b
zb*|r-USn!<c<>bF+Mq+iyZm6+s)XF79ti(Y7<h4YXTy8Iy113sDC(CBi}b6}K7PBV
zs)=n!Q$fZ36#JS5L*1**>S&KX&wHhfUPHSvis|)Ec}n)d7bEj(I*S<|bmr~D_aQ<i
zpt~Lo-X6Q79^ObB`SDhW>8+6(>QCxXlml5`V#CLtChUNDnyN>?zf4XsWsk?+Y;1aT
zenDJnG4_JDFI{fwSf9-$%^cNHm_fOS2}C^=SiZO+PwS|u6^{KMJ;g3(?7N@=i&&V7
zv8$4Jnr3J;F!DCW2dSqKSgFz#SiTYV?|%WTxmib=q=5AK`|{cZ<flc8VyRCR?#Fwc
zJ>H1Tf@C@SLkt3TGS|V#3c*YJKbTm|VazrxI@hzVkOUz>P4>xProSBMsW<1|6BE}=
z;fyvG9LbCQ7KPEw=`@dt;UFU7m)`Iux-cT}=-|v+B)BWEQki>LG{qhIk#-I6ddoS3
zH#X8B-YrT51@4qK_FPxs&Wp?!G5$!GBF$2pp&Qt{md=8&sN_i;je^Aj^VJj8v9xJO
zj2HAwgP0vhjEUf7#V<XYk%`nnAo$>p++ACPXL{rrTSuONcg^@fI&m7uDG{z$dnZl@
z8sLN78MPKKNgjYz$6G?(BTa+-oxR-87E3#5$T7Xq;LiJvHKA?h82bzF_rP(7Fd~3+
z>m|Zr?ZdHI-F)qL{A0fj%Ni%$-yRjT^J>0L{8~6t?V-9WfU8)2ShSr?aU{KK2U0f#
z3g|`bYA#h~SU2>Tj}1F+C{w+hf$y2Suqh@(sRCd$Wvs<@zs?Za$zGsj94bTf22WLs
zF>6wc6I`;podV%C&8(ckL@JHQv`_=d6hL^hfXbbK=S2xuUXqYWQ5FfOkf;h-|A4{&
zEm<w6tid`kW{t`6qbXtZ>T0LxEVJVAaUdwyx6#XGz>=YyvD{2I!YH^*<HtyX_O!&f
z7aKP0vp)c)_wj3TEG3rDlMFi2xYmC+OM%FTQA7+>q)}cM=odHrFYOBqJZflEAEOzx
z|JaEQb$DJHb65n5t$`@ZSoVU-mcQGgQGRjD4tVHrBfJJc?VEoeT;$iJJd!$_f7B@9
zzcFnA{p<x+Hw+`!?I!Dh9nskbHduRb8!)(v3)Na((<Pp4am_5u%q6|7GuK>@#;W`~
zfi7LJW*brsz7w-vWpT@|Q-sB0&1%N|ZvEd_OhANYI=HeoS4224h>LN7O9E$;{xq9^
zlX|0iG$-t^vq72u(1+Z^sBh|YoE`H$|J*^kaWU!t!D0ed-nxu_RNd=MZx95?RxNM5
z7X4frU8}aZZNaQ4ejsHtIYplm>nMhn%wqD+1l)#%1ixF~%FLIBLsCn$t9>Qr?@pK`
z&TlRLzqwEvGZ4OJTlG+8Sa@@!9+t*s3e|#SfhEBR0Sy~QRr}sUN%3vf{luAt4lET%
z1p9-%avxLkQWX1?8!gG#Z2oclX41Z++6pysTu~dAof>?%@Xbq$>)rNcl|fkS9>#l%
zuwUEYuf4fH0^02cu;tBI`N)t2A9&VG@`8}ff@~>|M%|fx=N3+7G3<X~`Yx`s6oAOj
z`EezhqXFrGO9T{|I8}3m^c4OjlZI4bygxJy7AM*h!GMC9xb?E(+|{e`6p%NSx;=RD
zT7!vv-(3PHnD$OVqvB=q_PwEJatQ&ka3HyYb#T3X_(+x$X1nDLr}Oi8S5(cdprvO}
z6ed=COubdtB}Jq1%_D^}$tkjXIfH3}s?oT}I1Cs{4J5!Kjh#3Fzk>2KjaU5K$C{x<
z%hQrgxS-}y@^oI@$gAP!bwNbvtRGA?|7zG96u;)7%mJHO#z_^);nCX;T?q@+Bcw<0
z8F|?HJ5JP^(jG?B2<5YZP4BM6jzsJxxqjr!&7N38nV5oy>c1hU2R3&Q*Kaj!NZsHz
z(a!n$dUZ4BVULEx-^je*=T*^+8_<J49+feK=-O6}nT#DOZ2rg7jelh%2m!Wuss{n8
zGHq7w-B}(NVJni6a%G=LZ9d=fQ||1o!GY7Vfu23m+b8?ybf6b_etp=+QwIADN&JH)
zl;N=r`?zA@R=2Fte{I%Snl0sjr9#w-#|QK(cSuL{X789UhbB4$Nhz7bB$ouco+?`}
zAX9%ga{cj8dXT^TiVqfQNm)mWE~!Z$+G`FZ6yt$J;E}m>{M?t3`<qxYhx~x=-yY@N
zD(xI?3ko(dS`Qs1Z8$O&@4q20MTP(gz+vOHU>*jn(H9)t%->$Kzn-vr*qKp#4IBp;
zWYPs|c)y93he?*kHtFEJAuY2zg*reMf7qD{5y$Z!vZgVbSev?!Mf{#1Y_rnUrT2Gc
zfh}k_7ITsplp1_}6VuH)uG_xB?FZl?EJnu#k7ZuUyBc2<<H>u)qJJO|uX9TsN=&z3
zoCO2fi*M!ofEgtDMxrowk849vkF7$bt}ZdCe5TbP57L_$3{(Q-tQV5tJT+QaaaTGP
zNU6NvEOL8L<DhAGoDdVkBivz03Fxup=xEX(Ij(X})fh%*B~FEjXRGo7d&%2g^}3@A
zIQTrT4M<Ja+HW2;Lf!g&dw#P>wsxn_Hq`C)HyK(StL+{4cShqmwI%nZi=DKvV_v_r
zpY_2Wys#bw1a$0Zv)INzv)o5_Zf1TE#qqqbI^bbUGZ0MVpwTjCwL)R@d%`t$UJh#y
z%@#a>Wx7MJ-f)t3*g)4F^KkFP*B8e)Bh@Npj62s~bjqxuX3&#TDx76;wsB-}jw_Ad
zkb?#Wa_(&uEY#ijYkU8rB@O@Y#q<9|ZH81tPJeEq$xA~vyMUCi!rxNDYbSXh|NE5i
zVJue7kH+>um3<Mn%Mqjt>|MGw%Jo5FuFd`1fk;4S%NPUTj8LS^AN&tjg#1UB?=OJ-
zYX?g0Q-hNz5i9L05C4n}9Le2H#?IbjrDIbuY#P;tdR3jEVxnKFkG7GnUD82-Ijy@e
zofse<-Mu`L2ZT|8B8#gRy`*=gRW^`jN3g+A<E#Q0R_>XFmW4$te);45s^weZ>+JC*
zjDoT2hu9^SArLGa(17GUL(9yxip_W5uhu162B$?_L&isq4ruj>&++_ede}(EX{)l9
zKu=p4wt1(^3K-^h8I9|JvjEr8g_exX0t2x_PY-u%65W)$1u|BkJ$ylNhs9`+FmC?e
zdp^w#HX4`Ir9=;7{lV_2ktgh8Oc18o(aXJH+GJ~9O86fQJn1aFXTAf9H2Q(OQJMF8
z7{(4;Y&?V%<gh?`rW0Zs5U>Rj&DJGC4y$Zapj0{=F7KM}9POv%QYkTkaIv<#vDF(A
zsij{j_4{aB4@%SvH?!cK^T8S=8q(g6=$S3%P`CLtDU{FWuxUQnNQZQ^C*^iSzVxoN
zq{%X|%6-SxM1%5Suyuh&e9YPiYhcMpW^EPS`)x_^CVN3out*QWYL<0A*RZ8t<G+)^
zLH{Nx?1uw(YW&;v99AqQEG0mu%&r&!ur7cZjVsLgJUWc2bMazIHUcBO>%Fhku54Cu
zwlz_QG1D!)@SMxyo`gtQjH7HDP=GS*Nr`^p<^>I{k_m_0U!kka|3tOa|9siF^wx`4
zaDVm0*u=~d@9*U&R`tRMcY+d{|IIS4572lAtW;z9V>HXx<E3#6ASc`>Ele4}Ms9bW
z&6rst{%-^}Qmp<z#b19f{*qSzoge#m`IjXuy)?h~sOEPgH-vHHn*wnUA@QukdHmIF
zDu6-lQ5Qd*_Y0V#ktkj;RUB+{gLj*_^JhqAas0dMEOM#>$B#Wk)Er2Vh`Mf#?<7Ca
zXEqBy4cA9qlHC400NjO={D$i(NR1UEF1NvcR${+Q03zbWv&^MCOQbesYS>tCe^IId
z#SdK`JLqKd=|=^egrJ<|NrXw%g-x#0%Wu4*IvW^e_7JbA?izIUHgw!7)^cntSvGPE
z%|#T>m_)CR^pI1^Yj2OPbLM9x3qn(k7&`Ef^>F!6jl{u=m$~iPv~_qlL@-0@63?JS
zU)&|3z`*%Xlz`o3hcc)Q4`2d+*6FX@y)%LGI9M2bC7k?5+;+lG(jUT0=o@k$^~gJZ
zyW_AzxtX&6fYy_tfs1|xoy@jcoj4)zK5_E;?|D^bOZr1`(5zea5s%qr=CL9TQnSB-
z7fJBx9Q6#kjF{-t)~_@=Sm`|ay32G`G7S`0GFVk!q+391(BXLHcnm#aXt(?Jm5J$R
zEQzm*b5{9n&JeA|>o|AOoLHreTee=FVbO6Auc{f$QvI>6g#H{naEyTk6Iq3g^;)9w
zB-c0&V4;R^L=wd9Y%H{7S=>qF)*Za}k3aTp{^f_W-`(?2to-o9-6OwT+?v{OeDhBh
zcefpTsQCTQKiPaI|2FN_=3}ptUi@JBQTxRmxw|J1?9lo~@%#^Wqm_VDyqK%6NyW3C
z!I5Z=O>vEA)ymP9Tg%0jW0ovw)6N4pC!H3v^s6BsF4dJOdE`(0aX&y}-Xd0x=k;i1
zML=Fp_Esy!>$mu#Qj|68+UTElgKTI=*|s{qaz<IElr`LMMbNJZt*l!b=Amql{`=(@
zLf^eQ{4TCc;+*wCgw^ur5n<>FO}~$`)mqa$;%LUiy3vHm18Y#dwQmwh2^C;-f_T%c
z)jKJnW@R9w^VeNxbZS8<aM#mbj*F{XdlKCdSh#bRFK<T6zSwX7+0fqYvz2N3=R>={
znQ^=j8g~Zn3RWpv)!ZyNSnDc%Op-a3yag2c3nC~2cnp52)gSkOq6@ug#!IDUhn5Gk
z=&mp8@NSE%89@oVpn^=u$;7eJ89ncT3+5_pZLhOS-SkHprqw8IZ=OPa4)Nu<lknHN
zmMny^xUKs%0_h~|ulw*c!?Z7aCxdpm9tAuC3OI9On?_z}H~qrm(+i1zFwhgZhwz%P
zLYa<~Q-p`NvdzVJhSIgo;eS{;fEHgKO1A?&4tOJraW{NVx377n*OLum64;V;y%x0g
zs`r-J_a09}-8&aR<yjNUTFQ>K9>o!8Ef~9;BvlG>eVnX`+gqw`7e{+!PrO?ymz7w`
zwU{8Rj3(dhK`PE^oLijBP`D~<to7#C##eKWu0ZM><ZwRv?wXQQ|E&0Qt5dAO+0X=p
zfVXa4{Uk@`rO4X<FZ1h2<|E*^y#BrrP;l-<Kghu`)DPkAOr?SlU1sgGmPm49YmAj%
zOIT4HY=EA2X*6HUUu*S_R-}FCIfNKwlQr=g<cXeugVw`@fnD}03m^mqRBzeRh#WcC
z(gNCERQ<+GEBX|7+L!or;+|HG{ezB3ty*8Mgnw}L0wN4a)@~U_#QFTOmhSMN%N4}w
z7p?9BF*7F8kynjpw}+Rtt%TF`r*Z7PaE7+Z%hY%}I@Hs~x+O0*8lBryzJ6dbjL*+l
zTDBz+br{bDhPH=Vs0V8!c!@d(YonUt>iSNlC1|yz>XbB!kRv{KmzRvwO{<?Sy2f4v
zjz$I!<~)puJJu45Ae*-+AmYM=jIYkahHf=B@B!#ebs^283HaRy5)4eaa3MO3_cuL+
zeR4tUlOWS(ay|6!T7A{Ygho$WizR0GYxdrQEdjdf_d-@ZvnRH?a%^wIcc2TxPwVeT
zazAS+=K~czmX;m1@(M_+>vzdrXG2$?TcMGY!L}YBim?&3-$9#8+DvPI_L%E6-!-v%
zC&OT@Es)u~_M7EI?3c9Y2wdu3>zKP%W=JxgKnd^~fPSscc!v#%riKy!MRXPvAdu$q
zK}sg5W?%~b6{5p#zL_T_8VY>1Z8|y{M(U183(o0zm#bw;{q#pdcwuVl`i#<InU}5k
ztV8%sL}!eIcV20cBKT&NV%0XR%06T>+ouY(Jniuz?kQ=o7*ziQRJ+!=h2S<)Y}&E(
ze6+lE{33zkAWydUsdiV0(t+C9TAj?FJ=QR+b$YhHn1fNmIZ;WedvQ)^iy01`tS@F2
zh^3$QdE=~qXo*FxxwYbcF4n3&he$6J?c(QQj(PT8Qia~MJk2{LSj=$UGhkm!xOJfA
zjc(01v{jgPZRcEF&jaFk*aZaMg)6>_5FM;_I~T|2FUEq2P16}ek%hbb4@SGRs)#KM
z_BYF`?&pqQws27KpIy&2{i$ob3O+XBLm>+Ui`Ist7Rj8GYz{bx^t)r_V3_+b{qZG;
zp;v)Ee}<+zoei+arE{X0=}6~R_ITIGv(ZY=wkdeK!2js^xnktii<5*od9ug*5{)rR
zJY+W-Sx*0?I^5%hn7Bk)6GdwxC#FCeSrY<asfc=uBc$)+S><1<9K<Z5PjzrSs{j?W
zWWmEC`99SS@qtMmGp}dX?}`0gR3kYhY5MlNUXNGuuE(Hu1PF9ioiq7AIt_#y7D4TJ
zDqp$|*0O%D=6Pj|9``WGYI-viBN}NtT@de<$r;zpBji~T&I!1>d37!(pt#f7&9%@=
z^oN%qt9r@$`~XoI`Wm{~<!Dfx&}*;l(_^<Q&H;O8zw@;)#*`I9b<lp$btfa+M_fH!
zq-ce!)9<vyGK<UYccK}6_;Qm+-N9@p9i#^;(uZo1O_5Y~0+bVacw0p*m-Vb0CI(M>
z-+wJ&S=;ZKNsP-}Z*l4ic?!@BD`g6jwzj$QHr?a4czfxqBmBIG<;tdker@>~+v9E7
zd2ZO5>&6edj?Xexp$ATTOw}j?%og*6CDfa&{ka#|R#v=t$ZPg^1IjqPJ9z(?eNOZA
zRu@v6lkk4sDb(~!_*W;`fU!+osTaiqxtBd<1MiumlTjjC%Pnyhoa@y42LATSrVn;l
zw!e;fWy!<#p#cnATSNv98H;TL)Fa$~W@O8Z1VdlvJexfR(<~*X;Lwp(89GAk7Jnzd
zR;4hl78uxeG*O7r`%Qach)sXoZP7$djkmZ8_3FQzpL)zbj$dT;$P$$QbOMR%zqwjA
zKK;tx{!7)<i7YHy0wMCB_IscJSP}Kvk5>VN*<Y_liTXfQK}*3Lh|Vl6@!E&IW54k#
zC{B{aQ!Y0)tsiJfi5)gO7n%xiz`@la)^tszOX<O?$Um0hDfKOL_k1J531UaPo)H2)
z;Z(BKC>SdFbWlGK&`}ym((ed(Z{ns8j6Y0Je+^HI9u|d*v<5iP#G@?(?jOUR7Cq?d
z(lR%hY+I;u!*?&bHg}yu<-DJ6QjH5nOcZ*!_<;zt`~3t*2C%5LUJR|%b+wL+W!aL>
z_Ww!iZIPDQSE$4uVW3x`pF003qgbpx`R*V7I<=>_IQm#yw5U7nsiH=6B?&flIu4pR
zB5e2vu{@S-S$#DtT9h9s$^@tfzVxH`WImd~XZj|vxX1rZv6`sbY?4oFKDa7kN<5xU
zTr=tma=ow*t#&Xv7$z#qI9^s4K>sxE<Yx{@oOkPyH(6S%>lrDuGNkrGDV{6InK;aF
z@z*85gFOapcv=fXI5A7xkPeM!1(bKx;k+oFVimdjUqkKlY;lYA{%z4+y+;f!bSGbt
z%%^w+E!$6*<ENqY6|2&i(c&uj{XZfEt^HBz8c!pJ)bl3VROgyaQVL={wB1EP2i$$a
z+uKb%(ux*T{qqL41sy)^F<GNn#3(*fV0Eo7p*2Dr<JJG#+I}C}TDL|VTx4}Et`2@R
zITEp#7Bv4jM*YI#$1>)LcMdsC)4Hm2lBF0)jqPL=zdytMHv>?%@9L7O;`Q=m(=Oa>
z4<L)X(2P$qStJQchu@mwW2vOJ^)?GZ>8Ad35S=+;K#hV^aF&}(Rnyl)AkU0E10o94
zdEQke(8XR29aPx<{wq~*SIgh;Xk);E7C3V4`WE$17QrHayl@0sP!`)insVmx+1TJr
z&eVY(TD==(+M*W{b`1cA<Fj;CsL^SUi5f+}q2e>XE49MNsx`0qoafp$>Fh#rXoSY`
zD>a*#AfLYQq#8-gB^^RO8uVr9%OS>Vr*eB0zwo&HEE>m5o1O1IQ0mAqJ0I6i@nV>P
z()uZmj3z=fLqzl7djVXhR1rb4xF)%h$u)9skKEz+pxdsB_W<|WGck9fBm!dvsd_*C
zG++;EU8PYt?UF_dJ60k=-&j6UzJ#gPF>9wLeBmD2UU^|BI;5XqME6T0eT1pDS3151
zv{zAkT68!n(<2Eq9`?%?d4)(#gdbx|l+L5>Ay9%|%Z#qPo+ySplr<5fYlH=Y&cx*g
z^)NRj<ooH}E_9s&t9eX|iA2d~(IW!-=PaOc5q4A(tY<GS3{83zv_ukM>*lCsL4b;5
zhll*Z{Mf(k%Okhe$V)|!x_4m(9{D{dsllzday3wS&P1A)CblOJMMGs92C2o|M8N#Q
zizlAyDl3#oV6-pGYN~x=f~w?1TDtQLs?+#wK}=U=m+@e5&q`X=<GgO7PhUhr+l!8P
z`q^w;Oe-<Y%*pR0?iFYpfE^E2#iNQ<fM)iH*Wa!c9hNEh;My_$7?7YW=NVDWcB+9d
zuJ83_`rw@m6aCqthbWLnPXcfH6?^HE1+Nr2GhXE`q_r#?(v<*-!FK@A%7~Eo0{A5t
zRGPYr&#of)5LiVIx*yPDw;!zi{CjmQl3aC8vQipE+6%3I>EmRP3;X!O2!82Y@kD`V
z&M<iLs;M0yQU_X=j7mh<-m=&0LY8&0*^y#Sg9kjPRllKHDn1sk0aFP3+@QybKVv!+
zvu*AWZnp(APNV0%ga=r9M`w>VSf-nXUm$L;H&(xKWecDjhK6FGG0@Y>x2P}9iL%T3
zk1+M=(mD7)=n9^J)O6qPuXUs#iw33`J$PnI8K4kYPV!|0eTP8<{NmL~_kW0r9$XSv
zBzpxuS=V+ErWTis|I{)2O7a$y8wtPSfa2enohyVCt)}6oUAUImlS1r`pflmS>;Ks$
zWw1gcdPEv2ZS_@q7&_lIb16(cvzr(&9u~JH`Kdr189`HsnjX5(G6{f8{&=zKeX*zK
z3y^3qLC-Sbp-}%9h}3B7t7`H$TgFaDYq))|;r(n6;L}Sl>((Oxr85Af6r$rG1iNUN
zJL%P1`XJ!?GM!}|8T=%Hmfz=w8aO4PxR^V#D{_f*ZT%7RI@x7_nCK>elb6l#J!soY
z_9{O80Jm(HaSOa%<r-5IK8-1}odup~--+hpn32bmi^I6f^!*6_8$I@!tSQF*g!)6X
zhU(9HpSg~Q`SF-`N~2C_D$M+(Yif^+np>6)Ty4Qskia8%*<t&OHHuqgvOC`1ljxda
zW71ImiN953=#tO{`;qE<)IN;~fJoiGf>GuU>)^#e{dOyTp;>$X%E+tNq4ZE^B2#1K
z%Z2qUNiRl_Saec|#`H+&_k26t)c}5=SpxiUQ^OElb84=afm!sVg}7GbnAq^-xYhV(
zLWQP>+plUuF@~%Vd9}Zc{(IK~O@!J$-yh_wFOq`+Cb7D(5BkABPefa?QO3EUNe{FE
z%a{W(GdwTxr2))bcN(QL_ee_xyqb5+C!y_n&R$AIj3e^HoMUT$XaAuN466B*B-2Gz
z<wmVS(PO|dRb+`?U4V>=he+V@(gQu5mi{iLN72tPPx`~I<Jb|YvApzf)k-fAo$sEx
znqi({hBN(A#gQkc-n2VYY_&gd_ZEZUw60Yb8=W*zY&enl(%19V%Qn;M?eQ$r@v3q$
z>r`$E7=TA-B!DG+EGnl^QrRlC5DPnYMNqsY79*`aX7*_bFJijPjAA-!t%od(KIciz
zLlXHbtRQ@1)<LMMm(^Va7?7!RFhAnsB|JDfUaK&Sgw%n|)IXmT|I`~1ciNTKd2`q>
z$G?5r_maqYl=(J$!gA{3l<}S&Xh&ay4J`HdZoYd3HD@yZ?X}?rb8PFZi_fC4;}H8>
zql_U!`S2$Upno2BS@5yb@fvOf8@LnlDrx}d#pnY>%!TbFKW72WukZp-@_tuca?|d8
ze^YcM53dvLj0n2%DcaJaB@dB{#Hlc1t6dh%vkR5tVTj_wxTof(?ml5Dx7aUW+R>-B
z@z|||N*-QJrT>yYaKSYakDg9Jj2ucnd1v;<Y}V%vE#;t>Bl_Vx(3ik(q=Go5z7%~k
z7TxI(Jr8_Y%2l(LJBpolYPyac^-v-cOwC?27WwY~*{jdct1zVuFpa7%=Q||&wa)du
zK6X<r_HyUA^0~#CTTnY?^qZ0L@b3Hk!jYYft|c}oIPcw9D^qJUzf%8(`fK0#l0hpQ
zf9yyx=O>%lemv)=)3bfW92T7n*!R(VFYp^chxMBt8%n+;#Mha$O9r9Q9)u61uJgl&
zBmq%i8m$-Rk4-gQeqo&6;eNSZOCK8pZm77T80!K3wVu~r9D*OXVXlye;C$)j)k5?C
zbWDhXnUJB?s_@AtlF<u`lNrF#zc(*6$7ODb8h+gU`dGA@aNL1krtU#_iSnCecaM7}
z(mBBd%xu>Xuvc{ty8wTq?)Pp08tcfW`161IWjO~cSx#fuT-l?aU54u#xhT)R&T^tx
z&cxV}-5xJYeU^zConH%;+SYj)1`pfH+DwhxpK3Xt3mwgH_<$MI-P7p#nylYKL0mP|
zW~5%WNkR2OqRZ(I9ruxoQzOikp-m;JYtK$XM8;(X!o`d`7`%hi(6?8beVD{(;>%>v
zdTeWB75J3<!m4xY5ym#Pj8$son)s#LunZaF0pV2_(c+gZAoxHu2`|;Je5}atQ8$TS
z>iWWWJ!t1V&J?kBj<>XJkIU18%lr&?Q*QPIU;?RI9-O?svZ^I{GQrc$dvtz}bMvW~
zc*JmVxG^^`^l*2GZsN5BNGtnB;!~O=)Ce0{V0ClqscubK{?pJlh7N*k*rHi9wCm0t
z-1^a$I;Z|(%L6UY%k^>|*1q7@Z$ayZEvU<#(_U_8^<4cDb0)Y>en%PB$)@2~TlFT<
zXBR(fC1y`hbgQmJVp9>hxuJ(whd><Uv2l^@<>1T&k5-dJGk%J2x%5mgULzQA(mlLT
z124UBlK%yRO#eSe%PJJNYnx|G_^3}NzQ~fi)WWk%(p(v~$reoL*&NW9@Lv$cfYYGm
zg<V|#v^@ZpH?az`@yrbQxusWaAi<CnY=*_KiSIdtrxVt$kCxt%)zK3u-8Z%DyFekf
z(NK-_yX5m?6A|3~i*+!qTCJ97P^TlXom~6Sg?qkpTncH<eU2-x15oafBI~W8mVNK>
zyzB`NyM!7s#nFB-$gYQ@?$r6J<29q~mFe)#(ZB<~-UhmfM{8#{&-M^R+U5sZc7yWk
z#z;p{lCz6*T2#bD7c%$R#K&KPzL9gHSIGz%&U)wli9bL?jA*sUBdp9P&vU3ct5*>z
z&12m^%{eYM-4B2vJ}1{T+^)Np6bD1tvY@s)Exy~z+A3qNaYvnyBe18tHzm351xnf?
zLX)-dXBQ20lYejTf+F|#4zJf`_`EUm?f-llF$@WyQt9gyt%0*Mqr6@l+-&RS^Rp*G
z<paa(d8fcP9>kkWx@Bom)ko8TOzf4*(^^Awl;C=Hq?)eLV0q@T+bexqUgbToqLJ3P
zMe-rb_jeRoOTuple4lf>UObLO|A<zAtd7K9ppW8uJYg5yz-Qc{5AvWl{xKT1=W15@
z9H>^!C%Y@V?F(47Iz@Ekt-jpR-H9*1Nn-elgm+P(?K-PKXh=I6jM#--`!#dwL@SA`
zFm>tVx`d7eh0Z*$KI1a1U|F)QB8QU`s_%&K^15*f+z<`Y2Ru{OPlu(_)YxYf1v7mo
z;$-8Dq9x70S1UN>|GNIhFxpe6#qnVD`PR8gSdrrO1ml)Agw?U+lMm=xMo~cE{2JaS
z1+n(&`Lg}~vx_BKFo#8+XP#Hq#2Ms%#`)eG{gV~P86_B)`qE*nWOT-**qY3agsTE?
zAxpL@@dYWE<wlK917>VP`<xn7K|6`;gHT!S(Yz3SVhrWPpK()KJpSZe0);4)SZYNl
zwEoaF39`wqG@a(Fa-m;IdNjUHtw*cBpbN1UhQRZfv7U%g$7neS5~SI9J+dYvdjbu5
zz=#IrBPQOeFTPDccsVRS&ER<(oLf}Tokcr*anltoleE0%GGh~9i5!dTiRE;x<nt}@
z+}!d|jIY9^Kq+-H53vSDI4&<*^<kHcJ<;##AG!NSJ#CI~rPFpDyQ!>WKQApwc%B-d
z5QSVkSSw)~6e(7+3v35h9_)y=Hg{KWIZY|AXH)DIq8vf0#<R+g^UZ&PI_c=tM*ZqR
z%{lhNryCc8ujD%Hdn&2MlhTk_iDYJrC@vy1p-+x`q}e#29|$+lK{lUWq@g?JO>^Rc
zEa}L_e_$>}+xT8!Xd9*zooAI7AH~L7`98Uo!G1HGHDPjK(ds!)e!|wKXnVS*Q$mcd
z4dvZ>E^6YQIx~^Bh0CUT_MTjgxBRe(zdr$=`#tV{sbI-hsmEY(P~WcZ95MeuV)^G-
zU{>D2eyU@XYs3Us4dNc|+jm7V4HYdxIh&Z4{zr|ti&`}rwf-F=M}1GT-U>aRO$^bx
z07;AN_sKC<D<B8A9vLW_BuD+~3Ac<n(|V*sJDM5vYrcI4R*#mmeLlG8eN?eeUyXGH
zKIq9b#L4fVE#{hfu;r_@c-`|Kj2UQrVY@!Q0|=9zUX6zz8O=V1ioIe3QQLvM)>+0(
zj7VNiwIZelL*iqPty~a?B`*W?9)(mX&TISDCIg3mT55Fw1g|$Hm^ua5Ss!e1x$Ng2
zToC6<J&91zigwY~9J9s>#aT|zte(?Ox3amsqz>f*#kaQ#$`WIRL}RF8tM#ka>AI%A
zg@Z2iOUYp#iu0W@7my1JgCF~|>r#+b`Yk3#Q76VqSNrd@UsFD3@dRcK_L`$DyoE;h
zPbT=CO|aL)=uv)XL0;B)5%nnJpqlCBEme;lL#bM5OskNxBbxosx-PBSGYFq!$EO7L
zss)_l8Rb`A*%J?Rlk;N{bN1$%w2*}9-xm|H=OP0T%>`ChZhPV9umy3RHqM2y%}>aD
z=KveySMBAkqq@($>;rZ|y1u%8weL&(;cdWQt3#-9=4M9(NY57e`o2t_Y=CD<mh5`c
z_T91|#Yo(XezsM$ZV(}9T$@b4pf{!=f<vR{ot|wSY^m4cw)Ag$=(=T6!*kLoSUemP
zedm&qn65Df9e=JpJ|2>V$!%0+CYT|t@aJMfB0Lq~BJZAWRc)SB*}7*ry?yu9Lp9(M
zwDf`ekVQpY`|sg!ip^7;@+vkDaG?AUz4r^OE!-725f&OcE>wwggl7r?=7=x@&%g+F
zIgeO^>9mhn>|#yx;+M2_`Hlh`x5@bB=hc=p|NQ2Nr@k;jn~QMR<gBvX`PVJ-yMgb1
zVxuwaYv+yALVXEhm{ETrkW6s}B@_#mYC*xsiNku6lEFYxx32Osy#+`%1eLbCbkhtO
z<u+Svk?)Y4n&AtJw#Ya@40wo<S0~r~i_2zek%tjhuZ7V$wvU(f!rUz~7SAB+EluM_
z8!tc09c&rZUhQH{dv(p<`3looWwWO9{6=ky=r)Ia^VruO7rcL4K3zuAmz=rm2d}wy
zAo*R3<lF;p!Bq6EtkSudfcx6h%5qosZHd~H6vEn6fxWP>ei%Wns}ZkcD4b=Z-#Ca<
ze8iWz(=`QFl!QTVhB50-w%KD)YzQKfaD`tv`{g-j;xmnp2K=Mrz754(4WFJwYu_o>
z&zj{#*l}(4&uz?Sop>V5#+cSHfXp@fB0zy`u86HTgL&>y?(*$QO|Dp!X*kJ`H%vac
zY_zXHv$$<By?`<9(dC>sm2R+PLu|VofrxuauMYu4tHPMCN7@Wjr(qQo_0h-DCC?_1
zx-}i^@YWYiYDlu4Z%1q5g=}d5`v(*Of50_9pjEN%fKxwI2V!3fKdzhnP0Bv<w34cA
z$y*1yRgtB+cJgqWExNl(TLe@bj{04oD;F#7eH7|Su<p<_Lwu+ye3Wrzgqk&7s;gX%
z9f-#1#2dR-5kPaS>}83KHq<&6E<$WRQ`l#w+lF73{9ffhOl|3pXGz|@Y>_9Po*}I{
z7OcH1(BrKNx7j%<nRWTpUGRlDCSI3KMG-h^X&i{Et}kaqU(J{hg7TkD6kpzAtMMt8
zN70*nwxXQYg025;?Xa#gBQWf?8R#*NXgf4-twOw-;yW;9$N*7_Zub_`6HDnDX5;)y
zL#ktwvZ=|NkHqj~Km~@*Ze<$XQewIx2WjY*MxUO?5106a08}5|l-;mKZl&J*W38!_
zI%;2K)o7K@K9)(brn(L*$+c%5+$RTcLyA?$;)&w($hE7jidDl!itY6l$*PY84vXs<
z#jn#@uX9oF7Uc%cEl%q?I&jbRz8fbhWJ6B6H;3mIy}>Djo8B*S%Z(-$j$ih><aq3D
zUD}@b{dM}1)&-FEW#!AJ*@V^;*2itwj6-r2|B$lZrQqj_->lXr2K?9s<oc{yQju0#
zzNbfKRLgKHPqb_ezlc?e75A^(bEh?duTiYF{OI1LWm)(+HhARvsz+dtM2-JX(KSEo
zBFobBcX}yBbQXygF>G|8m{3Ao-Uce~9<D#ylI)~m48PUt@9Wq)HhH>L^V~BRbt<*v
zMW?-&!ZFmTEZ`s{^~Kqa+Bp|USATufHG*7{uGA_!y;>M3LR=7h$`2{O;9AAgP%S&P
zx@^d;{i!A1>6zTwRwP?;jIC!2ufVja_s(s)E>K0^xW+QE!4Lf)X8Rq1<@k^o?-czu
zC%O~$hmN*0Fzj^4E1r-&v}YQ_B7qtxICJsLlhhcVIRaOehkrT=KiZPWCvtQ6np0*R
z(Zzwu8l0y<sJ#vQArPbR-nOdboLqyklz$p?pH3{$y1LF<1|haDd8t~R6w_60m)H+8
z6wXiz1xxY>vOZgBv7RL0=4*`t@!YMtRS&a(*Z%l|G*F#wOFF@sQ_vh<XNxClT4=MG
zTGT7zHO|&HRu6f&hpFgiL^T*6Fcp7RZJVW^kCAXd*-AE8p;iA49F5q<?Rlyt`7~Uv
z>HIOO7(6^}P_{?z=(qA@xrPWJTX+^BaFz&GwGxl|CYYUF{F;VdjktvvUM8mBSd$O$
zIlq*o1)CGcQMF)aEtAOxGCy|Z#~Hr%DsoTjY|lHze4E59Ei-jQ@PQUsQQ4@)xvb`D
zehQ7Bdf9Jr`%;=G1#xUR?iugZuql@qu9ZyiaP0az4&+X{ZoO>R7e^4pmB#LnC|IlJ
zw+<SMqAtO@waGYj7yME}7Sj#tmAGjVkEwGWRyp!7s`E1l3b5=6knSwJzFH{W4W!4D
zhbMIi1}&O}ibrh@w&+#zK+BNx(NBiWRm)zBAOY(yN*i_08<Xn=OM-*ZVSdz1p*3i|
zv?d;>g?4b=^U8J{=GB_~!RE<iJ_VJbnkp$4uFC5gRf`Pi=!t4A*tta`rNvp{b3Vl&
z86KDJpc=mVS@BuWAo5rezg5c-#}vGF%AgfS*A#Xru73@=iij*wtQMpTm>E0*I)CDT
zUd8}n8)A*i<TEoS%{-rnc47un>(3!j2cxee3<}!iG);rn#4ZW8PrHqx>wc!5hhd4_
ziljvdQ2Y_gCmMWQaplVFmFWV-m0LyaEK6q!kTSPHey*|(5|blxOrkF*hq4A<X1pt=
z-(d;xU7dXGXlN!!Jv;u1U?OAFL<qdzswE7I=GtgcFNLsTqcXc9AD515x&$=iF8yDt
z?lrM~2Au5aH7v~l*4Aeq6(&tu?;mjObTC_X*`+12SN+yrvR~);EAZLfX?d3ytoiqP
z<<)Q8?N7`z-~3zSWuGV6@_y|lzUcaj_*3aslghTwT)X}5^?mgozy9s~I8}C<eWs3j
zY0lLD-cn2ApKOdR+5KvF;nZ&@d4WsYy^r$5G2Dp#wuGnZTm0b~#gmJJ(iYcd9W%0@
zZKVA%UTVo-r&B$fyO(J^Kc$kc{<Hhu`R_h6Pnw-pc)vVYF5v4_!A!nocQ*ZfCwXdS
zySH(j+P79s<{){w$=CPvWd_YltKY2h+|1*5x8~{Y9KXmg;5h6wyLnFxc`w--oP56}
zAnrQf@+~{nm)(5x=Db>EK->0<=1b4axVNr8@6Nt6-nXBYzuYwCulK2*?|YRm1>ISH
ztM2QC5QYs)fEoJ0yKlWJ^SvtrZhoF@qm(-LGO#Okf67zIw5>Cryw%+P@xEee*$ejS
zN_nqO{hF_T-`gGjt<pOE@`<UMujBTu`u=&Q=RdX0chzd^PTgA9?>Y0}_ueUuJGL*o
zbF0>AQvF=T|CPUXPx<vP=~TnZshXRU<7XQQlx&_Nx##hj&ok6ciU(ht^mm=g{^><N
z>wG@V*E|l)3jCtge$%ou+uj1b_|HQ1-+8Z@FK_qXj0RQ<lU}R*e691m;fG(}k~K$b
zzD_y+6Bx1Q*86lmW)HeHVPEm2*-LCn3UBO_^_(gD_rMg!z1+t8-foGkyFTyo@9z$i
zto2s#t$xC}xBkE4({rN0WXRCadULYT?^@&31v4M!O+5IC=X-nsWY5VY75;Mx4?eX3
zJF7GI>ZrPjAC+AP(s5wYjoAlv?xpMP&^!I=2M^mp`5g}+i;8jz|4%u1+Szm6E@1C{
z@$#idp8uJ#7reMyXJ+q|_Wd)D{$AVXU+Hi!?bKl<(8A>r2O2^@{AZY--TXT9{|oT;
N8c$b0mvv4FO#tY{r<MQ!

diff --git a/reference/plot-6.png b/reference/plot-6.png
deleted file mode 100644
index e29d6ce27f53553ca57f16f0df3339275909c513..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 39878
zcmeFYXH-+$8ZL}ju%n`)BA`+Q=^~xj5JFXY4JrhrOYa0_t6Py4N(|K!L2BqVfDHo#
zh$N8E0xHcAAP^u3p?ouYpEGXx_1(XB+%*P<u@Y9+e9KebxqN=xRR7TaGy8dXcn(1g
zZkqG(><;7M*?E8O9&lzj<IMti*>m4W|0Z|^$J&i11bEr^(7-N`he!As_h(1a<RYAh
z=Nu38<~7T(>{(j)tG=6C_(d}_{+(YAbrwFBKKD%VvHaV;Z&^2-w7*+@a=6`Rm9}!T
zP4Bq&_Y)`l?98mM+4+sE*z+H`bF=U0u6sLDBK2+^am&6JvA(J=5keNHi$xz?h|w{F
z=#Zfj=@?RaRFk>Hj(=|G|NZm-A^;+QnGb=UkXKM3JCSgy71{HSYlF1%?HxQ<Yclc}
z2e~0SeJssV>`Bm?Jl)*RTGO&_W7c<In}_GMJ%JF+t@VG7!t);w{O4D#i@do1{92xR
z;eR^4$8IM$UmNOJrow$+`~{B!56_3h$H)kS8L7eo3>#^0o9CBN4I?2Wq7^(^Fi$Ak
zS*ec17Nlu8rT+$L6v<0rwI*M-AvsW3Uvz7wtgo7jLQw{qx6Fm1V+Ik|NL9$%dW?oS
zNfD|K^%VCg#B3Wxn44E5ZOlq|#A0rg+Q_1#n{Sb%kZy(%CZt=U3<Fdw28QfO3&I&U
z*vitQp=L(w#&jL)j!x^rCl?PSz}fzMP(FKRq>-n5;Tw!0IzY>L>TI|T$s77!quuWc
zoRyvblDjG%9<w;t$A^C&heZoG>hnY6Q*GxX(0A`&&`q=axe$TRgyA#sK2C6C1pZRh
zb4<8DOWnl0d)Lwf#szw@Iw2vZS-kKXl3`FVrMyE7tKD;3JSjB1YZ~5&OL~)KoKCBA
zB1Qd3$-h0tx4&TSL-OS-J++v(Pw?vzlJzfz97%r3`_A1GD(Yhg+q!WM2y|Igav(n{
z51l}M`)Wsxm#4c)yrLTCkq7_8{F^bYiaUlncT){`&O4JVbThUSfAKxh^@nmEVeFOV
zf(7iTq9v26DVE5)nWqiCUp~=>@?+}xYdfo7ch~ML36?_Q)0M9kTkgaa?y^*|3rX!Z
zdxMLMQi~`dVD01?QCA*S??7KIvsiD{>|#oN6w0VJoZ5IHmCl15)*RnWaSnm5hGz#2
zsu!qiZvInZJRt$X6AEER6k?O5$jW4<<`h(B!8XYucGUtsFLKO@xCf0Vq9o9*k3_;#
zjTh)`muH{d7%!2^nK2qlbKQCYZ}fhcH7_w33(F_?Dz9Bk3-w|Wbm~T0A&n(!;!8W9
zp~~bvz9>5#{>l>iamdIX^9S<6a+SyLN)O!_<<G<O<HiL!ZIHvtdw&t8Q^jji_7zQa
zRDY)3Fi$ZqpDEsjLZleyrNsnWcqY|}N(yWk_K7HOTGv7YbojrbV-hhcj``8zg;kg{
z%1K}9(&@$IjwYn16uQomCMz&ua!U{@h=Ir^Dm#S;SQ$-)1qY&Iaxu@yxE@<3A?V@}
zSu;|vKk2al+D0I>Scl}3?BO(Y!`woT9eO)|WH}`=c9{pgm0cx~_6Y7ao?RvnKl@L?
zS-KQgyToHvl}pLLJkHZK(dUX%Zz(@IQI`S7LWi6sd5}_M`-e#U#yu#fp@dMn8@svw
z)!ru8IIltMSC?qUX=%^bs)$_Ug|EeWUMeVDe_h68+LWBe_Mbe3)st^l8$f*70d{qp
z5q56n+FGxvFv<2)et8kl+BP>yg74e4LJ1Yc*Td}Gw5u@Eq@R5F!ps+SXY&T5Mgd1s
z_PN5=W6X>R9n_@4^VBOja$)*-JDW<mu&Q8F_r`x)CN?MA;i~JM%nSsY!1{vBMMWmx
z$LTk=HeeR_FUY5OKn1-X>1UQ>#Dqx)S+81?j7H_z2IKvLbWul=IBTHU*`}dH>(Y{K
zBe{ZLv2hNIa^;Oc%QrorD9FdfH7(lr<U2Sd;pz|l?jMG4?GF!qj(lk<vKGL(bJNf-
zHq<g(TT<vp_cx841m(m31T<uaZ|Q1TvL$u867kS3#5O<AL9ND>*VvTaIXR-)Bl|HD
zM~J}f*_kzs%x}=o40Ix03Px}Hpw@go9k0FUbTYhTSXP?aBj%*J6KC#5@WpNPrS02l
z*v+UMv<awwA|I}ahi?fwPPM8fgcs2(6T!-~N$*uY3i-u0x$eXbY8I$CecQnk%fI{y
zU};gjoqV2I+H$e+IhRV~_?|80q<N_TISUJOFO)}~;n43Cwi=Y|n17;uSrz^Av^+?f
z!P4J^X-Fi-<M)D{NkgdY;!}L3xPbIVFWerh?y=d9qKox+)FCO67^e%&=rmKbOw(d@
z@^XdoSXG6L>-w*iGug7KzJY5S5zmPGCrA0`UFaQ&;vQM}YFAF9sg>0kRC;%yrp<rq
zZ}w&nGBn<(&br#nU}nBzgd##0IDKjESkf}$;j1xFZZ_xEE+++_okdC##S+M|1~ATf
zae*Vt5e@I7i7rhWo--Ar&-Y>Srcy?YD&?-IScQyF;^%7g@^=Ty_h&?Iv0tE9Z=%*t
zO4gTYL-{&_+E_RY(e?2*e5mDM(~6Z)Mo=oAMIL;-+NeH^G(Iruol8c`kMr<+<cqP|
z&BN2*X7mCwaHoV9`b9kYqJVedxMGq+ULNwax^!-h2pQKH`uAjw&j}H7Ou(A<AVVFy
z39v`4oJvS=IPf`WA!0ca&Px|SO9?Amg;?~fJlViF6k-aT@^g+ZwUYUdAtPn)I^=fo
zD?*Ic4M|^=AGP0r$DIA}eva_)yNlfue~Ohbd?~xkmm92`|Bz!*ydhe@n<lb5aatOR
zm94l#Oa0kDq=F7kk1$?}@NOEJT$%|Gw$Y-VUO0k%@7riuR~d=I#i0j(Q99Sp=+)M|
z!MHe1Mbcab!;!F5`sYuDM#veNQUAsLIJbDRh7+w*ormXdb0=Req>9r8(Vr5x&Uqk>
zMe94%Z<(WB5wApciE=*Fr=H}l-s$i3(!1}nOf+B!nc6U8FZ9**Xi|F0F4woS`JS<u
z!p_Z_Ej6UIR#XcL!={W$HDIC*M<Fj9<;gW)dh*ZpEN`aTTGtZkGnSxauckRVF9TVr
zj+Uk0M`yTp!yoS#G;DMxy;b)x`+E(MgU3Y+v=feskgdg&TN)RJ_zOC#ON=HbPw=b~
zWMgB3c<D*THn)(;J(WVlP|A9Cwj|wdvbN3DmLx1smM<U02BYWu@d3G*CY}7Ta~{rU
zN_qH;wEn21%Y&qL7yCod9HgyvSr6Si1E4uL%HusRsCYJ!Kk6!IbhVP;qvbeJX4yru
z-eZf#{Tkc0^H<t4hhvSA`y%Qmv1p@A-lWUE(V1I6l()Z#gcE8c0>*lDbVzg4De^F{
z8CSl3!&{YbHCT)%VORf1Z6Gp0Xu@X#iIr0>4a><dMTX|C`a=c%j#O!i&~bYljo)k6
zAG*)>OS-&0odPMiM#0R4n--Y=)7sTPKxt~d%bU>Uv5<R4#MyI97G8x^jffV;*p5a|
zTkHFB?!7{a@)Raw{I@DolkqzFv~gDWRZ`sPb|X!3hmi4$>8CBb3T1ae8x2T1JDrm2
zZj!PN=TTEG(<HiW!})%$F1%vr;;UgIWRlM#!u00k2;N5yeK)GIXj~&_o(pxq&T&Oq
zQM@eNzFb1H{k8*&D5w?1evA)zYLIz>Eb~3)W&_$dcEK$vC`mFIKQpykgYUhUE&15#
zSH)#rVha*Tc_U|vWAksNIbJBSV;S!y+C))~^V?J!*NAU-9W?nW3FRX!r_4<3M?azl
zw8<SA>zNG1y2yvGc1P6{rdPyG601gn*83)xw9)6PY1^T<&DEgJgfyBfoRQrVe46i{
z9BMt%KS5vr;Sa?dc}kFrx_qNgC5LaU9&3PfMwJj)&mu8Uc5og(#By5xj5ygQ-=Xn|
z^0n-U7qd%els_@YU3p!5m!wqMvqgy!6B_*}l&fe5F=)`_GNe=@J^T{FID4)q?X~d{
z&6IJLAumBT`FH@-`Fx=63Hv2VTAvwdNS$F4DtJg=qQ3cg%Shvc<jTPR!CX$U!dDlT
z>}6L^dW9oLp<gto2)hzSxRT<k<lh;IuPthMafjgcK9Z?*dP7QKCpK9~&sa4JxEgvR
zD8rIok(=UQt7GBIYPVPh#OzzbIpzpkW`rbp&<`){ktl^KBgm6m!jmWd;U_vbX;H7^
zfpjY`{*b)y@Ds;;n~q)P1%GVu2hm%mgy>6R)Xzv$)bi~z735#e;wxzy-=6=;Ge#8q
zZj!9k7CU{FPa4e3DQ(K5<%KHe26s(eOj)*jOlZ2}rdWH^<W7i#;xq9>hWEEZC2}q>
z=cPtw4^9VZ61Z@3Emn|+XLkQ6GF&@4JbSGcs%a$oN6>Hd0mMkt&fFR5VA1XzEP&j{
z3eLDJ^<=qCIq6>0%A3r6OA>QZbdLJc34%A3c$#=8Lfk$iRmovA`iNOtcnKuSnlM2i
z<QPW%_&g>aoU+y{k;@U!DW2sMbniS^#CFJDlZ2i*x3rm7G!td&{ubzkSn0Kf9#9vt
z@cJ&XKH(A@L3Aw43mqfOXxAQXx<fjKjveD5bu+4UJA$J<<su>Ik8`4~_EYn6OZ7_;
z=yM&8Xng*(?3T->);p$aZBwD~*sEt92Pyq`+z{K60%>-a1b8y#s>}jTXgm2#&KW@6
zJ9jBLv}><#=3=HLqCb3%`-_h3uFU#?Ana15?K#P5!kDjl_2!oND}T~e>p1+F9dB%>
zZcEZYCZ_a3^Pg;9hx&CM{OrzmGO)LPQU|U4@F~UuZeKx3ueKyHkh$g&d?JKC4JTo4
znE(GCu`|-i|CqA7sH3LimG!3G+KC`PALY*}P*4B&BABOKU-i#1i`l!t```2b$FH#p
z;m8PRrZim^>LBiuwp@#mm6)R==l4NRAlKb=<2?k^F5e`%D-S#TTv)&0-ayF7S$icZ
zW@e7gPKuoMDpp0vYu}yxIdHZVk{bzUg+^s&OuI*tb5JwwqmWJE{P&fcsjfY7dk&Lt
zU94F8Lg(Cnv$7Vui?xuT%Ixr;GOEdaLD7{VCP^wHs&ABDj(24b?t$hRG+MZEyzG~D
zr~iqGkJd@K3p>*>i`ADnVKj89N?|l5D8>@@t5CwD9z!hkvA;l<hniHC`#M}DaRL;K
z=mB*#8?}_}(uo%0X4)o=7KGj;sZvi5npmQ|YF&G<RM!?2Pj@LV5Ui%8!^jN7{rGU&
zg^Li$xl~GZi(RLercUXf;SH;v0F;*2l9I_Sl{Uy0oAKrB<Q?T5sB#QNqbm;8pu9E=
zF(E8fHPd^Hf?b(;(P#4|nl^Jv*Yul=MN7XDQ(ENLw9h|^x>P<T{97q0FG6OBc4z5n
znyq3bz>l+jv^?7-K7m)%xfNI5wTs4gHH1#=tFHHu@klpXPsV&vsh@u!g~u?nZVtJQ
zc*(WqtRTD^n1K-ZSW<`bwTfI1N4MI+>CT=bKWHb|p}`GqMpGL@#LO$YT6R0jtJ<x+
z&hQLaoxMmVc|xFL9o4m-;(>9QQiX52Oh#pq-wb9@sgC*1H=u&H2vO;(=0_iFo{hL3
zuyYk1KQy#8pw*n%skaCI*ev9=t4Sj3hlCmF6J@NJT&@in&YH(r>ccov>2rwh5eVs?
zsMO=zXV;`Z1?OLITCr@w$d<HH?LD>?9<4wwk>BDhNq3<MrNMj>{ptJC_)*mZnGo^S
z&CArx>9Z)hpk%UWM|Fmdk>o4q2R=rRj8EH&S0Sd_o1Zw|CS8I?Lf0)NPo+wck&%yA
zRu7^@Ka-m09|TM2_;qi$?wTUjdZRX@psFR47nTOb8Y+yZo-HxpEv-PL{IGC=rYpC;
zg`5xkGX36Q>S({yHMQ`%F7E<b-EQXmuQO<ujes(hUvHX?MFb}`#q27ZeEceBH<!ZC
zWWei2qbV*TO($F`olYCjy*_32M9fHgX^cseJcDj!PpVpC-cOzPpECJ6O{wfL{!%&8
zmrA#myhPVVg_~MpYJA>&S@xhTKT|26e-Og>vMFAAdj(v?aB$V-Ma}4!O5AAB5ZZoW
zem^JD-TRX)+M0Y1E$_>7^_X1Pj*m%^aNi7+L8E1Ue+tSkE%)Q~;M;_Vu~%frbEtJ&
zW>sM30YCO379t=ses+*nt7gaE3RgHpI0J8nBtM12%{3W?#U-m8_<B@2W)ztjb_v_O
zB7$6o<GQQ`T4jz|(izG`1IDH`x+T<QkQU-7L)KKwVbls=^7vq!*GfRz^2O$iokedo
zVQnTd&wlhx1UQ4qznvn#ADE`}UjZX~LjM+VefOx#rK(L?4cUgg)!xMX@_Ttjf+ihE
zZn2v3Cp|zU)6GKl`dD~F&-wBZm(t0hXo=jSIX)x$L`s2eN6BiC*BW6W<~;E4f-buJ
zoz?QE6+Ch;>>R>Lk*^os;tlsXPnSeT3+gEL4jx{EPQ0RI1eJD;e`9~tc7fabNgz?Z
z)<dPrNrq-A;Rm&{5-Ubq7b{+}l|Su^{8(@dZT<7hq{wAYX?g&3JA*EAKfdc73FU!*
zR2E^!tb&W2pOaluU%CZ_bX3cGrCrrsMK0T*$6KvP5l|v@=9lvJ*zwaP#TghQ@o4Y&
zw37Dm5_S|~**3JcRt*NTcVJ}8lbf%hyTn#jJ7+~9$iZ6<Qs}^lN9m~d%J2@^fm6+h
z`w{_{yq_q`8QfsmY^lgOt)#goetAwA^Xp8TR4d))3&&KoI?u_U4mk;NV47|orQXI|
z+8mbHyntq<6c{9H2{Wwws!{M36}))E#*BP=V`WylTlvbi)X>(@*Ex@OmNR5yCua>P
zdtJeSs#Q44G@fS!dDmrGvV@ZxU@_4Ay!ikqq}?$ucR~eHD}xq|V;yf00TZmN`wv@8
zR9HaX4K(VUaf7@KP47CCP5vkjY1fWnh-+?sdleKS`ZSy+A}uFQPkS{fGOdKLovMI$
zG{rx)q{yOU?QN$XYfvsys#S*;{RfZwvo_*z9L1&!rt8m@mhf8nsE+8nxrYMyA3#ZM
z{w!8yrNGVSF|!e4$5BeK(c#L0^WBPiB5I)#ZF^Sd^r=ZH%Z)~vH35G2P*QZZ`WmkN
zd`|OdJpA#e-lf(_MQ1P6ihE#WGmAtM<7X4(287Kjhfk#~|Bc^BSN_B&1sguaw7tpj
z>DJQeiE5aCdmaQ@JjYLNI{pJOu};@PVpUf?OMmj`nWVL2=?OqTT7(XcQ7hPY?sHZw
zymD1Y^}fRfUb!ysbmOUdH(d01ANnA>Qq!8{&zKj=jz{Z9zZHIOpXWBS-s!Q{0o#T%
zBBTa4b2_5uLYtZ!9~elC?LcpJHIrz+%gbDw9wZx!{=SD64FKSi{PvBqpp}F!N<iV%
zsf>L=!B&{vmhO7Kzgv@+4h06SE4bP>wbkxv?9GIoCtt7qZf`Asj@dc4EG(RX_)B%}
z1Zs}8e<8ndlvnT8Bhh0kYq~FX{eAeM?bx8QAicB{zRT6#EGX^ttmGwugMrG@G7!)d
z0TXuEZ_1muVJA{0Q(V^H7#GV&CcoS1Y9IAPf;Nprr5?W+C`A|<85$;;<ipKu^WS^q
z3zxLlhSi9Nx~w{{#Tl2Ely%kaf_Hwp3#GPU8oCWF<jF%-g%|Mm&Cu4G1^V62Vlqjx
zaNc6JZk+=s&4J8mnpoQBwXThpzv$_(RwAonVBVf7LOUZH(R>do+MZT3u%r;7TaXbk
zI5HX*UO@=w)9azWHh#DbmvxenvrDCYdwpSg)*Q`zV{aY9PF}9RFm2#{WY|k++y_xo
zTpF(5hE|~N7!B&TFrVPivp*2`pBVImnfVzfW%@OD#n8n9yO8M#Jj|ajCa4#aXm9f)
zcbDYVO*Kd+FfuUBrF14fD%T*E3S-AWCoZ0#?_J0@w`qA`qIIE1bEB^S-O|De+(gYW
za&EFi^H*!e=0cj>j8w;(ONeS6*+h2mjh3$0@$MZPcQX$Almz=2kXb-3gcz~A34)7m
zFU4Sr<;GJL)waPWPXGPB;7gF!W`qXcj~SVO-YAUGXs-Fh9@z|<a+go6TpbT~Y^By*
z`~w9xbv?3TrPacyGwiYQwOa*QbY4B)`_t0?i;48>8e^}6%qL#qN2~jj!!4CB349MR
z#~TWnPjOtD3{u8z3+F_;Uq;U{lDBR&@J-~L_33Wan;#bSYW<+CD?NR+i$+;C63!n~
z3-u#3TP>SVv{CgXt9GcBTg2&QF@xGVs;t7|ap?e4$)y1gnc)X@_8W#hI~^u3l73<R
zx-LfwO>1C`LO64eo{ZT47@Upxe0Vvm`98r`lNU!XDtJq<_Ab5_si;&ZnlBz$GWxgu
zt!0K3`YzIjdZt%d_(v($!03~#<CKPZil`*cxOz$m^(~ms)n5OpM9oXIwJ;bWr$s}8
z;TPg!)=JG(qN68Q*NG;QWud?3n8r-T%`l(3iKy-=sWBT*Jmax(VJX-^D-zf->-IZN
zkjLC*%<>PNtax;_eOcn9SNsLF1XvPcRB}4&cGLDp3FYWG%?byJ<@AO(27IlI%{wW&
zW4g)o+BZ2Eo5{^nSy{;!l4^q!(INF+CPdHotBZ9?%GY{r<4acUhZZZwZoqvd^E<11
zf-0@1++@u<R-R<P7`=nwgZn-mffWmr>S{+<jqf_C9$!0uFLtC3lA-gOqKUTFzxOOW
zref5ESiuY3P97{MXpg>IWJie%dOb8>^!jE#XGZ}D{i<)*bi)9w4kvy;>jFor7tgd9
z{MM*UgAaIQ@2M?9?o&6^Vn@!XezdO~MkKGaNPb$AU<Qf{Zz^d*ZFJxeB_#)~!zF9=
zD>3Mix_8D?e5r}b_=n+r)jATiOBTLnOK7uJ&(1#C1lM)JnD%MbY@lMw(uLmOJV}eB
zQOrbVo@=7h#|U`f?OIF1)+q^R8%dziJo2&F4cK;DXWNQt{%Y_AT<vf|UKF-!l;2c7
z>*Dh=K7oc@-tu@cFhD4*h_jy4o@NmXgP-|4U5Tm3Ynzt#V+^j2t0q+!!|kaCTYPs6
zry~kFmVN3_#^s@aro(+7UGAv<cuGPbdHBPe13zt|Qoc!)i28L&%g)GA7i%tl`{b}^
zM-Sn&vf2qarFbM_aQHsoeRg8%v(SkV<jllhwI7!`l4$+kJN;q9XGz1K;>}$5)M*;b
ztF1@F*LygY^J)as*`;5m8hrc23!klsL>ea|Pk${)?0T2v%i-+w+LF8DF_*=uc#-BJ
z#g3|oV=Y^;>!@-HwL{9&L&wo4m*6jE<9w9y^ETmre$_38F+$tYOD>;}hx64MmadN7
z*R5M6UYe60W#cRP!VE9}P=qcL1?ohP9@}@g(TP~_E_c5S%QnOtF^;@!G1XLjb&)T^
zQCy&lW|nV!qGP$O*~CpKAL_{v!^TVy4!#O0$0UhwLg&Xgt@cdzyC1%@l%9%fi<VR;
zGm?sO@aWjDOI3{v7Ho?`S%H+Y^~+?cO(AUCs(_ILI^#j1cc>(5Govf1@QZ7se50)-
z&n$)7D%y*ev49-s9e94^$L`p^OVy)77ps!=!dG%0n0d`8nH3VI*)E<8$NXW6J+ng0
zx<JN>ks$J5?z~3bRMnvrk2NjRDYp+P*$(_|qtZjA=dIg1yedZ?`ynHZw3?Er+USo4
zaty1x&G+jhV4L$4q)f_P^Ebw0Q9wxYE@?HGks6dIf1E+<kITC9YU|vensaa7*69cy
zh~tl*P}X^E@LDtEYTdhcTkN4Rb`0#4YHM>B0^gCRGghaRH+?^n#=3ycK()5-vgRY~
zI^cD&l7ALoJO;ghE%bhu^R3k3R2J>Jmq*t$dkf7>{^rHW8Q9TsQ}lGmfxF=qaggVI
z3LytR4lpx<`Yd$dF~5_u2|*3UvT7F_THFY(>*cs>`3zysv)8f~?eCngp{=X*p_yVT
zuRf6ZEo$@LZn0+$<LCE$Dt;27%xJkW#d&Z$d&5c!;FE(404S1Gp4<AmOgf`=CZsu)
z9_?ck+@zAiEJd<Wm#^jPMmXjR<piLa-*xv*oj=-3_0}SU^2Q$I?tm7qY0fZVWr1tc
zszVhzkC(T#r~JL1->Z}MJn6p4UY5%Zl#m~R>ID7wSpZ=M?Zz-)xGyU5K+TkLG!5!@
zq@WotPg*+ep%WBuG?Z7m8dL*ge~GD%$^KRMbbou(WU?#ot3pg_Ez?4+A0Gwwy3SRn
zc?1U*89PUU4y*|G>BF&q$eBLUP+NPUjofxdg`Xje)Tg%}e0r=K=I$<0+gfn5lo}HS
zSyu`w*ce=D+k>cVZia<xSi0TK)J8vQ8zSryd;G$?iDvU)Hr5PgKKN~(!r7k`&XN%J
z+iX5J6&Shw)%P8h-E^^b)P-}^M5<}!z8A5ftD?dosWbtdkr*MP`D48*+hHpYZUQkK
z4bI34R^Dj_U3Az{kVOmlqmCXQvRQO+!zQf{mm+#%F%VmXLI7H7FnAaOi;|um|K6Fo
zCCb<y5Y@PgDc0>?Ua_jABKj92q{tk5Rviu9xq1sNiu3FU`gz%Q%8{wC)YlmYox+<e
z^M@X+-V@6;D-Vh>f+ek@mQyWe%`x=4IPtuKa!0N503=?!`-Df%L|>Ap%to1-G}*ko
z7Q=wwD#+-1$1;^GoNc);n!fyZjdtX@qYkSBH%~@>d{0gb8+(hs56OMS(3N>_@Y=f}
zB<i59-97Zl8`#cWQ?mg|lAa<>!S|JS;s<wNMJGpD_<0p#q1FAOXwzXe{C0Zi>u<S!
zk3zhIQ|IPZIte-y&U59`GF1C6u~|ysHS{AQW?~P^5H2(qNRjQ%bG5f=E^a^6czh6I
zI{dq4(^t1@RArD_LrcGjdo;pMSl43f?`~bObn+R)7dRk>d$-aiG`&M!%Vd>)XSuFV
zY3q?zmF2c^xqB+Cq|KS^P;^1g1<n-ba+$*m#93C!wgtWAuFX#?clSnJ6Y4*)ZNQEC
z8>J?A2||YFB%RjE(!%dg>7kjVX4e_wPeZf326oaei0mShpQlL?Z22cQwzfNd2u==a
z{e>t9vBc<^y}c#F3iGZWdU24X)O{~PJI|IWJTdx=`GB!$p(g@m6POXqG<w0p4Ycmw
z?Tchl=BV<{vbIVGwffz|_JroKK{NQ{L#1gE-!C}nrtFgmy8UZ&!tI5QXDCj7ulc>z
z(lazevhR~uU3Zni7X3Es)%0x!tuXIKWX`l?h3Q4o>U~`FxatR+e*_LYVJY-5nHM>4
zEgm*Qan4V*to2V^i76p0TPts-8csE7WHuwR8&eG6u?-Cr&YRr8wWXq!m{-QMpWn*(
zraBw2HjtIR=M>HrG}9b=(mUipgV3)_W6&1N8_JkySof=2JJ63b0A_uQY~S@&Ky@^t
zRj<b>_+`bYlm~uYsz10%C;VRPES|x!+s1osHp^D>Nq9{tndLLcK~ByRk(;P^Mot(<
zONc5TQ#)*u&uTVJVPpsG3R|rk&Dp~bb`7o~6H`4Tj%0Z3@UO0Anw6FrUk#{KCHgDT
z^Qv{3LQ8AIcL<xerg|T6B+GJ)8N8917KD9T(KgRKYKKe0J6&(RuXEs0O>PLUp)#vq
zekt+r-Mm+8;hKl>YQ}@-F48O^GfTYaALh0xI~2TgN{Kg=<w|F-$$NHGpOSlSpJ$=S
z#C@PsJA(M{F`4Vi_@&BMmlqQYsY-?AQyxDQ=U#rmy+nF6+?$q=E_~i~)e;+a2_ih@
z5PmA{`p_5n`g<+QDea0SXMIWO#nhVFPj@^LEnS<Q$yWODDkqhT$kM;PS{d=^M&#?m
zMk}T?+V2hh9Ex#Pd*12QVyvNlizG0#wb4W~6i|5RL7WZ}%lW5gc)f<=UuespP4j42
z_d^FNl?XTO`CBU8M_XE^d*6#FzrOU-uD!k$s-N#gv$y<*Zeiy_Egrltw8Ze=mwH#a
zCSr1bus7e9u1479;N_6V=i2xz%rYgK)aJ1x!;(?xSF+7CIfs>dM>(o<+g=L2G@*{B
zc=hf))StGdyP+=QBOA_fTqHzj@6cn@4Bt~qZ?<&t-vLI?WHzS5OV&L9^Uz|-YEa2=
zLQY<bJXuJVlyf8NavAk`UL~`j4A-^35v;+d?-K(oyR-BAh<_^Je{~!-goEX^d9~Y%
z1&_L7Y8Z*Gxp(~ou{GLQ$}iwJ46fUtg=!vVQ=Ne0ri6X^n4+w<YD#brTiro9)*yXY
zM)`i_Fykz$I))clSW@@AE5Y2hCuyh#*-+0<7&)j{6(??rJ^f<DyQzkPQ!<6Q=w}cu
zV{MS~&r?z57vfal*JJu3t_RHXDX*Ux=KE=~Z(=*-*^2HNDU;cc#?H2RqCYBy!P09S
zLJ#(;{q)5S#H`M)e9QP`%Z~Dm*8kU@Q^okrIyYY5-_9v<t-JFAVcZj7Af{JaQHsA@
z=koL07K}cnlq5V+Ni@Pz4W2Elf<0H*uw&sLnz}&;)q-;~x;F0%D|_J4(C@%{FJ4A<
z-IvnfkZRBt4qr=#8TDboIwbRa2&~R$ChJqM7F7IW)~u3A`ac2o&YxSwI+;cPE`Mps
z;D&guVT-_CF@4L?O63eOLz<kb-e)qw4z$Ldt;vhWaAm^oFc4Mo^U`FQZV|b<>-pGD
ze%F`^zj>s5%H1C#%Cw`GG8*>5T;<4KmdiI4UP!3?=~+@PUAV3+hE1}W==G8Uf$k_%
zdmtxOGRHUeBDhhNDC5)W$8xp*s|Tqep9N<V2Pu_vLT9k}MF&DP@1+clJrc@shKKqS
zORbep|48q0iMAn(RC`yDHqHgAgeP7Q3v<8jC2_K(!f*M$C?EDBIpX2lu9guQGK(dI
z{$z}Md{_tG_|q0H7n#syLaP{RIRRIc;Wte(HJXY#=*9Vj%nnLFz2Q+&)6;DY6HW2(
zKij-<lOm+${0JX4I52vX#o|L(I>@X1O;~!dy#1igJn(?!*VNxvZfRZ5Z|jqkHLG&%
zxmUL%0avTk>+<ufuPQqYuibmCEDSqvlF%E`zDxRK$83R|@~6I|%Ga81N?4c;LcT?%
zer0j&JZfbEHfGCsds3H6x&(L}O6F`Yl39KqYW?oaN(k(LVNo{1lIV|}{gf#1FcD?n
zi|XIq6OArbR35)d_?Vy5AS}N`Z~OJ)^Dj~6p8pyX$89(#Lf<e7;U3`38T7^S_{prv
zCx?HYzJ_KF2B7!sckbvz!|=MNbxNY7SbqV`<Oz^Q3+?5(%8ov_q%+q)CdI=;XSWz{
zrgn&~aHlbQc7^SD9ier5Dl)>dOof+cz-Lz0o(Z;0W-o)CldCJM)m<)KFj^7#sJvf$
zkl;_%<a6MvMf?8qh}*fcflp@O#Q(L2;3G$Oo~FEq>LcdnWXI`$A9dr=sJn1vFVGhP
zT2&j|_uuCuhbA=z{*6L=!!gE_<CBw`|M@UaY#(qU|GpQSb9@K4$+*V=^yvR`+LhKY
zof~OTwgesuY-tRci-5+&Y%Rr~lq(Y63~!CuG(EH@&Z04t8FSA&bCW&HY|K>k(rxza
zFUq91-_X<3yZiaCeLnm;%iou>uJ}agC!A50q<*;L5|AosGOxY-{RXjhJW)&2($aDl
zsiG{$?#`WFiBESQc9zzS4r!o0(3Q*<&ghYYdkTy*FBb;O&}ppfc%P2TtsGAq3v+X%
zD0pXA+^36hINYv5ztb{YFTh-b8Cny~v}WFFS{;oX4rWk%Y~%JEnkN#8h!(f*Otabt
zz4yJbu;!k|6^etF-$41BTi$18rbY&GW|Vo3?h320nCUAdpWxnw&guwRo*M3NPFv#R
z>F-w|YF82K#$qCk^K_$ObKl?G@K~-cDk{Q(pS$X3M}Mxv!D@+CmX>`Q<1bZAm$II9
zq)IVsKmbP(-Kx|PCHJ1}9gxjsi{Z}B&dl`mbX7I)f%5HajlexG&#2DSeoGL7Xh$sn
z-nm)k=k6w&>RJF-gj|C#f(U`~wG!lw`C_L#wX4wlPOmc8nRPWXwZAZXwclrq^V5oH
zIE7ELF11f+WDj}yZLc%1+kQ(^-EueH+)jA$;zb05QmXZXYz7flck6oP&=^El*1RGY
z9Oyf18hko-FE4K3;*;y=Ig+jRTh|@_UG0ZE@17cxqE~6FA<FbUhtJlxzzG=#+=#i+
ztkAJ&8>#-4BoUqUk4rsT%ob<FW{qBR#0mxG-*#d(YOzgFFe4-5a9K&o*=k=}W0iCC
zO26#ZaskAosHCJsDP(q_7bcg>b2l`lkl|cbtRf3*j+{u*ht&s6cD}u<XdBb3CrER4
zbkxS+a8b9z!`EJ)khRX{bg2*>bsqlsD78A^lB8PUQJA&kTqG+nXUA-eq*<=E%`rKL
z9X!L~Q)qX#U6q^3bfU0UjH9JxwZz~uVvXPotKV8@_=#Te9_XlY>y9`-xD4*gs@Z@1
z8|@O5PdMm%WM0fVJ@v_xCsw^nm%S+7!yg~*r1GCtUuc$dsEtx|ZBKr;#A1E!DDw(k
zn5!Gpm7(`KPK%xP4Ie#SxBKeUtjyTH#TgnX*|HvPnY<J?d8K%-E8Paxa5*BP*t)cn
z2-)7Sn$AOP$<zhkR<jXpGHlrW<0W@zxTzs@G~7K2luDVGm6g>7di=p)<)i{QIZfTY
zdpC5<mKIEPDljC$QBG{EB}M1{*T1(D&!`r8vu6jTu+FV9_Fzr55~*9WRm7HaXAe)S
zKKOojh5KErVZ*oUd}-gSJ;<X*5?G#Cnhj{oA5VpRf7gc04eg0RrK3;<AdlETAgr4j
zgVkUW;QpJQi0|foR76l{jk1$jd*{0+b{0JD;F?1|=k?~cv|_+S`&;nmPF4k`Irzt)
zejEU;7F{3BVcR5}Qu>wLou$SnK~K>a@elw_dI^TBM~TxOemUeaw?hehIxR!aVf0sZ
z&uJC6%bM%I-<f%lFda^7Yip@q)UwU5JNV8j{;D1JWlUT~EWZ{6u`3GSSj6W&cpu#o
zv$-z7S*EKzpp)J|si2#8i}Fm6b_+T_oIId_TbM4iC|HjHf&ck!aU36NaD^XSQ0~tb
zhZ)IT4fx1IJ(=M<w{IVrr<4K#^-x_gX5){12ga$!)2jQ)0YP>K1}u4_(2}>$^jIkZ
zRLgBK2xNQhuA=(nuwQF%^h?#_`*(YSy9`r2mCYg|wyGM)+N-qCu`NE4upe-47{s}M
z0J?d*)GSXo&aEeVP*FJCB#nAk{=|tBkBsCSCd#?%&&<n%G??aSCY?NS!fAtKq)=l(
zy=%h-S6ZFl7+FZocS9`_k{v{XB7)BBm4@Y7oPH(ggzqD&cKvFLKfqaKA`Hp-Cd_ss
zF>j>AuFBuk1KB`%p5bY?lu9uh-1p@Aqn}@%ZNCAHTlY0T0p)ygbK0QVveQd@P*b<V
zaa%H?J8w3?2d#ap7uO<itRd-?s%PId?otlXKZH7m{d@x9Kdn5`=~{O_C87?8)0T#U
zcq#NV$7ih7q%3ZcR&?g49gW@5i8d%WD3D~0>K}SNPp-eY%3;r8?P|U4l716G=?%Gd
zASFyPF4+YPe13AE0LPhc!dhEdU2Jsu^HBZl-e+HWx0$!|iR$AoW59Bn)Yq#jD=RbO
z1r_Fl337GI5$r*CNFD^S72TV(J`e}P0iqG!Vxd|`Z0+rb26EX9Z0nUgjliiq&ET27
z?JU2Rh6@pjk1bhzJPmA+vVIgsAZbuPVMY*x%jSnT6r_9AdD0q#S5i$leHLbO3eZyf
znsrTfFTwR}V3(3}H(JI0W9!xmR!q|Pm7+p#P^y&KD0M6H&%)2Kg2t@}9et3<`G)D*
zo*XTNqod=9-$=c`jx#{LZFRY&MZD-(9RRsITsX`L8S-GfEdY$L3vlGt^zH*EmlL*d
z7!0Pr3$5ZH91iwZx7GkLt^Rx_nmaCqeeT;jd{;h8&F^r?OdmbJzu0Cw=AE;>y?s@E
zef`+$YjNv`M6~}XwZ?tK0W9N##jH&h_zc(1tqQS_NTju2%i7d#5M5?0z`ul`h_f0w
zJIR>wSL+Q=!#7s|l<7wO@c<<l{5F5?p)hi`CUF8=vIb=o%z6)2N`(D*#0P7=-}b`S
z)PwKPndN#5O$W)IqYg!r{U)_ZW^bvj9ga&<RgMmlNq0)rcb2y`C%w{%3_*+cj|B3o
z&Yn5pb>Nv5Yo)G8dacdBc}1(5I;9sc0qstlDeWYgZ*HuiWMpNFQL|UX&6y(pUym~8
zTDf0?R~rp;iBLRctl4hwG&AK&_<M}$51GC%vzCA_9&zK=mE8vGads0}*sy{_Z3$OJ
zbj~+2%38}D8(1TNORu$pXrS|=SRLi(Z#*^j9;&9_f}R13Wj$1ny44gi=VnCW0=DMa
z!+G~wDP>L^>Th@w7p6SO`A&}SZY7l3IQqBg91xVpy{@bb_?~Z))f4sS=QI5)o_)XZ
zpnrrl?|zK6$Y!xSWwDL}AQ3ed`NpU3*xPH?1B&{-1dtuIzQVv3e)#Yq6ck6Uhz_E1
zZD;I^YuDd#%LDaa%wvKXBkEg*j#2CM0*;aSaIKFZXYlC_ex0BkaXIG}oHRP9t*tE*
zYJTn7C(21Vd&@r^acTCODp7wHLdEGo5NQJ3(Ksx*3_yAXRJ57)V*h}LRQAKqAGu}K
zO&}a_0yVf=YF9Ot&5a3w{xv;N=8;lo5A%tDQuC5B%f+F_aO?g6O8M=ZBZ5(9R6WyZ
zL1YsbHQ$jiYcA2Qjv=;5wLZf=xuBWe$50n34t_b#0D!x}7k&+X|NecPk&zJ}P%&ic
zW$%H9{ghINtYrq(FKt-YR0@LcWBP|PO6z(6C2cjj=Y~(s0CBL=1@iFC46+$?V>YS-
z#xN*EskVA-rseF|xhK1pTX6vS;h+x2>wj1K3{|iFY7=zEeU<acC#D3UsNSM^H?fG+
zhAA_*hC%gPg$tT>ppp(q6#5Qd@l=Z%U}DcCoGE%cA-LO1aOR_u+6StU8RKtMo<&lt
zSzqq6)jvA*vQQs|R)bRF(9dOK+Y2kc%+Y3q2d!~M8@f(y-3s>2(G1R2=kkz{R){wq
z+p;e6^L0S*hb#*@v6vS8W_EnjecHiuReKjMOalqjoF|?Gif#LpM!>{sB(pg*T=CRF
zYN>--U$G6=#7t+UPip%Q9-^CQDzJA~k@!UjWNd}n$)!Y5kcXql<*6(`n3n&quL_z$
zzi*#8xaZ3XgUas|G6*MdEBOu(pTqj1(LvupNri^(D!Uf9r&v${rtQ1%<CE8c0|$!t
z@8AE~Jtvrw`CLG1%LLi7K2yA9dLsJ5+D@LE8>_9`G-u>t^&dRcNA}137e4X%Au(JL
z2$;~MUV{ZLDr961L@}kYJ8;OOVAFYd{1cl2>Pkl=qQ#Pt_vXg4En9F@0xpxZC@}S*
zC*#j7O%oyZNZ*FZ4C31YZoyKx3(r^enH^7p=r-xap9Gp&apawTQa2Euk(GhdJ)(xB
z_7pJ%Ru<J#&Z)6_fvZ2Iq1-y^2Ks_i2l}|^lC<<*QGmKaf^zmj2=JHDb)@}!;)bVs
zaw5N!-}_>H{NzdN6X3_`;iuiQ)7@F27K*1Xje&Qz1SE|gFhX?}F%iaOe@9PBNrT7F
z&rja{BeKAqA8MFX8uZGsBxH_N$8`3P-Wwfo?|QW;1&shw^VXO@Ubs!kW1E73rfzG#
z2mj>&*~7{#9v5qC-7=ZXgrrQmaUw6>S?npbeN#DBJ(I2`0o4l~ef!i&OEx)AVK=YL
zgAoJaveRYj!b1gei0721DtBTo7dR9o3UN5xr)Q!uVc&lDzVn+KZRVubc=ktBa>c+@
z7kYec!nBOx27!zBbAgUrX0Nuh{Sq&(y<w_5Yx@Ee5Z|am4k7yYn`c8-*47m7v6jtC
zb=UapfkGJtxiS0v^Woj0YL7^~DYp#^G-$_jfaoojxOjYjzb$C%8KTNg9&spxiXF2-
zavAU#gz((=r-h?qc9&SOzXKjnv^J-A__qykE%OmF{p(Vv4n75PVmnjZBwKy))xc$L
zey#zsF9)yP9%C)hcFWaxk;qvdXps$ebg1{8K4QAMhbwsWp!Q86KGTx~6qS`qwba@A
zGtOsV4tTt^^~d7_*56F|_r(2_#o}0^F_iOAz={!KA)-)oU<YHgK*KqwU=UkNdCxdb
zX4-`!<e;-CP)=KUfMDH6KD|UXQ=l~^YbJo9K-CD$Nfbg;C5@gQ5~{u7{0`{ipGU=R
z8HX5VB?OA+Osd-8rH=NmWkLjg<v1XZC!QJU^UZN+y`j}V>i%p`j`_LQ(L|xN+6@VN
zCNP0-?+;eO$F~=g5RSoLkLu&hfl%2VBzF=BjP7iWB{Q8!c6Jc`J&L_-|8z4{e&&DA
z*ZNMq9$M5=_Y%N3bSZ#t{AJ&fdLQ74M^7u>{rEQ3RMW!=FpG?Z(>!o^Lk1+IgwbJO
zLm@XQ{AYeX9B8?{*I>Qb4=Z7mF?-%4d|4T^=vjoEJZK0$pj02KXWQ7@8#>?Q`9TJF
zykDXasBLIMnXqi5-$$U2{=x>%;vcK|4E1&S<k|oyBMfB@YzqP<Y3}HV`_i48w?3Vp
z<(s(zydXbXD|Fr;$ch-QdT!<8%Ll6SHuC9V;dSd-&G`6u2dMdUPfqJY!nP9rY?YFN
z0=BZKNCpACMZ$2kN0p>-FkJcMCEM~O@!+|k{u=&osb;$2{U&O(v>6$3C=mTQ%qw#r
z-X*TjH?NM8S(5}Z^?1(>8T@5o{ABr;!$!Z=B_c~xfiBGgx)cpT#K?L%OgFyQnM|{Y
z!L3bYh5peNnl{iRc$FoZT_{>oMM!?tr~_rL{`w(CYnYUtf8OU^9Djh5+gvnBFw3-d
zREINLWs?`p9<-1!aRT-CX_-!T&|d0N(T@UZQOeTFYPTq2W3iY602J=+{L6fv;noTl
zPye49&K_Vnx23iK*EoWLPNKJNun4+b1^Val0r^+o>;1<C?dyD_$paNGTKSbDfKVIW
zxXOIKEp4d+I&JTh>#D%{vr26g8{dT$l$`_!L?3Fqe(s0?VK0M+r|?$78936?-hPIm
z^yN2`_4)JXzjp2cH|r!PvHIr$43_I6{Ig(--un(e1>AuAhoYj+Rz_J<jA*#~x0k0A
zfEO(Y;qoFkxrdXWzN5H?Tya(3k%4em5u<ze?mZXx0n>}`elUxQ@TI9nB)37mL$V5h
zrfK0^1s`9tuJZuSbwmaV1|)|z%A}4Kp5=Dfo<nESq>!zy*kE%xFAsyPfWgJoEB{;k
z!$1p#sMpCLRNG$P+POM<n6;m@`r~z8L>(9DzdSwUpyWj<nTCKChD=XRJ`tybL4fwC
z$Ce|tJBDB&Gd3R6Q{Ys?6>q>4)Z7L(d=jW6gfAgOjtGFTna*wdE27cMy>E<A@^czs
z?hyxzoU5thW}yIS=LV!pouN4(YUf-fmv8at4h^1J%|uOx&*bF$n51&J*Gz&RK#}!v
zfVpgYU^d>R30}CkzNZ|}T&rZ}e0E@uovvq}G?WC0jWQ2(j8n3w$OZ*64c_qrGmOw_
zFrLYa!V8BUHzMT#tNoYw%1pmkBl*D~pyRO%Pzec%+-t7^0+zlJpR2%&w7wU^b`&|(
z&7Xu0)%(Bm+T{Y`(89rQjBRh(;C+Swv3ePq-@EryvXud~tmIa|go1*~oH__|40-*c
zmM+<C<1ExYyt{u9dYrm3EQ_`&A@-=a*jV|f1plsVHKXtoO@d92^#592Iv{VMJ~7lB
z;z%zFja;4IG+Ldfjpc$qjs;K~8jvYP4Gz@oG^iXK*KK8nbTpp6>el;Mu-_1b?M56Q
zKTOMmgfT+_&ozb@MMOc*9ob9FQe)OP1WwO?cedzrk<n70NWPI_WbW@@;f1)3o=S)C
ze2@ptlgE_sSjzBOhX?9)Vmoi!a!ar|a&=2-P`lc1ET%#{dV6!VB^R>EA}9zMaOEnt
z<7GXq$f`IX62s;bje7WZ9z`*Rk$30at!^TarNCVk5kL`JVc)*JumGIS;WP}W^Lep?
z%Ul96gsua}a@gNt9(W?5r`O+{LG}SIO`x}XRUXz}_-KW^jM$2zx+jlCEtx#)kxJ!y
ztftys8w+3fwG7O`WJ{Q@zw<BGB#+?wr~UwBjBX625fXE_#5Xse!Cs*{kKyS{-e=>3
zGN_Kic@Lfg=NsA0tQnb!Y6;tGWUhYPd@`f7Xr$80YBImGsOZC=%_?D-gAu~{DmcGC
z9sTOHQ<9Vzc3}GYf1d?#S}79;Uim?l`0luC_`d_o7*j!MDXy(RDVjBT{;L$<+B0ZP
zs|sM8BqTxSFk81b1=|4Lq(e!dPMjRHU!GFBe+LGGO>*nEk<}@?oxC=3OB~5f!;AQM
zQCWF;5f-?4{uK`Rvwl60)DYphq3(OnglmnF05K>%#99!t0<`#C6SJkQBR`sJJofM6
z{mJ!$K!5yc0ABAR&0xJ0F+;_QYJz-^W?;a)OL*(6fZ41A_;M^r66EkH#UD}ymr<|?
z`UNNu7G&5sJ_+*CBbQ6tmpcx=t=!(?pt!}~xbXQvD+*KNMGeW;4EBs4po3A+!pEKb
z2iZ-zV3w1;0oa@Jax<#G$VwGBiA@NUbg>{?d6*e-EPYm~ds5X{yD;Nl?SJpfQ#m9C
z2sPn_jU^)U2nGP^BW|276;uzZpQK;HHfZW1TN%!EURyUa9CC<)YL8+IVZbCL5a!mG
z64XZPM{;ABfN+jGwDy47{tR5xhOR9Y483A{sm=(U>iYUH1AqU;@s26*tm!-m*<&tc
z+(Gu$`Z(aXFo8~L1@B&pbHEfMYRHP=#)H<271U4zPEP<pWKfqEwwZEiy*CLUUcWBT
zHD)q$ax}3cdo2MxFjtQ{01wx69mxf)7*oNo$Z9aX^KAkV5>3N^Hdtr`T$`f_P;w7j
zp%ox-c;Hl5A$JfQGz=nxHJQF5!A-U)n8sGteE9Gne#{2U@XTil?*0D$#z=w=GWgrt
z+B$=^(gDIUW`a!Any<I($e$amn$8FJ#x{j!s<`*;1HPg)a-uy2+$I>yJB4;-YgGL_
z?VjE1_v!I&eRVl0;T0knLaw!lH-pK^PSJ?pZv-UhVDb^&FUJH|M%Z!Z*%=s63JF{%
z(d0LZ(~w55FM|eyaKLb7UI9A(ErAoMxs>_xu&`zt`**KPqj(PRm*eyG<B5mai@7T9
zrkiiyzBRJJ!f0*Gi0f}|cgG#pd@#~iXyNb|7zjxp-;Y}+8OZ^S<6B%(a>={zL+^_e
zF?~teukIw*L=SoFy<h)TdH`iwW`SymbmA%$P2bsp3gNkGkU6_JFmiFO@TF<ut^Y@f
zm~Sw~EX&pe?0kcDoqu^y%V^AC=5wjsKdDaN{BVokgvKHmrOm^2e$MuQgx-TO7IP$!
zsN+62RMQ&z==*yklpL6ca%bYdhMrZSZqCr$^P)qT+|`r;Hx>tIYWuSM#fxjr@%TeR
zSNdLE_AqC#=Z2HEKyzjpk$}b@3$X`^ekv$E#g%Io{{sPGQD^N!do-pUFNd=x#yCf{
z{``o0)|I73thf;~{P$6@Z`=lP9@-{msuq7bs()6gV+(R*?_WC~U9|#raFT1}DZi@=
zz)*U76WqhushSO7EN)95*CcPXS`}Nrvo3SATLzu%$lXRbVd}x1s{%#x0}R;|KnZcT
zCtSf0Oc7YY*rVKW+6u^M5^lDV5OV4eAnqjHIgnxP@z-&OBbde5mdxW?%>bgg)8Yak
zIj!9%JJYARif-Q_<r82)3oQl4c!p8yJxIDdiqk*q|Fu}5d6`5oP0G$U-@XaF@92%*
zaNCOeZ4ChP18Dg`rsB$+niBso5!;ckhP9^w%f=CPIkR0slQn?#5xHQhcgN-1%Z8N<
z2VcA7_G*A!8$g?G_RGD#0w(<h+~#_Nw1mNQKfHc@?&md=x%DQ*_7)KVTx|hpj(Nqs
zmOu)1aA_x+n5z@|o7^Wygbkx#0zSwuyLT*yIA@PH`*$EZhzk!acflvBVeP8&t1O0q
zh+U%tposokJRNLTB+i+UnQOD+7XLVj(G@=#>c;a+)Gck=m`h<)D|D-;)HnvB`P_zS
z3U;GbfNYF1!%gl2TAj;gJz(675YQ^s?C++lC#9~5tOn|?b@<7^HHYE+4;7`3?5&5@
zozvy2&QXx}4mCh$1M{?|RvPfKR^j2{>A>|nfIdU~CKnJ1U>9LN&qZN5Vr8brqciQI
zm}bcAAAeA@h{MmmJo?X$)4d-bf!J$a<tkV(Z`fMoI!J{sB2j_MyZwb0)7(7v=rVCj
zBH2JXw9!ZM|EUOH&=)r1-0{4IeEKqXe{KtaSuvTbf(u8l8j`rqdjn*xDPZHzqFN|(
zbaO00NDY>YLfM=KBMm?&t@$95D$fp{R$c@c0zm*{YcaAeu<Wo6MpFB*1BlSs0snlP
zbGMQE!N7t${2dbx9a*IA1l5=dw6HVpqpS7Y!3Mw9Y&mrdY&1cdXgA%e$^bOnnS60V
z_8YkL#3MkCF{0ob!9@`Ck%lFGB2fALfS<N)k)YmdE&a}!Y0Xt*rvVe<^piyVD8)8@
zrO~rotVCRLYFyr5aM$jNc%*Tl|Dn{q5O}wlJp^6=kogUeQDY(8jV8YWZmZ>k-J%}Q
z+oT#Wr&^AL4QF?_q|9?i0s_o+wRA2Ac@>vU8!1;VZl02ZHERX^{w`WzmKU=INSE^v
zpC}CR9Vq|Vxv{aqx_a&iD0?;$jIz*LVGBR>feustda+*C3frh53>QapZHPr9E9-+d
zN{5H8e*os{q<&V{0EunM1$#mH&FmpYhN>4ONFHba<g7~p7`3cAbFmdZiVGJEVNm3L
z-f6tkY@Ds`zYQ$#2&n9(CqVZ$OHkKRFra}BIoChDmpAO&G?&4hWFH935mhfvg!35`
z7^nF|{y*~GJgTW|>lekM3d$o*l~xrZomB=XN(2ER5;`r)2I*8f=|LnYtu!HY0W6ME
zRVZjc8fl_}Gzlbh0cnz;C`cO-LKhNHnn)lCT?i18yq$2)`OdrFd*hCA@B7}n<G%d|
zjO=dBIoF(X{pOr&t<4Spsj?6*pG5lUz|B`nhk$iVz}7P2vRnmd5skq@ObTLvR5COh
z$Unoy@kvbhQMeB<(%%MZ11L=!01d{5QF6k71Ve<TIC=g&)WwN{UzS=#RvDHA6AHg(
zE_h2%<2nZKh-+JwCV(4nl_Z1^fY}D=dNfwXK3_Z!0H9M=R-_~Vap*%0Q&w<gWgt>z
z<>yZTU~T7$!Suy))ayLo<Q8`zg^TqSCoKI=z6=A3tfr(t<p8v>|BzNA4G7bS$PrSC
zNc7g`ce%-eUWJE$dF8A+FVTLx4Ca_`C4Uv095a+!T`FFy8|w>z`fWg!lV61aIp?WZ
zvIXxxk%`9SLXCitg}@yh1l-ZKz+j}rB9Um2xVfA%XXx10R{UaFs_OauV{ShuR?2*S
z{;I+cYm`bO=Nkk(v7J0NwhffdC)|LP8KuUvIcp5)d;r(f+?}7FhXL+94rs({63~?t
zpkqT0IYJi8xGjSWfX%55zzH*{K&@2+@I`Y~lJV-*(Y(Qy#)cfVqb@m-bAJm|q;xPn
zBM8RYX*29!ZoX<;uR6qqPnx|V%Xs`_A!*022Yi;DA(s-aesDy-MiL*zl$PDI(Jg&_
zNyQ<NEh+2+e=@O7ppz<m^utr9U(NqV^<Nj_#C}ElJm!R4`=py8?l8%5!^Ymm^Sk}M
zSmcN=^kH2ta`18EkHI^(YYJ-z|20KG*YwJx)a3*luIhqL&MAvvqSCzjoaVo_7jj=~
z{Au5Wllj7di>=^bI%8h--!B^aK1vNu`*MNgN#*sZ{88W_Ue+u*>Th6x=f?eNvQx`s
zOFx(&$rdkb+dFRbKsA+S9ve?bUs+^lT(bSQhw6Zbeg}P+w@I-0g;hMt0OpbGJ#*C7
z#|d<l3hHUxMq1v#l#)?)=|849RBwqJW9yXRbZ+^K7(Vv}RwIfg+SY&WW4F@z90z!i
z7Rc_$`!b+!etrM-PrrAq_C7s*=jk7R{PEXO!>Y>KZ@%02t=w}ZRczc>-^acDSxN7A
zAH<pO;-!>tT+09A;ui)#f`0tbhD_FA+~7X@e2PNhk5qXxxE@Q@>(n5Qpe6#&!eV!d
zd<7=JlHr49*N5bF@4yG{lal2boJt~oXZ$2f!b7b`MPo`A3!_jvD32lEkdcVvBtA@h
ze+D|;q1#NcbnttMJ4IhD_QF3z1^m+;%smJ0_}RQ_=fOSqhWHRXAt8?Y*FL6LFkYw{
zN@3xy>xSVrj!DHF_+d(RspFu0b=L#TAsK8u?o08GQ5{LN=RE^Uc>}8yu;z^S-@fp1
zovorYs6b^n(U@*$hT>&l+XAT>3}o%RHmunyj{cl_O-UpEV+Q|q8@=tb$+Zh#&{tc6
z_<5sod`R+?zPvNzcK9jrW&*e0iGjpuK&DTO%F^||(VC3#J20&oHB_^SduLPiGsz}I
zwqbwxwn(-f@Y^S$M@s6L=AQZ;)(Ke|oYs*ZJJc(g)bcnLZjsa~0H2fmI(&cPDbNUT
zd!P(w2ZEbS%r|D6|8`7#s$*+^9D~!&>|fkcL5C<b&<AzJ1}r6Y<Ij8CmE3~W8YRk?
z;@`E~UjsL)8O;tWjY=lV)GmroNE^+z=~s*Smhv0^79gSm+7+aa-Y+?^9+{2b=3nQ|
zEXMJqxS8rtC7V0d?*BS&tQ#X+jfZ!5i0|w8!~MBLJN?T36JPrzo!fD7|DOJyzqDsi
zC$D}XnWmHa=-{5+-_O}o{{NC!RE1@Pr(PMqN^AG0e{(uH^@y12GeGw<ikPd?W9^+C
zT=5w{KqBrBqA_G=PuWUXC-Wt<cVDoWv5u>UR_c$Wm?Pvq7BTlz&Ph|m{gV1(UQ4`=
z``^3qe;sz4oG-X_yLw_QC=?2wLi}n|(&?eMzo2uQU$iE|NX$DFZ^fu(Sv0ANevN>c
zG~-FUYwFN9D_ESYvpzN)-#be#vP>@B@#1DsLJS)uF74fs_N72uE`zevD!k3>Wqw**
zzf*LT(3%iv86lR%ncrV?*Or;z8c~&89I@=HYSVM;w||ALZEXLsxD@2<c+)VE2T{b9
zSn)E&qD_@|>@Z7du1<zQ`O5atRiO{}8PgUqk37IH$ECd75iG8fEE`0BQo<O+dgAw=
z{NW4wHh-~T)fOG(ny66MMQ55)OpS1Ye*Zld?-!-Qg={OM<Wc|8ghB3zgzwyk;**9(
z#Lp)~yrV2&lBP95GsUE0quH8t_mItX7|VYYn3yLeb>0D>D56*y6`7BBKq-s*{{1|z
zP?^*SiFyEQawJ<__7>mow_j|6I4Y^=wL8TOZKihp!7w@n!#1GNT?k1J{~^`Mrq43D
z9e9cW4R&lM5h2YxH@I#Tv%;w))|%EURul6LbHV-#ng)g#du~A1*^S&L*=fcv;Lr*-
ztJ!&)dc*zhQg~x*Ft^dVx}iDYBA1u8w>{qdT+Jh+<Y^ahO)_RO{tpHlUE*2nCBWU`
z@3GS>N9V%axf@zPa&A$Yor9d4apB%-v>P_*2QXPK3>OguVL5X7u$Yzcxi8x@>M%R+
zHw<LDou{Wx8WAIm5Ol*cC&X8;1&ckVBRXB)Yh6!zY|jRc%>)|zxYMcy;6Kb%D~#PU
zKN(4nh#O9Qp#@x`pZtf>%>UD9Gy%7m*pwoeY}T*;R{UDOx-Z*j*GGf-G0W!z#V2at
zWL}Grcs{A>2mh^G2$ZL9n+47!{2DL5_36PCY`SB!sQjk*msFa8%6H=ber5eL-o6Z%
zM8$@#;`Eym2YA?OA7{7Nt=Lp8#Tqu?KT0jo)DYjBsykGxP6t$j`}~Rcx0Hzna3diA
znzZw}b?<~4a01~6P2$~^D2`MpM681&{aefDj3tgQ)H1x@B*nK{-CT{M{**o>J+k-U
zzI~jXIBqI_Q{kG##U3`WJtevFU+V63{>QzLco07Mw`Q#R`;!r0C($-0@R9*5-R50a
zusJDVrNfwGc-unE0<bz0P=<KQIm+n1V1#Y9d%K46(;viFbdC4vV^-OV0best?T^DL
zSE@_So&Za>XOuepKSAEVhRgqP4*%bI1^of*Gjl7r@41RZKTOmCMQDxWT2l^T(w)Jk
z9R{8lRFz&OwhCJv>lT3?^g)?OT#~`K*iF34Mk}e#aJ${~pJZlL3tzN${}qr~$u>f?
zw<r9DhwFJu47=1HKm&1hvI=laGHFM+xJBYU{>Z*$b;Tn3DN!u))Nc;~1NtgOqdzpI
zb>}TJlG0s)6P$y!F|Q1>QmWYRZ?xR2p|CJ1PpVcMwfZF8DrIX2ct5N6LJ~vKxx#pV
zlBl*dnwtrDC%jR2Qc(5^oU;*`oIS5hw{YU|B}SV21n{lNe$d?ml)(a)w>@;twb&FP
ze5(a^@w`RuuRFdZA=mbu{pbRxx?nX?u~sB~ylEQt(4&EYf)OsrbHXW(CjBa3OdBFB
zp{X9abZHFdTFjxP<u$F4VBe<C=cQaf<}>p!g(q4c!8SP*kyp&2GsfG>pL!@^O0;(n
z{%_pZ!P9`O{no`C*Fd}6)ci#F<4riri!~Dg5iJXu3N)OvqWa4PHCb%;_4f98xh4H-
zpZd>Qz?+FD7z6G+V|O8Q?Hu@v7*}rlIdI)kQS;8s^Fpb;H)5He7vpSww-wmA(xJZ7
zZS4)+Y}4IQS5^<j^U2f->$c!_wla!1G@+VVmB4my=Kpmd-Y9|gRo-TkEw*xA4u2;S
zQqHV43LkOYona}_`dZ_5OgStx1|CtB=@FxU##jJ}+2{eYaur`rKTmX5Qfl9nXKn4i
zm`^+}HTowYU(3lGf2N^T-14f&Q#ygw5|D*6-6im?nwrV4<iVb+oAzJOrE84(FIq4|
z9Nqr{_O<kF-E2)&@^fb2rXEVzu+E#!3o2PhD#lPK#DUZY3-7WWm^pk=^SS0aJ>k-T
zD9RfEW=pTdUBoh)l_mrVypK1LL`U&NICOW=7%p+4p??GXUgALimL^5nh+5&IUz{nr
zxwHFRp`FO?L2T<zl9bVB#hT|=^kw@xZ&V9>0}W~FnQUOVX<#g`7Qeq?!p=q)4+jk5
zO0FViH&*AEP89iam>y0T>D?;vG$z&7-f2AVkkgFIMaJEe8THMrDO~!dEVE*8dD_XM
zthGGhH|`ka5hYsj>iUJksdjmBd)Ka~9~n*d#v{}pghp|24iN|G^su)}Rv&J}hEtBC
zJ)Oy<5{MZ_11(Jcr|}q)CtfDCWVym|4%Ok#-dEH<kD~P}ecT1VaZgYpDJ>LF^cc8c
zQR235zuU)o6uo%8#mf?*+J9Fq>Ce4gPXFoAj{kSv(WFU5>jk&z%d{#89uoVQVG-{j
z5EBjzCDyX-+Rh|z!RsSrr9CWM{o&z0jt(Uqoh?%>0pcCp!@=_OV`wnmYD#dK8%7yG
zTNriIs#^oN{RLwsuC%xv$;<p6-}_q}$D0O&2yeienKcwm3cX-fugcQHqO`Sp10UqV
zt0tbjycd4j`d6m%{EcS2UCdk~2iu<fz0CzjmfJ(+)`oL(^rd#z(e5z!IzfdI!dzF<
zfNL1=%t*nilG`#BY~-W7;&a-FIDLj7xu|>h*%x%0(JbQlrnjMyCG;(ELZy}#@>KFl
zsod9>_i=)Px~Q(pW)aY&((n0A<oj7_l@d^y`r@Q>2cKB5x_W$*5DyHr-}Bk0j4byE
z6*mgs><ad6xIV1{?b5#~vZJvm8PH8a^1t3~{so<Lt&U@bR<!S~Fmcc3J)oqxOj|}z
zWymrm0MJjT+Dji^V|h&=h)Lpq&);<6)mUCG9={>OLAa@|M^l1i{&}O^_bY7Xacio(
z(kb`Wk5!aXbT*nmv2rH)5a3TOC9LLl0n7w@T@mzd)Db3WH)8iF4XG{J$Eo%{u=7%K
zSE2Z>zMx>iIHW|PN41SW|7d);F>x2N@o@9@poFq(9v5_5C8LDYDD>Eu?PCLUTHAXG
z)|EOB9ookkGMXjp?uBTn-+J8WXnV2KQ^11jp(W@k)xe|Aew02X4E!Ez)Fx;Oo<2J`
zeO1k~xb9Os$;(>5I*Wr*p9V(rDgT|MPg3{=)&>)fp8WxEn~5%5?Tql>U#yNZ)0Jp%
zAZ_TK7t8?6>CGJGIm&~D$)T?G1*UsOIN-kh<VFv!(A#iwaY3{YdE@CT0+ex4_fFhJ
zE)9sExE<Vuk3dAI+xq8MFQdQk86`(FC4CH}v=#17ke<joD?W+VR~dr4k~~@0VvYf_
zQK5N!L4W2ob)$*XhaO{Sg{VxBT~l1c)Six#-4v@WPoJrYr+RQ_D@vx)!lSd#@eJX4
zoT*_)f}s7#!nzy=k7w$05DDF3o*VTm>%3k^?++MDL}ZO{dSVpKFN!KvM@;HOFNGZ2
ze$<@&lE&Yp%z=}EcDH^3*ic8Lj>R^K(kWqpA>S+KlO_#9RmYqolq*Mh_NWW%+a%iR
z1^DJbSMBpRAGdciy@6m}GZt4^@?O4Kf6bzsyDCBmdyR|X839>df8nfMJexyX^u&R@
z-}j?y!y9*?lkBjRj<z7yKhqu7_Ia(VXJ8Vgxb_0D61-qgNAfGAX`6FXGK$vP{4+!j
z_G#9&deLeP!!SQvpXR(+zqOY%SXYM32nLy17@YU)_CLaBmm^Vs-t)x@tMKV<k{7D}
z{tUWO;%X5A<$q$jQ%G)Zu6&uDy)EGE<iB6BjD_xIGM9rHXYdl8JDvJ0_vTA-|JI}Z
zaLI{<aKOZyjApYGoF(gRQqvCqG9y^-k0YPi!@8&c02mk>xSyp_sU>Mr)&Q_wH>24G
z*}b$u$63IFgo#|MidoSWHM6}SNebJ2d0zxzOQfoBA3_Brx?@uEaCL3iQwZ+KC?vcU
z(ajW40x4z~{hIngX<6nT*TX76^Us3$`+TgbTFGZLB*IEnw$>YxJl_F15-h`!mqAO;
zcl-jd$Ua=DAi0LQvFE4%FOm%RJksDVN-+_YZ0BMRwYJ{%67veNf&l2h_?a~Q5PBM$
z?EipqU@vRodh1;(FNbMOv5_A2f`Y3ZuuaZNE^fL_YT_sSN|pb6-uYUrCtB7ORjG#q
zMnSL|%8iy1StAj*$^hUa2$(j9Zd5hfm4I(8n>34_(xhC#1Ud8aiJR5T5{i-xGqBVY
zZL=jPXDTX0O^q@mFd`QkTZ+%QMtHW9F{P>~rsL+zrz`)BKaW@+oDnQAEr+{h6{Du2
z(HhrQC({f2@0D;@m^x`NiG-~lHB|m{F2McKhemO@^`<n{=D@d`sKJfZT9b;2!0(0I
z37IW{+&&|!J~k6GSP6vno`l;_Q3JL8jzqU=FIj)pUNLx|^z!;N;KqAXd9Mpr1!B6l
zn+jG_e8>h?5Tj%Scbv&A7{gYT08{%G@UrqBBmz9?0X*Cm)WvsQZo2_Z>i-MT0QPGe
zJ3|OPRWWcM*HRL|HZv09(?$}Q{adnQQxd;ds{?3RaxAnx>@9H}+6DHgi{STfjo(R_
zxCs^AKOS__mjPhSZiTB$y9KMXe?3$!7cj;|P>RcEXEA-i5Yn=PUEE?*f}oqUTTpYA
zFC(;2m28OG^h<rzP*QPeP*<D6)sWb(b({2wv1f5H*%O`C?LXe`TwEWJt^7)yQ7lUc
zfMuKm*AH5b%=sK?BTe3s=M&bhEeA)<GTZ>XYX8Zf5&@nfGXiQW5}@25z`ed>+EdgR
zfP0NT%eA0<pAr7$_}0B0#knw+?}g2fowZb+Jf)$0R>?VK`UYY`Jk(_-3dGV7H4FA}
z5NMKlhd~66m<X|~pNzD?LxZGK!XL6-84i>hAlI38XKA0*$@Yj~)0F7tA1Ewz7WUJ)
zM#L89arexR`{UBrN-1GK(B6Du-bneX9bQcg7$k)L)?7)f+vx`2Z)lRWO#9e&CX--9
z6%Ea)tr$I-y3ax~YwU^cMa6e_K8Xb^Y)V**Npj#$GST06@1EmJ(~ZzPFUOh*xUR^*
zQpWA$pA^oi{_4L6IcG6vY_9+j{hgw?g<><-YL+-b`bj_%7J4VHG8DJlhg0TPX)}#f
z^0U$h^9MM|z8Wb#wo0_F@|ell<fcO;wq8bf0Gk-cz5wBTB@#R&n=Ry4#4H#o#`ZlK
zm~Cf>{C&+Mj5+%I?8NF&+tWXVemqTC4v9Kcr|7J0;vT{y9?#rvAto)}8<?9$i$d2j
zm+jFh!3P(7Or!fTXvV2{hNy^1<@Xj+!m3KF8U)Pk#>6xHBjH^uy|-0>{8V_~CQB{8
zl<W0lXk=5UH#O@b)$!%4tGjApq&vQE3iHCir;~NC&%#35?=^+eTE9!Na=D<|BEV>T
zc`~39_B_^je;CKpVpEE*f3rTOlhc&4eUD&2I<$R{=_tbtI5z2K;Rofb+rR$M4c^}#
zvc`k+n9rHZOi`6Xi7BI;;%n^w>6umvYRMGXjtDr$Q%Bd&OSw6duF@)dAPrPBpF7`Y
z8tTfysarN6ytuEm{CO>u^(TLCCI=tfw5+rlPqgR0V18l(yOXlfY#QjO#Oia$pR+%-
zgq+oYKk}j9SYU-<4nciL!yA!DFW|<BPj|3dwtG;=Hryg;MZJGiYiTe%HUb;&YEjj?
z@y(zasi&Im<Vh<|61%Zf&Du)k_OTXw?mK4Q6J5{Rz&ciShMIkN+2(64a2bDxxxoy<
z1eEiw#y}&2YuTE*#M*}rk|*WEzTL;U0w5TK2dw$4q}>PCeYvZP5^p~9u*MBQJ3YFT
zZda?Z8#ebI(}@mySTpo*>k4s>mC0oZ!P(pZx1DR;Hp;Xs3KhKv#0q->KzIxt5gc`_
zPS%@j;8-I>?`)`AN?7W+df0Hr;ssDV$YGX!&)36-n=}ZP_gP6GSI1Szr<3Gq`rN%+
zA`e7ULb$IwTL^<?X?^5iJ9n>%dqArH+L9dI>{N^U<ShD_+p-xc0h$ei;k**}vcanT
z0J!0~1!3dD9J!wsXWKi&#2;~yO2vhr*vBb`c5$%Dm;@($z{5=OX{l?EJAsggM$twW
zOK(U*<&Gb(*>;HItRRe{GpbtMg?9Uoa>>10Ofe37xx;3npgwJ3+zQG(B?O&VQwNk>
zfCJN*`%Nr7`!=r)Q|RJ|wzO&xrDK|1@?0IO98CMwB_q!1)Xw6-L>rg`EDDyz{N9M~
zP+j5)a2kRaoV1c8aHNdxd7Sr;;9{B2DEldnt{SKbQO)s8(duHPq?PI+^U)qceS$kI
zo#{Y1fUy89tdIE@^8p2cDP4Fk(XWnj6OPI=L|MXhm={+-F>_lcMPxtZF_~n>lmrWR
zlpAtVc@Q)7nA@~74V{pXT|jI=m|rhWkWe}P-RSX|VwgYIlVU?@Hr}7SNBeKoAGYdv
z&Jo3y4rulI4j&0U=YzgAic_4cXJ-9Clf<bV`Jw+PC!|_~{#jx9@5X5VsS?q2qIKW-
z*_Zd?t!Lit3+~8QrL1}L++6rTE4Bf)u-)f{*MizGA%HX0icZQ7x#j-y9)KUd6=%Pw
z6r7^hdaU$V+Jnz+Xz;21@p(N~uA;CZ``S}Jt`?y~_SFqEdtpD$Bj#A)8x7YiZy?<f
z8o`aGh$IME)e?A1<Sp1aJS^4Yf(sR+COz`|1GWmLv`M4x#j&jM-Q%B@F<3gGg1P)v
z^Xn+nHs*L~GyY3a|71%@gc+Pgb45f05WC15)<5J)O?b0_dea8@!r0}h=-*%*ahay@
z_tZSf7D1e*b9SuMQ>|1Fjmt7+<yr{(sd}vsTBvpKh~~HfsuE4cmT7@YIn&(oHd5@2
zg)4y!?s3vO2l(N%!*H8wZYzrY@SK|8mM36oCu|HS)yPPV+vWOU_d0svf1(Ti*Q_V~
zrwnE=^|{K-)8zU|h#D#8(F2KAs$`|Y1Ng&0T3dHd1&Q}HkL>9ykdfau+des1EI;E|
zwTT@U`KDh{5nnoND$X>Q3;J|bGs)^R_!H85<%{#Z2PJuqy$axj3xfR(xdreKzZ0YO
zAg<q^;u^#UpoEZnYm3HwTVq*Cn-I>cg<B+rV47-ncJ39dBG&f>iyJv~!5LnY=@oNq
z>1G61sMWGx*8*sjhD|ujU8iU`TX_onV8WjYA-|<u6^sB0Uvo*q_u6@KbxG4_tsiD6
zlr$yUm!V$*xuCZ41Uv2rCZ19Nc&`dKa#gH87VnZ6080?-ZF7|aR_}-;DXiN@bx}jC
z*;8ZFRCHVOTBvOS7u-@7UOy4G06B8tYf{V+CK%&eY$<Z)N>gNi7#|C6PuMy9+LUq#
ztuN;+#<#iju$1Mqw2kG&6M%NMF%O}Z5v1LXbItoydcO%l*C({0eF11{ISN`qq-1NV
zW_tJm<!_*P_x1ibrP)sVT=Ol;;$}A5(Pg^55-?Q2A+LVBqBmH=gg@#3gz$x}GpEs3
zxJ=J~Ja+NxI3<n~g@(T1mWa9FjH)4Oqa+1`nWJC+q0&hQ0x?gQ2#iCekT}`;PU4<Z
zAJ})a)g@tfMw#7(Qx7FwHj@zM86%x^7eVm_620uY8HvwImAC6?i$pxLc_*=v+h=n3
z3K;(zC&$G$;`d#+2dzR}ghlMM0~#Xkh1!uX+pD`W&>Ctl7rbS#lF`HVLPuZlu0JfS
zv%i&)2-Fr(hyT5{(DcvR!jb<!))w{@=#jXfc=|>+Fu$U)Rsvm8?;54S5yd_NcbE+x
zx0A&jLmyIn))h!`8I}mO_RW{=ON=~>qFYvVxFLdtr8HnNFdE*M_sj$+(T{a!xsc7?
z30XkSG;bkVf<>y4CnJTKXkiSL|C(tDRA}k-330H+m}y5BTt8zkgZ0(fPrC?5e#HIQ
z$ZS}g=$TC=e}XwrslkBlvbG;-HN+#rCH9uKR|Hig-`DW=j!zB1R;gdMASGPo&IH4`
zhbZ)T@XNmiopOg>|Kt5OeV9=y1B6(>VQUYRB&JUr2~&U|mdNa(lmLn4#GClioyA4s
zIRH7lkH1)<=p_k)b2EWtxkIsXO62$5RxcpsJ9o80*4UNm!`H3p6|S7BCWXsOiI_we
z4fTI16la%Z`ev5rLb^t)?Kvo)g~f&eK>hiz^ei`$(dC*#ll^gU1&Iri%FR`YADBL8
zG|R%62l;aUV8Vu3+RkM4T5U0%euusRst?OozXY5AeSOnav6^m36H=JqHB$+kE&u!7
zbsig-k)gOcMFXG`&!HJLbaH7%BId$WfNr|6k#NVjh*SzVkburc$y1?{dUqD!zDKXk
z+ET!-N~+0b;?_E#b^rG>W;)7`0<~=f<+u!*yR8NGl)S3;H_5+_YyZqN{eLnU<3A0a
z{}oh-747g{4qAN}$b1%=A}sy?-Pt(0S()D+z$w2}7YkBXO^n#qDaSE2(jy5sxCn}@
zIJO54K=Jy3EH5GasYj3_iuW%iD6E#C(f?J8xc?87IK_3~_qoi^%<<q5d_`+(Ytv+N
z2T?4ORC$MvHer)99LILEsBK+3n<+TM1yfWoaF;{|eJ`JL1ej#PO=Z5N)VRj)77TYs
zIHNGhbqV&+O^L3%r@;xKRl@cJ?=cOwkGr=y!t$XL+RY8{Cz2ua;^mh<CEga-q5S87
zv3sLu3zO6xDjiBJH4G4f`^^3JP>v7d6=j`4HxN@)`%#R2aw)jMP^;1&UHUY}(iu^O
z^NRAH-ry>_k*q9)V?mo%b*-$VNlCYM{VFbTS{NnUaQlPFKaP%jT2w%S7Vs$3{yi2S
ziK~c$pe{0`--Fyz(kgDH<KUk2@`iRPsX=ROu@&rgi}r*|oj^LMVR7AFwi2)BvpYoI
z1d`G1pG_9RdBn%Z?^e?-rZ-BO<i`8X!+J8kLm-?$hVRI<cidtJEz~?wadwI$?B+GY
z_Uk>j9gCAEELiC;`sx#}&xD(lt0j)bYSKiHMB|%bz*^d!*Wa6s#n4qm+ph<Ah-30o
z16?tT^Jb$l^sL?2R`88Sg(>XLc{eO;>AjLwk`E_&%1MR|4QSo`WJ}yRE4%VYiCA_l
z6A}!(y2N74tD$@yyYognZ{f&f<mzN{?ADfsAl@@8eqKTtyqdMNVT>Z&jK@B+wIr3V
zTW%}iK3n+;NV06`>_l61*t8<6eR=Ig%e`2ioxA6nJ0N124KiOY#)YKk-RT#OM6&!x
zOr3dXHehNPSu9wQF$#L<9`du%RfhVfx{8$DUOb}I>N-QMH!_bkMGG{8lk1A*8JBU-
z5=&AjAH1zG0<)cqcuq#!X6W`O?28>U2&qSe!bFC{3fHZo=T)f_4_y=G5sj0H^Lepq
ztB)M@D55|mmpt6scr?by3h~h;IyA`IgJrz28q6yRfrqv&ye1DepdKvzMch5#EgK!q
zJz-`^+Me$B_Q#LC*T$$N!dZ~u9_+Iaw4>xD9NPBXK6Ky0*icu)WXR{=nP@~CT!VCD
z>^+uw2xGoE_2kQZ@1{wS(Bv+~#~OvRBt>xOg}M{QjTJlN&kOw!p~5P?4BjMDwtTh$
z%~LyHV}w?8=uUA%aQsIg-(7I_rGlQSzb3kBJUlTOHmUJmKO+mg4ZUuhiFGIr07@oG
zm;;I*w>(|I`A+&Olx<3I1yt3YmYfM>?A%;XrhlIW;`$=-r3z1h${i}&tiLYr!Dptc
z#u4XYFWOB9LZ~z9)>=fEag#~Qz1&6qCpCMbD{j>O$WVl@{M>6T5d+ute)@RUE7+^y
z;FPbH&)XA!IFX&<f$#v*F2;>^hib70N&q1jkIyAO4~&Y$Pr@^soSQ?|2;BN{>f8|z
z7N}L_)b>PZf!E|5<O`XVcUq!bEDU{kk~9;JqZ>FaF=Q4M#O9N_1dGY`CAaA1BO+iG
z*42?7Tx`J+GfaT&vLu~*_*vkNOXInnP3FLHAHHO26qV#Kh1DYEMqlTOn&FJ058i>r
zsVO`a<io#z^EK%2hrf8X-~Zv^H%Fc14Bx3<{^F;yu(P)szw*5I;*a;KM_>PXv*X$K
zfe(N9@=ML%?H*O#F*$hloA3Vg`!zN7??Xvy?_&5_z%Dt{g`O@I=)tN*OR@)^7|o#t
zaS(K1vxpi2Ym(7r<6yh?E(f<X!Y-TximH2+%1J@C*W*{ZQ@7OJwByIy%_Q8Ni4uUw
zHZ65;LDFq*p!nooc?lwuLSB3Be`T|x8k5)zYPN4l+g;UJ;wFRVu>>X1wjENDD<oY1
z(L&;s#6tSt=eYKZN(Ni1y6PBmO1FonZF-_d*(TguAXu;=j?`i`F0Fmk;ji8Y>E}mH
zWAVzRL<<`S$F=t$xP8=Bq`-!lKR0VLXtOp75*Z+x8RZ#UpdLEN3OV=4hK-zi1CqTu
zSl6r>@6BNZHa(il2UV240k$WfD(AH!Z1n7+l96ETcsP>da82~1&Z?<V4=uqe1=8cI
z9Uq9y2(n1cgejx2gMINKX6L<8*%|NO8n7H&^ia8TgM?N`J5wC-@fsx2u@|<3@-Rb&
z_w!AyPWO;BTl~3ghmc-|0R(U6Pv@<Q=e|F*@v+l}ZE$uw{z<oK<D)f?#EZy8?Uq2~
z@$4j%9$94jv7QUaq==o)&(2f>kLIG*rRbY#YfQa>Mg>#Y2SMUGnnAGv8+hau9r3HN
z0QD^7ck}EoPb(T%MvTt}K`ozzH>#C_TEZ^<eiBQ^WSjrkC(!Wnext3^^X^;#lkiJV
zCv+*9F?eZxtm6(S>&a90aJH9a`JmE`7BUL^X_yk<@^x5uWnXmEmDbbc>spoT$8<D0
z@+j@7^*q*PPMy){1YYm@M*e>0l-=61(ads$H_E!%Ub+2<8!s`)?7jnDtrXT*)n=YP
zs)=L&a()Tvq)j7&an?S?jT)t}-l~10lZNsu>Za>MF{P}-w4XFz6@g4hA6`esyvP6x
zhL4Q6TP6FCrhR@7>PZl-hR8Ds<S)4dlfJ_HvEQR$m(Ei&uecfbmg!~QLMmP?v>_JE
zRsN7wWxt*@nDP`o`&V#cu}oZH*Ue^y6^8@`!N`{|Y@bv__Td^&2O7n+(IFnDWZ+Rz
z1X*Mg;=R|HWrD*sP7d<5qv|<xX*MN=Yc4t2NH6Uk?GJ*l_zo^A5$wtbCV)V9?0IWB
z?mt|pNWzC6>-nhTyC1zYZ5LHKq4?rZSU2N6Xd1gIxQ^s&w}f9x+T3I&`ZRSh%EVWL
zwBiGLWd&ws=2^0VD@huUkf*(oHQmwFZ(-^MHPYIs&lCd5jZ@dMDc73bn0T#b@h8ik
z>$tEmv7)_AdIa<W_=Q&xyLhf#hv7&fra2KUg+*c96HAE`?Ir61<^F^G7a%2P*a^N=
zgpvVTHqMcSo{j!MF^$=2TFcWiZF5T@v8}VGe(9M-?j~n+k2Ib0@pSvKg8hr-QjS}*
zq`R`i>4jrDIETrsA;|oQRNIH{VSs<&bCz&`e?9v{rN`>I0~?#idfAR^E$HRW3ijf4
zyF>+w#vi2ox!P|7mqb^R`7t!FBk4g)V+<do=gUUuD3qPeXzbbsdMeKe+2&$wlXSkI
z8y|LM@~<my+!JX)2lA^?{<v<JlOahv8i%cl4R6<8l5V%RKUlgW2p!=IAO#>x*%$q)
zOL-Pnfit&gut+rzZc(d=O;W6rk9WwCQrZ(I?Q8;=)71_sut;%dh%F)p@rvl<B6{II
z0DgDtdd#nQ$Lw=E8orNHnBO71%5IUL!Cm<lH($rPBBVZEFgl$Zp`g<Me3V93#ke8R
z@zSr;`iT{z39X0iKN$_SYL|w+sqEXyJ*(91taz1H_5z|5IJs5i%o;-p@R`!>HtUC3
zUHj-|@ku3@AGeunf~QVQwky<FI}p5wx0cJ>6WX*4CI<kkT_(hX8I+#T?Y>sVNKB|2
zy*l;Nx0KEGY_EpO%>IdH3`?uK-90R&WuB;ufw)6RXSYe!O(|vb#C~cwn&rR5pc&3*
z5^dJ*<`hJAdS%burH~c#i6OSn09mv(1vF+Fa6aW@(M|dxRfFdWv@8>y*j>dSvcuYh
zYhq*4=qhZeTx)gW;#xnG;Y%J>y+y;vE7Sxy&`w#lnc~6^mePuxv<K*g@MxdnYB%BQ
zzYIDKEcqv16-}Ol-PlKbJjY3-g6*T0z76DQZqG9ryckNAW&`~78tm8U8?mnq&B`W)
z6+Z^DJ;F%Xv&9^2Q_A&E>L>DmCRvl2Ly)d^Rsh7fM?R67k~e2|Vx;Bb@EZn6+thTv
zb9LQ=GqmR3J$A0}$r}b#xjh^?NAZS1ZSXLKnvnk!fnY=geITCOI<H@SSh~Gd-N8P}
zD}o<^+~t44um3vhVhc6mYPejaD{?95Hh9z;Ib{%N!1v@P^KzC;P9xeUEnl47mUW}T
zZ?@>KZ<bOvRqo}q*y9mCX%JFv*mNLqiF4ZiQv+4(zwXDYbT@+MmoQnjYt4jd?}9n)
zpwDB#*3y<>`?crv%>tdsYw+PB>nz8D`J`J&X`O~QK9A?_=zuZq+V-EENF-4R5(ghE
zelaqNhj~XcZ9X-4t}_8;2s7oL;taH%!1*wCze@eF)ys+Dh-n43iKVTdvc2zcG2V%$
zy1o^QOm7;QjxRK^bi{7*lbQxEkF)6$M{t<)Io`49pluJC?#Q$s=jl&+Mun5_Mh~k&
zq8*h=ZzXlPGq|t+VP3bg7O&{Q$FZDxZ!60j?rD_P19C|YP7q!`<)73B^Gbsw4^u)`
zG80rUu6?eipLyl*q>FJ&SyhygNTQ;EQOT{>?XJFkLEv_doQ%QP*}}0=SWi{R@S7m;
z90gx);K3_ueZX6Qkuhy|=F2I->>LxD-_APs`;MAc$Xq+_?OnUnMpM7I-8zDPVtjtg
zW^HYG7ur@Z=cKdrYYqC=Y_xN)-F+cZ31oUKP1!-kCevXJqYQcqo7yDPxh>R-?!4+0
zxh(Z7o4e?#n)5jrMCl~s{4hM)U+Yiu>7^(4o0LoSn4U5Xs>2h?$u&ZPu1L0BtNc$g
z5K1MVFjG{mrb=<j41ysY)Rgx2Y#%Ziv>JcTg>9h2u~!O=N%AS8460<u;%b(1oOlBV
zQ2-x&tv`nV^9*YT;}U^&hQfNRsd2~TBJ+1av<rcJtB!syU-Nv{yHT^I%!GEso>#1x
zcr5f!a_O+Hi9qolmy9`#QM&P7WYzQ8-3x{8Eg7#B?qrfMr{5Xc06jByqjBIc?8>xZ
z{(Sy<#F<IE0_F956SFp{>@G&WoQwk7vaG*Ve#PNPX#>(6+;dAu`UF3Lzib;dn^WMD
zcVnq#V@55+Lq7Wr1LiWJT{g9X?nx9<yGGiUC)=B_P7XA1SW>LD4-I_lKF|Z<vxIos
zI#X{w4>kadxDHB6%MU^k*6p`!xq7{g1_{^k)AKs}cZPI)VYG`;*O0^6NzZ?QC%Z;K
zyQ&%Qk0F!vOS6#|DyCGQ3Tf!Ki_M0g`JRTtdxteJz2M?ow{PLe_2g6I;UBi{H)QtQ
zq#6Xb5d+^Z7=)GeyIS1xBoyj2Jaru^6MGz@oXp{S?iKs!SRW->&G-v0JWcljex}_K
zl$*;u-&aj|x!0pgwxTFN_C<B1JJ}UL6)T%<R!U*M>7zOo4&_<o@E%O|VnCxx6!X?J
zu=$k_Eml~c33QDI^V9E$;3h3sO!$+-YM!j<bI92XIkdt#g3ETdBW?KY%~@55x#q_t
zPnn*Pw>maG`w4~>&MYqz>l8-Vu^wB1r@{}qSF9=~8IAW?Bm1ZPhht**l|KN~E9ZWb
zS}FEA@#6aFa%#|$o1fdQn)kP6*H?VDIb~iI>F(+K=<!9?>CmMa+u^hK0#Ak|m7r=k
z_7Odw@38V*1NHlZ2R^YrsRBSFA$Myzx4rd2^v&XEX7MS&hn<M$IzR_}0q$h3g?M|_
z`g&4YdVS9>7iyj74k#qooZT*To7XRRo|SP52Nra@u-b8*v6GG3o<aJA9|wX$2`?u(
z{W9~o?3zD6|C01`iKgXjwXNY~?ETjJ+t2&g^GlrY>t&gfH>r)QriN*+-tiykv@xaI
z*Ui(J#<$$&wJtZjsqpc6+Ou6JAS}Pr*=hUQyVx3lY2+g)+Cw+;I;8P0Q|{09z-VCY
zW6`S%j9GIdQXx3jflU}@2iv=bTO4_F${@iOPcBx;@x`0rM;Ex|qF3nR3i!B#=Vsti
z{EBvg?iHNXcUHKw+nz4-Rz_DGl>$ZqT~QG03KO!Bagr0z8ci8pG7GIZ(eQ@hLrHcE
zybWg;s~1}L3PWwXhhD^^tR1T+6GGEjje(<EMORpf=Q<J0C0>};&5$dt5CC0Pgdek9
zDsdz8A&tSN+|vNTV-`5|pxoU+wW#1!?`HtP0hk@155FS%{GJ>J)+#a5RdS|F{!llF
z2P;9<BX+ujLXyI1SIojpPkWNoxBjvdT8Wx#Hc7Q-w|%R7t>CYl7HS><R5Qg?Ke^D)
z2>Z4>rWs8D$4-q_LmxO=beOSbBTuMa$onBuL2#Br?}UVhxQ4x9ybBhUOSf}93i>QN
zsPm>`;O@EV&S+1dk3Dngt17&?MsFXChaGB|)5nxJNiQD>^Qi?Ry__lV5BIqzGHzYp
zs1HVIWlW$R*7S`2GJr1ZE)=SuP4wPHDcY$$nJZP^;LR6cN&E&_O;<Lc^4T>wLYv(X
zy}j{!RHUn^a+ulq^}n{wPcZX+oblU@rj6hllTQt7>Gq7%=?XS2rX#<|`7|k`$j)3<
zqw?h+ClVA+tRIy3F0o#E=+vx?f(>Al0I2n%sI^LqGqCi;4Qy;HtnUR(4*Ox@t2Uq7
z6g6V+2eAEpbr04$;l59FxMS0xT^wh`!u!77^al9q!EK0*6=B|FGUm6Y<cA1<sBKY7
zDS9bDr)SZj#N#nPIG2}ZM_pA8bjcoS3OGg1C-*(itP&>7LXjFp)`i2@eoL~<qa+Yc
z{B)j5`d#oqmop$w%_JANct@x@jUB+$EA_vJj<s5*yINF)n^`X9|3e92PT7F-_Vl5R
zRB!j87XTdVP^h##<_xn4?rP2LIb|?uY0EIP8FYve$RV2ngrL(CKe1DunAHTa)NUxU
z=3QF5pQC7)9uyMFeps`A2Z|iSzn{jcKCTf_MJgvo;?r_qyd7BbsUDW$;qP8vX-Ym!
zw_D3T(MY|k_sOXD7p>B;?wfQ2RC-g^Vs(y6)|7G2kc~p`45aZ<)Y!69JvW3~r~_~C
z?Ygse&Vqr7$;LMY#6Wv#i{y(1#R<1%7+obfOIy~+Buh{7vkQac;cl&`(>GrvX}0<@
z5|K$MGHJ&Xvv(_UU_UM$KCxu~c3~CbIbYP8u~iN5eC|HrQP@mEeDNAHenB6T?<8G~
ze_$6IkRzN8mtV3<^7Kx}kMrVhnVsiNG(}{08)|!c7g;OXtdF-F&0Bw}M+bc+m+`i?
z8+KvDlbm#f7Z;ry{a$BL&(hnMx=_VB4EuilTrfXhJ+Ii;)=hC+07c?nBwh46pOG-=
zb}O}+b-1{=hy2VfHtGhjibb{cTvM_|!qluI)uP%JCV+SX$Q`p)4Y=<;Lf0hWm)`Y8
z2UBC)8<dKU&iYAGXK52WodiZWnPDcAVMytQn3y}0_lMe?kV6qBp?t8D<<EKzdgN#R
zSevyu5YN`;7uUBAb%!?(&)3)h0aCBQyids*WU7eR7fo?Z`rwDs0|NG&M(4dpn&ny{
z<^C<7L!4yLL<9REB|kOPdegwVD&*lDscgzT#09V&@{9ftgvzspq>_1ZdT<y@>=k)<
zAnNSL@E2e;K9r(XoEY;AVEE7nL#+!Vy6+7|5eILFsn`fFaMx{>D33HZ0FV_*H|oi%
zNP;bw<$yc&HMi42j53f)o?c~EMvMc}$?4z#zbxtT=Ra~p{>i?*V?9E1jOZiOl~F!*
za{(BV-7PHUkHb+8Ynz~y94P!tnWmjm5XaFS#oholJ8nzG5OUmgfK*U7D5W^cLd3hf
zfehTNuo0RB$_&$tPN!#H+`hD3k-dvRk}j^j%5kr5qrA0ou-pD++uep;XgJIF)bV|s
z9MYR84DV0hnHwqX!K8au0Eq^GwRf2Ly~$Aj`(%~=3pfdd9qswv#<K2nX6*;{hf(U@
zkLJRyd;4d-%qp4Po{-S=%}~(xncnEl>bJ3hi>d{4i8^(rfv*tzQt@>`LqNGBwpcOf
zq#s_eS`0!tMmZprwBmn52548Oq;TqRFNAC@6=icJ4WHY-KZG4ii@p_bu^_%UWN?uZ
zMFLguntU7IW$HeQ&;(M3+q*%l@#58st{>uUOth1d40?_Yho&z5MYlwSx6QV7oR=}M
zig<-^Z<K|8lhLNAW0V!mdWzOy#EdJj#tY|=IgBOn4^w%8ur#^BMOnR<`ckz&>}Kgm
z)&4EhcC8g8Le4E5U?1o>sb~{5OW)+^)gQB6R?!(-2W=N`*@<VU9j_>Pm52#Tp<XO_
zk?4$c$~L#?8LNkxaCpd~)3`ylA#_FIRKY7VMd%7P)53<RN_IKa+&9xs<oqsKYHC-h
zK0^phTjL|mEysN7KR(4mUS=rIsGaQSD`S_uisp76VE6s96?&=Kvh{POR0nz;c38@9
z!u#=Die`%&w+%>YOg&tay*hec@a7K)PkZ|gNOYoSDSI~>8Nu5WI3uktw%A<F_F8-@
z7jCzi%jCgr*Am4Ezjz$ILeunmIs^zg3v4WN=juU9+S{EVZSYM}taGmc!MMjQk(FEy
zH;1c5nx}JjQUbYc!}aSAR_O1~@)4ca<10X<;`&k6Jb|QPQjU#e6&25^>3}!+P#oDR
zd+y<W=B+5KX3qoU?sX)*ynkia2RD$V9dD1k5VTgfpFKAYQVRU06IGdPwcK^DR6Xea
z&5&a~YT50l=C>3<EvGcg9WR<Hbt!KqP5;<qi}ZT9mXu=B(-*~37R~l2SUNDP2q`$b
zDB&Mb8APaVtALot6A}WDhqT7UHSKG)HbE<+!rDn5Mh6&djaybJrW253apSYuI?~Fe
zp(y_r{hq=LBk@N)WzoW=d`t6WJN_bmyIQ6@IvE)dJ1(k~;cse0u%8Nu#n-p<teKFc
z-z-Xt@?Y84#wH{k3#_fGDXg}D{pLy47^C`&%8lY2FDX~k?V?~!3WMrqXEm4?yf3is
z*;sz?6ZsAf;)eXYo3wj0vZ~f^9b+zqEF~5FCODEXkYE1~>JMQMB_p@Rb7lPY_ZxO3
zCN(L5mK%SFD3oq5)G;~%KN7cuk+mJPFF0`;I~Y;c|D8pfzlMYIkdO${D4K~Q78{l}
zngFRj6&8<DmlV+(`re=G1^><No~dQv=JHERpWL}^3VB#zrG)adX2*af#VK}zwnmiq
zx9<wWQc$tL5+$`s2xQ)Ps4>$`ZgG7xzziH18<!2?SUevzA$QvpU7^H}{)=1-<@S5R
zmb2pZ#C|pay@ffrr#v2MS|KIEu!H{r<mjN>(VE<glvmM_R=A+ml7b|=uy%!iv@%~I
zg|n|gyo*scf!(4K5QOSqhBf+IAw7qjWC}Jx6NLx!67J0<UT73iU0Ncv;zN-t6>>3Q
zpcaSTgr^!cy~So_SPPee@}=ud6GDf&-Sf!wa=GNpC-RWN^SxKU;C9fhon-z3BUOv)
zb<cU1i}GVrn^QV#26AvjRL;Sy2XccMvq<xwdsK{QQk(+HG%0tk+<NE3n*pQ4QYi}x
zDbw#p*VoXXEZe9wx$uW^QCNY-&z<4V@x$u^N=}?=UCm%C?Z5?9>q5pXh$;H!<=cOX
zWYP?SfV#(u>k|2;?y_;do-9$0nvi}YwfWDqdRQ&z{kQYXJ4nY0xgXeb8(xWi^G(*h
z>d2b3<7V>7MVAGF=!y86ejuq&_jydBDFkW}dOWDCh^%={WsAgwgIxJqVnKa>hr_1k
zc1KgPol!^PMZCzYg3bNBkM8?!bPMdxi_KqpA%_hw*EOge_S^ABZ(@5_#yIDZN<mBO
zFNYJFsVL3?!=#51DNTAcky8qRla?CVtE*^i<kD&VwI_9Qtg#aJ$D%tj^FXCORrBPM
z6lWMMNQp=%H5&sRtYW~WLJ$`VSZSJ6lDaSTonP_D0EnzIER1joj39JQqnreU9w~}}
z%Qiz{aT+Au9@b2<$mQL-9OQ(nVj@20*HiOO<wH~9DR@z(^33e=sgAZsQS6Nr(NMn$
ze|08@TMl-TY33>D@Ose$zN$_PAbP%^F8{ai6k@mhiUskP#4bR;_(`GYWSZ*D%gK+%
z?g!o^x#|b3z56$s4`Bqp_|9pn>Wl?Cc!_Z@r%KkASiIWzb&=;|5+%Q05z|{z*E!9X
zzL}R<Rk4YuTWU;wjL244X`&^o^<kW3Y^Yapgv85fYn+;1$F|=(`DloG{$gfOg~}s}
zbPeZX!>n?7Nl(Jr-h(ooUOsrU;MmY<eV2Lok#9;i@axtt+YRuoL@Fb|Dcw!5<D9mZ
z(GJzM=HnBfh-{V5`=rdqn?C~uFuTf<s<>@AZYZb)fs`$)wuq1Fl`mty%G?Q=pB<&x
zAxVzPm51k-D4TiRIf7U`K@c)6Td@JwkscBhq$M^93PUZ*g+~Lj3y(hd0OO+^qULqg
zs#SAe9sRtmBi%6Deqz2o|7{A81I=NuaY(Oh!~7j~j(f8xe@+AxOzH>%e#j&d68K;u
z1p0Zl&uO$s=-N9ld5=)8FfgOKd$qTEFp_z8n3|<Sox?}F;|YkP36F5;J5M$q@8=aW
zj02RZxIPW)Q*)_WY3X)gU;?s<f^D6p(|Ummvz0il2ym!8eKK%1BUJf*Td>{46={9y
zSb5k7(qv`96Z;sN@a>^daPLRTW{?Uy`pN;8Ptlx~P01s}G+0Wn{2O)W$3QhIsAB4A
zaFLlS)z`7*)r4+Z8>y&1lS8p#=L{M7;ylAd(VaUY!lY%g-GmmdI1phmf~q2#eI%{}
z5)a_R5Dh-rQhsBTi%?)LS6{WTb@x=?N!3S&(NV}(w7^sB5DJDaq^MD-F=t|O?+j0Z
zv(0q^a~V21Y4s7kem|@tkV&U|R17V_R#rXyMCt|2?FKnteRz><`PA1ZY}k(%q^E)K
zVI&Qx0qJKXS&XFR*h6a!FDqt}*B;18jZ@@+rKy;?Es(aem8kDK*zP6d60b@BlDpAb
zxlV&QuYB%;3ZeUo6R$Cm+utV2K8q@}gt|AwlqXS5IBnx82J{5)V{eX10pX{fPA6!7
zN$wmcAtfaaiPCWD*3^-Hoifbd?#^yE%~o71LDt{+4QZ~Gc6Rm(^DrHHv#KT&+e9sT
zM94l@lEr;M2{*U4XuD#uzJt8W{X)vM2>9ZY-{V~mJmMBIrSN$siY4IFK+U$wN83&Z
zK3sSbS`3R5=;EYpgr&`hvBpH}`Ws0l<1l_0<l(CXBHXTlNQWmqD)1_<SKwF}ex^1Y
zS%wE9`7d-?4?#g%+Zm^P6Id_ha?`d3mXk_f7puPW8(rlEE+OSRr)nmJH#%d>@$c5_
zd@gJ+M2EuFMx;L}>7frv)iNa_`?LYd!X+oiYWZ>^BV-!>W|gzL<R~2ZJD2R4ay#6y
zSMI$#Wi_;39ZF@v7writJkWNXL130mfsUD?NWZ5dfeN;3C@6C)pHf{ZCZIHO=4zC!
z-X+pNj2Hdt$pJ`TR4oozIB9j=Zh~NwjNjQ79BMLtaIQFO478n2k3|?ky3AT8h5fIA
zY4yhkft{ynmp8Sx7eN}8$-Eb!l+)+D$8OX%E!|SK=;Mrl$OX?I#P1Xb^NnvcVOQum
zuUZbJ?+R=QDUO3%UUeCpZEJr8^C>o&s#n7!kXE;z5oWLES{(*wZ>)ftIbpuptKd3j
z+b@BBJQe}PLCu9JTfS=p^eq9{sM|*UagQi%E(^`dtLUFoTbLj;Aq}a?Pq_8kfcRFu
zDVz4t>JV^?EgtdeKZp(8EUWH|D+I=1cWoYpGL5raJEolEopi8JS?&+DM{`p-ukga7
zQ}i5IGV1YMfYXRo1Q1T42rPn*Z(54`rDX+dueWy3iEML&9=EH|>k5TBJ0H<R6!~*O
zyLaRCV@1$*1%301jU}pn<g93G^F_5i|52k+BHR~g<*>#DIod`o<xCVK#_U`CXX2Hs
zpYihwmKVa^Ug!?pcHl5HY}Ycq5--Oeo7W1r%3n)pg+EaQjalZrdh#m0%v@yk_*ImA
z;#HYT(+}qsDalsZo8cbd-k-?baG-dRUx%e#Kh|TBNL9MA3;qQdYg08NeR|I<L3=|d
zkE6SRYVci~aD#GTvPCCUj!rylCg>i<u=3=qI8V|2rq;b9KE97Nn&n;-(Dj|8f2ZXc
zUZ!p2?__1e(Yig-={X#$3lX1d1vUYuRRA6_>E^Q5K>FxpBLCse`Qi}mQ#XSWld`6t
zm1r4E6waM+X^C!tJ>=DyHB#mxyA;CI3g^rYsg40b_8RcbtaU_`i!jo<w^5LAlevP%
zx&evE&mmpUshl=<m(42qoqbYr{=hz7wHE<ypYLS(6OnVvIqq;&VcmdZxm?_X^%)S!
zEWmU9woC8*VeRq>-5z~V{6%Dn@*0mcIHTo-7e+TC_NO*cIUg8K;nh6uw}A>}NqCXK
zsEK<w=jAEWL0nn?PJ3K$6V@Th6=}s0B$0EJUkJuVhN!)qh^_G+O&wh4y2{3qn^Uga
zL}=7EC)`K2<J@~gv~?Pu@G^h4jT8m*?X7M1Nd+ha)A4hA;MqwA^9xt>ZQQjt_*bFw
zLQc-S^5jad4}ol3fQW}7MHk&KtlQXx=gi>@r|(p`du7*aF}YVq4<W*O{T3)X(zzNo
zYp+3Yhp6=&l_+_$rjV;pVC8b)`}ybhHGU3h5MU8nfnnp_<^3DG;kYc$C3obfe2{4J
zSF=JJdUpM(PBrANYwwzjTX}z`pgRXv#>tM-OAclPA8CVVZbmE%lFlPZHeg%-B;!E&
zB@Kr)zK#KlbZ^k!-DCbUNdH|@E_6Kn3S;&_f(WR1XPy{oW|zF@^E9p!izTxYYfmO#
zg`SQmmpfuKRbzCm<l{ad`UA8K_|EH`zp|W{dczD+w_&X8&F4>Bu9DTrwV+Mt$qe5A
zF-u@CZIhKzv*QPLePAaN1fp#eQa^m^Qfb%Sp?~^?q~^z;K4<QMm#s`Y<94d|&Y4Gh
qw<!nbm2IC{^w|-tW26xcC;!yH6PVek)71vvWZ~)R=d#Wzp$PyS+=wXv

diff --git a/reference/plot-7.png b/reference/plot-7.png
deleted file mode 100644
index 1611fd3e5de59fd7a8e44ac4c8e3586020636377..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 24961
zcmeIbcUaWtwl<7C#==IU!~#SFQ4xqDAR=Hv4HCgdZwdm+P^9;gxD6-*2?_`Z*bwOg
z4!s#E(vdRsB2^e*C{hQQ?|w$J&wkH2-}~1&=X$^I-CS2Las171e)Bx*S?gZ+y4T9x
z3(99Ve7pTy4i1hDia#k_;^0{2$icC~;G5NWr8Vg382(sopmbINf8uBERJI5HSbOUy
zO&bml9xw9WGXK7DBMy!qI2095TycCn-0t#N)hJhfjM27c4cE#oD-|yLI~4KV3%vVm
zPyZ^V(i}d9nsJn|(O{{%QBd$PFTbdupgTe7=jzs9{6Q?{25<6#6PK?X?|Zh~n#Y*Y
z$;V3z5qVefW9n~};(47(jqbS-xihH|Q{lOd+kdNepHAI>ugKsa``La{Hp*Pb{_Tn3
z${hADcQ=hKC%*`sD`5Z2?|YYtk-r~*_Fr+qUFn%No9Ah1lBN_&&)YS86?VfHpPw-@
zlxGg7X8KjG(N^b8kLRO#%{VSC%zdb-`5To=9q-{|WfaNbne98>W?m&E_^P;eZ>DFQ
z#8bVG^prYy{`~n5O-<kX`ubixu+@ggr6B{qX8Ntub!tm-*Wl~usdgK+NxohhaO1`e
zZ}F-l&x_=?Y}sN}Dbc84XlS@FsN!)?@xh*ip8d<PCHl+tI09ydQ>%+d(mQ#%xC~4D
zd_zM+C8ifYot5A2GMs!?LE*9MMCph6`Zvm^D>*nk6MyI$`Au1Ns(v`G*8kYVPnqS4
zevO>9siwsir%s*PX;X2~d^&$9HjbaayFp#Q^4Dv4?w|g^bCX*eB!7=Dw(;0l!B%Fx
zPG-umexPzlBPUJI#a+rSFgnux=*yo&HgR!r>$_UTr8=n3HI>M1U9oz-x3{;L+n|#1
zI5z9|1K1)(X1`Q^2vtRHuGilDvZm&?&e>uqetv_h$I_xpYp{PQ*gpY!rBvsA`NcsM
zF0SN8J(nWmFmu|dIeEckL`gJLfKiKwyBhwtvE{$VYg^_@P$J!?n<>;0T(?zbG{aCq
z;n;MO;rKi~X+UcSE{$Keykn!|=MQI7eQqUWy{-<`ar}5kv$Lh81%GnWnjD%_j5PE0
za-B~<y`Yg$ojDK@Hg$6;u>0l9m&4=wi$m(oJ<O4G{n91%cKcTIbo<5lJ12MXI(Gds
zyD%WnhbOT6{OQB)-_~#2RKn-s`tze7pB8OzO>v!@$e<@?ch_H5@ik9xmvsL8iED>t
zj@3@rvA5&5`4)cAa~YvZmh9c?BPd^88>Q6RW<1_qB-nbmQ8h`Q^6j0I#}>xkSuM>C
zl;FZGZu6h-W@Tl4YE04}&SQ>REsU3x*fr{M*S$Qi9Y^)lcc1Iam?#tO@!sv{jpaMG
z!zL+r`d(_aoXltzJ<*Y#klqqM5)&P9&qplp$eH`!bL}vHbMf%u!-Z8cjGfUD&ZccP
z)w0QE%_&uRD-%NH+?-tpWIFG{?2C$bSniZ{{_O3+O!x4CVeaO|7tqMGHVPi~2rMox
zZn_yAo-(tfq(6Ttto6HZ@667Y$VZrq<LioQS3A|ey4dU<f&KAzwB#Ag9?cR~mKk~T
z{h{kWku^)}aHb!@slwrzJ?qo+bEKTCXx?mhlcB%tGn49w4@vs@v%Ny{+i)ILva2Q~
z7A;JbO#B=9VCI@Gz3=}R>0s=Ser$b?mN(Pk(&+4h&EA{MPZ?=<=oC4laFGArh7B8v
zPCk8IoCj-sKYvDH9E+iC*eMj@$WM2uI9T$gY2dL<Rp~r7ZI_E{Us<3a^Cms5$4@3v
za{B!ryHf|EmBY0ji<>B0!2&Gp9?MQQJ)gyry<t+bvAOrArl!nk4u7r{`i@&nFQ?cF
z+os48R0?{N$>?XaGC3EI>E%<$yM4sArZ*Y**fSVutrirUiV$&=tuxIgk@o)0O-&gs
z-wUfJ9_X+x7b*H>#hSECH=M}v#OzVi#_H^sJ!yie#+Ccsdw28Qv|+7Sv+;m(_))4Z
zlM%13??%<o)C?J6`g<%G#x3Tv78xdP%;^zroi{fw1~*JEO-@rA^@KNWKm4kpag?sl
za-Mdf6y4iW^gu-}vtbG5ryeH%t|h0W$Uv@y4m%E%)xUhXZivw>5pu}roMbUp;OnAF
zi95<XdHQsheY;&Fc2ezATvA?(LB7*uf5LQNXXiq{6lLcO?6SP&EU(k`je+)UFBBgb
z!0P-4B0cgN+*M?!uH66r;4@wE1<|Ze&yI9^?{1Npv}wz*N`ALdrtsHQ>soYe_V-su
z^1(#t+x^4*9vySCUcdeDUS9XvkIKH2t({AfemLBZBq(ZOZdmM{XLnjDEZ_h#>gB?i
z<@!sTZm;3=E8%sCugt@O-NR88*`xgUQR`CU^|7wEf=bJFS`~7Q!<yeT&6Em7-cRya
zq&F|#-7IvbIzl!(r`X$j?dCm&IN;pkCRJwC&Kay{#>d}wJd26hw^3&J^=y%t`v+4N
zF`bRFpKIEND(H@G{UIhP(@RVSAAB`MwixdYrV5=8{SLl8Ja#S2ktVUsa!vTHKi6t9
zpL+3{CRg|P%STYX;48QotG*Fdi@%Y~%CPE5neK35<f=_wQB^&ita0BUS>sYGRxqU`
ztTV!8G*hjFIT4({(`T#5Jza-(i9Ps)b=;!&Pd@z{Ea!f(p7TCG>HbIY44n5JJth8@
zb~)ocB_&b8dZF!nM{Zqh!@Zj`@L}HA4LLs;aeOs4aJ$z|>z{(%sJWRr`uQ{+`(`5&
zf4se$ZoS*k(;sH(2Qw|cw?+8bg}n+-FL*~O1s4xDr;><xU(92!Jo)YF9r0E<!TRn_
z)TEalQ*rvf<n}?Ly4&F&5stcX41Kxv9CsD_?%-at^Vab=iP4N6zk?QS8IKMco;jeM
zcEE+v=}b@fUf$hxJ<rj@beH+T=#quGNtp%b_(L}R6-_3QZYiAW=06W^mg;ac%vqYJ
zCtLTG?fvfCWXJJt$4^haQaCmE`8OW@^h7fF@%wkUZA#Lll%!<wFKc)+S`KlMRl2?L
z=-zBrBx^2GkZMm&a_ti)U$Y5oVp#TYANif!53bL#_m8s%HBU}xc2;{VO<G}Zix+1)
z`QSsW>ds3Q<v2|eTOeNy$I553%CUoVR?R8SBK-DYaQoyjz387q0;pb`d>v1nou`Jj
zNJvODFE*!|INaDt98=h#itVFYBb@p}hAFvTeD01uZp!kDpA8hgi;IoDH`DH*Ph$;6
zN9rXoWEeRQWILl(A|>1#GGz5{%Z(RqPL<=7{d^(Y<7L18l%IF^zUar7%uG$S3|Y_p
z<&M%_uoQ{Oo!G2<i}O?E^BZ?a#y1FVNz9#Y2{<~M&B$w*@6A;>Wy9mP*g}a*NGmJ$
zm~J+akx+YjVedwsqi=ef<{I;t8a|E_=ic30`k-|o*5&f$%j19Fto(kOIb!LN-LPcN
zlYZyq(+s#;6F0f}5B_QM5t0%TTT1S4doU$GRe!m4W`%w2L|?g2t6A##v`(8U>F5_P
z7pGgrTt=VV-Z)efu#7@q92nuw8@nSh>VTLZC*t<w)dJ6MxSj?B*2$+Aj(x5<m6~Vr
z*s6Ft{YEw8WsZMie@THzdRy&BD};1^v&U<A9iP1G!Ezp#e)P8WL|kGH?|8=@rw;iC
zNg4-?^#_;gqvf{jK6C$l1m$KL9gZ%9>JNM0=+<yV1uNqNr!Bf)xY5<ymBs9=Zozk9
zY&U6<oB#Cupk+tS@RICI+h#;Njh+mXa67e~*6(*UFIsvqTPc_4oh?5+JFV#g1Ta#s
zaXDI9WG-~R@b2cGQ+)Fmyh>U<Sc`^$e&&<2aq;o@BNuvwnimn)N_aX<6bj$mtn2=8
zme1GQ8|IV2$YW~Bp4l%f+!LI?U{v4Ocvhj%SE{3_kK4`1n_x$HCF6l2fA`PN&29z?
zCuZ>PQHj{e#Rcc?O8`!IWXU3P#ve!6*_YXV{lWR}-Ulk8PQ!4J-g0j9Z<_L-J>$<~
zwlj=p@aUcPv3Xd?!MbRZk@HgRZ{GEoB^PbgY+bL1WxTYzYvc)@>Bv~YirxG;>Euz*
zMMO&57;o+6v8%sGPI=+$Yw!4<+3x#Ibre269_#ZdaesDu-ZV8S(A7LsFfJ%u!wBF-
zWIDSqNIPZvu297HctA@VR^_`Mxa0#{MRl6`C1aYfMyb;(?h_A`MNW*vXCI>YZ<mO@
zKKS=UTB=F)fyXXAzETEr6Ma0qboigD7H*weCq54~#8sLT^kfp@WNgp!SQ_v+n!}=g
zsCCY1)c0^J&C2FHHg?<$zMB3bqCamUFiFAA&MwX^RLr3!+kr9r{;XWBDXX9FWJRRB
z2bD9GcBZ3*GHDkt)pFzcM*q}i<1nKUiG+mpSPi=HGD{)Xe(}iC7IM#0o65$%cK9Hm
zVmK}zZ|{K!7ts#4nf8FC%2PFdiGm@~)AbjCW67Z#P1@I`Eg>%c?6U-=5-8`2b$)`+
znW$bw|EiXOa0flJ+$ow>vr%x-WffT}?Zh;3@Y9+m;epz|<-ub9#AQ1Tgmc4sy(=YK
z4-5?rJxfbV>z%Gu5;JPGHRfqs!{_=VMJh74J&4m|_MK*S!n)=CgNyWWNsp1Ftf5%D
z?Qf(SBPP@4er^QdimucwW<)Bshxq*7=a!SGReqGeW#M?N1lC5=MYm!l;I`+nwT?2^
z@)xJeTPjzyczeU|i1T-Z?Mk2P4}(dT$S;i#&*d*oC22CJ8uhD1FG*Z~|GR{{uCuNw
z!z`meb(MDLU9AIK+33waKUcpb5tfgPL@t|ECY<z!lKtIq&Scd=_k~%K)OgQ`9-Z>F
zblp7-tl4zqx$iTTNI5H-rb!X}jdr+*=7FtBd82s<+;&-MNI4pH9dxnz!W-pgyE5j7
z;(9)M@%c&H^cRwll$@O0OKS@i<!$1v6bquVmY8ROngD1O{CQ=co%QFvsIDH|9=fKY
zQp`I&(f<!hQm$kgwmP7Mq%lVTC>w9`R8md)Z6=M>Lp_eyHDq-=n!As`ty)=KDgRiD
zI*j$<cyY3zbuHg~jTbgTlAk}@z)e8`UNNor<m8lX*qR`?4@;Vb+&k)=D!qJP67^Ik
z-~2RUWP7VcO6eZW?75$FN4wzr?#)dOl#o=-i`V&v;-_gK$U>|r-|n-1_w(GC#S9M?
zb8CqsixO56z(KMo*Xh*Akt-V6X#kXmB_xby)Z;a{l!6Z21-2|qF)BAe%93nQbZ?k;
z?5gN@-!|LRo%@Zjv!&|H-TVY$rzA$^rZM$`srE(VEv_SNtJZC`b$l$<=_*c1H(9$4
z-%TU(8%LR^DJh5ij!nEDUaF6|yqmxK4!7R><#D_Hj_|wWy?nWcEWP($U9a@HJ3eyb
zcU5d}R?=ITD_&67nyOct`t1#mX}lP&Nlz25Z)hk&@a+azxr?0O5!?p>9ad%gq7_tC
zRTUHt=w$pTHx;A8c4`f29xO`hLbo~3_@C><3cui#g9Zu<<NhA^0c$155xZe(dJg~x
ziD&jmO1Wrwi%DVDhHaC!I6#4}M>ca=)hY^UzO&mfA*Z+d+4a>^kELyn9#fvAx~jYN
z5OUzsuC9?6{mU4{2YWj*%Esb`5w-;vNgjE&(53uPMYxO<mN)J3DTRmTBUOo)E?&HN
z$Y!oDSl^=29)a%qO|O3ODXrRwF?%;X?NqGP>Ww?UCzz?tvBx*rcCeOYT&V;3K!%Bu
zr;?vNd)Dh1g48M6G-L2(Sjx0j2@m%JAr(Ku1RQi6>mrAstgNizWEKGhrjcBX#hsOn
zIuVlQT8iFn9om!YChe&EF25{1&8_>eUEl6$3YPN7rzf|M+U-gi-OuQe;5Vq2ovCPH
z0rtuqtBU0u;5VQeN#V<0brZt3^>-UK)MTpX&tt&>$@A}e&jj82)Sm62$m*w0%Pr0J
zj8Bw{ndp1*Nt$VhimQfc^BE^ddiaKVH77W+R;8}qupLed-{X2h#dO8$$9S;p>2{q=
zy?hcOSPPSeO|P#t8<O<hbn%Se&Je0)&Umr%Nb203WnwvFe@dFr@+PZf1XJT>3|Wga
zmdSoo9{}yo(dw2;RopjWEvjh2OMvoQ0TkUc%e0gFoN~@QT%98l&Nq^L>*JpRQ!bTV
zt3Yr_EbmYr>dg1BzxYO7&q+|vb-c)N_M_MCXMIz)>222K2Ocf7Gu+j!Rp*c@N(P6j
z>E;L<EaJS(RZpcxF2?IV?JDBdf6NAB-KRTA+7jQs(0!g_4fCNk3mGgrFNU)2i@2>?
zeDOfJtOf3r@0_u0QbND@-_;XD+C!5KOY9x>84fVu>m%#La$fg7Q>Q<GJw89{?{{Y%
z_ZfvkxFcQy66$HRV&H&O>cbL09s<9`Ok(I!$NU?;y#at_DJqYTB%k7Yf0)%}Qs~Wf
z5gArYDA_XI&13eLPM<!#(_>-0c~Q*u<2udks3}`vVPV6HM+dpMtbjC-bBy~8XZEKq
zmoY3CNv~*8kzceGHMB35V-9Zy9H<=SLspZZS4crBB}m^S#px0zLqOE@DsG(%)=v{>
zvXSyAiJCQrCb}*LiRtrl>)Kw-J~>LjJi~n^d)Qs6{K_!0Qo&tg%hv4t8&apC3+9z2
zbIsN6G}BI!5~b%&C2Smt7Chy(u`(}D9x&;9S!dhq;Cgb(T><+m6M|?uARwC8_W)lU
zc3-6Hn}~;#90&f!hUS1wUyz%@-%0Ej0p4%dsoxX|UAPD6?*n#Kn%w7eXtUx;#IAFc
zoh}Hw{o(mplac<^X+pJZWhQB*I}log`)12EFCOcd=()xqkFnc)g-CU@C${w3C4i;5
z?$&sg&+V{%(>1S0N30AaPo)9J+}|eNT_Ik5AY|^C>+D#-59yix*D5-frqc=3F>HBr
zQ?sTOgdDtsJRux>>y%{@+NbQN=njKX!DsT_7rQ?__xEEm^5r{RCo9b=X$VOqod640
z1Vqit$eC1DPyl!Xz!^?%(Le?$Sgwcjumk8yyU-zTGMr>NitQw}6C)<M4_5B8DZP-_
z?ehms<~`v=ZyXmfLIfcN;TrFt&n-+=mw-^&x`xO0f<;nEY-A5UtID{b1JLOpuyRI!
zD8-=o{x(6rZ{jtQ`Hd?>w>WgVIo#hav41ru?_&}F+Q0}9e;H$cu1jTd%FSLaHDh>t
z8~RD_k#R>1xhC5&Q#$FW+ecWB792=DdT1!i;^^?}ceKCglkn;3R!i>P&w8i-@OQoL
zyQuH6IMpyufSNAzDCm+D^F2R5sxgR!32mj*>FTX#jf{+vu?-psjcp*M)LSjU&1iPU
zT#inQi&$D@nzX{#rZ?o_)GwJzQ0;g8c=fmyxYGnbr|!E#X)ZZ9Z9#4N5B&1nM@S+@
zb>5)XJ_;UAtz@w-a_J4tP*PGd=srpPx<#{xYez@qXELj|IJDf19$vye(=_y5{92JT
zyW)2DkakfPQ$I60Jw`gbsc){n{ndcck*nkvC83#SrfEg8pLvp0>uR;OM5J%t^Yi=p
zNavv#^Jz;{)BR2f!3Q*J4f3b+zOW+uflDnK8SO#fY=m0v;r}R95twScSVdRm3EVez
z-GPMK7UwJi$lVv`1CLFEsy&+SJVd91MjF)^8HdvV9RTuIBS`ZO48f?)ojEe1)j37%
z3#hyVoJ5;x{tk<bU140gdv=(;K5+~QpVd$dmahuO7C^8Y2*D*xBB>UDHUnU%@raL8
zAYtc}#B#5s>=|r{MG7g;nyXGO7tMatJOelHPf5(&Gc$D#`6w?TL=e=UwYNTYpC3GE
zRQ6DZej`!iK(UbgT&bDl6jFMfl<&%!4Gj%}@g*)(4eBL-U%NRgKJU9etDM9`usgFe
z#M9EcwHhe3u9D;F`HLgUjU>$W{m@le$`pHLU$AV=n-&)G#l4yNo$ky&t!fY++Yunn
z=vWnPNybZo@xN#*8y4Tc3oueh&=9y3gZ0}E3E<%k-ffV&zhdK&A4u8)`xft}X<Pb6
z$hxFS>w!9N1286!ePpsi+%U{M?JnN(sV-V&r&I4AR`4msi2eq%V;?E1ChB?az(1GS
z?{FAxBcZtak)hvR;&ouCg!4o?)4fiA_9S8b`m4fQN2wv{Ev9j^pvbl(!0NctwL9<q
zvHNWCZ#yi$4-5?afK~q<A(Jeg6-ce(Fw4AcSgSbS4@lA-K@=D)%#K@uPb@6>WyNzK
zh56}`9VAllEsc2{beSH`==ybC3e{vXt#nUVAwheOWL@laY$_$JKuecw)i&b-O1h}8
z&xcgq1?lfL!Z(&6c6LY0FWy|wdn`lA^sXc0_FYx(Q^T$4mc)e2mxi@1GL7!<kP^+7
zM}G83#=#=E;nA~WevPW4xu0v}7bhcGIl4@-tv=26Bv!aJb78|Y)KOa5sAIpAgSH!|
zz~mS*ck;=^p8>{ETUaDfB~Stm=?+hr{(;SS*Cm71VjQJR*!FfVuJRxeALQ)R$|VxV
z6%^L*Jo2O5WJtB`O><mY6G%vW%TF~=n_Flk?>Q_fnV8G~OTwjcY+OM>q0P4Le1a$o
zmvYc$P!(wZlVD*iz(J3tg^cMIiuUkjWEF4flzqHsXy+^`xwg1eqf_%XH+jzw7mdf3
zx0ivh=7!G=&r8VSP<Kcob)cWZJilIn`UrkJ75@1x<dy%Pf)Tkh$3JRPu~}`aUKn@e
zIQ5DCVi%7b=c20W7OoDDPxoQYXWZP7Lmbd}{loShJIZyqUc}_($vICCZ{y<f;TH3Q
zPYufC|LHQI{O8ej8Ge2evO)Xpw>kZLBHYVqcK(qAQpG57buIP7nH{HbIusOgrfAw1
z_kLHV!$q%+?Tq5NhTX!?e@#7%-09!G-hWHU?SK6UvVy%);M$hvsy&Jy?z`L#uG|-(
z`7G9jWYGr=3cZ^U_fVuuYcnElh@<yq>&hHbtRhvi=~<*RhC}r+Y~SFq*yvI8GR#8j
zw;}fDOf;G=?>H~nEFi~ht}eF79@!dZmZDf+Ur%zoipSF8OU!(h%uAI>UVYYlV-K8w
z_q}`ftb^I5#+sY$#v3{|D<9dlQM&i{@4>DD|D6GhyP~dM5~OXh4Q%YqD{5-RpbHIX
z@tS?2TgZq0z}>#5%ly2qt}Y?9`T4hs>FM;FetDenU%qkNv13PSNH!>JQ5y;Nt8<?u
zDPw6b;G>DNAw&oAz@m$1dA5)Mky+0E((g}$Ej|#G?R~8B(ZStSAg@qPf*qiw1|Ab`
z#H6t1>8aM7f|hWHj%`Ti{k2^aJy=VGf^DN1FVYqs4H`$Ui@#hY_`W)JvNUsLcu(h^
z_g*$@CqC-TU!@4Anw?L7{I<2e{))8+2)l6tPmo@Ihl*!{MK(%Y8n3UO@TrzKeAsrV
zXpdWm$Jh7%x^N`&2R_`@%jPci!^rXuaCnCZdF1UV_9+~`k1+TCWO>Km@AfZjN;Z5R
z8*A#@+K4>z%ntMVZX1Ndrz#iZJY9cXJ$Fm=gnLET9oO0P7^XC`m~!3a@iqio2H&T<
zcHe##R~0?|<86lxB#waVw;C_vE&=}i8v=ws>Jb9dz~K@WHuV?Jk_Ds|4-8a>HExMY
z;icn>>8Ax?oXd4Kyof`|iFiUTHf@aD>yY}&o3Tkr{^Z;$6b1+?84T7&86|#Gdqr25
zC)Z`h&wD1XDE-EVXSlHD;8NtbbNdG7>QS`XXsI1y$U{$n*&G410)2$i+y1|HcJ}zn
z9IiifZ*3qO{Un?G6<Mx!umhaszit24WsFwKe4d_SUmp}fMbzRo{*dheU==s6@Mg{W
zv%CrI50qhuq{1lvJ94id<lDRXDR2nv)B#1uv|>5JUL&UkBO*(?hH^fBbfdLb<`6QJ
zkRrUuJ{%`Xg?f<p68BTQC(LY9t3|f?aPCaU_(x8UBe8MWMgFvp4fXf{wibYW4*tf~
zF3{!c-2`mhgDt$j+y7WfTNL~L@tfNXHhgiwoc#Zlg?aYf|KVr<2OQDAK7#FL{wJd5
z|E)_MTdsTD6vr4ojjn+(LYrPXRRSzzGy;7>0P{Grp}zk5$k31}OY+1TRkfu~o#5Em
z{R5kRKaijQgM3Vc%JL3sJ(3jb;N#nzSiAU;VD5ETTO$G}cN)<qv)r>+Rb8Da!}(KC
zr~dDeSE_}UF|<OZEcG&#uJRE7Cs|j+<m73@Er#oAfXs>9Bt-bv!wrp%CLWHWdO1f>
z(bMX`_rrN0UV?<Zy*2V&oepTdc~!AxneBa6^hx(JC%GwlW74};t-m1JUjS^aBlId9
z9Ge6QtpkMi@ZTe}e_cS<#Z%#5>+pZ&I;iT7E}xhn^#CB&9%vFM3&`5;r+2!`NTNQ~
zWE8Ax01XXMNj%JEi}gHp?0h^OKK^-!5a2+%ZJ9QP-Gz4#T6GtGFX!q=v`t7jRDb)H
zeefn4MV5B}hOo=sU}ch#D4YdtL6THJ3N9kd&~fSuB+2z`rKS63B4NrbXCK!0r_(iV
zJCMXY$!&yk5G>j!q@F~gfFHv6qj<_g_(Y_1VCx_@yV7s^<x@Z4Q`1OEpPwH(KDMuD
zAp6VF+p~geZ0omo;&s*4)r-%uUz?~Cp>>&;PgctVdfi^V;U2H7^P83#Qe;OhB)MpS
zpq%J%`?6d)nk7;`&1}H_`+>(>LiKx7Ad2d3Ti%kCHP2ddMIn8{a<lD)NsnMuSHqQ#
zrz|Y}y70nT&W`En0OgBSU!XdW95*O?Nv1sXRpamYmR=VPNOrnTW-v#yZ4&(7ka}aA
z;PLdDnwkrvo1SRD0sx$!pSKQPRvH%*a~DiFey!O>(4OLy>kw_3CA1L?H-^hNXtNSR
z3In#eWhyQAwCDrlC+x8HbyYjGJ(>Q4I>4&JRe~I|KMbWUvxLq=I4S2)1V{z4J3XK4
zR4g0qNqu(#`G<>7^%RO<@V%iXAp%bl6H&JA)J|-HuU6OA4s+wNN1arV{(QxY!p;@H
za}I3keV4hCjX!BADG7r??E4e?feo<7KECVub)D#q87+U%xV}$Mc_og|LA}$by65hV
zyrr`4BA;@NwJ#D-H(uu$ETX*?j1_?H6T<~r4g$aB9xMGGX&s>wzSdFLCf+dj_5L=(
zF6<;yLt=5w!%Ac;Jx5<?3r1GHYbV5a-dv?+kPZ(!BYz?@D5-S#JD;KqXbihS46*e&
zOFi;Sr!v@Z(w5S0u-bC+mP4?v{auLs6u!Rff|nWbnyKvfczZ*W!#HSM@knBtN5<X^
zyojT2AW^h()EH%*E@;)knu3T_izLN6j!zr$n#_%wCuD2Ob=lVy9BqKHvCVt-hyPX;
zDS3kb(X9FZUkK{ekoEBAKy|&l@(Uk1iEOJ7%<m?`slhuue9q2yJg@ihj)`if2oQZ1
zbW0!!4TqbO1*`qRi7A6b4JaWxoa$qfKbuh-)SQA>kUgm}@5@tohl-71?F+?&H)~%s
zfl5+}2!rg)^rMAB;h%Tb^*rZWyaA=_A%-=12S-*y+lmhIAet`7wgjsehfU<61_0N3
zKrK;c+q$g>ydg5#&ClO~${>+<Oo*~A@JVZ#Y*#q6bhHk$)XH-a*(hT_hm0<7z9H!h
zoHQwKuXEeY4-O(Nr=0CNKc!i7I_f(QMiCDgOdWtEZnD49=d!Z2;C^A2Wjd-Ny0j%|
zCjuO#+0WLlHoA;~lCYNR0`h5f1qC8774&Zed>h#&ZrrQ8kz95>J<z2<=2llhd@htz
z@r$!P{;2R1g1R;z!kH1MU;F;)%50>ZMk8#=2<QiCetr}ebZS`%=^bS)I82rueeBP+
z)y`a#14g&B1@2iZ2jx&UZ%G&%l-2HmxDHr~GoC)9z}8e`6B!hTr$t=Wh$v-X>>Z~+
zDHruugluN-P-b6HvyDEprBRLUQ7RaW5)ozGeduyz*<bGyp4pZ|q}FhhC8$_2J!E3y
z@9Qs31ji-hGjY2T=qk2wajEOtsg|y4WIx~y-BEGJ-5uFAsQ1(nI)*L%GVl@6T5pNk
zh2funU(3aH(7N}1;$F5-c+xr);3>tA5-2~4q4lPbpj1-h>^aHJYDQo;YzyL`pTTX#
zGJ69^k=omv0kN!Z8atgYSfh*Wn8};Ty<ui{-+S7N%G1ntJYBb|HnFeZxT__$yhAnB
zfnsV}rb9G>9LJBA@s8zz)MKlQL<|MViySux>$m5K<;@zvZ3dL6=^hZ_H^A4psM&1&
zo6g?v3`W~BNoH^2&tqG2B77g);JQ{sU0%k(wmuR`stEwc?w(LA1A(o~WwsR*tU1>l
zd;=WJH||911|of*!#?(hm42#U-$4X6RKe|}T1x6sTnFuj>R*AEv1%m<;_B@U&n!Fh
zlA(#Te@WCb9G;vnl*K-bxx<n*3y2=p-WW{=_|f_h<C!j)lrwRz;}koCtoaoOXxWO@
zy|M!A3q&-;L{q*v<d$-`)r|chEe<+O^kzU{(+v@vA0fP`pv-%rNca>Rz<yI!py*mh
z5bnH0jZL(Gx(AQ3|KY=`WemcVv!H?H28~+9y8~?f`W=!)Lo?CALoO;<L(Yjrqt1gn
z*+kL7mSdqdHIPdr;`k~#ZDJ|*Aoq<&5Q4u(rJB<?jQGR|ssrlAIbL7Z*aKb%qT0O_
zbyd|q*)Na45}xsNK-pH)1tKvgPPV}muyFbpL3Bk@OZ@pFXGqTl2ggDZmg4R0H6^4B
z;O)(>XkaUy80>}e^nO+4Qh;wS`y1*ugiCat4C}lvlGenZn9+UjNSEgdgXg}I*pGW^
z>3pc8bqKQEhOH7#;;G-(6ZOUL65K&IDttF*ra>$*zOG(WTks9`6I^Y7k0pn0Fm&d3
z_6u)*z{3+|fy+<qziVB+%if#+O;(bre|-eiRT6DCT6zE@_G+A8a-JF09o?i+0!PKL
z`5QytR%g;ChLCCpww7ybI9@X6!U10KSJxxK?93HUOtce3%k!=^iLs1P7AVY4gJDnX
zG?<<G;uBzuM)9kk^!z(8($K);f76E3(E&bDO`oV?k;M7|)foAY0C`;mOmf(#OH`;L
z4r}8LQ5V#FPha%%qk~Dd7ZuUXdUKg?_40=g^4EN&D%Vb2u9^fPpZBh{pLYx@hX*4=
zQUk1Xhk=h;L3X?a`C<dF3;DnI&Jg5&a`WE0j{z@~?;SmKTgivna6~gSRQ%>;ixEGi
z(a=E6n>NxbjPEFkJ`V}~^yGbjbLN>#{l9wu;(2RjmD2#HSjPp;m3oeuoRQV~)za0A
z$Eq1!OqqxQW(Q;O=g0`C7NzqMAyNSPd(p+9pYUQk*_W$sb+5hRr$$<(h!__35kYE(
z5QV*-`krkBYA4?<?+}Q&vOlx?o^XHw>gs3RVLgXazBt}zhYxq14}cHPN@JIXtnRRP
z`@5GPmoWn9qeOJGAs)>lASDlM;B?g1(lTxRij^E$+YxCqM@LcX{!s7*wwKlBbng<5
zPDmJW^9_L%zyc2ZF_g|!hFInoUpUbq+ckpL2FU+1YD+fYJQ7(X{cD`bZWt5Xe$@8U
z@0M<K?MwvC+NY=9eF=hl&$OG2EkXIONTJZJONasrCiXLHX)$#qt@R~n{wYzmCw|f%
z=Tg8FV@-%n$$jyfO4rV_zk=t<uNn2XsxyMygQBz&UPeU;D&lE^uj~(`+TA4I)6@Q@
zZ&g<~3TKI#{h`B<8Qrw7P;N}rt)o&JZ9moQBAN&E!-PY27GJuFyoIABL1OC|kpwhD
z!e$NW|8)eBn-q#m$r5Cn{e_!V(8l04;iJD`a6XW|Ej1Pie+X{3$eZim>HO&_W46Pa
zYxf~2>F^wY6dv#4rX!HikfIO~wFmSfmg#TqV}IHkB|A<>9HDU7Lj3epatUu3@f(93
zxlEbj$}lN0);J&Q57H2!7(kT3L}m@lqx$vGw?t)$efZI|{uR+*DIP+g(}CE4vDV?i
z3rb_+96ntmEVs3#C6u+qU?qM<6eDWwD>}j~vSQNQz@$fCiaFr0K;$H?yyC_`=YH9t
z8|g#5V}5+{d1<n!4sY@6L`)5GKEQtFa&@?qx4*8Uj^ONW;ja0}5Pb9IP5sv}ZXX9f
zd{%DakCthadDV`%BPvrz;1C=C`2uKBF0?M8+?e&}xRRWPd8r|LDK+x4*5pJo+XhTq
ziS%X^@<Jm&k_sqG_dSf_Wv3Da#O!$B8cDlTOCu2jLoM?Fy*CFR>^WZsvEgK#{^EWp
ziLav8nADQ-K>6rC(#7+SuNK-^s3?Xug&h1lwU3+Vr`u3%D&NTpsdn9?j(T?JSGdp~
z%Us_fa1)UnCCo_37>uD1DBn4g?C*A9rqjd20(y|cQb~Wr6U80JlXk7=F+)UFd2F`p
zb_mQ)53(83_0nJ}K`ON9U<w*-&rT{WXIu9h3dN|!Y-Y4Q6<NEs;g?LxkM{m6kcfAp
zTwl#V@q*IIH(Rh~NOCLr5YPNLJQFDn=w{oWhb`L-CzmANT2K@Uzp{e`!Ew}A@b*R<
zre%pk+cLdY=<(}6$4ht|bj)24#HPNY*Y;SK0v80^#xx?ez$Qb2W7<mG%0HG0R-hbl
zEHkYgu6K(c1L7TXCd8?y{`mqFjwTsGjMv?D>iYx#d6Py!CzvzgkhDI|LQ_Rcb0L=d
z1#-6}V5dnn+?)v^i1x=>qT8d6kOrK8etzf|RA_0{(gTmT_{pP$ZQ2Tdc<Y}(-EXJ#
zA>Axn#ejt&_53nbb=^QyYF2$Rd875>RHKmMQTyRX)c~mf{H<GY{-LRfot}cwK7jOG
zuqNeM4~Q-sNy9eipGcb@QEpMKh)z`!=*anNdSM}vboT)-PI~6TMEPW&n1{6Dp=*Es
zo%Gn1@AN>&SwDmydYhxjW^rH>BHv~=<+GUi18C#vk)idvr4xm;6ZH9$4N2~H|Kvor
z<@js>?jVZXy4u<Rvd08t#WtAEPtk*c5uT-~vFpAZ!j`_PbA0RuYX#9UQ~^;!WAPV%
zLR&{M`x4waXrv<#KgXYbZJNCEH6Q-&YbJT&udd2}UGP7@P^K2P%cd?RsEJc(@kAF}
zLGT)^gjJ5hatdT)17V#D#?QYvidWh8o5z%)W+B~dE_t0a8qM=dahaBbe?gaq5m9cU
zBA9~s**`ZBS&jcXKc$fU=b$~pH5;MB0e*~elQnO7vs#Vqe_r=Hy=Nnj|7aqy?eTB5
z#n>(`Yw%^%C><)(sv|x3sXAP~w5RVnJJ&QLU)Zj(_F+Qi9D`Lp*g~biPFKV&)396O
z^_;I!y1?hWb||{uebZQ8dAV5rV<L@?VM?~!*4Yl#-xN_V%qsNoc;yf8E}6B`Q?;M5
z>=kG3%BiQpPu$Oq2raX`=67`H)tO)pO8cve&xhQVa|`Ehoc~N>w69>7TQKB1#%O`m
z8iJ1bC5rksBI+wY$^3N-xzC&Gz4U&azj<BW9E}5UuJN8Ty3o-40dnDh4U1^X3?z;R
zyn*aHYvm2@uUG%uLl3>~-o;+Kc%OfNc*x4M;&hUN=(L@-wsydExD08CGAW(%BX8<b
zkm4jx&9mP<l+eC1r!<B8q}~Lx!PL$?@XNBFp@w;;D+P;Y!tL+)%3Em}el4l~1J`^l
zsqFn)Qh9RWYt7|f7ksHb|7#uoo7Ul3)2E8Rv;Z9cXPaL`+o$E{>e_}Q*V823))43M
z*U86OmXDgg9HJ_9AxSTHI}{}jpLy6m;p-_sC3a%*VGGKufuF%X9?n{+!T3-8$JB7M
zO`!jlvBVdEFz^=t*=75vk;krok$e2>lzxG+|5}HCtpj4<e{>%bQdyqPVQBM)3ituq
zWRHNCuXu9eb_x{4MvUAUWr+0tf<0mOTgK!d36}1;FJ4QdsWCf9$EQ%F>!I{^d+jQD
z)`BglwO_WdskM5QY|Urakc>#<9d$HXkc!8AT_#EkHt4MA4W%S1f?p6P+Gw!vzM@3+
z%JIaa<$Qq)SizT6ssvHX63O68gl3Q|gP6r23Slo{C(wOCI)9!kts?Jw<zvq^rWV)G
z29`iE0Ga!+EHm4FBy<JmF(l#k!%bB7TndhY&|1;u9h#}eeaL~%1(<;$P~u62_)1Wb
z7s6($8rwt)LhD6&Q>H-#2`Qoz`T_7vlwUKtJ_7`*@eQr3CD4+QgO)v|_Qv}9hZ}iX
zPVd=x`hx=-V8M&IeG+8Dp3A24R8)@9pQgCP+bvqTMsDmEfl~0MwJj}K6+g_P<5I8x
zT@yiIZD`7l(-e7~A~*Y%^FK#GrrgUIsFXYq0DGd>|448}J17jJ)j9Mm$kHYLBWOA+
ze?qziCAzkSY5!zLo{p%S=H?0JVwPD`@UD+hkQBAkM_rn7?GW9+zrP>ch`@L!pW3^C
z@-~!@_coYP-)=b(0=npe%kwO3hPk=<hQx2muEbn1o#Db?7#aE;YRc%qd62Wn@%gmR
zR*8)5dxICQEvC&n494EzUS^3l{|5qOUPk{TwBrdmWorgWxgf2dse~>uyJ8W-I*p?J
z`x>D4)NrXJjT{>iSC=(F=Z97=LB$EI&;0FGY8Qh=b!+jWfTEI;(un2Gt9^&1q@vZ6
z^b|Yt?nYo##BJeUt)~X-gi2d2gECBhRVC6It@lzURXZ(vlT5a<M0K|qCGK1h#XXb*
z2*ptDI<N?m>3qQY?|;0S(7-mK*L$#6>A344bO8aq5a;O)Jwk`5kRnMu|H(3zcbtm~
z0NCB5=ss7jpJ&1XA7F!=T}V+5a?hlO7GjVbw1=?BJ*B(6b_U>b($s=VS{3#_ASc&&
z6zbY=te6ob+r+4E3xBa<fBW&@sl6k2UXk2a9Br`nXZ(4?k>ckYvY>PgA*z54vFyUa
zlO>#!FahXStIz-ve9*A?I*!B~TD^i!++J;jX0U23au`haOz=14fi6gm><1ayk?S%*
zl=x@~7dxEwrKl%C5Hy>~d7a1gWzNt4H!{-L%nz#`zPuCztUyZ%3JS9QG&VDHGH3PG
z214zD;!%W;dKr9m3pFjx80{~L0a)}wL5<T5rmA8m!Vio_nNnkeuS$@sq4|z=ONeyL
z-``N=&{}Yb2V*ei&<pcWzzKbDKs0V#8e2LSM;&b7oUr%vb&<xX!I|DVec>fmzZb-q
zQov)hwMe72eTNR$!$Ghp<xsx{3N)jYh~l;?K8F<P@i}Eg8qIX!)NqPKW8tKi{K789
z3GAFC*yU@tg`a&F^rdX@A6@ld4C{Z$L_o1<ncH%`7Ind#8gZ!l1xiu94cN$&Bep!Q
zo1~wRo(2gtLp73${_?{a{Cn#@57vpJ-Xu`!CHMKM8QN#b@jeXKMw($1132BsZa168
zC2Yrcp`hC$MlFsU4f_tKfiuA?ittUN51CXw7AXCF*Qr_%ajVr8_WhNb9oj+~A2|Ns
zGo&jv&)0c6hyHJaPTl^<Nzkw5C>&3s0%?M}<sb?Pr0a_Gt-x!K9P%erB<a2Ai(o-+
z@g8gWQ8ji0&4I))w1WnlQbWP-4ZxRM+A6-LIWE47#TRix69{|4328nd)beL={DjjL
zht8cWg|8?VZEZqAf)nOrh@+kQ7(~!csQuNowLQkj7?S9gL5vfQ(J;9K84Z~p%Y!Uc
znTI}vjp%~-R@<Swvi_ZT^|1$Jd=K%J%~DWBJoZ?c8*Iy*JQr?*n1J~qhCwpMAhRk%
zOd>MtIp#1z<2{-vbJ}`HZzt;4lESLoW78YvNy|Ekv5LeE5|-?(>f4r&Q=D<eP$fw(
zey$@y)W~nHn>+sz*X}d%MX^`gN!xB8T!41}l0|z~A6nmj>dbSabZWOVg|5!2MTu?B
z&6NA`cCh`w575%kPuFJJ4CJ76E)9b3(ANM>g1GunPT)DgcfV~Y7eaS3=|ahMM#Ixa
zo)k2Fk>(!KN?nZ78>y3^L-6yVY8!Nj^%tz*)M@A2tnnvw=aU$n(Et0Xjis2H)b`2x
zYQn_4mnxZF^mE7|K`^M5<lB<@O_i0EArP*UQvCwH*?UqJKu^&N{d{*{QSR-gsLGWQ
zVbL&3Poy+Jj1c@mKbEk9u#RMO047_U!ECe0s7o;dF`bK>U)#pj2H)+_;nTo3|A>yK
z^Ark&w8xYF@AC*WmtwBeLQk%9>+LE+GG|!z9Ks~1z@wIsb6dk0Ot4?)5jN`(_~F>A
z!evZA^vumS=0^w+Xj4Z*`=srsCk;;`(CHKgqUG2-FV1ST4>kVvWgWo#3Mi7MaV+)5
zL9EICz;kFx2_f|aA~y)=MMseN^^n)fM_?p}h)trEo}4~h7#bE<m+!%P3~)4vUIS=z
z{s1o-V~@`FNY5zkz`UP7_s<_6QWyvOV?JFA26;MS5$l>z89@$<Zc;89FB?oBA#mze
zZxDP0Whxo1weUGwUN^2W7Zhq87LqWn3`djgv^9FH*QU5te`Q#(z_uf|x-ONF?p4JO
z_uZ>B(#ZFptdvp-AaIDmJTjy5unS_NGUyB*LJM&?>tSF@tXs&Z%IV_;O5<pr3{3BE
z7FIOPujnuz=cALsXn3Gr=X~MpJh6wSFy|2TljbdsSPo#Y7C8-%(bSkorK^M<-iJ=#
zYV=w+038@5=yKCm;n9Y$BEN0q2^A<`VA0?y9tc1Twx|`i?%W%+H@B>R3zD|mjc;J-
z?k3WHPN-u!v@-UQCS*trE-QS>{PKJS+jG8Spr)V_0KaGw;`57``}PSa<n$XU^00ix
zj*UU`W4~}7uxSH19Y_W@5TyVrKY<|lM9CIIjb=H$<@5~-r5yT}U~&ne>$b{A+F-*g
z$aE_-DCP}30?95Tk8v()^IDM6oI>9!<A9w$zzswU3?uX;B(}eJH~%M{Ppy-<FjC|~
zp{!ggZ`_xtYKUEelC;K!{lq^4s+%fFa4@bUs7(&VC81w65mh`2Y5tEw&56OBopE8U
z^nC+3AB9$wOvnbT0qgrQG2-U{GvIb<bi5OSGpT;Z2&FskhGJYR&WIW3%AD^k{{ddR
z^)Enm^Am2Y8}V%>j=K(8#1#X`kTUJjCc()1V>l<IvO?zS5Fbs7j4;_c)5aB3YIrmZ
z9_V$*XX=u!ToP~<55Xmpc`w_5F|Ap*Ua7^!y>YL5cG@Q>lY#G{j0GWQBv4N5VHH^m
zQmY{WGZP4Ik_s%u!_2ay2a!JRZ3hkhJ|I-EX$iIe0nC3pPt{7OoM?q($J)9^PbZJg
z7#fX6KKX-GKglSMxiK&K$D|BJ+PldUkwG|4=*oB?K)f63g%hL~_s-m(p&4Sa>iT<}
zQ*-||b}0vp!ye70l5c_*?zGZqX)Vhv!8bxonnSvnFkGTS5~FzPt%w^7&ifoi1ewG`
zMqMh!j!mEp(H|R^SccBLKge7?$Upi`8&*sm#?oO(#)?cFB8ue`l9r&==~N0rTog)?
z4EixrW!wlYLRP~RL;$@AJ(U=#wU9^goQ9nY4d9^Y3`J*=P)6LS%|KOy$(Sf(V>Fto
zBJeSL;8xn1q45*$jL56HB#NQ_XM)xpO|%n{c50$0b;3-e3Cvd_jbK2m9x&*9`1o5f
zku?c3ZXXQ3n1u?=A}MojBzU~r&N0}yRNpmDX7_Wggnlx11rt^dnKiw3U7C#4?{A87
z`+J&py2YHA$E+9W!V(I%y@b(J!8hw(M!}26sdCr9BZrXr%_UTDt=p_h2x|@zl*%c(
zR>rMsq`%6HsFVPpc4)1OH`ilbkauXeQwKERdjV-oCc3W0p|L_jV1oX?Of>3FVmQhX
z^zf7u<q1Sq5?MMLLg-DFhvT6Hq%-131yBb!AvGm3e2Iv0%fpYS9-2{kElMW{2a%Ih
zrbFNv$v`vod<r^rIGK>zCuyD~E$xlxM82$srR%hJw@AdbgCOyZ{6geC+ym0zbv1<K
zYA2}Kps&53jI+dOxzHr{skpeD-$qH&x6YG0$zg?WT5tD!dzPIN=@hzqdit#I_CV+i
zcB9T3WEZE&a9lj)hB9vN8u{510lhArOg%DB#df&qAYw4yS1!L;C*P;fn$jn=Dd4-b
zw+(-0#bEQa&5}>USy)~NRIMUCU!?k3zEi)NDWnlhk|%5msTP=aY6lv9KXlEAja>fN
zhK+S#bSMTui)J1|r{@6yC<Z`86hC3Nm9>0fk>L@Eys;k^s58=BKnh)GTAn~_vN^SC
z#T1(_Kq6EPSwk?HP)Evf@W+@<MMhVU;Ytjbw@c1)KVO~bEmaJ_@f(!w;r6P9wN@f%
zgbHM0*=rHo0`nx}eR7tb%0X(DFW}_D(^2FhpVvz2xMb&?x_((>4fvVWe8=e3mHl%A
zm{3G0gpePvog`!3X6|clo>8g!iQqR}HbH8A+YtH!(p|1$a%HJJWU$6q8oePa%KR>h
zPi`b)rm>4)SCk_;te8MZ;7kslz8OXK4sm0kU&R1`celXA<X*7?wDw7p<V=^hUXlRl
zajxUEZP*i3@NfEIveiyZadN^ONgKGr0WwX1j297ThA$U~9)QIBayRGUP{&6F_o^CS
zIaFRVADj1TNS)2xP`=Kbo1s0iS)`y=BuRx<zqmOb=+S%*HvNd09a0Jx(<CNPP%nnu
zly$wmO=5LlgyH!%Zjq#pEe+44T7{O06~$fIzfIcuIvETxfC6-%ON6K#k^<7itk@9|
zn4E%uE($draXRF=CLu&m70u^}J|CvGE<)sbRo1IZG1t2aevw8hV1txm0PS?Kva-{N
zH%bAn!-jn20IQD>`9@_un(8O7E%!0mBvwgGjZ8ruq$Ox0bJY!|zs)LydomoK+(;&h
zQ~>bDSiY8ruT=~nL&k=6HT;mdXtPKH?NOa4VZ~ZWQKD*9;uub5cP@&O!@l)^#_s~#
z4&S&y`aj`OKf8Iv9@MmfG&H?$s;_&|IdyT<Sgj~Klt3>rn$KDy5PlAW$pV4TjX-M=
z9Gil+LA={GhVxK1#t2QqnmuCpJTxY_yReRvn7GDtV=Z?@Z~Fw=(vL#Jry>c`r{(BN
z5fPCP3`ZRK&F%f~yPDBUtv;#GL5oLj-BM>r#_9bWaZDORQj~~qv+5}}f?p*QZFgL1
z?cQ-qsyp64XX)c2Y5iHf{`=p_$UaPPbtc2SNIXV3KZj6P?8g%p37IRKzKVa_aMG|6
zAq9Fu7h=+8U`N=*M7qhJ`JZc*$h43kawinpOEJzuRmQ?@6rSA*Cf7(a>x3%S2HxBl
zqu6fD3R$XC*>L~WmAD4e=RTg!jsWG>sCSFmbrN2|M>Ok344+UwtjijLeNl-g&d{My
zD7<R>1N#z{Z17s1p_4b^%`od_*qYLTSZrM-JrKW{Gx=B5gGd80(r7oSpxYpZO1=24
z$gSmTs<KPs_;X}kWg*D*h-QS)Jmou6u>%9>H=itCOQ#aitPM0K)+M%dnh53>ev}#g
zMQvbdEo}=K3Z@4KO?o4h!Ubr1<67A8{n1;cY%<}8?Z~m|WaTj-!IGHCbf*Vv7#J&V
zakTy(t<q`6MG(ijjKl-7IXKKAN9|0tw6qA&H^VTlxk4d1MEl;|7QB14;EWHU@nSun
zT+W9mj)HGVPXO_UWPiX~iCupAGl}K3>6V>D4WG$b`pc9eiC&oDFo0%+DpEEEW+i+B
z$x2ch)U?^PV0gv4ww&Rgt{IUKJU-f-M8D_!*$5|ASo}-R*0<DQ>1L8QbHvyEl}Y~>
zD+KH&0K;BjmL4%C(us;xo}<9|35XQv5?n4OsHFlnp9x-u7Yo_*>Nq@4s=`}PKb&Si
z+3SglE;d>9zmgLz38fE&qQBl(558ES6q;mAhI19fAlRRXq6UlD2d3{QZS7VWKQ|mR
zG8qWnHuj3?7VRmRzt)QbZ-FE~wSu0DxK5?ZlFs7<b1dW>L>(ch?;gU?50!zc@Kns0
zAW0?3^Hk$z+M#!Q1n;|#%t9q+3~5O;fKxN{>1^*Ml|Of$V57i=;N=`ot`LpgllIWV
zH!qQ?ihC57kiPX}{HPll>KWxH3(coB98xqwslWV^BN>c?K%8TugT!wVdNWYLl1|lm
zaI9j3n3)vNOJF#W#Me{W;SQv67R14$^wH!o4+A!xFg=CFlK{Bw>(CRI%US^CIE}qs
z6D+fu!*lvAS!de1MogzT^sZfi;PVN{;r9}yiQIIuzlcQJ`2ZqN1-{hjJ4&!mZ`W;3
zy=xgv3vgUWJ@l`kd81(dG-h)bIDdp9fo+g@Y9o;GP|C%LJg9m4uGM2}BwAkx+CPM+
z)3GiUbQ+;Hf{YFYtTjeVo~!YadniCMFfueVF!351`sZqc^R6)G0ShZJk0c00E(C}o
zgvFvK<wu~6MhX(TLK(ks%&v+6DI`z|Zz>M?4VmA9*hP)(!*EZV#QddsqpF>&Ii4&g
z6HWIBEkNn8?MRCRsE(76u`Whf5xGBcAB}lS7&R4)iFyOFJ!FzGVL}kL?z*7=D%X&8
zFX+Y*bQ0OXg!U*B?uW2$45kSgp__)OA_Aq6OP`}b?RL5Gt>ZY(G$NxJ=y%qMW?YT=
zXc0iv`fOQ~A^95_Za|<=fB+e?bxxNk?qly{lYu6A*xsc2aRR2$gA^cu^iw|us<<2Z
z<Z#AwlO}hgQSC#svUr+c=|n_K7>!B*Q6okrvM>>*F%8{b^<dTmkLCg7WI{pxK!DgN
zftX}S3u+E$C%{c5)n{^W+>rlrUr-^*6>dI_U~4g3^1#e75dO*mZ!iF$eoEhMs-b1N
z+lW_<_QzS?5X_O@k7-C|m{0qVOo(kJP6qO6q_9G8=A$x+E3g1UF!!suW9}vuiNh5w
zGnCS^`tel~X_)&*+$YHZUMh!&0_@9@(avy;6H`eZvSh*{8RXFL{5l86w?E(>p46*Y
z^F(tpmOajm$UF$HCUQTtjf8-j8+hg;dM<$UD8a)Hnd-0L@U)90{in2FalYfz+={lS
z=NO)`lqE_}(!el6N3%m_;=me?t7vW~pYR0hG*@9F_dtN4IroJgK5PL|5n?cGe%n<?
zpnU2G@`nViryLyL5sT-bl!D*XZzVHZZNTddK+Rr73MDzY$_$JKoP_C;zOZt9l;#e3
z%!a{qxXAZ}9z}j7f+EDLJGqv*Q)<K=61OkuPIP%n@)#rGTaR)!(E%Ssl0gp4T$y@a
zfWS5h(<^6uw^J*|F<q$7B~J>ifcJMwTW2;7f~+Snr9pTrhbR9d!m3j~V7yciUXO-~
zR@kl$+Y*97qDPS<A{WLmLUQticdo%keOcsT!hDd!MWzLiu~c7cP-IizdEMuZ5`iaP
zB>#^9&T7G6od5)zmjLusL((Y+d2BCOjOq<iZNH!5ls!x+Ws>xgR)AN}{mOv40zlFo
z1b8>=Qdz@M&~hb)#vIROJNNi481zDZ`#+ohpZCy_`6VgNOI4FRtKw;8g~XFrfBk;|
D=YN7!

diff --git a/reference/plot-8.png b/reference/plot-8.png
deleted file mode 100644
index 17e45a61252267f965ebef123f2adf35cbcda259..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 27928
zcmeFa2UL~WmMx50rDYbCBqEB6A_5`;DsV7>h=`yFNLG*}Npc1)O9e?rKoBtjl0|Yh
z5JV(PPRfxS4>|MCeXP3ozJ7h*|96kqqx&D@-qAg<^n~x*-`;D@HP>7dPsKAbt5<AV
z!N9<<TK1&WIR=Jb9T*sv=>E18f79syd<Z|5>dMJT;V1q#bv)A@KbBuTsb<Z<z;T=W
z|02J3nmz-=Rt8zAqvsu-^ff!$Xz8a*4s|~HA|LE~OpZr?_{Hg)ja;gCPqC?QG+wk)
zWvM`->e6LbkF`h039$UKk@>ef4vcS}DIU>Jdbs8sBag>57J*--R`4C~ZVERG`thWd
zQ+S5^fPCeTSG&aKTxve64<722aF(ohw9c3*$!rZN>KvJ(U)@&QxB2|^KV2(j99m5N
z^y=X?7V>jz*e_Y+Cj-yUZ@-eCCwUeLke^3REq7Q%{&4QpyOreU-E)hz$<J#K*Z;qN
zgvDo<ZaJMh!^6Y71qA#a3mG!oG;ZKNeKKWkcBG8OrZFi_%>C);W{K~eSvK93ujAr;
z@zS*Kx7gMDACz_KEzC{$;V&lZ3Mna%cz?mS9zOmqc)yR2&siNE&attvdu5#~Gy22(
zYies*Y%)!??dQe24>^4(o3aj_(Tvf`6p4zCruHS4nI|i7Kk)30&!zPA_EPP0XSBXN
zm2l=&U5<}5#ho4zn7{=UCK*?8akVv=#QVH``?h3UqU+fW-7kTHjEt5I#u3tbdQ_E2
zM!chI^CAlYn*uh&9RW8MzOQxPZFlI%J>inNSX@7kQg~A1!ZOLpqf$~?^V7Yu`&Qu{
zqo)_QJWL;NFxehjCPNAJ+wObtMKxvh>ebS@n>SBNa!h6S^OuQ_oc<ClD!|3XpZ?>%
zh~c2f)F&Cr`j59(dwF;WxO`t}K*J+DvX=YXuU+{$QzaJM{Z(hgVvi3N86?ttEL^s8
zaWM;M_bt7K?<>E7r{)xC-{QE-Z6dQ(S}O1QvMsEQ7P+psTb(~F-?gKE(4|xCCjV-@
zcK14R|F)eW_T2p=rD7IdEtzxUjfGF02iO=H6~t*$Sz|RCzsAyEx@wWdF7f$EYLWbx
zAam`cLo7NjmX)blS>oDmQ|%?=TDfknmGR->+*avL+^#vJVFEi{gLFii>*@l<7rb<4
zxg6)qkFPaLSO||l`|^Zsmvc$G6|L)~UqO23Q})dM6pi!B%K4w}ZKwWtpy10UWRRq;
zx9x{jdZ3uSxu#8pwB@(!%L?9IU-I(B3$9zM*w)Vs=DI&|87{J<`iK_HPK}6Vl@3YH
z)BMtUqM~&xU&$tXTqCxlQEcel5^B~+saNa#cxr-a{ad?Ur|Iqp>U}wW&jh`~6ASZm
zFEcU@kGD7v9;A8;wPB?^#(Ha6+SVm?hXrWg8FU?V7QFP~57q-lisbroXNz<5>J39V
zgbYi)Z{50O9{)Tihk|DtfBCfQ_mw(E6B6l~{J!GDpMFSe*mb5a#rDOn(}8#UI;T2A
z{XBenl@`WRbA5!$#Dz}yZe~+fR!(R%&&+r13}JDfX>3(*E|n9=-WZ^1CF{*5#63OU
z&wl*YDo;H>36U)h->$FmTPT@pNeNp=9n0vAlWU+B`1F>$1&hs%Ry`QMv6hmSKe!Cf
zH05`34bze!;VF+#GEzCXyU^lxkw*KvH<x5vJwCNMP<?{?XRyfJbK$2SS6&HS$)C2h
zTS%|K`^OaSh<V0%(4<UaqUmxYK3`7TeIUJa*o#G=)cxBMP9tkEysjwzLW1r}U+o!f
z$M#3FXT<uyejjX)%N)33?YxqO{|l$pr)?Z9+ODp1+>J&?MxR<O3pu8~f4rSKGvCwG
zGi!l6UE>-bJmY9y`a;@km~(#Uc4O?N$)WF4oll+Pjtkr5<E%a{Ge|6Q<8<nJE;6Ho
z{UzJ_{>HLS)p+e@v*c`h@myK~`xJ+LQ$h)Kd@SBwV#>AU&86e>NmXY|5*P3g>z)r@
zl5;D)$r?Z9LdnH*f2E+NrWPEk6f9)Xl69FKx7OG;u>`BtT(+v^3<*B26V0MrTvtp?
zsAIJ{7Rjymcgl&R$4sf_PFEYG&J2BAyQ`n-Cz-?W$gZWYK2a~V+0F|mAfd!fL2}N<
zn?o$+qL5<%zt*Oify_Yzzi$5^A;U9$X;fc9`++8Yvrg<Vp9g0S^Dy(Ne>HDyXlMvL
zV0><Nwo`JjlG57i=RT|F%__+C)S=DOZiZhyY4-`AO-zXfC3}ok4+`t%c`z3<oDMp`
z)8aPMuQEd!`>I4Ex9uoyQWMkFm}-$)Q+v6<n=PSmi*w$Gn=2dp?5ElT3f&hb+-3Ci
z^jz9DgnDS0r-zlyD^R8`Z+0E6_%%9Fkt)p9hP^D`aGS&K!sY#s`C1CLIQQL4GAw(_
zf%vKwM`Hf&y*TzER-bufXYlDJ$z`GQqY<wptbTm*sAIivTKDF8^}JHV8Tp1~oGwGm
z)k^2it;doNPq}zC7?eE?5pSLCy~(0&WbLdo+s2|d)>EA@m@`vAR<Jrs`Cf7Ri|5a`
zX}gY92kzHDRXIV9Nq1F*#o6p&>w5iw{Dx$c?OZQ(a@{Ov`cn&WXL<;Uco$bdaejtv
zZ(E6It2Z)B0ZZ;=epgk5LUiMFq2zo*k$$(%T#qU>pGC*pwzo!_KftS?px{EHfqd5N
zaIv6m&)0*qLmwMsEU?kNBb6cwOyaYddu!tre2d=d7wb-sb|<vD&Fpk7JKP&{A;ZR)
z#qZMzY;Eu3YYy#LUb$?|W>(Qw7i$t5E6U5);lz_|gXNK%pIpZ%=r%iPCqNyHDtfSM
zSHN&>ZEYc=%;tom^^DOfv3F?H!d-sx=^a7Lwsxrk?o-7HcKr<pC%@g$YsqpXfBw{N
zmEblvJ^7Jyex2ArnqK8ZlXaPob?Kp9TO5DfBRf^YB3syhuiAQ~!iu^&R*Nj!i6>4y
zk+KVT`g@-uCi2%_=%O9QMt?Q;e68ziCE@C?U!}1BhbOUKeP8YHSg)C%&#V6!Igb3u
z9l+0N{Y5Z#{7`R9;xOYq0Un+rNi2`-g5gs~g9kahDlsRuIS*d_W*`tT@F}b%;;GY>
z%1+#E@6g28+56)jpF-WHyQi?gCF7E`ybTX>WH?=IZckMCYuVNBtnU~plWgtoM<}1g
z=B;fDoSz$X9~>PwD1Guc6lWweAYD{Mq(k9Q`@_q*t#bqRjeV{6LJj!?im{>I%jVu#
zb9i?y)<wi5P~6d~a=}}u)M$ad??pyN#w?E7)4q}4SMnR3Eqi`@OHh^w*8U71%bTzk
zv8QE{3$rBy_O!PA;JvQ<uv2I8Y}NY_4Ts4ER!$IXkmx^-FtG(;A}5~}r&a85c8H)}
z!H!19(F&Q&^Bb^KiVg-tcIT>^+qa9fxO8O;F^^2SSanY$s<G%5`pRpYn5Fb8DM&m@
zlAP;Su<x<zIksv4<&#Z_v-e8O-8?i&^qP6UjC1d&lD8`H+964;3-c59PGZ=5nqG2W
z4NHQUG6ynDB2w+^F5S+Tk({oubLX!&#jQ{|cllWW5uF)KEJBK-(h52KJUjqT945ct
zei;=NRkFz(ne%l_jJJnJutiRKqs6R>w)PgpjZ+21wzXOTdk<N)KT<7ta>$yCOLyqo
zd&`Mtdy6regiTj@+H_B}XlCm~_Mj_4ILQtJSWVySOO~ZHNIodabD!^Y=k^=So@n8(
zR_&r~N}zRIl}|yoI%8oN5n^|$!GIvZBW2sa^}oUu9c=Re1|i217pjq!bM^a&9Wk>%
zc1g0iPLFz4yp)lVDh^`}YO-&2m33~*^K4eOyY1R<qGMudY-bYWA7XS)L!+^MZi`*r
zCD|DqVatf)RD2nqu|nX!=lPY(n`13nb7T)rj5k`yXlZE$q}U@M?COtqY3Dtwq*OKF
zjMZp+VqGe`WuaZbz5Cp1j*14SEb>G&Vlrm{Gfw2X&1Gc!PfQO7<ZccN3o{!x&S`po
zZPAi89JH&!bHNLXbh|6Vf@kj;%Gne;nC1A|0vw2ChxDhmQj9IkT>;}6o;m!Ir8N|(
z;JU|Ha=Pm!pSDR3|KPUUo7Xck?h3Iv2h=z8VP%3rNf7hgAU4yh=K8VyD<iiHUrwxN
z#hw^Wi#}=~mA5lM^L~;5URr0C+T8ZGnBOZR)a|fJta|nLen3+(WB2jI(0(f|d8A7c
zi!8t2WGSdi(9^>$``x*72XNYgHr5jz&pp`PvndTw<a7h)!epV?Y>xX(s=Hi6fAijW
z#g@!179)BKH#lc6MlbCzvArxV8rtskY1{22vEs?>#MhZtOWqE!#><<i$w+0o#~8~=
zOG{S{3_SjZ`+VoWVaV=1IZ;W$%aRxw5rPz?OrI=I1E`Xc%DX4C<#E=0r{uyR<6OkR
zqLMVAne_VN52@k4rIRKU-nZ-I4)kH`2;$+Bh~?oU`gFQLG>387=Br0W2U{hVB7}y{
z^%?4wJ~<Q|BKWwhQ$uqg2l2}$J<f%~Dca(o_x0=7yt6VKOsYDrwqHWTM1gNNvx}H|
zOixcU>y(%cggbFm&l;$a+;v`6RmRrV7645ekm~&T^E-4(j(2p3Hfp@Ncr@Tj-FnHX
zA_ZNt^=VBA#f&=Bl>sU0s|Ty^sg;!kY&V|2?KM!tfpn<e<no61yY+}<(*eF2Qu&N-
zUfjZ8hm-UYIG0Q0)>V0-j!o?B4|cKy#YfCsU+c2faS+HWFe$i3)yLx?@^WVOfbZ^(
zP4B3Metuzc0R<>4+OYl18M9(9ByhP+SA`a&SyIVaqy@Lm8a$ZD6*c0|yU##C-)y1#
ztW3bi2EIqgiwOyWj#-yJ-nx)!XGW@-ns^;fq$t1D>>~8r)8r4xvM?#l+RMjhW~t^b
zmG{{2soGb;<gl5Rpk%v32Opz!yMfP-c(H{^DPX*o<8YFkFuAFlyT0}RZYsiF$m#Af
zJPG)X8Y4i*?a=aeZuy{xS=uTp%rYFd=cSS~Z>LVJ4x)@F_}QA27@s|Rmhr(Z1>a>`
z9JZp8()<45ChgmewFTV~p<8zD+{wt8aHTrx;6NI+4d~Q_)-rL+y2PTXhD{)M)=JZ<
zJFJKb#9eTFt;ELo+<B)MwM1^L<o1AKX-cR<&NzSwe}AHEf<;O#TY#3GnwO-yl$1+{
zP-uc%ufayAED>XhmlI-nNXwa#WS8K$6Sxrur!|W#ScKO)5S-}Y5h!e;GTcPa-;0Ii
zyPm9LWDn4Hz*EO}>?}ScB~?+gyT9mdz3D1WiZ}(OLLsu@=H8lEujsgpJ&toDp$jfv
zY<C`?4$rrA0s7BKS|FL$IMk+6R?x8Y38_Sxx1aF#z_I*fkQn@jjYZCsLBM$Z4unf>
z=l)0HPbJ;mytbyl)WK>x9_=wM%W@oP(`cPm8BXgyp$|w!paFlgM@;ITQ3o@tcZ-(o
z=wFy?UFaXR3!1;m#2CCcn>?S5q1RHqD8u*scJvd_#K=gC)|uPGCBpxG-^F*=nCF7y
z8)qUU=Xh`n?mGhJ{I*DB<Yzp-spw&A=^TVuoYwWiOLw%ZqR3i7V(h#Zh2Z!G0$2jZ
zYiDmclDaK3@rr89g?llPn!k|)6rDVJfLArnG@*19NR%2#per6CzPUDcUYI|l>xJ&4
z+LugLcnVvFawaBTz9i8~cTQV~@B_~}8HemD%AX&95d4Dg!kew6EFfi@R8r2<fl{%-
zCrz)r@Hw_5wmJ4!h0fa?y>cbvsG8J{-o+CqfR`;Q-3E7dg-(4jygxoS(aK3`p~Kms
z9f>!BYc$P^lWVmbyIqqzMynJQ8pzdq1#6FO1nS~G^MsXPDi04m-$UG#nfkI5j*gCB
zf`ob1WTr?Z4=Bl!_;pL<U21aC{LElz5|V9oq~!RQP)SK+KMpaG&!0bMXc7`c@d6I|
z=ga#ohl_UY@*~jURda(=2GTIM!qd%+jP}j8s!dkKd_^WDhLU#q<gSuJ9XlTXZfb4v
zIy&0R!nId}BWNzeyg6eM$UH>c-0gUzhX*R%<lnBX1)RE&WOO#7=Sofp343Pa02wD-
zaGL{P-9!v?=f>suzw5lcEPMLN!Gw>Z@)~JZFS(3G>+l&##HP31;k{p<-)NSqHH+MJ
z+TgI?=0p3<@t#+Md7c|In$o;J_||gYSi1Voz3nn5@(O}SYz^w~b_f+H9PWO_UrmY}
zo3)E97U8o6jYUI44qo~CT4hEwwowBqx6R-@=j^v-E6*v6$R*}3%ncS2xYoA6_i10H
zoPf`^<G1oWcRgkG6CXZ_GFq%CBNpf$51%Q~u*?_FN>U}}0wW{iqeqVf4ZZ|2Hx^yq
z&sQS?pqHu#lt8baX?POJ2e%+VCtQ60dm5g%9soda%m(K+&t2U8yzyE4`H>8PmqdXd
zr1J{K^mSZl`3Fb72<gpE(r!v@U6|DdHp#in+_&9NoKA`mHU~#8`LZXf>Z2{FBinLj
z2E3MU;Lh`4+<rgoct$ZG857wo!wwx^9`hO8laV_8c>m7F6*-qaF72h|$ZRV#v1ZY{
zdbIK>5-4+RwF|HU3MGwMH!Rq*ndJ^EpE*8j_(9OY+l|LjAk_Y`e8UR6sga}0I3D&n
zqk0j<LK150{{8#p2gs_DTV)dYi1@l6{#Y&ujJ2DW*XuO*2JE8gi9t?s(!AsfQU466
z_>JL?w&cz<58f}tPM&XFm~l@i=FgPE21Y6N>GJ-wYHDm^gIU&3u6|f<2{O#$Z3sD=
z31yU-9e(jS(>-;@5g{*Lz1p$=@#%*F0fnHQ-e2zjTIM=n&1ded@*9a%yPi65v26!{
z_rOz?v_nPp`t|FRIRO9>b-=HB;7Mqow)<^Nx>XaSo*X6p+O<Eqwg9DS^Oe?M^9&yH
z4i689`Kiv-&tKXO@O*VJi<^pR#Ok|w^*AE5Qsdp{Or%UTIbBe{YmAbmyV$#Ki1{6W
z)qQqZVPg}C=I6(67u?-;TwUBA<kQBOr#N$D6Noxe@U&fBUBA}WPJF6=z{JEvlUx{)
zELzn$zKV6<R#NR81jrI?G{kDs0C!R@_}R!kdiXGJ(0yTsdD~!zre!{p<=j}UeX)8N
z=XCjP-HI2|g2t7v*5r1^B$nQB9O+=n^7rr&it>{fue;1W1~^fz>v?}7>h)`*ge`;B
zsyZA(24A}T8!DsLQIpe=Y<Di2-Jl?zP^4aMW=wfGnb|sLezb?g<P{3<G`0?AKX3Sc
zo3q7WLFvq)uzV_VS<<sHQclx=OA8;VYO~$m;$*aHi+I|z#zB7UPgT@z5;Q#Og4ooS
zO-kXXJ;@mr3HCt_VU=*QdE!1lg>u82q$#l^a~gkU-+K4?kx(jXW)H%Y2%0ygb=Ah}
z5CpmJ;t}sH5{FAsl&0Y<Qjym^D=RDIQv|Pmd*^@*st2$rsQ>vPE5*fuyg}UIDmnCS
zs3i+f9<b?5eOA4nY?%3Rrez{$vc$q|1Ib{T?sH?7<R~V3^78W1dDp};ideXOFf($3
zH`xa#yjd&wk~M9sm5byTFI&51b5*#!$%Jz&h%3FGuaQMm&kef@a6%jT#=?}h6_{8D
zXge8K&QF!Ow>;kjE?_tZ?2UbK<#w#(kxQ3{L(-EEGelY<0p?XHNOB;0B<SYf!QM1V
zTmZ?kg%j*23Ko1cpZN~L(~j;%xiM-b+H9jFB^9HSOBp~kvYXKXG04a$h|M?oHG)zd
zn@k>+)^?Dj?RX%$m(6-t3S{jASP^ZUDB2b3y(`pZKNpvI_E?Qwz%{&-;f~z9m2Xy0
zb}P7V0)$Bb4?S6=;096*Ae2<U*xE8#*~xkM+-q%#OaElWAHQGO=)t1xunk8qAUDW7
zCpq>j=?RoS_}}>Qn!gxzdZzqOGU@+eGX9j8){%z2YHwfF?KEX-Z*ND5w4vCa&r6eE
zwa>~a2J`~JzBE=6kaW|gO?NcUmQ_b6gpxv1zu2EC^WhdVoZz$S8MK#<D2l+ed*LO@
z_>{+|xmDsc)%wmkB_TIRNlAfT+N+b3^{IN)l%yb3<*#F7`RsTkRE&*{$vNiYB4rGS
zAL+Q)ou3U71&zv|UC4H_{aRhkl=);!^xL=l3DL>On4FxPfb_|w`qdg6LK@YV-qdhA
z7>*0%nsZ!dPU!3F<C-<o_v4AuFGyLLiIkVX+03}+SoekbvjDG*rDcs(@jAHxY50?%
z!_YT`Q;<fp3uzonmaXwd1wp8w1iAV>LGyz(Z1hU@zq}ptAAX4c-<vXF`TWns_G)Jx
z+yWfz&=VP<3!qHPnaCDcLz(<`aQM@9J<tTac8+ArGyGZ+y|@MYn-q0Tc8z9yPjvt{
znsAma(FZ6%*~T6gmfh1xb~HoE+3Y>tPejbrPlp^9J|!nd1(UD4`RcnxBt4;rLF)*%
zC|I&qT(NIGIeQE%!fSuYlJno|30CcdRNiXQrgg_}uaVbAWbs1iE5Ns?#tf5JGOUQJ
zZMtmMkYtP&%Pg%h^*4L^CpblGGY)l(Oa&gYHjt9qj4U7>N57Pdy<JRl2gu_WF|Aw2
z$cWGNL7d?sfr@Y`r~(ULpMBXDsORh1Y}Xk3=@GrlwebvhDE6<GlxjA3dKzL9h4c?d
zEasf+l8GG`Lr;TBm}6@V`jTAzxH>jR1ljv?NUPaVw2rj*iDxKtKag$kRi5#V9A{JF
zsI`gN9Fc-QbP7f3c}MyRg&!{HbSz|3>K|cC)~CNlqag4~s9?2aV}rvLib%@&_`vrX
zs;U6<<Q$SOw(LtNEFfiy>&&<hj?E-!G0s{Ych@H8!E6F8>f9IR8k=fR!fCk1;stWc
z#C%ec2UB@BNBdenIb531!&(?g-5foB=!8ZA?MkTLS4&^z!h>fxIxpCIQ~&-S$z_5)
zfml=jhYxXHL51RR?JGsQ<3}=5ap1m7$G+Co6!4`sttT~tCXjA$vkJlk>h#>|+Uc?0
z)RMXOiyG5zVjp!{QoIGLDSA=VFOxm2_~CNG7Jp$`aK3O-eDI)Nlq==s&sSV2P7)43
zLz(&!tGU|vD&U~Sc(Pf)(3RlO_Zmt{50ZABc~W;J)RJ0|Pri61^IlTd11^~)x9&o<
zJ_+V1z221%A<QH{oPZ)i2Mo6tqn4vb4N$_Q9%W>&{oUNiXt$NLglZ_o^#y{7Kc9vQ
z2~UfbER{0uF7WF>L^D%b2sDcf3JO9X2+r>L-G}S|Mn({*2`*E^T&h)xt2Z6kjX3op
zB_*XUZ1e&MB0<zz9(4}YRXx?wB-lJEG0-c2c9gI?KduOgC}aPVC+FdjY*JeQE}8^U
zWkjF3nF&88UKJ7;6N<YNNNhtm`&k{oj*k~0mkO3*w7Y7TKfjhbxzzc&nO#%YM@CZ6
zdJ=eBA2VW|a#ADo=-*>iiAv0C2YRfkUtDwcK#5_~noS2@S5vC1t7S90Z<*51lW`62
z;uat~lKttS9bHC~+BmZcPV&!N`<<lUvC~hQj8q=BKzUEEb!REAzuCT3B7K&P{-P2m
z`FArFer^ZIW&26aJCI|qmK%*G7ubA`pZV8|{SE!3zEX>S{8=4*qpFszvZllLo0hd&
z_W4@&&B?N%07yGYSB>jDX-R=Y&F22*;{k$`qYb=~qc{j}+Dz*Lr3J8`_2AWu1lWLm
z*a+0R+s>1`h2hsK#3>IC5PO99J^fTXy)KOY4>r--f2QJhDJT9Djp_fwNc?|nk6*Wu
zk^V~yfE)S0+im~vx~YHN2-%bWMG}ht!iNgXu+?l2va;3(6#zX;BhaQ&lu!XHVF^`P
zS=q`s6pFs=!F%}ld`@#GmdviPRS}tScejsz{d%|VP6-{6VBzqTYkJi=sBZzCeU;dQ
z+)o@*F)O~jpOFzL%s;9AJX%g(DI=vNCWi-OUcdGL`uuUlDXycl6Ofc|OVzom&Ie_~
zDZ&e>@o(Q&cAjTHb?THiPs#-iYW|dCq5d!>ge+~>2|}$W$UR$@f%eQDtz+>QGrlvc
zgDgpqtK0LrxcDVB2MGEGw~nwQSp|G~tC^rHec-@>jp;9*Mk|C;2vR}P<pm~fAZ6*}
zZ*l?r!pEPj78Vx1S9Dpo{qv)I{xBJK5cE<~2Sh{)kY>Cz<;z3>R0*|%Uaf9GrQRcd
z{{p<7i|a*P96$OvT0ibM#eqd7sJA*=m2@RYyvLHhM5_X8prOGIa*>zBdn^QDz5NAr
zwjk%^^G6i+plQ3m$^_pHHH!}UD8xC^!Q88rk%SCngfyZMCvw%TC551MkziZ5Qf6ab
zLw3YU6wHM|!X_!Y()m@R?y043{;Es<x9a2n7>=;E{|Z6>KRmhrdZ_;?IodyqrS#&7
zPbWvbRD9&ir_PazREyThd;zz!efLTDAr;NiERddYX>E5PXYugxK&P8al?c+n8fV*i
z-b%ObECA224(%(`1UWi1>SE5`|4j=;rlxg?uw@>j4C$Gv^;wZOF=R-`uW}#~074Iv
z#y)6-gyO&6$N$Z><FTPG)Bv<c@h&^*A$kpY+mY06Wl=Q^$gErzW?KCS`3lKLRis%K
z#4xnayzxoX>Zp+WK$mOr2%=aMUE>2z5_$sZxNub73xgIr@b4t>BEW-W1ZgRNpr}3)
zy#|HAoeQ#M0VM`Lk$m4SBnvz!l9HA6q<*~xa`RP4i2+Bn$bpaz5B>d5{Fx(01luR#
z&Ez{f$w&IK^JP!WHZg?+*B<blZAp%u3fv<gF!8wMmZ?bZ>@~ic8V5_a7R&5C-{wHC
ztS4<OEZCc>M$#sj9l9-PI+n~~SGSF@yk{)Le1JYR`qrO=G%KTS=c{A#@+j6%t}cwy
z+Z^Ew#hb74GBBQE*ZZCU@4I{STtn=99Tm05jxADAO|5%H)*8f3OE3dGcFQBb%|4?d
zFR0@4-UsT+C5dVSVOy83=@PD5O?YO~HG1MakY17NzMvaG*m8#Qey99ZEhtFRMl0p?
zetfz|@7tIWS(G=?)L4uqqn5akYK}g1C^OjIJ$v@(&P|W;A2mLXQpL(xNlPoFeQ$%J
zy80$m5=72{4Du44$sGaL&3HpaLzID<L8YMLxTAH-ujYXDzR<FnITINyCNj_ou%Zca
z-3dXZYq_}4BqO31I(zby6((I!>_OdtDw34A6RkjpynO!r`6s3^&C)Wl@R*prYNa2h
z92<k&7M@tA#7<46x8|lCl8g{)m|oRTQ(Bg3RjnHTn*(W4Ur4voCp4;-ww^Is3SLnE
zStP~_pze}uRPe&~0Vzi+oWnd<j*A4#FK&UHW=cVp9ZVxh`$%#cq=M-%ITWvzS+o2y
zdr*z7NO(hy&HcN@?VF?R2&!;(&BjIoQ@P`tnUT@en1;J01~2jCaQCw^sjE>P$-c`3
zD}Pjp9sBSF1b#KK8l8Si&4{vT{ZJfgz2N1)9){py4;Tnn|J0XBNe2!|#>d9my0zR2
z7k2RBNs-c^hGk{B1Rw5f>R8f}JKHV*RsqsA^I>(RgXyT*NMjQEPSD7{q#qhxO|MJJ
zIZzcfzRJ$MJR>JcfAqnlcdktYnZ(uinb3tbc<os&Ee`UDsk+MKB@E5H8hwj1G^&EK
zw6uch3fgB8-D7`BW-b2G&_-$}EVa<EYqBg%1LJ>3i_T#jc@OT8e8zuBN&C-2*S|8s
z|J6+JZ<q7Dv7lWBX<04I&rl%)@pT=q-$i<kDBtsU)0NU&lUHmRibHpvLi1FBG6EGZ
zN3FKAnOAMZnUGCHDdp99<;oQzv-5!(Dy+PnJbZ>x^<Q;fT&jst6LcQ$3zTrSBa9lK
zwnml%jr+T~R9=*FblbBVYwkdgdF;=SWWpr=EgKi|U>r&ijVvfc=JrA@oZ0u5B&A2Z
zss(|9df4FPI&X8jrTzVz);e^S1P3IgLea(?uLY?c+0R5C_Y%z`T-^O~QLZ+~H}}{a
zqF-FL%KIfPP+bTct0G|a6A=TnkZtI``k9U|*kvNk?dT^`+nfCKGxV%>kHfo6#5k2u
z<fVlzjEqD*;MEDr@4-0wW-47j)1Fi8@5{12-d?NF0H#-VM!|g=Tz((q2Yk~IslU=v
zQd0WIS3+iRvC<8F`1|9yCwwZX0iz1{`nUSBGcP&kjGq6QG077+^HlthF=@w=4r3tL
zN}{Evj65!r0txLkFE8Z0or^LSslb%6ncU(v?dYUAevJ;o6!xby@)KzZ+G?hKXetz$
z=*%?KigYi847%*%5{b)q#tHuW`@RV|6c-m$p|$XwK)ZAy^?|7f(O}8fw1yy;i9?Hv
zt~o?E&+=E-(wEan<Gta~3Dc~K1}LNS(4itCKhnEXUu{A9Ey&?$VW7=3@N-K&tL^c>
z1|=#)1-@bfbq1kJ)*O76X}6YqSa~m#izhvhFr@hYUiNLS)IMheDhhlhWL)u$E%fyD
zCYIOX-Hgt{m=00OGjgmo5gXY;-hRz2WalRG2ZnP|GQVWWC)yVy?`x6<3&RnojS}Cq
za$FoM-S8B3@=lSzWT*_2Ga1Sm{>+OOfQERjO=6OV$J9utj7X!?ttZ9pZQzc4c;nKz
z9L>?G)#9P=hl|0LVb|=DD^L5Y8kH=g3|;4IXaVnA`{?IBNa2yzd9h7bZ{6%|$=Pjo
zt*#Tg0d%JJ?k26E_cKJ5_g3m7nxyP#jB;^7i=MulhsSs<8wr!7og2&6`eL`)h7oZ+
z?&(}WfRRHyaLSHvxE$Baw?Ap_hVrm$sls3ucaL(@v6YcD_BDEZOimBT8yS*b$TVFh
z%65ipe5yliS@I;)LXSjyw`6S|B-$G+&?n*+y=+oaERfrzwUhoKqkEdy59I(BSav<X
zT>yPZHAgW0uD51v`=*6tI@_B&ZvsM<yUYia)(O8Y9B6FbY~di^^G+oC+oD0pi$nj&
zB$7X(rYCLUV}m~l?qJoS;FxACl`IhvGY`;L#pVHyjX?J3xyly9P}ddNKksfn5}$<q
z5tt-*>eT=8mp_UgJQxROx$+-hiBlGwgbt5(R|$QK3@QA&)<MRbl`$8y1c(|?&Us?(
zQp+>Rbuw(_I<+$UeQJvnFJ<)k+I%SCH?_`oh|*j7q+h=Vf`+cvTz7y-kyc4aD~|=x
z@o+gWq7=s=tIH@+cln`h$w@lww_8ELuhd#kKKV%e`Jq4eVBO&rlp7P!Gz$FV10uV<
zJQ$xgxzN|_?D9l8L~>~G%MRlpId-F?KZbv2iT%X4^}-7EU+>8U45(aJR=KIKrMU5e
zMc>-z&pZTvWi*kN(mxw1aEWKJ$`SsRi*GW=Ft7NX>o=zJ&yGcoJ`GR*GCSYJtGBXf
zhw-k*PkzkJ_sVuWE)q2?6)$^S_T&&wS@Rny`Vt*^@n-|Z^UqnA9e)kMygvP0areWK
zMG9j5NnL$?O>Hv^T{4n;^n}}MD=N-*uU{&Srhz8EsP23#QF7rVgLgMHG<-PKejo;Y
zS@fl*&kKn3D>XzVY4Wo$CMb74x%VvU3U%!K%A3EZ^bir-kC7?VB9Hj>KlN<ll<HD)
zJ*%v&?^5m+M>{IzFgNz1cPv8TQHpVBrAioE=!6f=v2{#@B02Tr!Lr27>uC={XI~XI
zTBU1lTyc|8h|=XWm?AyFr^kUfLFHUH*wEORIyt|%<@vUNVzwjW@53!@#Kgp8z8p1_
zBN{sH2W*1z&-tpWL%B|r(hTHmvmL9>gqa;aS*{Y+Zk3MfYKZ*9R76AkjB#`&Y8Tq^
z8&6xho<bvjLz9Vz=t?u^Tx)ydU%vb}ne_)yURw&y(HJ+ixP^jy=N=n9%l&}%)X$B#
zS|?+1%e~5oTj$?pBDV#YPNFXbcE+c@IDo}Hb)v;t2G~Dcj%^)?>{y%SH(r42!B~%8
zbB66w>79Ve%&lE1NlD&;H%k5g<Z2j4HZ`0h8T!UFSou|3<yeIbPG8>7<{HI<yu+A4
zn~jb=6&)GrIX*s4Abgk{+gsERcl1(xeb=}~;l|isp76O-Q&WS9iQ{3#+qOw$$Kg8(
z&7`ib{^h9vp!V~F6rj>M6?w|00dmIEHA@q+h#Z*j%w7ogVrh=c5ti?j7>w~uhn;ti
zga#5{$5<7S(UNKA`F2?rBtch0$2C$+)t(@2Sysu5=0h~YQj2;~WoI=xDGb}?^ZDPV
zkZ~N+MU&a$c$swSY4a1gt-CJKUooeca7|Q~I=z=nw`}KXLp<;xim^v~e}*I;rqvGH
zxyI}C{MyEecH}eD&o~=+XIWNoT&6Qb2dZ!A$MRazdGT*%W@dcm8aI9-+QB_rhRibt
z2E<yYN>`}-e4l?*i+wj*<E;^{MD~-~WF<#N3wF=DRhiK*=oGK>J5&Vh{r&xOIjs+Q
zl-ee#eggJCam7cK`~V|7e_EJ;D^EvBFiX8@{?8nK@<hGi(EPV$k{+d{rT1qb6|XZ0
zHq#{|7U<dWV)2WA!smByRL8GPNX}CwKhNEzf1}OEpx@|!_}eLs>XSP#`bu%Pl!=Ln
zloTTI3GC+>NqV5)w|8~dsut3)QC8JrTE<~RhI*vI+@To_j?G8*0e3!8abT)~?Gt@r
z7;eQbTIm3pL`NOcJZ^SU3d23ykq$*OUD0K=#57UiB#laiAg}w#BYKY5XS({+`qiYF
zw%NYPGI-7$Jx(&dgbVQ`Pm0VkkU=o?nkLI{Q#LX(E{t0&h^+gG4OhwOuXVtn3&*Y}
zSC7#;{oF=h(g+qopI)RfJ}hh}T`NRGw@{Lc>x?)pFYvJKl>@@Uk1}_Vg}8Pp5(_}O
zcoi)zKuw?&{h}ckS`Az&g*$Z_bZ+G>ZRzjp^9D9cY5Ma^HSncGc*2qjr6U_OkU!>$
zBGuA)peeomC(gTAx5xQircuR<wsB;qmfm?_Pr1|yIu3r8gguS@li#n&U+cW6-u+>D
zLxcLj&xC&~7XOH*roE*z8v5A-xmN9;+w82Zt<!%#e%6%FfN&v)lhfv)wt7=LCz}1H
zA^aacIO2)bDa#o$#sxuCVxcR`-V5tOr@uyIWYa4nhOkEx-<mPMLnV`cv`*uy@O<ai
znHDlj*j6FK=|kos%rm;f4+8PEVF-<Hz#ap9c^Gy%`?pJsy6#YJ;X-96LL=q?oj_3w
zZGg!5!a2yQ))u7Z&h=maiO;YMGpo>-LgdD2mEQ}&BblS3x(zxMinT#y=*MoJwk;}Z
z=?)WmR%cCFdO+E=HnW9kEoJf|QSzegw~2Itk3C;Urt<pGd&U^azPs*pkaqC``$OYI
zwARTm&jr1or_p=q<v(%X(w1vqzL8lwOJ_(t^&9E`e$9P1u8B0mNe>hJG-bNNZ#-jx
z&G5<DrUguy$o=C|w?N03-%rGD*#quS3oxg4mWF8jZ6)xqP%AiF&X`#^YsAkC;y?2H
zF^Tf>ANLBUb6-LmDMg%{a}yF0(&idl=l_?@1bHdVZ&}&+{17%%;|x83D5}>lX#tqe
zNy`Qf)1I)fun;g8C0q9=EWZ@pw}{B;&o#M$0&{9!AT@&J40gZaKVJV9_gPv#UHXsf
zr{%M-q>_n>fPefDbXQ-dq=-Q4gU1fyj`xM)Zkx=(wQfT<(mH}nK;mOAj)Eua{}1o7
z!0{jlCqGmYwL(6C6qT?&dPGb?1{E<y(RTOv8c#BLnbt~1m@psux5E*&$zz!WQ9lzw
zf0^5OB9bk%b$PP8LcUVx@obErV7P0^Su#${@a~QBI9rw$DN*tuw#uvz%8Iv#2%?ba
z_zwaDFN^w#ZZDY}!Su=FgO<9peTI@=$Vv_<iP|6y(?D(~{Zw73rFr72>8!$$h1Evy
z8pCks+@}NAVc3?{Taum%_HnJMTTc&x46FA3$yEGPe6#ysFCfq8?tjNRL`B&wYJZCW
z>44op-+5n^6!Pyd@#^-~B~&yuZN{vf50YYZ<xcvMdRYDDx1ZS})Mc36Z2{4%0FjUd
zsa*#x^Ft8jGNx<BE&iGLAGYGvLXS@qB`m@4MErr&K~f)*k;9LBGP)4UCLqSAVSJgI
zw?QGsCY`JWgTCP2nr~*Nrrw0FCX#4~${{CXK}|*`6^WXLj46Y3`?y+^fZdzX^u(~+
z?-JwCcn1k3Qc~ElM0j;Gnw}<i`!UjmK5gSr?d-YmvkIN^_P92Ii2`2_572anfUUH?
zU!h;YXT1xWmt!B|I)5sITrf&8V7q`hU`aP5E_>%j%vH2OZ^>ukdR$o~Q*g!tmwW8F
z;`b}J&|Gr|67B6xR$i`UqUXmZ6kZv&$`}iihWX?R?@!&z0lDZJ#R+N{_#1f|{BDW%
zZBzaSMH^G2qoc<n*Jm_bLBP#9dLnmpG^1xzsOW(Mtk(WR)3mAZ){e{81)Z*m#-aNu
z0wt4A_|8V<0_VSujI4;5zIb67X&><COtxWcR{C@(MPj_k`nMqdslT)UW!i&@P#!v<
zqRhxPE99{E#<yDq9D1d!s;Zjt$7Q4F95&<l=dsUYr%#BD$*VfZ;>ypwYn0JiVt%Ue
z0PD4!u6|qnFZK4??yl4eE%psj7D*fCDm2__lyHk%{cQ1yyYs$07NU|=h!OR?LQHg=
zA_d#Ni)6~+e8w-gp7q9M##QcQ*!S+1{yUA`8+E}=(T6)o5JW}Z(#Lx<j9mH}iP3U%
zi(<?uLa=G&fA})4QRevZ<B(-ls=hl#1TJp(lfbujb#=YIDt%lPpul^R@E%CUmAlW~
zN`lyRqv2ZLhtv}PE8wlrJO<W^wBb44Akt{MfsrvJBqYIUtY?L)P?8Vy`K0#>Vx81M
zpWChS<39nylflnDnib`biY~v1jN}1V%z`g@#DDvbQJeXAAbB$CnSAAu{<5xsr_<;0
zZ$#e&8P&4%7fGaKFudXHuBi@={c@rj)p|PY>r4wSOG-vYCMhXtm9F%os<9R`xmB6h
zQhA$Pf!oM9IT7i$IVrmmIieNF{o;UtmtVHU^IrViIkNQbPE#kVK_Du?*A*&4Nw&$4
zME00fQ680buHGtF9x<(;F&3w|+bCka$lN<kjYl=Y^IuRzdAfB+=5{HWMn0XEtDO;_
zi%hD`?G&O6sy$cSM|sxj)GIK3zI9Y$>h!zt?*ogLZC=+I>YigeN>0PIOD1(+zVU5b
zL4%4{sG}^Dlgz_wUVZ=T2QBjN01?bW4x+2=gM)*qA65!%{K+MJY<Uy-eg?vRGNp%t
z0Th<23+B4JEL*{Gh!eT@`#tnDaq-<MuWR;7*?c*sM{Vezn8wPyXb^J`n?VbhgcXy&
zV_-f1QFpaA`O~#~C;wF9{C7mne_Nsd>l%JG8UD2n|8v&ib<`=AyFpP=_bM(EUQ=W~
z7c<8ns#j?NAE!y2n=bH&I2wACBqc%(X~iGU2q?Nd8tGQ4l3i?&2*yR;4{i)$wj5>P
z6rv;ME61i0hJ$@ro{36~MS_kQwi9uhbIw7=dw=DY$lz4n`mkxQ2^ZIho70i_U**nK
zQHWyakx(Fx!H*r@NwwuQ<tnZ(L#DX{ijAWwtio5$BUndO>hE|CwgbI^Jz52KRUuI6
zUVrkmMd1}h%ji@Jcqn^N9pGbR%Okq{TBf}~KmFz~V~m-YNBox*$v7sUG$#DQ3)!+T
z+d92fReD7ZhKuZHW)RVl7d@evS!*S2Yz*yYX*nBDVK84cQ&5aXYTSoMtIV`4q%br<
z-jDS$1Cc^X+1S|FztX_k1C5Q1FVR{3z_c36#|;X710NqBUZve_b@HS_=K^U6BNZK@
z+#VQ3op@{mb)-^+BQ#z+yJ1z!dK5`q!6gMS7vR-ev7BctCl$k%hYuenspiIov+brO
zw@)r^aofbRWHc`Rc~WXB`;kfkiR=@E1ybDY9avlL-(d5uuA<@+gW53^@~639DvaH$
z)oJ}P;@E%wLwMm1GiCK!)liUX$s6V>HQc+qx(Ky#H0sv(XP8qMdt_Mnx)Qf2_|441
zLX;x#oL7vj<hi*p#oesd<M+H`=m_a=jN`w~@&BB2{I6&6Kj&GHn5t+_rj&>V2eW9s
zbcl_aC*Q+8uMhUU`G|=^<qUB;LV3%kvS^Tu8^LDA9Q=CQAo0p@z`!RJz6M_Lup=O{
z`xPp5LQ#?O(ZKVTut`lGta!LW%L+*&i_R$dYnXeIiD8URq+Wmw5%Z?emHI!cv645-
z9Z1^<Z2@mMO*NBoQW+pnG)m~&XW~1As)=}sK$fnYasKC(0a{r+S0ok=hrsC_{Qie%
zlQxD03O8GymXZR)bWzcS-jiV6{gI2_WkSnk8yA-y+G^$#hoGefmJ>xQ=UFwq=-PMp
zR=#x2JeXogy+KYwG;f<(hH>)_4D&%0af+fMCnrauZ6j$CTu%JRwP6qw4xS#Yf(rue
zRZ5|#Tf(D5GEzk69+%iA`%$V<RM=0SHx5qmN!b${ZHD@yTFJe8+7*@KvFG~&ciuZl
z9f?isc4+F2KR~rkk+PVR4($m>?H8w=eVA*WvCYm9)e|FQMNJJeFuugn_k<tyW6^8G
z93r(w!=m}~m8#0GU(X}cd?7Xy%Bdr3Ng<=u`orpwS@kjie<C*N-hW?FNeSWhD2k3S
zxvhfP*>MHF9AYOQ?bBop4Jbnuq13(p{xgLN&g`)OEH$|NO|Pz03h!Oq0tVj0<G#XV
z_;qti=%C=Yut9XaGfoAd)ZeSe^I1N0@vZ%ZyIBunjbo!4CM|4YKQ@LSY`57(MmAeG
zRRxnmpt(f^y;|Ks@h1KKk0CnRB)Bk$Y*6ou?txA%&$i$aGTw*(Iu)@071|I?xl237
zRMjzdLC+F|cj6M!zI$HdkqZ9t=CMnOf=hyX(c>|lC8dE1dh=Kx@XNefpYLMkz-xm*
zlCrT@yrf=frFYp(T<@6S7quWH+&|ksK1no3)z-T8xaj`rpi0uJq*92$%EKq*<=4Ju
zxQ%i7!kuG_g-(vk6tl`>e-xLbC4%%?a3u2&6iaPiXhh+2FhCJ|4(m+j_{r5nAItRd
z0W!6-(zy>rUGSi&ODUUG(#=K6bQKj>g~w!uo%p;kd9wbXZ;yITWOlX$Do3U7cWx1B
zEcthoR1cs0v9Gck4?ZBl2rUNRddHW@z^r0+rnm-!ysV1m^`JZR!|YJX?Yhf*2^;0C
zs;VKnT`&w_KF)&NO`HOPOM39_qUM@Jy0&peLljG+X`(!`b0IyEe||>VTsv>&@5v|J
zgog(t49SJ>OVO{ys0V~)Gk-rF(ENP;x)Jw>u?$q+<`Zg|&c?72oU3lA&>7WtM>gS1
z;z(eG9EZq9v}N*EBWY+7Y7{;ny;_5tzBZ84%dqq_NR{nezL+t^cnKNREd*2e0#k61
z4?K3p@GL4z8RFgqbCcUEROtku;+BtXTqca{ER=;F)!a>}?A%bI?*jXhSl>^2!!F=7
z;uUS_-_R~<N+uyXIy#6}gl>kxugMypJ+4m1JAEnBU3B{u66&PL8?KR1XCakG2od!A
z99-Ze6K}&q|Ewif>SBIMkoHyn_S4;O{a=!$zx)8t{;7nK`cqK+KjH<-%H504Jt(@o
zN9)5gG9JBqBle?r;DVzZ35EbuQc?-oyXDy*rGQ&D0<%0HlYMF=gEtoJ2lRh%rwJ`H
ziPi=O<uR1Ot6xQ_uBdn(d#Y=boUK_^c4_Gu*C;<6u<-2c?8ye9GrLKYJ8O*CU4f?V
zteW%K8TLmt4Go(Zeb3H44{-|Mu?d;G=9US=7kc{e@TgBTltEMDCNGyQ@ie0n&5wr%
z5!zt<5=Y{s8lWOPp|G9MN@6CnG~~6!qhWR?cLBSK4BHU1g%zq&J?3aRO_Hs_$VeQw
zh$~*)Rq0m)y*dj{#9Qc3$%n(1Klp4v^F)BE6UQ7rI;IxHi|hl_y$nM2+*-}vCU4=S
z)P&|D7?MjL{v?6_)-C(1r~OY+%^oG3wMYDm&y5kw38G+UWF);D5Gg8YT<0e6#Xpyq
zE?6?H>AY&1j>d);Gaqaj;QLdEZ&t<4qhs4k=NqJ?yxB#Gb;AYnx3FYtBc@*-9>gq)
zsO8AbKgC$|1@uOL>VpLB`s(O^i*!xgP;gM;h6fds7Wo9s{)60?LZ6t3sK`wo`-_QC
z`bgjM|9tx@D`zmAtOhlthc*`tOM95~HDTQ6{nGdFo;6~9Z=jylm<=S8P-MRHDQva~
z!oU)X1O*3^+2;hvy~Z3`e_hF(hiG<+G7op3^~4P#^|i=!e3NUQ^=Tc#dqNr_4ql%I
zBe}wOx3Op=V`EQf9k%kPef#}XfXweH9ZkQuCWBNXCNiipq>$f;c=!>&=O1`KaeSf%
z<j(3tq%1_nWb3f5)P4(h=hML=d@T@XsAfpr3bgVGz$gSW+Kw{fu<#;|hoo<*`}rY{
z<=3+^`A~OCURzl7Q4I{OIKGHbz358pBGyMx7%-t_;DM?31LW3Grge|!6qs#S4OiT~
zagWw7uQ~`Zi-C60;i*}6^I$Y3Twvkk1(S=hUO{@x5|bK5u7n;YbM+#dpTH31%acQi
zKjt7{sjS0WXOvkq47N^tbOz?oZ;2N(>FjRq#M}`h?*}J;y_Xa#A$pW7*}=iAkv1~5
z?co6(T-rbeSU;vnoNPbBLBjy8l1d4k{jp0A?bRC>n2tBs_y(-Fd5qpCA%8D1AO!2Z
z*HQZ;eCKCsQgfI_I!bzJ<B;`E!h~_|)>=mI#uFkk>|GK62IT>NZP5Mt`Yh2@=fOg5
zGnr>arpgBroE}R<!HI_7dWh^CP6zJWhIEQz@&vF6X$%WmcYGmMAAb$ik53*%0qnFH
z25TBK7H^rBM)}#$HY=S19w6rvwqToyO@KHiPie{VuCbN+_g-6t1V4L7Hq-+J3L6oX
zH@S?IU}F;!m)!SySZP=S`sKp`g9#9W5T<F+v9;lZ8f(IQh3=J!N2u6s#m1Flmt{jd
zG9e3R$BSw8Wo{yC4tG7nm^Oha#NLif0WY~A@exzAW-Rn2CFZL{8AmGK#6%%&=%h&(
zt&Qm3$Ob@DniDiRPfJ;v*<}pljaCtV0{|D=fPJnQnfoLk4|W&(=(wg;(N2tC>%uUI
zY)Cf=bIe9Uadb$~arg&G=xF#Oq7(2<TmaEV(1FUHbWsnBi9rHnT2q`+`RBrYhP@AK
z=M(EQS)CGto}$^^@_-7e-S<ykNH3Kye?VIM`T<(ZL@lw|_Ul=!+AqyOnCMH?s2S2F
z7MPhNQ$J{@5YvMK%1qw|Ms`EfV*#;j0%Qyz4&4<Gk4ZQrun;o`GMEUB3mGm3THFLv
zfbjZE(*$DWNCuA~0aC(fsuYj{I0w3+*9$NHI{-h$0Rn!c>z6pgmg-}AZt<;~{{H@f
zI1ohdzPUUXarKUn+~2^y>Jzneh!@|^_ors}o8@mJFsi07++|O_48zi3>kS&cq^}6v
zbg)0EN+kH^FRsXdFCYit;zF;lyq^O9#j603`xbrvUchdFLjxm)mYAVGK-eFr!P`h2
zv0J<Q+_XM+b^`1pkcJSFr)_npDc|Ma$-RXri1eH2W@EP@`yyQy@rBuuCDWp+ahhyI
zbndZx%Xyu6lXX)lCBkFeovV^HXEN=0Z$#X8{pI4DL!t@7Ti}I3w*M1IW$AF;>ZBr;
z2D%DqsnJQwMSKAu7qHkGb*)u8H;-)_z__``Z>@(<HUY%xjrY}Cl4ORyHQjS;{c}TS
zXJ^Zf0q9&S<7YOduC~Y#I&u!lo`gaAxH|C*(yINf3B%IF{vrV~B75gtz&$ZSI8Dq9
zqTlRBgC|`2<OpdKdw384>*safUMp+@vbf%fR^E1c!^!9Q_pXYU&P}8$xC5a9eb*aZ
zxpIfdeu<@P9TajU-fX|zsE`pae5Y7ZYrP>$8!8i*xn3PVqFZAvfA3GiH@$BcYvIb9
zgF%u4jh115q>+ls%JmSF8`4F8mTh@$=T^TFCQdql7+4cHtI<V=#QKc9gp7r`=rmM$
zTix;{$%s^&Y3S0>eQpR@&&5SNvmBsvU?Hw*xfGYZsvYGIkbC2^hEEZ%j*9wvwuR~V
zg@(bfI`76?qKgEMYop4{oga(0gt1=%{8P`0KR-htB(Q5B{7B=Mr5t<yM%v_yjg5`z
z_y|;l+uJx7Y1wD2c}RdzeZ880v{DLMFo;AbIf<21k*jF-??c2j6UtmccupFboBIm5
zcFO_uBvy}=C%*66{#5Vc?yj=*<=TUYsNU#dAB5B79<3FgjgY~%5w|~0$dbuuzFT4=
zx^z?mT9MaEgjenMg0vZ?Mn#(mv>}rU`Hkut8dT!;1CgYFujKsXB-D56&-ZkD;nb1L
zOU2TgJh3cbqmkYz?vin>#HgY;#BmVX!XAYPgo!&{c8(#hYR8*M$%UyhT6&o)KeC7A
zct7cJ=*$lXcop-FKC)~a+VxHUlRCf%F^wXqOxFlAjbeq6XRwnqN8nbOpRjGuR|s)>
zRBFEx*-cn(>6`BBlA&c5w(o+=I;m}50=^gu&x0;O56aj+yIeByACIR)mYuX<aY_&$
z=sEEYZ=44)!ld_h(7;Vl9#d>MGR{~gxbOVvSko`fj>N*4tN>K1i8|CnT}563Id%P_
zFG(+qBrAl$9+QXCwdYJpkwCY}A#MZ2B8aHPVIfB*z}Jy4A;^aunnE;~6u;l;_CgpV
zj}XO%k%@^HJmSQOzL_*9+h7QI2PHLe&m_T>7$ckK3aiC^`97x2xk*4HW$#lsf!0B$
zIxU`1h|I8VX#b98<cs|Hlir+?lrs`zUrB_=pL1T=+G;IN?#*inkze~xLrm$}ztg1L
zGSNTK>jU33+S7?Fy7J!eiRFpCnrZuBIjzsKg!Rahdiqf;4<$ztp@KL?C*YJ6px3F@
zZN<72s3C3=If<PpZHNmUxoDK_cd*!WT|e=;B$>&wT3J;ZI6bU*r=`{gA%vLCupcGS
zXro0mav5=nMZCO|_&IkZWbk~{`4mM0Ic>_1T9zMOCJjD?07Ucjc7M7z4>plE@mV0w
zR!1ZCiIk0)$#IF~+}1?KC01yt^)?Vm5HSdK4N^0064^ibbj9(?a4&aw#=^IkdiTOF
zS-lSy;3k(|=t}(m7};;+#6Hgc_ZqZqW6Fpwow^6>k_Xl-(c3rkDI7uS;_l9L&wJfi
z1TW=i6=F0UG5DsM93WzC1cN;ix?uJ!gham`I#o19XGbI#tY<sj=M9rmQVQUaga$m@
z8c^&gA`on9`zZwZNJ(KJ#+}D4)9(O41#Fz=49aHw_kL7dgXRY5BTlxl<g!8}B!dN5
zNm|MNS$>v{pLCZB9Iqs{jySv)M5XUW^U}#9DH~<QK)NC6F}FY?H(_#=X2s_~0oMY~
zYOGq-SyOBb(mYSPUqN=Od%Rrj^+|vM52D!vFiUX~8G=Dq8|Lgh$pl99mfMv}qy20O
ztyisjb^V~I;IKG^kiY>@er0`6iCa5a8>hvNX1gX<p;x2z4go*qJ=^(U$*aSqOFf*W
zrT3=Rndr8N)fq(Zg`Jd%L*JKb=vReNQt)c?iAb~<%u|z-rEP7;uGLCU+1S|?kkW55
zcVQ+MtQ~CN!BsdwamT(es<5z;EJx}DdzV*?Equ~w0t>?ef<$T<#njKhu<9ZGcLYF#
z(e~!!nqE6eYRWyCcg1%f0;~{C017-o(yq_Ld165(vTuUU<EQug3x}JP(22&^Mr-(+
zydMb<4<{`aC?yEE>jQ4ENGn5`1f4hJOWhN#y-~@%M_h>e5=t)}e>Gy2>mv0>eE8}&
zh}IIc%}bW(KP#2@*39-P-XnbcPLfTD?_Q<`yXKT`(O)jf`JkU~2g_KZL?YAE2@phO
zUkb&EDj7HID()PiK`~H3yp7X9wcaVD8;$@^O-Q3Wjz2ng1UioM+1b^)O(Mg;Ij{`_
zUJ)RDMsjYHfSb0p?vux9J^rNV0@Fg}7^K_#k!)waHmhBINF1XP2&7a)$r&|L3R|QN
zkD$V*qPPL;mH6v)G-3&%LkZgI%UzhaeE%!U!}cSveimRkF1<2pZOrw7f4I%lTCJaZ
zO34H#v3w%a)!=WIChR#^ZCi337IK1cJgV=%-9)koDpDLKgTZsm#61%}`mLT|x%kM%
zi8i$e_gcf+F2C1$-zcfsK>bjO?>G1yUi7`Nt48cd9DXJHV0tX}`}K@Z4ZTb&ZnkA)
zk<)-SQ>xGD4#Niz5o1eAN-RqcbrRx<6|OwmSW!v>Om$H$u3W00FhV@xv2FzR5O2yh
z&x?>&TK=7Difr`KB-0R<sHpP;az?{Mn^L;3r=_PKA`ywg>7feb0fx+WmY8()wP~Zh
zMm~S?iy;MgSn@jF$C3J@HJo$l3qJ?n>!<31tT&9Rs!nqM@iM-^k5fl;xQ7f45Qp7-
zHXFQ~R;)9>pI#S8<-yiQG2t+qHYp=X!U2%pO>lMFm@t8vWVRI}GF!QH6IZ?w9Hnn$
zQkKwT0nvew8iWcT7{aad63L0%&CK>B6vt_>@FCu4yL0@!>*{jn;h$dkD~skj<Wx<f
z@<4{lsCV^gJdr{)`sXu#t5e4T^CZIyG~f(DGUK~w!I}8(wc(dO434M3u0%Ko#|Xbc
zA`gLZ1F`NrASzlY&7zf5Zzhm4scR{bahhAn%Jm6QmrG*&d^{Y8NQHA>co6uJ4C&_A
z{;^axeTEdZSIo?Ouzx@1hr%V=CZtD{<6DSrjT$M7oXdN+^J!R{{|*~`3lN@`&9=45
z8JpK9K<!2B%a~wg*<%?BAh;J6Z+khnC?+LS!^rrRnqf$Cz16B<_&GjsdA%>M1V{ui
zGB+RczncEM;8SXt+ll8~L$)qvHKuP_1|ZaS%q(8D63tUfgp>k=#Y4N@<MI57j|(w=
zxw(>sZhKB=JNzhz2}v9~|1c(X%vpOO{WMY`!H{H65I|2VfeNr+q@L>oy0!!#QHZ^5
zrYPp|xfw}#R7RoDPu_iSL~A-ax0CzY`*JX>WP=fbEKz$CT!Iimc#D_m>FGbD@3GdW
z7-S{E)PdN+(@!t4meq!l43U4g5pp977@p<H&TnIhI=6{+I^AQB09Yz~_<X^dQb<|^
zG}g*$=d<u6B{lS;mA%89;4S>ve*Rh$&$oXXB>Yz?9`%P|Zz_y-IT=qN7XGI&wD;~e
zMKG3-;-rB<qqnB_Lojpn%b$jD=l>*C=nU1I;lFo<`%D4L^@w*Hm}(kSJNQ)|WD<{L
z7d;64{6LEwM0H1<?v3}uuQ{l9C5O6odZE8r-s4d<=Q0e=%oYE-%x9=Fc=yMXx{7K=
zirjo}ukRZ&V{|IcKXS2d<>XtVswj`&1CI$Ru}J^6C#6fCh5zC2>KhNYdH7#Et@gvH
zx#aarkwfz=6+5fgvU&9EJ4Vz(Iw`p$v!&Lm0Rp$JTiwKP_KmMx`3&1j%!$SSfMV>%
zxR}$ZQL>JzYVBPUG3$=g6mU@wi`#LoA};Lhz?i}(MA)~OgX?ixh|JBE<a8Z93w;;H
zE6(j@w#G0-J9N}#7(N?8Ty00J7j{!(VED>w2+#3-y0&(92EZe-_n|D{!3eiBI1#J*
z1_D*Z?D|eVS3C(yU<=UO6A9<>`x!6={efQ`l8y{1kUoK`I2@-y_ASifh|^X;{}`&j
za83S*_;`y?W1)v__W;*b_d?*%4tQ}=L80vK2=r`^h>h_lOwXNIDhAq|l?iR2gGOn=
zJH+e<zTiLv;|=<NbOEdFAG|sjc^R;J!HEMD))}a?-1Dzagz{^q-9=93y@eSoZ+=3t
zpSXzwA{=&(2XoG40GRvt?NgNCex5)p3(Y4JUpO`aIXO8f#(uyUtOOXOs<kx~u2cHM
zKZ<1UXFE@H;G3fzS2s5|JHjM9s+Roz4Y)~xgnGPc)v6yzp(n%66%90JCOXjCygJ55
zM}NSlSN1-j)-H^m7J-fy!k4-RaJo%y3l=fkWYf71lP7-LaigpW^KMcfA@R0GLk0=@
z`+eZwnsz{jc<sT4WM06coL2)w-#<P^J#}BwzKsGBi={cYl5sWI-_I##jrVt;7pekE
z)v|gJzcAt+@tBjnKbX5ffm%aVf(5Fv;x+VjK-Eiq``!FLYOLJDZip=c>;v3IUn7--
z;F!Fl2=0+4J?>c2D{0P{7t}|t6D|ow<%Eju{tvWGBaQL9jkX8i_ZeL~`<+x~qb;pB
zTBW_1SF6{Sm)|9HfUFf#bI|kxKtMGOg4)XRa>c4y-Go5-V3D1&WcO)kB!WqEfuU&R
z`8~oCD{*<pK_$O?y$>k#=-`~~Qj7Z!Plas@Roct^V8@arON7TH-DVwZXKMx5ChrL-
zK4jCSi1b!fU;h~Tz)s|%Q<3NKf1fdEVRsfTTqh_*S^e@u^*~^(tzm!l^o)*<rko#v
z2Urw0H?VFt_V$K_z8sE_wOfM?A-}F79YowPh(b-jol6)LmyxNoeXEFvWe+@8Ki1A>
z9eaKk)fy&2Hh`p*cXD<ffhixiEV-2CX4eu4%+d{Rz{z>HFA$WiZ5D3Q%WhPAEpmez
zXz8(O@_S9<%9b&F%atY6`NwBCq##~T-$z{D1>)zmu^aee3R1wf8{e@Zd(t8zI$QZR
zGF;2xLO8p2X&4bmwGcB6#pthxSme%$%+1Zo&WlH+rlyMOxQ;15#gNkzife{?Yhz<D
z>c;~J8udA@&PWESSL(=+iH)so82(aS61EMOoOl@+%wH`uccGpA&WP0PI0Kg<FOt1q
z5-e&6BKJm3yjo(%{j%|a=3w;x*VG#}rCF-Ix%fVrj~SPBaTykR)k|;(2DmExHlfq<
zBg9aaScl@Ea=aZt7rzLel5rQ7{Xb(ka6$p>i*3L>5~4MA4{)SmCUE<Q*HTdO0+PU{
z+og*aC%$|44%kr-eGY8dfkyIyY2@av(rhukm>HmXdthJ!_ZpuBo;p0$-|nZ1{r^A3
zz)|8N=KFjM4G*srJgoX(3)!r4_z)<`11IUMpwn(Ef}zuGgHPZ;`^)fcD*MFpz>7{j
MUHx3vIVCg!0PJfH;s5{u

diff --git a/reference/plot-9.png b/reference/plot-9.png
deleted file mode 100644
index 26d26629e62832bf8ed7ee21be58977e035ec7d0..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 30774
zcmeFa2UL{V)-8%{t8Eq)Q9)6R1VK;%C8?k^qM!l-k`zQFOO6Gc_NXOMQA9;RQ9!cf
z3<iRN1VLgE6hv|;G6e;9?y~!wGu}Vm`|lg?zxR$e{?lXh0Ia}Q``df1x#pZ}ap$bE
z!pdcvm$9+2tyKJ7{yZDoA}2Psg$BPZ#=p=WJ|D)H#Rf_W^7xAXrsZ-x@MY=s-?ePn
z*tXmz|1a?Gq#LoZZDUiEKc?pNw7<pq0k=_>?6C3flwJ3wkF8^W&-+KePC5_oZ@(=~
z-s!nOOP7tl%<C7+SKL9aS$jDHy}hW*sO-Ytdp92aZKojbnzLf(e%-m|ox$nj$3+#d
z_A!|E4%iz9JPr+gJlK=)vCFpNsj4*7VZ27&s#eV-sy9Anr(yu>;<nkwl<Bg*V`J|4
zx`=%DVebNA^7ZJ+rA`aTH=?KBEGJ*@p8G|QeD(BMJNygz=E?(4PV#lzi(hleSGG;N
z{`0qJ^*KD#pYkFhL8!O4w|y{YNT@eA>P+y%ikXjl!UTJBd-V9~_XNIJTQ^!Aq4Xjt
zNrXHtxlE`p{Y=2lq`rille)UPZeJJj>{Z+{aiB<JDbwr~zg^?SCAi|lOPvlx=H*FW
z)Yg8HmUbv$Ctq)_<PTa?)5R~-)hXu_RQMHk#dWE;kK9jk8SlMjZtmmnG279S9vu_o
z#WOd2`*Ty%=7M?VkYQ`C^NGaw?vYdc&plSE-xSjB!>8N3XMymWHIv`(s}!#kZ(hUB
z9;D}D@};hBqG3XM$XTU7e`b)Ko&MS1{a_P~wj|;tUh>T2*Ic<OzNn`Gdi;#rva`>d
zJK2ottcxuT4Gp;+o2~QG=PMlZ*RiuN=M*}9<1!n!;R3d^kAKbGWZM%FbTVM)y`75O
zc2&n((oM|FY@_$8-94U8J3l=dHqSlURq0Qqj#kRb4mIX%X`N}EEdKuS{tn*u17XuX
zDNX`9hi5p=%~|hv2X{xVK_h=8(2ntfhk5N-Vst`+pRmVt89RHD6Wu`E7kATh+3v54
zHipklS9%l=dCbpHsSRd{dF}C56tlO9(mx&<R`(=&F#C#xR+5!iOIE_Z#lAkG#yHgZ
zYM9<kT}J+)bed%jmFhJ7c1cHCy`F%n1)h4;`ZuSUZ+}~|^>?e+pRGs3L!UafI-Ah4
zv$LHBv$|Z&a~uYmQd_z|zt|B}W>NK{Kc!NcW|wcCjl0NX3cJ(2$0lcn3Os@)ho%~x
zyF!*T;<9vv2DAGTBszQ!7{tb>dwqL<yW|0%YJ1^26`#YlKkT#(MZVi)WzNsdbTr#m
z`sPhE2^9NsagL9Vi&(c8HrUtec95s?%ukijCky8DzKRTQ5Y$}5sqbd}>5<5e)`|3t
zh*28(@ShG;`UG3$AAYz`yUtA4Uxsn*smI)OQk(ls(IXMVaO_QohRe4lIv=PSO!PM<
z;gc8${*fC>nP{BypP%x7BxZ5ZheuZC)@q)uLoW3>?HxSxPR*3bmd|x{7e+Q?lN!zp
zWT?!1F_sy9tfzm!th4>=jn&Hf2PllO0N=|J7EP&Rlu);kk5o-@hxT&uHtwiU7t_;^
zMR$>R)V4@df9m|>F}=QEE{!%Vg+paMH{Df&6SZ}Rf}hWwb$g2MDQw-8Vq7VdIX{`&
z@<Y{Uj=>!AkPVUf>bY>4OIwA^M3R3*O-8GWS(v`;Sqlpb>vv1{yezXEEf@FK(5D<4
zGCP%IvED4SW&GpAhYvI9#atabc&2|hPj9>`;x;wh;6StJzzy@mWyL(3-q2K#?K*ET
z)WNy$;(GmgM!liG^aQoIxcHHTwE<aKp@P{F%_M^p^7048#T}<z)%qz%Bl@GfqqDLO
zTjo#yU@c@{e1;#nqf_*bj=OTJ?pS%u(C8y&!kkix=f~fV2ioM|D0tJX3Z!M2Jt`$R
zZOlP(j>&%deosNni6X*~6Y)<vl_#`ZS1dxJ;up4OK5H4~wK3|7@;lk@9TXGWou!jd
zl74@WiZbn_{0QzzW=Z;L+1cw&gP9$Eed*;<vdo^;@~4MwtJ^Zt((IkXTHVRUUipLa
zX2Ft|m9q50bsyV{yzg6Nw1^G4IkZtGuS}10QZ!xiT(SQ59dkRjEb@5RmslV(G`nHn
z#RvU4L-~s;6Y*9zoKw@%cGc^-S|s;G22U^;Se)-af~|zQQzlYnTt`X;5_h@|<vAuE
zGSBXb{J2g<>fG6t3zM#Yd-pOs`v|2CSFz7(#){IXZ9-eRZud?BjRZk0i@wROAbt14
z&NH;OP{%AvkCOhWj4s@-@8qnvPPh)FX&eiCF5H*3k$Ffl;7Vh%kx6h<Q&a9tbm*rr
zS=o@11xvWAy6PQ`nd8;>)nhN%_w-2)q^c$x6nSl>1sTxxPDs_y_v5t424GM92$}r!
z_`v-%v-%$)R?nHS4`!w&CMLcM=ue+&bzaFOCYaUvP&?p|S)JW9#e;Fx@l;ZnkX_@1
zy?!gFc$=GTbyZbWpYf5Nm-hoABO~>?G-BKvo`0d0dkg0`bR2Q-etx^Uo+CweeaA)t
z#Zv|39u~=vdpP^_Ep4^0uWwR~vLrT;s;a7>PUdTxgVIC6#;?gg=9z8t1J?N&ZSJl|
zC}W=$IG<O~1&dqhnWxoWpyL>Sx0vjH;a}`NGoX=rz&S`rcQg5j{$;6&2tuJwSh}r~
zZ65QUKWuBxh!-v0^K?yO?#QV`i{K|u42pd=S=Xv-tn8MqpDdA`6A`1Cr;r%uYB7Tw
zR#+i9#BJMEvCnj6A08;rUa@e$VX==BbG$ysV&=mR|H7dBv4}BxbC#{)_Ye1W#V5YU
z)-njTNWXK!dqbziBqFSU`$S6E30GHFZuhBg*5kDrT^94h?^l;xzP;r_>Jjf+{T;kp
zJ!a{4j5iBc`^3b=$W$8Q^@>ql<-wfs@$m~2V%K<c3j1wwY!(^lyS>%rn$*aL9U?Ac
z-IlETH$LHjQ&o&j%!TFpNE<BXvhtChY(8-D;>Bg5j@!Mm{8a*W`pM2sWq$o*`A9`9
zVsE5M*w!Fj2dx6r!85yXMh9Z`nWi|zFP<InGcJF^G2}j(Q*$6<`MG{~&EA0@cyDjV
zygps&gpQXrk0S%?czzzhpEpkwDFAm|`Bc;p&VnUPYOG4GXACi$*Ptb+H#C1+`IAEe
zA+5$?6oy*m>&x$X6$5?`c&TJD`lDRQqiwpYCn~H$vY-;Hlw{Lc+F%kDVt@oq$H_S<
z6y%+rp6>MRCTBZNuM$mW`ulzBk}Zy7(XF96_-B&msnm^!Ox0#bf(m@F<R_Grl#m}j
zblN0cEq~&K)$TyZ^ds*sVo*r6>`J@z_K%f9O7rhH^Vd{Ljh>FvO4)6brIVVGQ4*sb
z+m8L>t(9`MXr^s`E_WP3z(K^UF3yQw#A!Itl-47vm2R%pj7vx=6UupMp@My;nQhm@
z7WCnW_lA95i2mX$wMahF&~rA*ooF&cGT4BW_T<Tvb^r*a`RO(6BKEys^n^QjUOhfw
z92qo{k)B?BY{{0?w-Fgu`O{ZVha5RXar?L{$SykNi_BzJWl?W!tYH{s;?DGFR|&pH
zCM{}8GwWFGF(sGF9I#TMg<2PHvS#!|m4rC9Zo(;|13^eV6x3QfFfj1BwRNj-!K}?w
zBpQ>MnZX?4t=NSN>NWf4j(E(?+PPFpPxfn^d7<F%ho~`TZS{V~`u7JarPgU%<=G2r
zzTU=L7w?ZK?08N?quyfXygAp6q&&)WnZ93LUEL!oXPZx<(vmGD>8YuOH7c@O6t>u_
zHBA8O`J`0J<RELrJerw8vK31xm}895LtO_qt(G1;n;jC5W#5Xk+~C+o0jxnP-`8ub
zrJprXQ{;dfY`3&ryv432bG%MFXDkpPWH;Hsspao(F7K$4TfJ{8fMGE4{%wOL&S&m5
z6O%jf8LdYKb4M#k2EzRxYe3K_n9q1?=JlCpv5U{4jeW^gMyj7W1lid5KWq!#C-qWl
zZmOdv!c2kNT0X-<88>n)`3p%r7V}-Q^UgQY%YuYe!g$0jb^L?kHrQ<H#)o>Eo;8#|
zvuy5ki14^K^}8X@jAnIfYpdZ5a-o=oE6@VRtjx4|QqhJUsW2?Kd*z;>vRyg-$x1;+
z3}De0ZKFB31@~PCv!d#i?$Bto`}dWP@KM^V+((0*X`IP2wplvKl`=Cg2U<r%CwInb
zBvxp@M?kKbh<a*sYKZ5$=ab;z`;PgeAw90{jGC~)^ObWmLt!%F*izxT0U?zP=Co=U
zL(P9E$BEv)Qb;ibcPQDsG1-#VWs~*&_4`#yH22vKo=R;ctG=uTk}-rRUz^Ml2PRpm
z&Q8V=sL#{%RKTp<!De77-CMJnQ|;cP6|w7SC;2im#%I0*76ipzOp7U*n4b=tZ;W8#
zp5$26hi%{NmF90{X{mhDcXMF6<hQ@Cb9rivsAq$nY4tju!_4EX7t=25)-~L&9#Qjm
zmN^3tGkEmer#p2Y_6%gSwHngSkG+4}Y`IE$tZM7+$euVgkBk}Ix<`G<ca2ozShvpG
zGb0Rft$N{i)D0_Fy~27iA{>8wdaQ_?(KPpNxp4C(ciSJI6vJczAYI(r-Y#?6qx|$y
zYHI3ti+Mb}o~h*0C43BLx@XdDt80IfP@42agW<x2__N5jr`sH>Un)A&HYCf;Ph=EC
zSeTlbjrz$oT^{eP)h(Xp`j9NBrqV%AkR2QZa?~2R81T}}BlMuj`9yt}dLFAu0O1d`
zd|RI2{g;oorsu_Xogb?YacGL(`?mkH0#8Uq*AugZP1kj$7v*Z`Sf2<oI{RVjRYZf>
zsjR$sp5n0$?YK!oee~JU%As^!k6&6120DhYr^lv&B=jZ^?c2BS<KLIAJGFLr;>7cr
z9-BqE2%&w+K6(HFdiu7=u8{+it*%2`X@TK9tLhw7yK`q^gnurgCTsuyxrp00GosVe
z#a12Zen!=F+|y2qBHd8^)NAMBbh1Qwd1a>Ft}dps?=V34^QI=qiP&W#?N|5o)I<fv
z^#4Jjd+Ybxm^e5dxfRk~_2tXKuanZlZx+_{@G!d-nEAG&hw63iE2{*H#h2}R6d`8w
z{X>II**=`|@p^s#A3uI9C?5(cm^C`WL^YCMV*R@$myDa!Q+GyBPe;jCSM#HX)`#cN
zof0$Ky@WajTd@m8OuxJes87zXqzu}gQ&qj6ai;6rs$#z_Tm(+FPLNP_B+ss=`j4UC
z+Cs486&(V8n+`;_jXL*fw>?pwJeWUnbf@B4{n>WTnz^Q`sx=-sJIk6@KQy3ij2jpL
zYWBzu>sC}(SJ$!oe4?j=e2%<|jI%{eC5~G)=DhW14GtbW_&vlJAlj{itF2ouBH~;>
z(=JIDE8jDQ$iRJn_nF@DLv4(@jOv~xJZ}42-A2px^i|{BB{XT}NNE?Ua4#M(FTbuc
zdR`W(S)#?prnX_h5Sv@K91jvFDjuOq`>jD4hRtu+DNAhU<qgGpXr`H}2Tg<&u#Gw9
zelp;Hc?+AKz->n6Mty39?EI`9<Ed@Mo%r-7excmqcTRvt%Nc*vTt@or8QUX6zq6pe
z_Br4;fxZ%5Pi+kfm=oR^`Q!2D^!0h{?d^5l^^ioz4lUfWa1vQkISR44>JR;TXTf)^
zmM_%=`}|92&W#>jl5Ly5Jj+Wt&KT8h9eHTy0#H0K9(m>c?bSmgPTRz{37qg4*k~xi
z>#}lWBswusC2D!r;j{XJy4eT)aJUXixs++Qu0>cgpQRNrwT~qV7p046(zHw3ZnXBJ
zl<XM`*rj9ul(4A&flAT@eYMo^(Ib=(+i$NQYt8eumpR+npu(x$)!8(yxg!jbBEPJo
z)slbHxm?!brkd0M=G?ZKy#4lK&`qIWgDu9Wvb9FSs!!^X+VTixAg#Wb6k%g@QsZLB
z@BT_d9Z020Ontpk(=|U9C0m3FJ0Ha@o?0F3LW(dXagE+;SM}~q9wC)pg2h_fB>P{>
zk<@REWmX^%71=R=wo|wSo4J#!nIVOOs6hf*r6X^MVwGWRv|DklzwDe7!wGN>r(Q9j
zkV8X<-8jVlLfFvNw)%;_KDU&(g{!6ZGe2z+PwsJZoT=!026PtU4tij*nw-84f}{03
zbFu_PK-k>ivm2}U8E3*7;dfJ>#Kg`I&FhB9#h=IG`%*WG7|1tGE?z6+xN6)*QZglR
z)#2}VQ;SIvv(=^hxlD+Kj+ceu>-3BYW6!RXxe<UClc3;VCNxAQ=!yY$SuPP(*}GLj
zb<XDG^LM#tKML4+r(kY8v8SW8;`P8;UEM7xPAoP*$RDd#N@{VSbyChGF}_{05D&J@
zHaRx#kgJk4^Zrso+3}z=B7P}sY_Fs$zHaf{m(eVmmohUuJz8_%2S}5e?VKcwkDY!)
zu8T5&oc%=EQ6{GYoQsfi%gyeA{-X0wj@X-$onx8#?UqFU>-X8eS|Jnr3nXOnu9Eob
z`3RA97NNQ2>v-SoHANfOenqtO_n4pb;2B%9EvU?X`j_VF>b1z6ezmb0B>+kUw|%az
z4wy?#Pmh!hDLBhH^Qp|u)YLRvr;dl+x;HxNVVcVLc;1qfI)n}?l~i~=V_qBr4CyeB
zxzVs5yX`?6CQo#w+|c)6j?2_)VIlWtWE`)zMZ91*?hcpGIA!<A@mbwaPG4AU>-~zb
zs7Iw~VM6KSL46<Y?O5(Uushg10H4#lU{6u^?HgGwV-Z7Y$2~TFvTqDBI#9*RYyO^&
zY%#I19Vlazc}NWkstK4VLf_%)r$_f>^tH8v_1-l(bGKr_`9kL2ERi`|{UjwkKVRl*
z#Zy885IClZ0N(7<7f(;}rwHC!$+dx<oghkX`!DC{h;apFX7#IujuamsOzaAE6%s-@
zFh)ns#EB@cje2YUh4_mcfQ<LbI(fAmYu^1Hu+stieL<e_ZJS%+Onabkhv}#j@`Yf!
zz6JQS9a+CVr@+Ap(c!VOuwHIv+h~u}L#vY7--<?1w+po>2<hibk=mM2LCah$_|kpf
zHn-OCL`LnwmRtJj6o=nL38O5w>f4S}?3sbKvbyGd!u7}B<*nMHvG7Ekq<4Wh)+=zz
zG3GpjY_9yxAFMa9u>~r;kvtl*u$7d{yu3duLW1WKMl<c%*-N->Ps_(~J^TiCLIW7k
zPU1FoFKuH)U>wS8F`!8a6eb-g>U>bY78Y^pKb7&Q(=xU{Jer8AtIWo2qS1KC*qrSA
zbY)Es3Bsg!oWfGC!THq!wQ$njetfZhyB=k{mMjWhBxFcuVr7R$Zl+d+RfwDOMG%m#
zQNXAjvo7&1B^*%im*|Dr?2TvD`{$Z2BSQQhMUcg~tIU@9Ot0?En>pWST6eTZ23vX&
zD9BsaQfPL2ivyLAF7@>!L46vCuiECy6cJfsJ9W2JO<mnStFWMosn?bA&O74iOpifq
zOiZs$QRN?tmM?SB6$<JT<v!Y#GUa;S$6fP|M-+dA?1~@R+|>N7BrH1rn!9iep-wzx
zu1MIOoqgU>gHDbqk|`~M(q-++gJ_}o`qD9%3!ebcE{vrdyw%=v&&<q>#kS$-?_;*?
z@RzENC<m`_S6Sb=GjO3xL527y;P6owX>AjbCAt;4n_g8o<__v8MQ8z!2U$mYjJ@C~
zLQ#6(@9@+YRh1&Ftaic(|3#4*v)dI5h8Q%3=OkkabxdxUS<y<hfBOy?VqH|2yNi6B
zWR7jUuH(K}KS=SBmZuAdx+twLmzzZB;r+keFhX&p7)5~6A9^*@Vb6-r4<a?&%cOVD
zs(f{;D^S$hZph0S%X6WrE&6Tm>YLn#E9<faBagEJmEbR}3!VVLxu~Xs@8t^x14y{V
z9pN){J1`ofV<%4(OqW}cinPbJgvZ0}oT;g~`RS+`o5Ppt`crF^d&U?+&pf8y^C(Ao
zOy;!NA6^%kcqA<C{D;oZgg`E6i<RV=X(*}g8Lf~^yk>ZPgDQ$z0I4Wr=PlNS?DRV-
zZoEkA27}F6JyY)FB_zjm3a0ULq!c(1KC`AL%eHHxFHAbb>RB4t$rBb91D<`o$^@<;
zLG;M7A+{Q#j`IN=5+CXVtGA+Vf>cq3Wx@qdGeE3cw>o!4_{&azRC)OF;ol?xcx!r=
z3+KMB@2o|Asz1|I@1>`ZHCa<sH|_Kl44KKXZynQJVPVQcxKWE*eoxe7zOJ5A4osFE
z8{cRK48!+blJGXkzd0>=6%)g+EJZIIqesb%N84rR(CuPlV-11H+1ahZla$nI=!eME
zPxqJntpT^VriUD`6OX$tU9+_GOLQNPT|}8i#eHV>hj@V^Pol?kJ8HV-PVbMT1b6BT
z)L^J<bWP79Nte?cDxMxbv0B)n(I{wa?c6S<hlOK3HOuo&n!-<bw=dV9I)1LNr{V;s
z?t@T~@rN0?yYTM%5yq#_AfLOVW{tFnQVG)wP)bdSd&Eq84-(R_^Z_4x(qLS}oQ`|*
zv|uKV{pF+MW0CkwLitneQ;hn8&^b$_V1JZ`c_sA^SO?Mqq>*zM1x0${3xAm>GBV>|
zRF{WNUOG%ruz<>i1f5M`v)}jRG)u&$>{M(AM&naFt44O^02mUH%kTa^M)%otV8l;X
z3WbPJ$F$?B-&|F3UQIlrPBro{eil9AMN(vq-|bw{fWrQ<eBokQQnsWs&xp5tv6uk|
zsMpq46E<hp<c?)&NzX!QStEu@UOCjM{mvgh3C%>;_B(rAEW&mwikeiv96JQ!wqTjd
z8I`&O-QbG8M}~fi1RQ<`GFL&gx)0$<M<^&xGl}=Hq@!hWxft!G1=r@y*}8SuUphgZ
z+Pb<SZll3%r@^UL89B%4=7>3i8@bPAnYsONJ9S6(WZ(6&eHnI-D;tf&LP%0Ev9bdF
zq==#|pa5*eN2n#$11=Q2$aVJ9&A3YZ#+%DIch}a|n$VIAOLohm_6S(Ce8bN2@^X_L
zR0~%=hoATedi*J2M($9rpg_AB@9F+T*26uhat6V~o;C;op-dyEkxQ`4pbsV40b}K+
zxd*DsJ!uq_ph1;+Ad0r4tP2I*CYghZwq_c5Gq%1mHJ|Wm6eczhW0o<vK3VTb{TB9*
zICk|Nvp7i)OkMkC8ztQ%Pd^SI1+mFI2x=t~CcgiDopN`zkban;o$UD-$}FLngwg<`
z^FM+PcsM~LBrQsV(?OpZvLL0OT^!-7b<eIc(!Q3H|47=^A!u_(vrXA|0=z*F1?<$y
zSbY(ZcZBZmk*kzL7`!3Ted@B@J_C_e?%JgVH0I&r%|7;8ACmbdScy>XjrOlrx%A@2
z`l}{bWnTdZC$7{m2ouMK)IS7$K%U#Va^*^{Tm^vr^S}_0vvz~&s9BsC`_eN*SQfxx
z&^nZqe=|}A;|o(jwh0~6u#zRa_dY?|&xE}wOBWH!=p`Pj42ycOU<lAC0Cy^K$R03t
zq5--{+Q$vr=AS#4JMtOr`+6pz9TgZ!l7CDnwrJ%#Su>iDH{w+i%cH8sS&V#H*KCi9
zM-vKubFPR-#$A<wnjX@d#LP91jW%Bmwivl3(7DP9wY^OtyTbZh9|Bc`{Jq(SADYVQ
zR(Ogl27CuN8c>}NCiDghzuZwlb=lJa)mGv+CuryAS_X$iE=1QG=`U=}pZyuBL^p;%
z{|#9#xb^oKxqQ0ao-<EU--cBAAO7(uytW1<p?IrP$Kry@P8*^Uf_5b;PU@t<Lk5E}
zRi;0;R)L3NcWj^pMJE(4V{S-8(&>+}(x|(Z>lW`Ah`s#&R#Dz$E0^S8reUImaCUe2
z)VI|hngL0tCULPOkP)tVQNa&vQ*Xco&b)_IDxw2OWS`EmAO-ot?YFtBji>?Rwe|Im
zoBStT+h+8@{Y33dGg22}m>EOU@QYTBIw@axW3@~}lv0Oe+lubsw%M!adV9P<zDcv9
zoj>6ur^3L^SkeF=_JSV%a!#R0e{LjbqqK`%s{_!5rW7(jPs2bUiQS-{+6oVMJ*_dL
z5mf_qQsqm3Bz)Jr1sAsS8v&IO?X3Cw`<07)cYrV{DF(I2)luRv>@-(2csp>!5q0hJ
zf)dRLy`Y<Wi&#7Jd1UjV6&!oQ+-yj>{5ucXVU@Hwki=d}LRmX0xk#~`X`$~|Neb!#
zR8J8sLWC+Q1X2!4$?z&fTgog#l^==ms1`O5?q3S!8WD?1hQGZxq*-M6V!0)OkJ(9?
z_vOo2>&=V?1Mew>#PyJs+S(a?iT*6c6FlQJQ&X?cFBD2HXj^dP3}%~gyA8jUuCQ&+
zw5CsW_?G}waA@e-t2cYhGwI;NiU7(C2<gh4Y?JlFEhSVZ-v%Yq`)9=Nw9QRqu%1L{
zyG$@H9nWuX^#bF&3v^LO%aFSx5sL}U`v}F%w|sktzq|m6`b|sP+r5RAt=38)3!uc|
za#?9A2AuKeklBSJp?yeahTq=ZO!~Mx_(5Dn9N$Tgf&M?13sHft4)F3OWcS8QwR09U
zn5{%HxrkF|4SQdmg~IPr4dfCD4NfEmln#xR?qUSZ>ra<zh+Wh6vUu@@Mk5Nr{A`zO
z2N4LsnH&Ok=28k0@+B3IHOlQmF!7;&W*2)%?a($|;n*IRQsG4^ud9`zTU2D3c0L<L
zcqdvM+lco06iJ4pG9mHtUa1ScN7YN+vl~b?a+}B9iLk<!(26p_ml_~|6Yjmiwo<yo
zbKxo~0lHgka}vE;Rtt0vsqxLodHj_0&Uukys5FsZkzs9S&nC)}7;*3pZr;4v0KgKo
zMOX$C74w=Q%8=_0gO-&ZN7{Z5n3E)V*TV}SWZj9C;wj(p0Z%7hJd8lszDIUi@v%Tt
zW;cu(`^PqhFStbQ`>hJuj|-SmH*o-Itr=ptUCo&eIUdGYMKi;W04TL)`ZE_9Q#hJ}
zhKC??%iZwgyhNoL1?g6ca8}~5TmuG68hF22mYZmc^76A-S4U@|6G)(Lkm$Tf@ywoN
zsd;-6T!2vm%p}`ppwn~xjjSeFUb5~$XSp;n+{x&u(9P_9#q2{Cg4W%`fr)Or@@HCx
zilGp)3tWh!LV*KmQ9L-(S!Re#pM=V%h*Z2$iCG9fM0e`s-^9jt>BDu?4R20fdiVFG
zEv=1tDX33K<P!azL*xO8Qgr?Kas5RZ9+j@G>mbHBEERw;putJv2B+9sa&u8ytelGK
z_@T|EWO3A*D2-kqs!N7>umC68X&h|(w?Q_xe`;T+WsSuvA`-h3Y9>S6Um`P*_(V=5
zp%o>1U%fEES>bccBFpCmC&}Ws5Xs7PkAM79UtfZBwS|y(SxV12<+6BwVS7^?!><_d
zviSs_&c;^dSp1(`P5ukqDu2F344>lyt&}St18jo!?<HV-tow5kSaYaIokCFtzMBv6
z_xAPe;3a#I<b@wz{Pmzm!=FFj4)ef;!Sno04M~PCBCob2IQLXXz}E3tAV!C1D!2zk
zk;qbtzP4_UvJaKqOQ#NB@lu^LMqIM)Zjj0em4wGN4;?xr;?Unfr4n2mD4_8$E4)-4
z%nKr|KYl!EYipZT7OjbVE&@gsPeVjQf`34srmf8dRq_sU*bdYJWxC2mn-AMu0cR;M
zUx;)ao|cyOi9<bsDDQ;U#Wi*cMTGlqI<OlNYsWjfvlU2t4q!09N>_5Gkr%vlgt}&|
zg^8AhY?x$P8)FkBy7x$a&iL&E|0|3`|8;lsKeaEtdSb(Qb{WxB35T%N`TIRhpcM1u
zkGqpVQWm~l!ozLlT&7*ByOr$4qvuvc6UZ?IjIyO($M#H6oAAZdy$Q%7nQ*GGO7|yE
zwvQ(fH<hRFm%FEqSRt^Jfd1vnmqKINnZ^bLGO}5qu(hKa@@-5p4$5Kq$2`L~i7pr-
z^k-e17V0GwHbpNL?^k!>zL46)`p0AAhp|%RIy}R<M8&mokTQ6A+rbkJt{r6Av{tCg
z@d9-197N9fNokhp5El>Io39(!${i$F9^jjo*8sqo4nGIW-J~K#mj|O!Oyq5%fDt9#
zC}BN$*)yjY;Pq17!lQY>$<|;<iVE+n>p+#@%`^LTNlUsUqx(e%%YER?Bq<t7XdJC;
z_n9XT{6=#`$ky3jC4CAzpFbo$&Lv7^Tse$irzAP~=}1k~GGiX~jg*+{47JPsmCNy1
zg-nbU)Z89b$_v&Ln~Uqhj!gRIZ(eIWEQZ4&UzluE*1kk`>KHffll3Q&l0uMbF9@lP
zSVIujMhJzf{iyzld+bL@J@$JxB_S!v`si{=Z?W`B$u$qvFAGM!)4Fgo>FF|Kt?>%c
zUaqoO$A{;>3C6zDRKA%Kx6JS`M@2HfPp6thSqvVyd`td~q<O<@DwXKSaC8(AkC#k6
zU}8<=8E;q{G8w(GEGmhZsiZ4nLBs*v)eP9aIO*HeJ!kt`gwi@U9F~#)ISMPlQL#z?
z=I8&X4}U*VB?S5OhR#Z@VZ`j@k}k1y%9oQKi-+Kysv*LQBkc#tow14eNAz5JB9MT5
zv7Ibwl;ImZKC9h_Z%F6YV;?QPV1DvTZEX==hA`7g)e+JdwBfINb?yZr3wi5&7q$E+
z0Spcju3Day{LMg%ND1RkwTt>rKa~*S<wa;A@446ZP@u)^6#a8^GxwP@XAnsYYO&D~
z0g;i_RfuiH_C#PTNVj5t9-i0>@uw2S4CVe2w>0xy>>;{Gl7NIyu<^jv(}Exh|3ln@
zz_tmI$0*^I%}O2_sgph%*^xB&Ir4E>z^P#3{&h;K;M|LoC+zLhI&2v`+hnc*%n|U8
z-?+KZt2C-3N83Y<GszFkdhq~i{Tz#$hK5JVj?H&j$75R=?y@$75W->0%i|i5jKLcZ
zR<n-%wmv+}#m*x8<qP{8Q*tPC&|2ZjW67_bmE^3jX^CC=rIqA(5|f?2{%r|4$30WJ
zxNs4x*4(L&D(-99KX)!R|LfVxM)9$+-tfanQ~*X~&N2Er08`tZl`HNESx*#(O)(3}
zbNA+NUpIVGK6k*v$~!h8o8K~Typ9DmJA?H7tUHQ}z|L(raP<=vC3;-*E><L1Uw(I)
z(+*x@A3I=N;RAcY!yML;_6*o8YWUOCX>R`i!p!-<F(&`7b*29MgZ!78eE-`lr@HFk
zFUgV5pFeNjwZiMk6MNkk2%_*EBLbQ=yuRFHN1;%%n;Xs|3?fuo9HgD~P}i7mn{NSJ
zEYno%j7><Wi(<S<?$rfZyMjyXlTGQ1QJK=@QYa4myu3*AH6~(e+S)XuCTru$BYu*7
zfSEyMfmu&VYp17j6A}{246}P1WE+FJ>Zh&r<>YQK)waIBwX!CvJH?GzM0frcv(|MF
zi5I83K}YR%T&ngAZku#m?ows({uij8ALRV~7`!{21dSS%Jv=P8%^y)S!dQ!_Gl;*X
zI_Ek!Lp=moB&oJk_tXh4Ne6S6-BU1Vhit!Ff=u!PS9>8&^Be)XzydpX!=X}wFYy)C
zP;DeZ9$ek$FJIK~_ysJ5&lX8BUS2p=3_D8$N&HsV&>)Ua@*|Rl4ieG@IXO8W?!)jJ
zfchfRFL@fc=wdi;+ezL6PPPeBO*-kv{W!EI4Q53oynRS3W>dr8NHIwSOoFNbdleJK
zuDgY{krbDm-9$f32@p2sf^P?UUok9Qi<81fdZq^Jc>cGR6#pp*{dd6opXw+0Zxd?%
zCweRXPp(A_7`+G8Tvoa4#H9{14Kbk;Lrtg9QWAfjUAtxeCU~bJG0+gtBRe~kWTS-N
zSdV4i>R+=G4qOi@=2(q2jNJdgB^v`43=!L9fRfOi8>^1&T_rR1)h39AfX=R!6aAX^
zJwPcORwZRw<`^L>9S@T?WZNb3{g+mvBZ<JnI0f#72k^2}2u&imk}J2qoT(!W58>$y
z!BqHyvC7jQb(-!BBC6KVMAofaw;G^)FF9{;^~;yIzRjS=p?aez?~F;9E116rp4`Ft
zv<4!rH|m&Iv9SVxo$;gUOHfD~#l0LsK<iVbT+$j(z%_x+xq7%$1oeU)eCkcfr7t=j
zJ&TOo{WDL*+J+@FJL|38^hlV``drH>=}PGC*HVy8Zt1MgQbEz(7}a%7S9f5_Q>Lt5
zpF51x^7lo#@O-n>6@n7oD?14CB&OUJnaT8wynTdPbFY;2MtIa3Tq}w`nxy7kSKc91
zgJ8L;$*Oczb#*PYIG!aWR4lmiH)I;-ju<vLcKksC1m*agzIN@}^+kFM$geA;vUdt?
zU&qa}-tx<<bL87GK1O)P+<9EoqZBr|A9nAF1C!8&#!szJLb-}WqYL@(LGxXS8?ezI
zA`JZ_QKM7PXzY)UHUXTrlU{<1O5d&0M*xWrJexKKQnx4E4?G#Ve*`F?&&S0?)z&uU
zo8ZW+=;%901e!Sx79^;nnWM+(5?9$=WC7TiaI7(sNf|OzpM>px#kDkjVGx2O2SKvJ
zKNs~%$bb*Avhs)IBC4U5g?mQIZIr!uQeJbXS{&YdEbb{h64%Vk1dv9`=+2i(_WdTv
z9oAK!#60}tQ{pH+Yd~1Gq-#tepHfxq;3ji7HXk(rEME1+C@R~-y@fUpZ&V_7&VF<(
zNy`*~F<MYAD?qLhxZpyf-oUfH{`l$OSKXAWzQVhl313RszMo}Mxc$~LXE+I0Z5!?0
zSg#dUn%J>zdr+Cyg}hgZiT)_d%J}?op$ETOn3<)-jl4n|4>l_3potUW{`d|Ln3%91
z>^uj$2TKHl^XB=39|-?v@`g0p`0ifcO>VEknKQl=XJJ@esyjxp5lgz9?R?(Gs(j|0
zW+>_Eww$#8l=XQ@F3bp;^72n%9_Es&XID<Rn8xpSr1uq-It7COC(5tsXSJ*ob#@sT
zhJh-YM=~>ti`@2&r^vcM{v4ZX28qc{nuG|OK-Ao0ainvO?O7^1&yGgh`+StFN^)j3
zn(>$4UHy|ey2bxj58BE*B2J^Ew_z)^Fr}<A@=~_FM~weJbfEp8jFSHiQMS6OpEw0V
zV7p_%jBYOK*`y9X$zq~o0I7BEW%V3xOYzz|oS>I?WR>jfFxYl<KD1XW2@j0Sf(U4b
zcsn?fL&TX5f7z|t=1C`#%2*p_W3rcw{Cm<`0rv$drwNaiEtNs6&F@;MRQNLv5TCTX
zwzf8T$=CT5$Dwc58uoK<WNj6nM4r*!t6>tcI$p=(-gz%Gjw?xa12XED*jtK>IM%1d
zB`^e!zg40<@YY<w@#VxKo^xm4TwK$=I?(&*iRJNMOaEAO&LQGx^w^Vw<widy70MgX
zdfBA7+jzu<IbQw=9KgrNqalY!p)`Y<d!kPFoFr0%$qXd*#&6lAiwATbRBi?Qht#m0
z=v3_Nm9q10aJQva|BRns;=GpR!rmrqF+`Yi9Y>_75&+rubzxKTurLvFgm283N}96-
zbXSqXakM<f>CMbojY?3ug3rctO>;z?G^<bWoy?!iyH`FpV7VInd-Viq06g_)CQaer
zIt=;N!F8InLE?bl_L#c8Mab&fwY?VeP-0{3ex}t!vF!_nNEc0$rKP2i#eA<uffto3
zw_4inHq*Pvz{C~AOW>bb_NuznS7|VTN;L2-{1%|W+Q5yVg=p~gypnL9T9XVk#(%wL
z*HXTy2kB_J_~ZmqlSP|C_AG(sv~}9e{Wde_>IC?X(U76Izph50ZAOTBfZTV~G^9wN
zlZG46qo?XapiCq-ct9#+PqNt83PmaUpG#W#H6HO4oO!XC-sqEp7DX{G(!rpgG&cx2
zkLRaiLu8t!Z%gXQnzHRpY(L<%Bb`~*MEE=+YP^Vu*rmtp)o|o$6+>l2#C?#f@JcKW
z|4FGlIkatomAw4!`*+Z*^`NEe)eFDPuX0%N@|81(emH9fd%itK-;|5)PoIj43aoo`
zmyGzOjGvo(5I04Spe0Br?3C4Jjz7d$*YZAO*Jjo?YY*-Gr8S{32WeA1;aBoc+qS5$
zcB+rp)@X7sH`7xsI>mZkb+VG#FtMPV6WrjK8P``*&CBbK+o2#vqukQjyS^P(Hy_-R
zRJ%sBKD`$Bx%E~E<g6bi-q46nj%-^rJWcjFK^2-vc)&tm`{UFocWzNFjtvvJO*Qg#
zv;!aOw|#7g*p%y!79@d(*N6<be%Vd<W?_j%-D>YjJH2}yI#5L1XQ!Y~GzE&XUUMr#
zS1<fqC)#ZIzJ`cdG$n9vCI9Xz>cIDPpaR_{sy%Wn2CK>!=K+$fRm5r*syr%wtI$7=
zwO&;o9AR&15h*DL2=SXakWMo|IY~^CdF@LYvqup{R;ekne%&FXqrbGG)vXBP_zz+}
zT-;8Yw%OPU*?!IC5;LcQPE~-M(vu7)o2VZ?3$f2e46)AFr3RcYS9F90zH3>yN@5!X
zo(*WeREZcO{oibt8e*sX<Cw#@Wd$uD+m^p$wO4yCI8XO<L>FSj02KNQjGw=5R=sFp
zE7=|CurfrfUQJ&AMxUG;^pu{W1&dk4jD2H$xT#otZuI;LTeP7IHc*B|__KK+syMb6
z-bn(^z~<|VOCh&1ZtQ1A_Zs%;Q#`|@UkOBC@l8&>jhs5ykL5CI6a7eIt@SLuq=ay?
z_dh<!V?rCy?)ljhZh9V9MPzL-u1X%H`-u2CW!myJ1utapB-iNqM2-IE9^A-{RnLJG
zc8X^$>mf@uTt37-e#-CUXSVFicUbG<DgS3X`Px4Vncx5F9zXZkAW1Xs`+~|NeR19W
z{mnfzBF-mVIuuy{`SZDfwTn;S4Cx4oo6og{1j2fH|D(LTys@#d+LT=kU!%ELa$}rK
zt@pspgLHo3pt2M4{ncg_*Y8$Z%}Cc8B)y7vxmVfs;_P1e=VxCz@X1}MpW1djtk>I3
z#>%bAJAr;oUSi(mv;AC;YNth8RI0jjaTmqVt<^Elw-S=)*jvHd@@{bL4bZc7ggA=~
ztQ&k*XSe1Ie6bMFRFIc<gSYh#n23=4JWt1K&;l0UrMzZxwcs#c9(|c|U}W0uis<9Z
zwyOr>66V&6xP@9dT&NEgA5d$Zxbdrnx*nW<$Dcn7DgAIYC8CQD|1h}m^y|O80K4z=
zQLa_haJ3w$kTAR(%e#nZOBq3qAs$<{Y$@(C0*fD*?ps+_);0P|s|?;^_xRYk-S_FC
z{_X<x`q{s<eyfhUrS`^k>f7xgi>c$!*AUM$8aE#bNy6s}qAI^SVg<E4SX_fJKtB!t
zU|-D7s-1-5*yC%dq*V6Wxw?^bjD4=E+HYR*JS4EPOG%{{Y|i4lMu7RdMIZZM?EyWh
z4HJym_|jO!RjdafX#!bnnJ{7B#v@0Lj75tkH<}Z$3qWqzR~KI^bo=jPc6z97PJkgp
z^UWh^{bE!nN*$y7wA1^mzrVOB4#FHO8d~N`RL|<dKX9d7VL_mg);F482n*LQ9S=e|
z1ys)?Mv3XmqgA{@-b3%#xJd!=@4{cUg|wWW4i?)-v^d>UF|4C>D>#Pvp$RQ998g0I
zb36B0-666Sop{FgTxRWA=Gin)aRVAdkeFi7mICtxl{lVi1zF*H8=v}0(CDkAhTl|A
z5VF6QgB6_aK8jhAOH0g|fx#rzE0Fb8FC%&vSivk~LfiavMgm1++|zN!-mz`?mbhBd
z&oh(y-L);kW7*B)VG~*zmbx%~?W6x~wc9hh=?NO^T^?xAJu`?tNTbbAa!C6jd}BI2
zHmnCtFVqHI!|kU(u5{Z?EXg0P#_Iiq0;2pfHztV#$j-sxVGDZASgHV8QM}|<9o~sT
zgkP6+IFFXcV(qNvC#(WxSSOsV<@IjXXF3}GryOu6W@_z9X}%5KL_uL^M>&1(<CVpI
z39L0Wt@f<=+LG&Rs{>b;HS9zuWWxN>lSEpx&BeHg3}FSMPm)PZ<lh8YuXuKKrRJL)
z#4VQTV<FGfQK{MC_viMmcgS=v^5URBUrpHyeBe(6Jc+YLtPlDnZh=<#<&Ocotg99G
z3|wy{z>hK3gX?HVr~4h!n};<-k9AQ_>v#f^4o>IcHGcxBmk6LB{d`naq|?SGT>%TH
zfL=C(^=!RdfU<C^a(=dw*g3bdvqyzET!gV|4>1t`GbjdKBWlbGRF!a*qxBZYv0Q+_
zOnlI(ruzH~Px5GujhougDZ>A+P>i&gr{-Pz9RA^osL^Br2#ZrRXv+zIp%{Xx;M`E6
zE#ba?G;v1dsYDWu##;QddWgipWz`OFZRdIRx19pX8UL@ZgRkgflOC$##F<3>g8Woz
zc#8fN)K$P9mdZqracn&^WBzj5fE8<$$5CgF)lQWdp-?>mzJmv{*?@H$Oj5Ekv)@)b
zx@xnQyA*rE6+Q2TB^VS?-G(M---7v>0)KEkBE-Q;Y|rgz!u2B!LqvFeFcA#TB{x*A
ze_+7QCGF1$3DkkSB)G5OI@~mcFx1p^%ndn}NVWr-jvRgP^RuR4uZr{XVvoW_$w|nd
zPcUGXFy{t`=7d-&9T;~<7|-1}SXod%UR`-DhEXh@FR~TCumO874)p_=gmpy*BCt6o
zHdqxfiBDsp^Cauz9I`kYS?&k60q6FpSwHKSs#SRker7%T<_*`~O}X0X=6n|Z+=r75
zW>}98izB-sxmTXVzz$Ntl6|8$8HSH(GLND*@}HP;NiHc+O=ieePMitDWQK@=B5zKO
zY;N-YKMQ~fnw79c!JvI&pZz?JLSXt3MDu^WU30KF;{vuwrq?skdwz-r9QAv|&tm75
z7Va6a-DJiMV^>+X-S>MY(pi^If`emX-UPpeRZDhSJriA{O0<q*-`IcMD1GR8sGb0m
z=F&1x8iP${R$4j9{p(w+pm(}@q6;0xdXoj+;ohfcWcI?p0(40rZLmUADwRydNXQ*2
z-MhiNWpJk4%KZR|uv99DFmGZO`^WtUma#w#fNwB=Y?8o;M~7{%x{*?VL*PPTX`n#I
zEiOwcMp%p!A3@+m4(^=czZMZ>1MLa1JFjD1#KD@a!plp{rcTJoEgf^j4gb`k$z<U<
z7N>yHXKF>O&^(=Io}GN0Rc0gk&7}E`*b6)0vflJcl(l50b)KKT4u~3+KA6J1mh}*0
zS;hzW&yeFqfbPRc7oABwiz;AaSRGy1Dn#^Y=aKL5`Hfs^-a_8y&FlS_ln%1qaR2MS
z!sMU4{?5ND{l94Zf5g#$KY_Jkcm6l71Hb;jFZ@PK#yFi!fgn^Jy@bc#zOSwb_?I-`
zh!|HqmH2*FE=HG^T-?zUweoLrAfiivj@GxHh8n*SZRn0<c#8;{he`hnif?{37JGfx
z`4Ic?Rnf=9&_jm$kX^D7^3s)*Mv#f#a63w14#;}aq(*>-b(2YydX5hHJ)T~#?i_N$
zK_~6)z67O11Vfob05AKFlZAx*xF0rqMP9xcTH5EvMy*a$R`%q7vHO?S)?7pqZz|Ow
zX(Oo=Y={pL-A^LuSULXuY|~bhEF1!?r@wl7nSEG>)ig%TTN`2h3XJ{vh|iO$ducY1
z`G^S<#?X8IZZGSZglw3C&AfTQp18|MXG-|^O7+J6^!|eAXg;qZgTUBL*(1{hA;;lo
zslH%j7w;<#0>d;&^d+jeY6k#F4uMUKk?CBS+SPz-+T#SaQz*<;UE3u~uZTtrkclQS
zJ_+gYp}n-a`&n2zL_@b@6eJIq?L+zqeR`tp#}x&D@f1p!sElys)YFi22?bbEFroEP
zjAxMx@gdi6>e`)))6Nhg=@?D5*Ap|eU>*_FPLHGY>5a!|5O5h$#$64h6Qg@vG*!KA
zMTYG)jcc*_pOq;!9D3)0z}F2=oU8W?9!X!l_?fVaWFqEMRZQ25hnCGKYmfUngv+}0
z3AdJvohi?6=kP8fEp(&@8nYufhG4*s(FKHi<2fDVF&$(Y^k$768d@}Req~#%d1i4g
zVu=vgV3zFz?Jm~f>idTS-`SAMCCz9V^;j%TAiiBnTas~&EL)*F(?x%QfFX$X2$cZT
z!s~(d)f29Dn<Da(O#!+%Mz{2?7?!)b|LwOCx0IBV{#&;^KD`@HRvS2fE0r{1@$w##
zlr%&ZP*1pejp(%m7NJxnFGUe`N}sfBd82!RU*n*ik?3QxVTsoeOj%ml_U+Pnb6um;
z#6gIeE}9OW?mf{ehu{{3DjTO?;C@%&loD>SiCE@jNhfhXq6VWMW}Y9U43Wu5FJ9~b
zwGLgix5s5D{eY>^*MVh@9%u?LM9pCE_K!so<1viz=;-MF>?2^N;^iO9>1n)C+q-_n
zwz7t~qCC{_xcXSl@~XPp3r(+b2WHfspWW{%C>Pc%Yf+nnsxG9oVQ!%<3a0w0l(T!U
z?O{Yj&4sHZrMUD8>kf^MMUCjqg;q}G8W$OSp7M0>i!bO_rPNG*Kd^vyJBQfWbgsW$
zRQH)hWSCwk|MYd=25)+VFolnFXGj2ub`S`Eoi*L;mB}HI8;&#A9T6umi|6<^A#%*l
zj+z=Q9wJ8f$reX__<IK(3%pg84@Z<x3Pu7G3!#E9iDl&m&jqUM77R_eXuCE3(M?`@
zH2&2cPe+(=tudD6Le2{EZ=T20*O~p~v{xwnIcDwuzF_|M6Udrv`)eNEzZdGKr19Tn
z9XexmkD885fBSp|gL8P(nFTkGKQFCaq5JrBWU90dqqw5<Y6^<NR8`^hAywV3?JZ?j
zL>X$bCb6wXDOv++#S@iE6$K;Np_Su&jM%8TYWq2a;Vs3>tAaxrYGT3rT%~Ebv(ow?
zr3?<q1y8#5``9pxT=Vm5jx!HhHfNd!?SBO?r|Q7k05W9BH0bW%zd5WLSY}Zx-V|+R
zll3&c_XY{nh}JQ>C)dL1L%5bEqZQ7`l`#%WcSdG(nZl76n;~Q5R^@0|;ur6*RIw_m
z+rO3JO?Pg?v?LP<%e6w6&w}DPfy%?YDK)xeQ!Wpq>h|I+{`12y8t2i<&~`^uMKQ}}
zi4PnRMksY6HXV82IepdG{^o8_H0lXguJyU>IKOwj0ZgeR511jI6LtngBw=TSf&?hf
z8o@4OZV1UR$jc(gDr8y?o@E*&cihE=S=lv8rfVd3*mf(cP+|lQipU6C&CsV&8<&6G
zM+P*4*)ea;&1#Ccd)H>j6=;UU|8(<4Q7VUeWr7=Z=huO#Z=y$s7@JSIy5=m{$*it9
zd>)1)!lpfylQf%IV5p#Qd%=}WH;->;@kt651^ImImo-KE420bWC8w)T8SL->qCBQ#
z#f-pvBR4FG-fVgdm(wRu8Tz7BVEwWP&y2jhmZ@Byg#M#p;j-X{43e&WYE*E)^3%)t
z?^yA3ROi3f;os|k-1$%P{Vyy*f@(y(eu2yeVLfilCrC7C$+j0kg-kk8xzW}K|4YoD
z4G%?M)ZWZsh-M+Q2u=iFyGH1)8;D(?8N;85s7bm=Vl=mtG<~%5%cr#~QK5LDX%{R!
z6)0(!CAw0{$e{M#*hD|Y`@4%k%W;r_b7^HPA?wwnCrh2IU}=UR;3J;bbB2sBQvmt|
zo!$Tyisc6aBfnZ^>JssaH%0v9bL7wbN>DeMZ3Y@I)TLJ-!3z8XVW^<CqD^}>Ybq(m
z^+uqA8t<zwsf*V}-wtVs8nh*ITjUGL447lFES~Jrwpz$b%uZpCJ5(x$x9osIl`%sm
zI-*n}607tDjFThdEL|~J!P`82w<ui?XqrmBJUV~=u=MyQN{m%@Q?!}uOhMy0jTxZO
zsCfAO12ZVJYq2BQO(!Kk>r0OZC6uK!ny>dRdMkLRa8uaOlYx>BWGJ7?#!Y$SoOlaU
z-mfj0)`s%4`w9mt(B1%pWqkZT#6i~;e*_6?Rd@I&<@Ks3dXYg#oUv!h(3}{f`^%k1
z)T@p?E^{YIA=4YIBNP?Scm^67=9oV-(|EKDY5Gi4+9r}0qJdPj<>jeY9K~IMDd>Dr
z&-we#Zq+dbE9Ne+a=4?5qo3HJGNuth^4!FNHkqn@>z7F{&K+w(d$|4pDxm-)!4nZZ
zs&CHzIJ|7JGH%B7Y`Vt_vn{QIgn>>kJ2QaBt~_`$H?`=ALjyejNT*Nq@#*iQ(^#;<
zxPhR7?(Xhk`w$`*pxDC?M6|gG9f7j))3gQ=Oiu4cBS<*r2P5-ui-zVCR;iGr*?jra
z8^7%0CG0er{q+|*iLd~P2-+h`(kn(Ojn#$ANwqL-{7EYmIo3YJqxfm(rm)En{ID@>
zvgJBgnB%x_$hmb3tm^9@koN7ZlW25Mk~i<;?z}p*74sK~U^eN*ggePGxipm+l~C)D
z)VjFI=-{D4fmi*&k2T)6{ZlVl10j)Id=qZQsc#>VFZ1~L2)6}jG8V0eKH078N&9?i
zE_4n+d`kiwv9-S4m?jZpHXVud+mLFa8Yr+i8w)|2MKPe}yu;de*U@5f_F^&V*E)E$
zE0ENT{S8TkSHC=Z($Uywt8{ki7iD;+jwblr+o1s3@&)~vFxr~6m+AUcmJtA8E*7gY
zZh-a-Tt~R3plPgLKN>%Y$c%}ZP;G96v$T{u>Ru%<?y|EB=T8~<-w5H<bHdOmGGXVR
zX9G<zz(d5FM5SVwj$=+Gcx*Dp3rFZ#?ILpe_us|dEeEi`+EA&Pn0u;c%|spA4lWe0
z63?gm-nhu}481uTeaBOrmo>*1A~CvVtw>KX%PUB$`;>>-lmu81n*@<t0|i*`|B6Lm
zP%W0TM)CMB7s}j51_P1CStto)E`y3&`#SX)mLBqK#B}#pUn*(3)q!gZ!;jlp(~2X4
z^t|{~BMZskPcp&-+>U9`Pi{x#5C^&}$Qa+&`MJpwGIGK+h!t!0KQq0$@+a{AcJj{&
z+5f*F`(C-FE&u(@?)%XWOK%WXCP~xh^zH-Zb6p{16YgEVjGaAVB3#Nstrf8ma!1}+
z#sq(*vxXj?2yqVL>+`c!7xKXUvVY4JFrJaA{rvfPd};S)C3}mSyow5CN{qix<xC`A
zX*alO&x86}G1TyoF>spqGUM}HqI^ru(4F*K<vHf_5SJi6hdjmtW;({)644&LYas7^
ziAtWudw!CrGIU7yXlf^h73{S{n-eeN&h9OQK!d=Io;7RCE7l;-C4JfjbA1JPE@rH@
zW0+qYo@)ugh>k&WG8kJoa?hXiFza=C{ocZvKYWwyNBlz4%-dtX#-VPL@l$p(gaUgQ
zgtixk7?6t)$5f4mY<shQuwvF2*{qSnZTk1WJQ)8$&HgK){$J&8wMb3D4Y9hvzU(FZ
zI~k0D<AiZ)AX#E?nS&VxObh7fyLfqt$?n5d%BqqIfeqeCZqp-}U<%3(LS}p^$GvQj
zHdE*!=VWIm+#L3;8m=(|gwzz2rj6|Eq>oes=|DY!`h^J+k4!T{n<wd7B((<lU5Gv2
zkW-QD;KRZ!_5||2WX1q)G3FUaU{FVl?y5t_|Jv57@5xza`;XH7Umq6ef$gO!7*=wo
zrfL)2IUw?SlQV_Js1E3OzB)n~H6?Dyu!cn5BeyEszK;YiRvSFVad6vw*e=~VL5!@V
zVX&Vrob^3GqWjqmhZr$NVl}1E@U%Wsq-KFTMG8}1=}>x$$d(}G$__sXUNR3_BS&g&
zszBnIDh=}M;Tt&ssuDr*2$3gDgXE9nM>2huo)X3sB>nFwlUVJ0WKuQ}xR5Y2ns$j6
zs+s1dVnrd(NVb#SeIh86*`z5~%e~N?Spxa^kmvk@v^Y>zBsXa0yE&7oW4;K33eY}3
z-rrLqQo_BCj4MQ@fTY%u{!KttaD#WP&@bml|B0QxanzIHGh~(w@t73&Wr21k>V~d^
zo5of6d^nTI-v|gJ<S+&}Wv^U9X6O<YvXQ2bHa}}(%!w6ps5m;t?k_@NtqRd_G<0Mz
zMqD%V`nRu~^3@LGX-gRh_Qgc)`)TWBtxCZhgJ0tWKk3^e3^bW3RYGA6cZ9aOB#Hl4
zw(b^Tu%0ZHgjl3pn~mx1MfWO$RTbuS$?!P{dSgyZ;z%P}o`8(}eoL-aDcVt4OfTmn
zbJbDQDIe~9AOZW4AxdI8PGunm@T^yn9#_-BoSr0%+jZX}e*Ytj?Ium6WbKeMd4tPO
zf{i4ya1z8*)(U@JB1^>t==hQjG8WN`z`7iURr2!1G5&V+jjGCk2SgmlD9tUmxov)D
zxygY>HF5;ksz=xF-rGfX7w(B5nMb6Vd_@slr-*p-bvZIu;&Vg8#;!03nAnK$j1cnF
z8bc(Ca9rO{L1<*JiBj>$aQCRDPEq;(UMIv3OD{5K08eJMyOKF-FiNZeVN2!?lW4a1
zwTo#B**Jvx@4!%Je=@cOX*(PnXAZvfl0Xf;O)$uLfnyiWN9a+2?W3h4)50lbvG7<8
zCg5!j2??P`is$hWi#w6ukQ_+@)Uo-v3z)ne=bD`GvA#^0h^?Xj7b7wnlKjr{$`BfY
z^c>RBX^3tb0=$8#`{@#maW&aZe=b}F!T@Urh(pv(Lh4lj>yW0zdpXd>;*U^N_lFZO
zY}1KMX~aaB1c$>ryNM@)N+o|kK?Jv&2!FUP#1~)OpM9&)s7P%PGhCP5U+QFn`7d;I
zKnnBn=1-(07Ljv7ngj5BIM<Re>7xXKo5h5X<Nz6s@hL=79N?gUzzRHN+D!(X!|MS*
zE&~t}12`FVFWF3%=>FQX>p(P<!O7mlP`S`Ww#XpqV9Iw4{h<_%!?zOgIFKmGCx&M1
z6J2$qZ=DGiQy*(kQ$95F+nY<HO@#^B?Op-|Ojf{rwB_pdV<g0oh(e`e8v@XftprSY
z+Vj+(xdORJOhD>#*~5L_WWE8J`fnOkExhBT4;4->BuDth3M|c<FVKF?-AP7iqc4j5
zc?TadADe7s5~U@cO=h_>86>I5%l9abcV;iXM<zlDxU|s{JDb|lv^PoB_U~Z3QEs*_
zhuGLi79ayU;aY#*)Oq(A7KWHAsNHnR8|YccVQGn!foHWR2KYC4li>>|<frP6wToG`
za+7)U#KX<%gg_n)mG1`UqxU3B5`PZCy`?TQK7G1^O%a3(>ES1BQ_X<rpU1iW#+uU+
zEG|q;xt3Owpu`X6<02<#<|5AoZwO=d8!LIt|D!f~QF7vAj(fx;^6TIvGmykM`NH(7
z>u{e-4?1>*I2NN{tfSIn&en%ZT!`!t)_%feo0a=)`t`<V1*zgUXFu21?{><94}Xfx
z-zHQ&-?JMbq|2S9+hMqizY3*SmHwnYmo=3eW2MR4;x_pr+dOz^>ZzNAK|4fvqf&8L
z_mC)08o;GsMI}U5^-Oi>)dvUc$N&<uvB<C<qU+JgR1IWga$Pi9>jv_x3uN5l)P7Xm
zhr%YDJA9m&TOh)1njce{|1uQO;6pPF_6*vdh`2gGH#A>NX2d}MQ|MV_Pdf84R||E{
z0T+q^(zs+=@o5Exl2sm7&6y&+a~rDcqx$Y7|19DJM*1FLa4gvzuz-G0kU8(rHsEp0
z35HfOqdXJC7!4pku>3t>6CJa*x`=H=K3uV45!2b?kDwV6oQQ@^219{{%VbW_gkeh>
zDkL-V!lnxMIAoH>4f@<<fh?Ld`ovefqN4UyC`|vzjbV8jIvCOy(Fbc5YZN;2i=;Bt
zb0^FJQH|j5@rHwE$rih_<S&tsP2_|+2zhakif+LoQuwp8leCB#bj3)OO?|HEr;Jw7
zhzV*v8El8N^YCh$@}vFRWo8;HA3oIX7U)QC+Vj=u!$k~rYlf~$?2G5Gi9wl^fJRA1
zT>1fdO-yS-YQ2L$>Cta<2E9V;=jQrW&}8YH%Wmqfa$0`#IO!xxvTVtIghBsgT2ISh
zJt8={6I;pIKprJJDy%(D7<}AL#yXP6{Dd*Mgd`o^nA;<Zll`51{`NdP5s)EdPpA_w
znUNrY_%d@S^hW3`PRrm#@Hv14I+;lb*>3z0f4CW6I++AW{12?_-7@5)K(cANWy|IQ
zOe$)RiB0f=^`qq5ZJw=U8wCnP3d0FV#*;-Vhit~JX6evmY7q$|SjH1bTx6gm8KZ+K
zDT5Q*_XA70^#(hi)1MqN!$u0FViZzb<vLd%+N~uEgypJnjz+6Bb`pz}#OS93hKTS0
zb*^e@2?_eg2V5=>RZOV7$vL9>44V_9%Sa3n0gQEmJE^`SzlD-#O*RDVCWkS&Y%Td~
zGGr(*>RxBr#rr+~T$g9EiD!mwu=ds%?OBZ6$+PF__2X$}&lfj?`LTuvwut!XhzsL%
zsMKM6MmYHjfs(l}NM@dckTc}Wh_Fa=rS|OIS5c5lkm(RU2P*dJl5q*-76Z^$$BmHQ
zkXru05H&av$&`3jHpQe?GT=trGHcgE{M80B6bIhi5=<0d67i7mMkGP9(^}<7^FRXr
z)(a{8+A(FDn&{RkKzX8N-Yh0H6}n8WdcxboA3`PCXKoepe$@?|#dPFR>A@7NKDA-P
z*g$E23ywLVpkV7;NK}3jZyj7HwZx$N?rnm-<Hn6Q8WqyJL>4cW+od18F_es~N0YDI
zQ?Lb$RXK5Jb0PznaE3|jB6>98@39BbPg^8A*CAUB!MK$SW-v?0iY%=?znsVjP3{Oc
zbptXO$EYJO5BK|l2&>6^k!~_PWK_gA>$b;68~Oi7ch~+-Wx9rInlY2gutOTBt3sj!
z)gZ#qRF*@ENm5S9p&T>LEM)HzISiH|$DBeA5w*&)LMLNsLUOj0kn?hGVP%>7TkU=A
zeeFMB|FEyi51;GeTkHGY_kEt{z2E!3AE<zMCL4rzJ3+9TA?$r-&<}z_N0Bmpfe%xH
zsI-v<Tv5qs&VV)`L*(o)h987Jf{@jHa@LvA-{YqR<OTFN!eZ~d#sZkZ$mF0h-+ssL
z$QzGX+)}f66gb5Wly7B?6<2o}XYfpns^RTR^O9hQ7i+zdj8#_pjqk&Qi2h-e3`Rfy
zj8-D6A|2l3+8Y14k80@lfRsc8YmCt*3v%wzEys+A8?NKiKGAhntZ4TjJ7NQV1wPR)
zz%Enn=c8TZ<Y^JjRNUBGVN-R9hR}#IiLZLELOxL{4To@da^J0TRwtPuhNl+>>SsC<
zjeF%z?ss=kO6ZS0<?&-knbJ-}`@}s<EoK32KT=g1|G2goeop1`p))si5`u5-IY0XR
z#qur}>lx}xJ>BF-@}dFj1q*I3W_f%=V{^g>P7Efg21~K#W;!0Hu^$hUH64em$?0tW
znpI=Rri?@xoUfOdm+OL7l#3ABGwicNv%vyoYc4!UEw*mms<3yjnW3R!=8zNBU-~Pn
zC<D0!l8%m!MNKdr+vt3-KNPjKt(TUT^bHIWWY`<?V9hS><<r>z!*md`#q7euX-CI=
zJn})HV^%%FA|j;t?Ck8reb&B~w0+Mvi0*EVk<_@Jmz2+ROOmlN61UB{5;Grex6vjy
z>MebJpL3Z{m-)hPVdY-z7zz2xNh{xR@vzzudzDlX35jBOZWKIze73Hxj?O?Q3$}dh
zOw*{>7);YzLAmgr=u~DeT!2_2^5)Hiw#$+S3#21R?R<BqY`n9WAIabODYNkW$VUdi
zh}x0;f3dP+WiV&44&!s7A$J0^o&Wpqr{<n&oL`rbidSVtx3DIoU5}?{z@MLc0T5JI
zA10{=Zt!CadG6BGv;^U%1-{RTDJkl^w|uyMozJ)6kI=Cl&bdu-T#u-=QIxnU)jAFD
zn#8uGl$0N2Wn~+hnvA=;x^m1F#ZI%+)&T&!?B>Rr+vcmIb5-UN*cJ1AFUknv$UNed
zqN?hd{{DV?yDxh=O*8OeV6Z@x_TD4|ADSgVtqve#Z@ap7jefjhpr_Z2W<Ht$&VIRa
zXKCq;L!PmX;zB|#!vWo{eG%L)C1Gi4>9ZwfuJ4{$9y)X=09rp`DXG$@Pj&O|+=*e>
z46k3m9ss7g_Za1VMFmqe?B&a2lbz*9RNZ=s8*|bI$3O+pH#B?;+GL9}vt-MKqPDw)
zacs=K=Qwv`Fc>C~{>L`ZB?Q9xH*8n{Ag*7v^!0vu`H`B>E`ByLuZ3ciL(7#Yft%kK
zvZkX21$-G7${&SScRwN1_F7gAvUG+$)V$_LPNCXyjL$`};Pz@PB0^S9?iiQLRgjl&
zzPj$)-Z^J4Z|_mG2wMhhEn8UswToED?A%;pT3RtQpNdLKC->3k<FO0fV1MzVyt;aK
zj)$+>3T(yIJ>G{xDA>CiOL5Dv##hxhGHL@<;0<V5QC+<XojrQ0QyXxC`m*Y8wI?wy
zqu4v%Ehso?V)Ew3)D&Jqce&m1%yBrs6(T)og-5ITOeF6=aKQe71=#4NQsdBBG6!04
zdb_s1zTwrYiJ7qwAeaOOyLMbcRP;`|hVN`jv;0=pum7-%(irkk!$zPtGyOz<mHPr!
zW!l8Vq!Pf%se95<_VR4GT2tn5YE4azfYf2N_O`1!K<Ay2W-p_u1QiQ33V-hBw}zVa
zFbRu|5+DD@PVM?R*Uqx5;rs)=*k&XD+)GYQ*3#&UN(8f4M_OQUX=xM+;mI~uHuEFX
z&w9;f-JgC6y<lf2A}+3>z^Sk(>8848KH)Aqi-?MjP1UTnBNQ(H;&4XRJrb<1(z&Qf
z)lRy#-#+pai&=!ZBO)Tw;6`oi_izG#88NyUxak6Rhph#YN1u^nJfiv|xTVrmZu&57
znv&0!TO+m!3Y$}AX1~AA&@4iQaQFUw6XfL-71xwvv+>Y`?RB(Gd30f6;aoE3)2@B{
zjNxMor-kUR)~$Qs)Uy3Wef{RNt+kVM>-r5hUu0_rei{MJ`4$QX85*-RCPn~{M@~85
z7=t8>p)1e{U%icWKT^j#1p)HI_d@IoeH{XqlmDbT{x2_UmC&>9cYuFEWb$b<GnMu>
z(APq@EWR^8LjY6B@-U@^7~o|Up^Tc|KL4Hv4&(ao`@s?DFh9eCZ1Ma1LjhlCHU#HJ
z=)W#)o+|9wllbLRyoVSfBoll4yoatmcL2ICaX6Un7U2St@vWm1<*o(>H*u7qdGy#Z
z1|pNKtu1`jsgQ-YPjTVX&g3X6DvFo60SaI=G{pVsV5+Bg<CQ~2SXfx)<u(N}P#e~A
zUujX10U|-Pi%P2Qks}Dz-XPvx20yC1@=}6;v{o^ehyUU};#qPoY$UK|Lvw^ei%O?B
znZi4Fj$x47K?hsjO&rF_W7M^@v=kH-Tj4pZ5dN&CsP}E5srSr>qeSx!dix3jBmesK
zaWa;VJamZ0Air9p=M^0v4*cia;#ELbL&L+v-1ArfaNNC{m1SEyNlXao;6>Y7f;gKj
z_GL!gEAyS|3w~xvQuiFZp~YRagCVIiU{>;VT0AVxa&f2C;ea3rPa<Pur4RPf5m!xI
zTqbVwxKq^|c;epbr?1mB?90t5si{R!rZ7gms=`Uh2S*(${5YQ`k%&ye7I@Rzile2F
zQG8KI>XVQ=7tkY5KmK*5_C2`mw+*OwPWsT%IgQm65fM@4(H{o2$)K0I_rS|d@XH%T
z?6m-75KDLzBWLX;-M1(qGBvDj(@x}bs4!)@w8~vs`6K}yMD6ouf&^P28RM31J3H8L
z7w#;0g@K72>`5pT${;uB^Wqhh0oi9BM@I5ryLOFK)wP)o$0b&IaM`vWP=x28my$_5
zB`z+j<Q%nc#=Wq(_$|O_84{X~nK(I(ND>76(1@E-#B!SEhbj4q%>GgtEKpjKG)`Hy
z@K#ead`p4jNRi$^4LZ@n1+$+e;4AGvC?vacXJF+DHkE?|13O3x3JSAM1XthiJa0)(
z^CQhq_%ciCrm$;R-s8Ccmave}HqtV}xMp_<oS6gtfr}ztNx8bKq&^5)4JO57YF@mU
z>oe6?0{jz{yoQ`9td-j5Iy~(6o}i#$^Hdc}*-y6NLixr;O^1rJ@p!+Q-K)|epZl#?
zQXlI~RZ@+HhK6v?m7}k(PeDd0D3Cw9TNFP)ev0aY@eP$`cx$azvG-E!S>wR=zQS|)
zPsA8n&=+o-?4cS1<GSB7$!KhB#o5rHhuS<mubdGZm(jc-JcmKtVGR#!wh`_PSzTcb
zMMg$$3(1U)jn&oFoyX>E*MS43*aGv4E7#sfr+jGSKREce+Qs9%2Z;)M|K5f@%VN?3
z=SL(3Ab~N_*N^1Pp#st<eh1}U=u8B!o2l40ctp7T70VPpdSs3!O%ca?k|>lB%!74&
z&k%E2Ly$YInw`SU#rSzMSxLVynB&Tx#s2UC*qbY;dy!h=y?#yWgJxeuSa<{_b~AEa
z?+`(+3|jI;)KEZ)0)UED#$H|+$D%cLaVe&KMoDyh^bBH~&F&TVVnuw$8b1QkBub{9
zhpq`Sl860He*3-D1U1OP!Qoy_x^{>a7LSyN_S&^;_4V|`EbNJY{_aZvJa!7th_VEN
z3Sa+9Cfs`}9h;<#G7kIKdg9#D6e;oc5I#alS&$bAb6hZBNcJX3G#@^ExL9sZ%rhj6
z&Q9~Z{QO3wtfr<WdNf|URnW{)Vu3Yh9YJyPi;c}~fGN3%lbH4=P{yYtsRNZPR{rcw
xyeK2N4iDpZ|H>v}BL5MfxBoj<|9@OAhcry)({;~lzb1x{bo9i}rMfm@e*xWgo<RTr

diff --git a/reference/plot.html b/reference/plot.html
index fe28b0d0b..d8026f9ec 100644
--- a/reference/plot.html
+++ b/reference/plot.html
@@ -9,7 +9,7 @@ Especially the scale_*_mic() functions are relevant wrappers to plot MIC values
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -291,192 +291,6 @@ Especially the scale_*_mic() functions are relevant wrappers to plot MIC values
 
     </div>
 
-    <div class="section level2">
-    <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-examples"></a></h2>
-    <div class="sourceCode"><pre class="sourceCode r"><code><span class="r-in"><span><span class="va">some_mic_values</span> <span class="op">&lt;-</span> <span class="fu"><a href="random.html">random_mic</a></span><span class="op">(</span>size <span class="op">=</span> <span class="fl">100</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="va">some_disk_values</span> <span class="op">&lt;-</span> <span class="fu"><a href="random.html">random_disk</a></span><span class="op">(</span>size <span class="op">=</span> <span class="fl">100</span>, mo <span class="op">=</span> <span class="st">"Escherichia coli"</span>, ab <span class="op">=</span> <span class="st">"cipro"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="va">some_sir_values</span> <span class="op">&lt;-</span> <span class="fu"><a href="random.html">random_sir</a></span><span class="op">(</span><span class="fl">50</span>, prob_SIR <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.55</span>, <span class="fl">0.05</span>, <span class="fl">0.30</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using base R's plot() ---------------------------------------</span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_mic_values</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-1.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_disk_values</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-2.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_sir_values</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-3.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># when providing the microorganism and antibiotic, colours will show interpretations:</span></span></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_mic_values</span>, mo <span class="op">=</span> <span class="st">"S. aureus"</span>, ab <span class="op">=</span> <span class="st">"ampicillin"</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-4.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_disk_values</span>, mo <span class="op">=</span> <span class="st">"Escherichia coli"</span>, ab <span class="op">=</span> <span class="st">"cipro"</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-5.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="fu">plot</span><span class="op">(</span><span class="va">some_disk_values</span>, mo <span class="op">=</span> <span class="st">"Escherichia coli"</span>, ab <span class="op">=</span> <span class="st">"cipro"</span>, language <span class="op">=</span> <span class="st">"nl"</span><span class="op">)</span></span></span>
-<span class="r-plt img"><img src="plot-6.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_x_mic() -----------------------------------------</span></span></span>
-<span class="r-in"><span><span class="co"># \donttest{</span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mics <span class="op">=</span> <span class="fu"><a href="as.mic.html">as.mic</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.25</span>, <span class="st">"&lt;=4"</span>, <span class="fl">4</span>, <span class="fl">8</span>, <span class="fl">32</span>, <span class="st">"&gt;=32"</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                                counts <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">1</span>, <span class="fl">2</span>, <span class="fl">2</span>, <span class="fl">3</span>, <span class="fl">3</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                     <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">mics</span>, <span class="va">counts</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_bar.html" class="external-link">geom_col</a></span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/labs.html" class="external-link">labs</a></span><span class="op">(</span>title <span class="op">=</span> <span class="st">"without scale_x_mic()"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-7.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_x_mic</span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/labs.html" class="external-link">labs</a></span><span class="op">(</span>title <span class="op">=</span> <span class="st">"with scale_x_mic()"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-8.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_x_mic</span><span class="op">(</span>keep_operators <span class="op">=</span> <span class="st">"all"</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/labs.html" class="external-link">labs</a></span><span class="op">(</span>title <span class="op">=</span> <span class="st">"with scale_x_mic() keeping all operators"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-9.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_x_mic</span><span class="op">(</span>mic_range <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">16</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/labs.html" class="external-link">labs</a></span><span class="op">(</span>title <span class="op">=</span> <span class="st">"with scale_x_mic() using a manual 'within' range"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-10.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">mic_plot</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_x_mic</span><span class="op">(</span>mic_range <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.032</span>, <span class="fl">256</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/labs.html" class="external-link">labs</a></span><span class="op">(</span>title <span class="op">=</span> <span class="st">"with scale_x_mic() using a manual 'outside' range"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-11.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_y_mic() -----------------------------------------</span></span></span>
-<span class="r-in"><span><span class="va">some_groups</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/sample.html" class="external-link">sample</a></span><span class="op">(</span><span class="va">LETTERS</span><span class="op">[</span><span class="fl">1</span><span class="op">:</span><span class="fl">5</span><span class="op">]</span>, <span class="fl">20</span>, replace <span class="op">=</span> <span class="cn">TRUE</span><span class="op">)</span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mic <span class="op">=</span> <span class="va">some_mic_values</span>,</span></span>
-<span class="r-in"><span>                    group <span class="op">=</span> <span class="va">some_groups</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>         <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">group</span>, <span class="va">mic</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_boxplot.html" class="external-link">geom_boxplot</a></span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_violin.html" class="external-link">geom_violin</a></span><span class="op">(</span>linetype <span class="op">=</span> <span class="fl">2</span>, colour <span class="op">=</span> <span class="st">"grey"</span>, fill <span class="op">=</span> <span class="cn">NA</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_mic</span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-12.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mic <span class="op">=</span> <span class="va">some_mic_values</span>,</span></span>
-<span class="r-in"><span>                    group <span class="op">=</span> <span class="va">some_groups</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>         <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">group</span>, <span class="va">mic</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_boxplot.html" class="external-link">geom_boxplot</a></span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_violin.html" class="external-link">geom_violin</a></span><span class="op">(</span>linetype <span class="op">=</span> <span class="fl">2</span>, colour <span class="op">=</span> <span class="st">"grey"</span>, fill <span class="op">=</span> <span class="cn">NA</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_mic</span><span class="op">(</span>mic_range <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="cn">NA</span>, <span class="fl">2</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-13.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_fill_mic() -----------------------------------------</span></span></span>
-<span class="r-in"><span><span class="va">some_counts</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/integer.html" class="external-link">as.integer</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/stats/Uniform.html" class="external-link">runif</a></span><span class="op">(</span><span class="fl">20</span>, <span class="fl">5</span>, <span class="fl">50</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mic <span class="op">=</span> <span class="va">some_mic_values</span>,</span></span>
-<span class="r-in"><span>                   group <span class="op">=</span> <span class="va">some_groups</span>,</span></span>
-<span class="r-in"><span>                   counts <span class="op">=</span> <span class="va">some_counts</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>         <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">group</span>, <span class="va">counts</span>, fill <span class="op">=</span> <span class="va">mic</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_bar.html" class="external-link">geom_col</a></span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_fill_mic</span><span class="op">(</span>mic_range <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.5</span>, <span class="fl">16</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-14.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_x_sir() -----------------------------------------</span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>x <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="st">"I"</span>, <span class="st">"R"</span>, <span class="st">"S"</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                    y <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">45</span>,<span class="fl">323</span>, <span class="fl">573</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>         <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">x</span>, <span class="va">y</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_bar.html" class="external-link">geom_col</a></span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_x_sir</span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-15.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_y_mic() and scale_colour_sir() ------------------</span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">plain</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mic <span class="op">=</span> <span class="va">some_mic_values</span>,</span></span>
-<span class="r-in"><span>                             group <span class="op">=</span> <span class="va">some_groups</span>,</span></span>
-<span class="r-in"><span>                             sir <span class="op">=</span> <span class="fu"><a href="as.sir.html">as.sir</a></span><span class="op">(</span><span class="va">some_mic_values</span>,</span></span>
-<span class="r-in"><span>                                          mo <span class="op">=</span> <span class="st">"E. coli"</span>,</span></span>
-<span class="r-in"><span>                                          ab <span class="op">=</span> <span class="st">"cipro"</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span>x <span class="op">=</span> <span class="va">group</span>, y <span class="op">=</span> <span class="va">mic</span>, colour <span class="op">=</span> <span class="va">sir</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggtheme.html" class="external-link">theme_minimal</a></span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_boxplot.html" class="external-link">geom_boxplot</a></span><span class="op">(</span>fill <span class="op">=</span> <span class="cn">NA</span>, colour <span class="op">=</span> <span class="st">"grey"</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_jitter.html" class="external-link">geom_jitter</a></span><span class="op">(</span>width <span class="op">=</span> <span class="fl">0.25</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  </span></span>
-<span class="r-in"><span>  <span class="va">plain</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> </span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> <span style="color: #00BB00;">ℹ Run </span><span style="color: #00BB00; background-color: #EEEEEE;">sir_interpretation_history()</span><span style="color: #00BB00;"> afterwards to retrieve a logbook with</span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> <span style="color: #00BB00;">  all the details of the breakpoint interpretations.</span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> </span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> <span style="color: #0000BB;">Interpreting MIC values: '</span><span style="color: #0000BB; font-weight: bold;">cipro</span><span style="color: #0000BB;">' (CIP, ciprofloxacin), </span><span style="color: #0000BB; font-weight: bold;">EUCAST 2024</span><span style="color: #0000BB;">...</span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> <span style="color: #080808; background-color: #FFFF87;"> NOTE </span></span>
-<span class="r-msg co"><span class="r-pr">#&gt;</span> <span style="color: #080808;">  • Multiple breakpoints available for </span><span style="color: #080808; font-weight: bold;">ciprofloxacin (CIP)</span><span style="color: #080808;"> in </span><span style="color: #080808; font-style: italic;">Escherichia coli</span><span style="color: #080808;"> - assuming body site 'Non-meningitis'.</span></span>
-<span class="r-plt img"><img src="plot-16.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="co"># and now with our MIC and SIR scale functions:</span></span></span>
-<span class="r-in"><span>  <span class="va">plain</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_mic</span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_colour_sir</span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-17.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">plain</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_mic</span><span class="op">(</span>mic_range <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.005</span>, <span class="fl">32</span><span class="op">)</span>, name <span class="op">=</span> <span class="st">"Our MICs!"</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_colour_sir</span><span class="op">(</span>language <span class="op">=</span> <span class="st">"nl"</span>, eucast_I <span class="op">=</span> <span class="cn">FALSE</span>,</span></span>
-<span class="r-in"><span>                     name <span class="op">=</span> <span class="st">"In Dutch!"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-18.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using ggplot2's autoplot() ----------------------------------</span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/autoplot.html" class="external-link">autoplot</a></span><span class="op">(</span><span class="va">some_mic_values</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-19.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/autoplot.html" class="external-link">autoplot</a></span><span class="op">(</span><span class="va">some_disk_values</span>, mo <span class="op">=</span> <span class="st">"Escherichia coli"</span>, ab <span class="op">=</span> <span class="st">"cipro"</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-20.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/autoplot.html" class="external-link">autoplot</a></span><span class="op">(</span><span class="va">some_sir_values</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-21.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span></span></span>
-<span class="r-in"><span><span class="co"># Plotting using scale_y_percent() -------------------------------------</span></span></span>
-<span class="r-in"><span><span class="kw">if</span> <span class="op">(</span><span class="kw"><a href="https://rdrr.io/r/base/library.html" class="external-link">require</a></span><span class="op">(</span><span class="st"><a href="https://ggplot2.tidyverse.org" class="external-link">"ggplot2"</a></span><span class="op">)</span><span class="op">)</span> <span class="op">{</span></span></span>
-<span class="r-in"><span>  <span class="va">p</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/ggplot.html" class="external-link">ggplot</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/data.frame.html" class="external-link">data.frame</a></span><span class="op">(</span>mics <span class="op">=</span> <span class="fu"><a href="as.mic.html">as.mic</a></span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0.25</span>, <span class="st">"&lt;=4"</span>, <span class="fl">4</span>, <span class="fl">8</span>, <span class="fl">32</span>, <span class="st">"&gt;=32"</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                               counts <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">1</span>, <span class="fl">1</span>, <span class="fl">2</span>, <span class="fl">2</span>, <span class="fl">3</span>, <span class="fl">3</span><span class="op">)</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                         <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/aes.html" class="external-link">aes</a></span><span class="op">(</span><span class="va">mics</span>, <span class="va">counts</span> <span class="op">/</span> <span class="fu"><a href="https://rdrr.io/r/base/sum.html" class="external-link">sum</a></span><span class="op">(</span><span class="va">counts</span><span class="op">)</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu"><a href="https://ggplot2.tidyverse.org/reference/geom_bar.html" class="external-link">geom_col</a></span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://rdrr.io/r/base/print.html" class="external-link">print</a></span><span class="op">(</span><span class="va">p</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  </span></span>
-<span class="r-in"><span>  <span class="va">p2</span> <span class="op">&lt;-</span> <span class="va">p</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_percent</span><span class="op">(</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">theme_sir</span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  <span class="fu"><a href="https://rdrr.io/r/base/print.html" class="external-link">print</a></span><span class="op">(</span><span class="va">p2</span><span class="op">)</span></span></span>
-<span class="r-in"><span>  </span></span>
-<span class="r-in"><span>  <span class="va">p</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">scale_y_percent</span><span class="op">(</span>breaks <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/seq.html" class="external-link">seq</a></span><span class="op">(</span>from <span class="op">=</span> <span class="fl">0</span>, to <span class="op">=</span> <span class="fl">1</span>, by <span class="op">=</span> <span class="fl">0.1</span><span class="op">)</span>,</span></span>
-<span class="r-in"><span>                    limits <span class="op">=</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span><span class="fl">0</span>, <span class="fl">1</span><span class="op">)</span><span class="op">)</span> <span class="op">+</span></span></span>
-<span class="r-in"><span>    <span class="fu">theme_sir</span><span class="op">(</span><span class="op">)</span></span></span>
-<span class="r-in"><span><span class="op">}</span></span></span>
-<span class="r-plt img"><img src="plot-22.png" alt="" width="700" height="433"></span>
-<span class="r-plt img"><img src="plot-23.png" alt="" width="700" height="433"></span>
-<span class="r-plt img"><img src="plot-24.png" alt="" width="700" height="433"></span>
-<span class="r-in"><span><span class="co"># }</span></span></span>
-</code></pre></div>
-    </div>
   </main><aside class="col-md-3"><nav id="toc" aria-label="Table of contents"><h2>On this page</h2>
     </nav></aside></div>
 
diff --git a/reference/proportion.html b/reference/proportion.html
index 12d978f18..12fda0bff 100644
--- a/reference/proportion.html
+++ b/reference/proportion.html
@@ -9,7 +9,7 @@ resistance() should be used to calculate resistance, susceptibility() should be
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/random.html b/reference/random.html
index af4212b71..7de456f15 100644
--- a/reference/random.html
+++ b/reference/random.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -99,43 +99,41 @@
     <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-examples"></a></h2>
     <div class="sourceCode"><pre class="sourceCode r"><code><span class="r-in"><span><span class="fu">random_mic</span><span class="op">(</span><span class="fl">25</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'mic'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 0.001  0.005  0.0625 0.001  0.001  2      &gt;=256  128    0.005  0.002 </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [11] 1      0.002  32     16     64     0.005  0.005  0.025  32     0.5   </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [21] 0.25   0.005  8      &gt;=256  0.01  </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 0.001  256    0.025  0.0625 0.125  0.25   2      0.001  32     0.01  </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [11] 0.0625 0.001  0.01   256    0.0625 32     0.001  0.25   128    256   </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [21] 1      0.002  0.5    0.01   0.002 </span>
 <span class="r-in"><span><span class="fu">random_disk</span><span class="op">(</span><span class="fl">25</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'disk'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 16  6 43 36 43 50 45 43 25 34 37  9 25 46 39 40 45 47 17  9 22 45 14 47 41</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 41 43 23 17 19 28 14 38 18 30 42 11 37 40 39 16 41 43 49 45 44 28 36 14 10</span>
 <span class="r-in"><span><span class="fu">random_sir</span><span class="op">(</span><span class="fl">25</span><span class="op">)</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'sir'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] S S S R R I S R S S I S R S R I I R I I R I S S S</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] R S R I R R R I R S R I I S S R I I I S I R I I S</span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="co"># \donttest{</span></span></span>
 <span class="r-in"><span><span class="co"># make the random generation more realistic by setting a bug and/or drug:</span></span></span>
 <span class="r-in"><span><span class="fu">random_mic</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Klebsiella pneumoniae"</span><span class="op">)</span> <span class="co"># range 0.0625-64</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'mic'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 0.125 4     0.01  &gt;=256 0.25  0.25  1     0.002 0.25  2     0.005 0.001</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [13] 64    0.001 0.025 0.025 64    128   0.5   16    2     0.001 0.125 8    </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [25] 0.025</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 0.25  0.5   32    0.002 0.005 0.25  0.25  4     0.001 0.005 0.025 0.01 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [13] 0.25  0.005 0.005 2     0.01  1     4     8     0.5   16    1     &gt;=64 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [25] 0.25 </span>
 <span class="r-in"><span><span class="fu">random_mic</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Klebsiella pneumoniae"</span>, <span class="st">"meropenem"</span><span class="op">)</span> <span class="co"># range 0.0625-16</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'mic'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] &gt;=32  16    8     2     8     8     1     1     8     &gt;=32  16    &lt;=0.5</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [13] 2     &gt;=32  4     &gt;=32  &gt;=32  &gt;=32  4     8     2     16    4     &gt;=32 </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [25] 2    </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 1  1  1  4  16 8  1  16 4  16 8  16 4  2  16 4  4  1  2  4  4  1  16 16 2 </span>
 <span class="r-in"><span><span class="fu">random_mic</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Streptococcus pneumoniae"</span>, <span class="st">"meropenem"</span><span class="op">)</span> <span class="co"># range 0.0625-4</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'mic'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 2      0.125  0.125  0.25   1      1      0.0625 0.5    1      0.5   </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [11] 4      0.125  0.25   0.25   0.0625 4      0.25   0.125  4      1     </span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [21] 4      0.5    0.0625 4      0.125 </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 2      4      0.25   0.0625 2      4      0.25   4      1      0.0625</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [11] 0.25   0.25   4      1      0.5    0.0625 4      0.5    2      0.5   </span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [21] 0.5    0.0625 1      0.0625 0.125 </span>
 <span class="r-in"><span></span></span>
 <span class="r-in"><span><span class="fu">random_disk</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Klebsiella pneumoniae"</span><span class="op">)</span> <span class="co"># range 8-50</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'disk'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 47 49 10 40 23 16 20 34 45  8 42 10 20 11 30 36 40 35 14 24  8 38 10 10 15</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 25 11 13 40 46 37 44 42 10 41 50 39 39 40 33 19 48 39 20 11 32 20 50 38 46</span>
 <span class="r-in"><span><span class="fu">random_disk</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Klebsiella pneumoniae"</span>, <span class="st">"ampicillin"</span><span class="op">)</span> <span class="co"># range 11-17</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'disk'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 14 12 14 11 13 15 13 11 17 15 14 13 11 15 17 14 16 17 17 17 11 12 12 12 14</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 14 15 14 14 17 13 16 17 12 15 17 11 17 17 17 12 17 11 11 14 13 17 13 15 15</span>
 <span class="r-in"><span><span class="fu">random_disk</span><span class="op">(</span><span class="fl">25</span>, <span class="st">"Streptococcus pneumoniae"</span>, <span class="st">"ampicillin"</span><span class="op">)</span> <span class="co"># range 12-27</span></span></span>
 <span class="r-out co"><span class="r-pr">#&gt;</span> Class 'disk'</span>
-<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 27 15 17 23 23 18 18 17 15 18 27 15 25 16 21 15 24 23 22 22 15 23 25 23 24</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span>  [1] 26 16 22 23 19 22 19 18 19 23 23 24 22 16 17 15 17 16 18 17 26 16 16 20 22</span>
 <span class="r-in"><span><span class="co"># }</span></span></span>
 </code></pre></div>
     </div>
diff --git a/reference/resistance_predict.html b/reference/resistance_predict.html
index 7d2120b4d..b002dc5df 100644
--- a/reference/resistance_predict.html
+++ b/reference/resistance_predict.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/skewness.html b/reference/skewness.html
index 5e1228130..4b4e78919 100644
--- a/reference/skewness.html
+++ b/reference/skewness.html
@@ -9,7 +9,7 @@ When negative ('left-skewed'): the left tail is longer; the mass of the distribu
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -89,7 +89,7 @@ When negative ('left-skewed'): the left tail is longer; the mass of the distribu
     <div class="section level2">
     <h2 id="ref-examples">Examples<a class="anchor" aria-label="anchor" href="#ref-examples"></a></h2>
     <div class="sourceCode"><pre class="sourceCode r"><code><span class="r-in"><span><span class="fu">skewness</span><span class="op">(</span><span class="fu"><a href="https://rdrr.io/r/stats/Uniform.html" class="external-link">runif</a></span><span class="op">(</span><span class="fl">1000</span><span class="op">)</span><span class="op">)</span></span></span>
-<span class="r-out co"><span class="r-pr">#&gt;</span> [1] -0.0199978</span>
+<span class="r-out co"><span class="r-pr">#&gt;</span> [1] 0.01041273</span>
 </code></pre></div>
     </div>
   </main><aside class="col-md-3"><nav id="toc" aria-label="Table of contents"><h2>On this page</h2>
diff --git a/reference/top_n_microorganisms.html b/reference/top_n_microorganisms.html
index bb3171357..a863088b9 100644
--- a/reference/top_n_microorganisms.html
+++ b/reference/top_n_microorganisms.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/translate.html b/reference/translate.html
index e81c8add5..af18254d2 100644
--- a/reference/translate.html
+++ b/reference/translate.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">AMR (for R)</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9153</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">2.1.1.9154</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/search.json b/search.json
index a8e14c456..88d96ba69 100644
--- a/search.json
+++ b/search.json
@@ -1 +1 @@
-[{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"How to conduct AMR data analysis","text":"Conducting AMR data analysis unfortunately requires -depth knowledge different scientific fields, makes hard right. least, requires: Good questions (always start !) reliable data thorough understanding (clinical) epidemiology, understand clinical epidemiological relevance possible bias results thorough understanding (clinical) microbiology/infectious diseases, understand microorganisms causal infections implications pharmaceutical treatment, well understanding intrinsic acquired microbial resistance Experience data analysis microbiological tests results, understand determination limitations MIC values interpretations SIR values Availability biological taxonomy microorganisms probably normalisation factors pharmaceuticals, defined daily doses (DDD) Available (inter-)national guidelines, profound methods apply course, instantly provide knowledge experience. AMR package, aimed providing (1) tools simplify antimicrobial resistance data cleaning, transformation analysis, (2) methods easily incorporate international guidelines (3) scientifically reliable reference data, including requirements mentioned . AMR package enables standardised reproducible AMR data analysis, application evidence-based rules, determination first isolates, translation various codes microorganisms antimicrobial agents, determination (multi-drug) resistant microorganisms, calculation antimicrobial resistance, prevalence future trends.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"preparation","dir":"Articles","previous_headings":"","what":"Preparation","title":"How to conduct AMR data analysis","text":"tutorial, create fake demonstration data work . can skip Cleaning data already data ready. start analysis, try make structure data generally look like :","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"needed-r-packages","dir":"Articles","previous_headings":"Preparation","what":"Needed R packages","title":"How to conduct AMR data analysis","text":"many uses R, need additional packages AMR data analysis. package works closely together tidyverse packages dplyr ggplot2 RStudio. tidyverse tremendously improves way conduct data science - allows natural way writing syntaxes creating beautiful plots R. also use cleaner package, can used cleaning data creating frequency tables. AMR package contains data set example_isolates_unclean, might look data users extracted laboratory systems: AMR data analysis, like microorganism column contain valid, --date taxonomy, antibiotic columns cleaned SIR values well.","code":"library(dplyr) library(ggplot2) library(AMR)  # (if not yet installed, install with:) # install.packages(c(\"dplyr\", \"ggplot2\", \"AMR\")) example_isolates_unclean #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria      AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <chr>         <chr> <chr> <chr> <chr> #>  1 J3         A        2012-11-21 E. coli       R     I     S     S     #>  2 R7         A        2018-04-03 K. pneumoniae R     I     S     S     #>  3 P3         A        2014-09-19 E. coli       R     S     S     S     #>  4 P10        A        2015-12-10 E. coli       S     I     S     S     #>  5 B7         A        2015-03-02 E. coli       S     S     S     S     #>  6 W3         A        2018-03-31 S. aureus     R     S     R     S     #>  7 J8         A        2016-06-14 E. coli       R     S     S     S     #>  8 M3         A        2015-10-25 E. coli       R     S     S     S     #>  9 J3         A        2019-06-19 E. coli       S     S     S     S     #> 10 G6         A        2015-04-27 S. aureus     S     S     S     S     #> # ℹ 2,990 more rows  # we will use 'our_data' as the data set name for this tutorial our_data <- example_isolates_unclean"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"taxonomy-of-microorganisms","dir":"Articles","previous_headings":"Preparation","what":"Taxonomy of microorganisms","title":"How to conduct AMR data analysis","text":".mo(), users can transform arbitrary microorganism names codes current taxonomy. AMR package contains --date taxonomic data. specific, currently included data retrieved 24 Jun 2024. codes AMR packages come .mo() short, still human readable. importantly, .mo() supports kinds input: first character codes denote taxonomic kingdom, Bacteria (B), Fungi (F), Protozoa (P). AMR package also contain functions directly retrieve taxonomic properties, name, genus, species, family, order, even Gram-stain. start mo_ use .mo() internally, still arbitrary user input can used: Now can thus clean data: Apparently, uncertainty translation taxonomic codes. Let’s check : ’s good.","code":"as.mo(\"Klebsiella pneumoniae\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"K. pneumoniae\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"KLEPNE\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"KLPN\") #> Class 'mo' #> [1] B_KLBSL_PNMN mo_family(\"K. pneumoniae\") #> [1] \"Enterobacteriaceae\" mo_genus(\"K. pneumoniae\") #> [1] \"Klebsiella\" mo_species(\"K. pneumoniae\") #> [1] \"pneumoniae\"  mo_gramstain(\"Klebsiella pneumoniae\") #> [1] \"Gram-negative\"  mo_ref(\"K. pneumoniae\") #> [1] \"Trevisan, 1887\"  mo_snomed(\"K. pneumoniae\") #> [[1]] #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\" our_data$bacteria <- as.mo(our_data$bacteria, info = TRUE) #> ℹ Microorganism translation was uncertain for four microorganisms. Run #>   mo_uncertainties() to review these uncertainties, or use #>   add_custom_microorganisms() to add custom entries. mo_uncertainties() #> Matching scores are based on the resemblance between the input and the full #> taxonomic name, and the pathogenicity in humans. See ?mo_matching_score. #> Colour keys:  0.000-0.549  0.550-0.649  0.650-0.749  0.750-1.000  #>  #> -------------------------------------------------------------------------------- #> \"E. coli\" -> Escherichia coli (B_ESCHR_COLI, 0.688) #> Also matched: Enterococcus crotali (0.650), Escherichia coli coli #>               (0.643), Escherichia coli expressing (0.611), Enterobacter cowanii #>               (0.600), Enterococcus columbae (0.595), Enterococcus camelliae (0.591), #>               Enterococcus casseliflavus (0.577), Enterobacter cloacae cloacae #>               (0.571), Enterobacter cloacae complex (0.571), and Enterobacter cloacae #>               dissolvens (0.565) #> -------------------------------------------------------------------------------- #> \"K. pneumoniae\" -> Klebsiella pneumoniae (B_KLBSL_PNMN, 0.786) #> Also matched: Klebsiella pneumoniae ozaenae (0.707), Klebsiella #>               pneumoniae pneumoniae (0.688), Klebsiella pneumoniae rhinoscleromatis #>               (0.658), Klebsiella pasteurii (0.500), Klebsiella planticola (0.500), #>               Kingella potus (0.400), Kluyveromyces pseudotropicale (0.386), #>               Kluyveromyces pseudotropicalis (0.363), Kosakonia pseudosacchari #>               (0.361), and Kluyveromyces pseudotropicalis pseudotropicalis (0.361) #> -------------------------------------------------------------------------------- #> \"S. aureus\" -> Staphylococcus aureus (B_STPHY_AURS, 0.690) #> Also matched: Staphylococcus aureus aureus (0.643), Staphylococcus #>               argenteus (0.625), Staphylococcus aureus anaerobius (0.625), #>               Staphylococcus auricularis (0.615), Salmonella Aurelianis (0.595), #>               Salmonella Aarhus (0.588), Salmonella Amounderness (0.587), #>               Staphylococcus argensis (0.587), Streptococcus australis (0.587), and #>               Salmonella choleraesuis arizonae (0.562) #> -------------------------------------------------------------------------------- #> \"S. pneumoniae\" -> Streptococcus pneumoniae (B_STRPT_PNMN, 0.750) #> Also matched: Streptococcus pseudopneumoniae (0.700), Streptococcus #>               phocae salmonis (0.552), Serratia proteamaculans quinovora (0.545), #>               Streptococcus pseudoporcinus (0.536), Staphylococcus piscifermentans #>               (0.533), Staphylococcus pseudintermedius (0.532), Serratia #>               proteamaculans proteamaculans (0.526), Streptococcus gallolyticus #>               pasteurianus (0.526), Salmonella Portanigra (0.524), and Streptococcus #>               periodonticum (0.519) #>  #> Only the first 10 other matches of each record are shown. Run #> print(mo_uncertainties(), n = ...) to view more entries, or save #> mo_uncertainties() to an object."},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"antibiotic-results","dir":"Articles","previous_headings":"Preparation","what":"Antibiotic results","title":"How to conduct AMR data analysis","text":"column antibiotic test results must also cleaned. AMR package comes three new data types work test results: mic minimal inhibitory concentrations (MIC), disk disk diffusion diameters, sir SIR data interpreted already. package can also determine SIR values based MIC disk diffusion values, read .sir() page. now, just clean SIR columns data using dplyr: basically cleaning, time start data inclusion.","code":"# method 1, be explicit about the columns: our_data <- our_data %>%   mutate_at(vars(AMX:GEN), as.sir)  # method 2, let the AMR package determine the eligible columns our_data <- our_data %>%   mutate_if(is_sir_eligible, as.sir)  # result: our_data #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> #>  1 J3         A        2012-11-21 B_ESCHR_COLI   R     I     S     S   #>  2 R7         A        2018-04-03 B_KLBSL_PNMN   R     I     S     S   #>  3 P3         A        2014-09-19 B_ESCHR_COLI   R     S     S     S   #>  4 P10        A        2015-12-10 B_ESCHR_COLI   S     I     S     S   #>  5 B7         A        2015-03-02 B_ESCHR_COLI   S     S     S     S   #>  6 W3         A        2018-03-31 B_STPHY_AURS   R     S     R     S   #>  7 J8         A        2016-06-14 B_ESCHR_COLI   R     S     S     S   #>  8 M3         A        2015-10-25 B_ESCHR_COLI   R     S     S     S   #>  9 J3         A        2019-06-19 B_ESCHR_COLI   S     S     S     S   #> 10 G6         A        2015-04-27 B_STPHY_AURS   S     S     S     S   #> # ℹ 2,990 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"first-isolates","dir":"Articles","previous_headings":"Preparation","what":"First isolates","title":"How to conduct AMR data analysis","text":"need know isolates can actually use analysis without repetition bias. conduct analysis antimicrobial resistance, must include first isolate every patient per episode (Hindler et al., Clin Infect Dis. 2007). , easily get overestimate underestimate resistance antibiotic. Imagine patient admitted MRSA found 5 different blood cultures following weeks (yes, countries like Netherlands blood drawing policies). resistance percentage oxacillin isolates overestimated, included MRSA . clearly selection bias. Clinical Laboratory Standards Institute (CLSI) appoints follows: (…) preparing cumulative antibiogram guide clinical decisions empirical antimicrobial therapy initial infections, first isolate given species per patient, per analysis period (eg, one year) included, irrespective body site, antimicrobial susceptibility profile, phenotypical characteristics (eg, biotype). first isolate easily identified, cumulative antimicrobial susceptibility test data prepared using first isolate generally comparable cumulative antimicrobial susceptibility test data calculated methods, providing duplicate isolates excluded. M39-A4 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 4th Edition. CLSI, 2014. Chapter 6.4 AMR package includes methodology first_isolate() function able apply four different methods defined Hindler et al. 2007: phenotype-based, episode-based, patient-based, isolate-based. right method depends goals analysis, default phenotype-based method case method properly correct duplicate isolates. Read methods first_isolate() page. outcome function can easily added data: 91% suitable resistance analysis! can now filter filter() function, also dplyr package: future use, two syntaxes can shortened: end 2 724 isolates analysis. Now data looks like: Time analysis.","code":"our_data <- our_data %>%   mutate(first = first_isolate(info = TRUE)) #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Using column 'bacteria' as input for col_mo. #> ℹ Using column 'date' as input for col_date. #> ℹ Using column 'patient_id' as input for col_patient_id. #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> => Found 2,724 'phenotype-based' first isolates (90.8% of total where a #>    microbial ID was available) our_data_1st <- our_data %>%   filter(first == TRUE) our_data_1st <- our_data %>%   filter_first_isolate() our_data_1st #> # A tibble: 2,724 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 J3         A        2012-11-21 B_ESCHR_COLI   R     I     S     S   TRUE  #>  2 R7         A        2018-04-03 B_KLBSL_PNMN   R     I     S     S   TRUE  #>  3 P3         A        2014-09-19 B_ESCHR_COLI   R     S     S     S   TRUE  #>  4 P10        A        2015-12-10 B_ESCHR_COLI   S     I     S     S   TRUE  #>  5 B7         A        2015-03-02 B_ESCHR_COLI   S     S     S     S   TRUE  #>  6 W3         A        2018-03-31 B_STPHY_AURS   R     S     R     S   TRUE  #>  7 M3         A        2015-10-25 B_ESCHR_COLI   R     S     S     S   TRUE  #>  8 J3         A        2019-06-19 B_ESCHR_COLI   S     S     S     S   TRUE  #>  9 G6         A        2015-04-27 B_STPHY_AURS   S     S     S     S   TRUE  #> 10 P4         A        2011-06-21 B_ESCHR_COLI   S     S     S     S   TRUE  #> # ℹ 2,714 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"analysing-the-data","dir":"Articles","previous_headings":"","what":"Analysing the data","title":"How to conduct AMR data analysis","text":"base R summary() function gives good first impression, comes support new mo sir classes now data set:","code":"summary(our_data_1st) #>   patient_id          hospital              date            #>  Length:2724        Length:2724        Min.   :2011-01-01   #>  Class :character   Class :character   1st Qu.:2013-04-07   #>  Mode  :character   Mode  :character   Median :2015-06-03   #>                                        Mean   :2015-06-09   #>                                        3rd Qu.:2017-08-11   #>                                        Max.   :2019-12-27   #>    bacteria               AMX                    AMC                 #>  Class :mo             Class:sir              Class:sir              #>  <NA>  :0              %S   :41.6% (n=1133)   %S   :52.6% (n=1432)   #>  Unique:4              %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)      #>  #1    :B_ESCHR_COLI   %I   :16.4% (n=446)    %I   :12.2% (n=333)    #>  #2    :B_STPHY_AURS   %R   :42.0% (n=1145)   %R   :35.2% (n=959)    #>  #3    :B_STRPT_PNMN   %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)      #>     CIP                    GEN                  first         #>  Class:sir              Class:sir              Mode:logical   #>  %S   :52.5% (n=1431)   %S   :61.0% (n=1661)   TRUE:2724      #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)                     #>  %I   : 6.5% (n=176)    %I   : 3.0% (n=82)                    #>  %R   :41.0% (n=1117)   %R   :36.0% (n=981)                   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)  glimpse(our_data_1st) #> Rows: 2,724 #> Columns: 9 #> $ patient_id <chr> \"J3\", \"R7\", \"P3\", \"P10\", \"B7\", \"W3\", \"M3\", \"J3\", \"G6\", \"P4\"… #> $ hospital   <chr> \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\",… #> $ date       <date> 2012-11-21, 2018-04-03, 2014-09-19, 2015-12-10, 2015-03-02… #> $ bacteria   <mo> \"B_ESCHR_COLI\", \"B_KLBSL_PNMN\", \"B_ESCHR_COLI\", \"B_ESCHR_COL… #> $ AMX        <sir> R, R, R, S, S, R, R, S, S, S, S, R, S, S, R, R, R, R, S, R,… #> $ AMC        <sir> I, I, S, I, S, S, S, S, S, S, S, S, S, S, S, S, S, R, S, S,… #> $ CIP        <sir> S, S, S, S, S, R, S, S, S, S, S, S, S, S, S, S, S, S, S, S,… #> $ GEN        <sir> S, S, S, S, S, S, S, S, S, S, S, R, S, S, S, S, S, S, S, S,… #> $ first      <lgl> TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE,…  # number of unique values per column: sapply(our_data_1st, n_distinct) #> patient_id   hospital       date   bacteria        AMX        AMC        CIP  #>        260          3       1854          4          3          3          3  #>        GEN      first  #>          3          1"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"availability-of-species","dir":"Articles","previous_headings":"Analysing the data","what":"Availability of species","title":"How to conduct AMR data analysis","text":"just get idea species distributed, create frequency table count() based name microorganisms:","code":"our_data %>%   count(mo_name(bacteria), sort = TRUE) #> # A tibble: 4 × 2 #>   `mo_name(bacteria)`          n #>   <chr>                    <int> #> 1 Escherichia coli          1518 #> 2 Staphylococcus aureus      730 #> 3 Streptococcus pneumoniae   426 #> 4 Klebsiella pneumoniae      326  our_data_1st %>%   count(mo_name(bacteria), sort = TRUE) #> # A tibble: 4 × 2 #>   `mo_name(bacteria)`          n #>   <chr>                    <int> #> 1 Escherichia coli          1321 #> 2 Staphylococcus aureus      682 #> 3 Streptococcus pneumoniae   402 #> 4 Klebsiella pneumoniae      319"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"select-and-filter-with-antibiotic-selectors","dir":"Articles","previous_headings":"Analysing the data","what":"Select and filter with antibiotic selectors","title":"How to conduct AMR data analysis","text":"Using -called antibiotic class selectors, can select filter columns based antibiotic class antibiotic results :","code":"our_data_1st %>%   select(date, aminoglycosides()) #> ℹ For aminoglycosides() using column 'GEN' (gentamicin) #> # A tibble: 2,724 × 2 #>    date       GEN   #>    <date>     <sir> #>  1 2012-11-21   S   #>  2 2018-04-03   S   #>  3 2014-09-19   S   #>  4 2015-12-10   S   #>  5 2015-03-02   S   #>  6 2018-03-31   S   #>  7 2015-10-25   S   #>  8 2019-06-19   S   #>  9 2015-04-27   S   #> 10 2011-06-21   S   #> # ℹ 2,714 more rows  our_data_1st %>%   select(bacteria, betalactams()) #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 2,724 × 3 #>    bacteria     AMX   AMC   #>    <mo>         <sir> <sir> #>  1 B_ESCHR_COLI   R     I   #>  2 B_KLBSL_PNMN   R     I   #>  3 B_ESCHR_COLI   R     S   #>  4 B_ESCHR_COLI   S     I   #>  5 B_ESCHR_COLI   S     S   #>  6 B_STPHY_AURS   R     S   #>  7 B_ESCHR_COLI   R     S   #>  8 B_ESCHR_COLI   S     S   #>  9 B_STPHY_AURS   S     S   #> 10 B_ESCHR_COLI   S     S   #> # ℹ 2,714 more rows  our_data_1st %>%   select(bacteria, where(is.sir)) #> # A tibble: 2,724 × 5 #>    bacteria     AMX   AMC   CIP   GEN   #>    <mo>         <sir> <sir> <sir> <sir> #>  1 B_ESCHR_COLI   R     I     S     S   #>  2 B_KLBSL_PNMN   R     I     S     S   #>  3 B_ESCHR_COLI   R     S     S     S   #>  4 B_ESCHR_COLI   S     I     S     S   #>  5 B_ESCHR_COLI   S     S     S     S   #>  6 B_STPHY_AURS   R     S     R     S   #>  7 B_ESCHR_COLI   R     S     S     S   #>  8 B_ESCHR_COLI   S     S     S     S   #>  9 B_STPHY_AURS   S     S     S     S   #> 10 B_ESCHR_COLI   S     S     S     S   #> # ℹ 2,714 more rows  # filtering using AB selectors is also possible: our_data_1st %>%   filter(any(aminoglycosides() == \"R\")) #> ℹ For aminoglycosides() using column 'GEN' (gentamicin) #> # A tibble: 981 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 J5         A        2017-12-25 B_STRPT_PNMN   R     S     S     R   TRUE  #>  2 X1         A        2017-07-04 B_STPHY_AURS   R     S     S     R   TRUE  #>  3 B3         A        2016-07-24 B_ESCHR_COLI   S     S     S     R   TRUE  #>  4 V7         A        2012-04-03 B_ESCHR_COLI   S     S     S     R   TRUE  #>  5 C9         A        2017-03-23 B_ESCHR_COLI   S     S     S     R   TRUE  #>  6 R1         A        2018-06-10 B_STPHY_AURS   S     S     S     R   TRUE  #>  7 S2         A        2013-07-19 B_STRPT_PNMN   S     S     S     R   TRUE  #>  8 P5         A        2019-03-09 B_STPHY_AURS   S     S     S     R   TRUE  #>  9 Q8         A        2019-08-10 B_STPHY_AURS   S     S     S     R   TRUE  #> 10 K5         A        2013-03-15 B_STRPT_PNMN   S     S     S     R   TRUE  #> # ℹ 971 more rows  our_data_1st %>%   filter(all(betalactams() == \"R\")) #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 462 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 M7         A        2013-07-22 B_STRPT_PNMN   R     R     S     S   TRUE  #>  2 R10        A        2013-12-20 B_STPHY_AURS   R     R     S     S   TRUE  #>  3 R7         A        2015-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  4 R8         A        2019-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  5 B6         A        2016-11-20 B_ESCHR_COLI   R     R     R     R   TRUE  #>  6 I7         A        2015-08-19 B_ESCHR_COLI   R     R     S     S   TRUE  #>  7 N3         A        2014-12-29 B_STRPT_PNMN   R     R     R     S   TRUE  #>  8 Q2         A        2019-09-22 B_ESCHR_COLI   R     R     S     S   TRUE  #>  9 X7         A        2011-03-20 B_ESCHR_COLI   R     R     S     R   TRUE  #> 10 V1         A        2018-08-07 B_STPHY_AURS   R     R     S     S   TRUE  #> # ℹ 452 more rows  # even works in base R (since R 3.0): our_data_1st[all(betalactams() == \"R\"), ] #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 462 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 M7         A        2013-07-22 B_STRPT_PNMN   R     R     S     S   TRUE  #>  2 R10        A        2013-12-20 B_STPHY_AURS   R     R     S     S   TRUE  #>  3 R7         A        2015-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  4 R8         A        2019-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  5 B6         A        2016-11-20 B_ESCHR_COLI   R     R     R     R   TRUE  #>  6 I7         A        2015-08-19 B_ESCHR_COLI   R     R     S     S   TRUE  #>  7 N3         A        2014-12-29 B_STRPT_PNMN   R     R     R     S   TRUE  #>  8 Q2         A        2019-09-22 B_ESCHR_COLI   R     R     S     S   TRUE  #>  9 X7         A        2011-03-20 B_ESCHR_COLI   R     R     S     R   TRUE  #> 10 V1         A        2018-08-07 B_STPHY_AURS   R     R     S     S   TRUE  #> # ℹ 452 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"generate-antibiograms","dir":"Articles","previous_headings":"Analysing the data","what":"Generate antibiograms","title":"How to conduct AMR data analysis","text":"Since AMR v2.0 (March 2023), easy create different types antibiograms, support 20 different languages. four antibiogram types, proposed Klinker et al. (2021, DOI 10.1177/20499361211011373), supported new antibiogram() function: Traditional Antibiogram (TA) e.g, susceptibility Pseudomonas aeruginosa piperacillin/tazobactam (TZP) Combination Antibiogram (CA) e.g, sdditional susceptibility Pseudomonas aeruginosa TZP + tobramycin versus TZP alone Syndromic Antibiogram (SA) e.g, susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) Weighted-Incidence Syndromic Combination Antibiogram (WISCA) e.g, susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) male patients age >=65 years heart failure section, show use antibiogram() function create antibiogram types. starters, included example_isolates data set looks like:","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"traditional-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Traditional Antibiogram","title":"How to conduct AMR data analysis","text":"create traditional antibiogram, simply state antibiotics used. antibiotics argument antibiogram() function supports (combination) previously mentioned antibiotic class selectors: Notice antibiogram() function automatically prints right format using Quarto R Markdown (page), even applies italics taxonomic names (using italicise_taxonomy() internally). also uses language OS either English, Chinese, Czech, Danish, Dutch, Finnish, French, German, Greek, Italian, Japanese, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Turkish, Ukrainian. next example, force language Spanish using language argument:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems())) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) antibiogram(example_isolates,             mo_transform = \"gramstain\",             antibiotics = aminoglycosides(),             ab_transform = \"name\",             language = \"es\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"combined-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Combined Antibiogram","title":"How to conduct AMR data analysis","text":"create combined antibiogram, use antibiotic codes names plus + character like :","code":"antibiogram(example_isolates,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"syndromic-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Syndromic Antibiogram","title":"How to conduct AMR data analysis","text":"create syndromic antibiogram, syndromic_group argument must used. can column data, e.g. ifelse() calculations based certain columns:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems()),             syndromic_group = \"ward\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"weighted-incidence-syndromic-combination-antibiogram-wisca","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Weighted-Incidence Syndromic Combination Antibiogram (WISCA)","title":"How to conduct AMR data analysis","text":"create WISCA, must state combination therapy antibiotics argument (similar Combination Antibiogram), define syndromic group syndromic_group argument (similar Syndromic Antibiogram) cases predefined based clinical demographic characteristics (e.g., endocarditis 75+ females). next example simplification without clinical characteristics, just gives idea WISCA can created:","code":"wisca <- antibiogram(example_isolates,                      antibiotics = c(\"AMC\", \"AMC+CIP\", \"TZP\", \"TZP+TOB\"),                      mo_transform = \"gramstain\",                      minimum = 10, # this should be >= 30, but now just as example                      syndromic_group = ifelse(example_isolates$age >= 65 &                                                 example_isolates$gender == \"M\",                                               \"WISCA Group 1\", \"WISCA Group 2\")) wisca"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"plotting-antibiograms","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Plotting antibiograms","title":"How to conduct AMR data analysis","text":"Antibiograms can plotted using autoplot() ggplot2 packages, since AMR package provides extension function:  calculate antimicrobial resistance sensible way, also correcting results, use resistance() susceptibility() functions.","code":"autoplot(wisca)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"resistance-percentages","dir":"Articles","previous_headings":"Analysing the data","what":"Resistance percentages","title":"How to conduct AMR data analysis","text":"functions resistance() susceptibility() can used calculate antimicrobial resistance susceptibility. specific analyses, functions proportion_S(), proportion_SI(), proportion_I(), proportion_IR() proportion_R() can used determine proportion specific antimicrobial outcome. functions contain minimum argument, denoting minimum required number test results returning value. functions otherwise return NA. default minimum = 30, following CLSI M39-A4 guideline applying microbial epidemiology. per EUCAST guideline 2019, calculate resistance proportion R (proportion_R(), equal resistance()) susceptibility proportion S (proportion_SI(), equal susceptibility()). functions can used : can used conjunction group_by() summarise(), dplyr package: Author: Dr. Matthijs Berends, 26th Feb 2023","code":"our_data_1st %>% resistance(AMX) #> [1] 0.4203377 our_data_1st %>%   group_by(hospital) %>%   summarise(amoxicillin = resistance(AMX)) #> # A tibble: 3 × 2 #>   hospital amoxicillin #>   <chr>          <dbl> #> 1 A              0.340 #> 2 B              0.551 #> 3 C              0.370"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"AMR for Python","text":"AMR package R powerful tool antimicrobial resistance (AMR) analysis. provides extensive features handling microbial antimicrobial data. However, work primarily Python, now intuitive option available: AMR Python Package Index. Python package wrapper round AMR R package. uses rpy2 package internally. Despite need R installed, Python users can now easily work AMR data directly Python code.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"install","dir":"Articles","previous_headings":"","what":"Install","title":"AMR for Python","text":"Since Python package available official Python Package Index, can just run: Make sure R installed. need install AMR R package, installed automatically. Linux: macOS (using Homebrew): Windows, visit CRAN download page download install R.","code":"pip install AMR # Ubuntu / Debian sudo apt install r-base # Fedora: sudo dnf install R # CentOS/RHEL sudo yum install R brew install r"},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"cleaning-taxonomy","dir":"Articles","previous_headings":"Examples of Usage","what":"Cleaning Taxonomy","title":"AMR for Python","text":"’s example demonstrates clean microorganism drug names using AMR Python package:","code":"import pandas as pd import AMR  # Sample data data = {     \"MOs\": ['E. coli', 'ESCCOL', 'esco', 'Esche coli'],     \"Drug\": ['Cipro', 'CIP', 'J01MA02', 'Ciproxin'] } df = pd.DataFrame(data)  # Use AMR functions to clean microorganism and drug names df['MO_clean'] = AMR.mo_name(df['MOs']) df['Drug_clean'] = AMR.ab_name(df['Drug'])  # Display the results print(df)"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"explanation","dir":"Articles","previous_headings":"Examples of Usage > Cleaning Taxonomy","what":"Explanation","title":"AMR for Python","text":"mo_name: function standardises microorganism names. , different variations Escherichia coli (“E. coli”, “ESCCOL”, “esco”, “Esche coli”) converted correct, standardised form, “Escherichia coli”. ab_name: Similarly, function standardises antimicrobial names. different representations ciprofloxacin (e.g., “Cipro”, “CIP”, “J01MA02”, “Ciproxin”) converted standard name, “Ciprofloxacin”.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"calculating-amr","dir":"Articles","previous_headings":"Examples of Usage","what":"Calculating AMR","title":"AMR for Python","text":"","code":"import AMR import pandas as pd  df = AMR.example_isolates result = AMR.resistance(df[\"AMX\"]) print(result) [0.59555556]"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"generating-antibiograms","dir":"Articles","previous_headings":"Examples of Usage","what":"Generating Antibiograms","title":"AMR for Python","text":"One core functions AMR package generating antibiogram, table summarises antimicrobial susceptibility bacterial isolates. ’s can generate antibiogram Python: example, generate antibiogram selecting various antibiotics.","code":"result2a = AMR.antibiogram(df[[\"mo\", \"AMX\", \"CIP\", \"TZP\"]]) print(result2a) result2b = AMR.antibiogram(df[[\"mo\", \"AMX\", \"CIP\", \"TZP\"]], mo_transform = \"gramstain\") print(result2b)"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"taxonomic-data-sets-now-in-python","dir":"Articles","previous_headings":"Examples of Usage","what":"Taxonomic Data Sets Now in Python!","title":"AMR for Python","text":"Python user, might like important data sets AMR R package, microorganisms, antibiotics, clinical_breakpoints, example_isolates, now available regular Python data frames:","code":"AMR.microorganisms AMR.antibiotics"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"conclusion","dir":"Articles","previous_headings":"","what":"Conclusion","title":"AMR for Python","text":"AMR Python package, Python users can now effortlessly call R functions AMR R package. eliminates need complex rpy2 configurations provides clean, easy--use interface antimicrobial resistance analysis. examples provided demonstrate can applied typical workflows, standardising microorganism antimicrobial names calculating resistance. just running import AMR, users can seamlessly integrate robust features R AMR package Python workflows. Whether ’re cleaning data analysing resistance patterns, AMR Python package makes easy work AMR data Python.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"objective","dir":"Articles","previous_headings":"","what":"Objective","title":"AMR with tidymodels","text":"goal build predictive model using tidymodels framework determine Gramstain microorganism based microbial data. : Preprocess data using selector functions aminoglycosides() betalactams(). Define logistic regression model prediction. Use structured tidymodels workflow preprocess, train, evaluate model.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"data-preparation","dir":"Articles","previous_headings":"","what":"Data Preparation","title":"AMR with tidymodels","text":"begin loading required libraries preparing example_isolates dataset AMR package. Explanation: aminoglycosides() betalactams() dynamically select columns antibiotics classes. drop_na() ensures model receives complete cases training.","code":"# Load required libraries library(tidymodels)   # For machine learning workflows, and data manipulation (dplyr, tidyr, ...) #> ── Attaching packages ────────────────────────────────────── tidymodels 1.3.0 ── #> ✔ broom        1.0.7     ✔ recipes      1.1.1 #> ✔ dials        1.4.0     ✔ rsample      1.2.1 #> ✔ dplyr        1.1.4     ✔ tibble       3.2.1 #> ✔ ggplot2      3.5.1     ✔ tidyr        1.3.1 #> ✔ infer        1.0.7     ✔ tune         1.3.0 #> ✔ modeldata    1.4.0     ✔ workflows    1.2.0 #> ✔ parsnip      1.3.0     ✔ workflowsets 1.1.0 #> ✔ purrr        1.0.4     ✔ yardstick    1.3.2 #> ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ── #> ✖ purrr::discard() masks scales::discard() #> ✖ dplyr::filter()  masks stats::filter() #> ✖ dplyr::lag()     masks stats::lag() #> ✖ recipes::step()  masks stats::step() library(AMR)          # For AMR data analysis  # Load the example_isolates dataset data(\"example_isolates\")  # Preloaded dataset with AMR results  # Select relevant columns for prediction data <- example_isolates %>%   # select AB results dynamically   select(mo, aminoglycosides(), betalactams()) %>%   # replace NAs with NI (not-interpretable)    mutate(across(where(is.sir),                  ~replace_na(.x, \"NI\")),           # make factors of SIR columns           across(where(is.sir),                  as.integer),           # get Gramstain of microorganisms           mo = as.factor(mo_gramstain(mo))) %>%   # drop NAs - the ones without a Gramstain (fungi, etc.)   drop_na() #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For betalactams() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'TZP' #>   (piperacillin/tazobactam), 'CZO' (cefazolin), 'FEP' (cefepime), 'CXM' #>   (cefuroxime), 'FOX' (cefoxitin), 'CTX' (cefotaxime), 'CAZ' (ceftazidime), #>   'CRO' (ceftriaxone), 'IPM' (imipenem), and 'MEM' (meropenem)"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"defining-the-workflow","dir":"Articles","previous_headings":"","what":"Defining the Workflow","title":"AMR with tidymodels","text":"now define tidymodels workflow, consists three steps: preprocessing, model specification, fitting.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"preprocessing-with-a-recipe","dir":"Articles","previous_headings":"Defining the Workflow","what":"1. Preprocessing with a Recipe","title":"AMR with tidymodels","text":"create recipe preprocess data modelling. Explanation: recipe(mo ~ ., data = data) take mo column outcome columns predictors. step_corr() removes predictors (.e., antibiotic columns) higher correlation 90%. Notice recipe contains just antibiotic selector functions - need define columns specifically.","code":"# Define the recipe for data preprocessing resistance_recipe <- recipe(mo ~ ., data = data) %>%   step_corr(c(aminoglycosides(), betalactams()), threshold = 0.9) resistance_recipe #>  #> ── Recipe ────────────────────────────────────────────────────────────────────── #>  #> ── Inputs #> Number of variables by role #> outcome:    1 #> predictor: 20 #>  #> ── Operations #> • Correlation filter on: c(aminoglycosides(), betalactams())"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"specifying-the-model","dir":"Articles","previous_headings":"Defining the Workflow","what":"2. Specifying the Model","title":"AMR with tidymodels","text":"define logistic regression model since resistance prediction binary classification task. Explanation: logistic_reg() sets logistic regression model. set_engine(\"glm\") specifies use R’s built-GLM engine.","code":"# Specify a logistic regression model logistic_model <- logistic_reg() %>%   set_engine(\"glm\") # Use the Generalized Linear Model engine logistic_model #> Logistic Regression Model Specification (classification) #>  #> Computational engine: glm"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"building-the-workflow","dir":"Articles","previous_headings":"Defining the Workflow","what":"3. Building the Workflow","title":"AMR with tidymodels","text":"bundle recipe model together workflow, organizes entire modeling process.","code":"# Combine the recipe and model into a workflow resistance_workflow <- workflow() %>%   add_recipe(resistance_recipe) %>% # Add the preprocessing recipe   add_model(logistic_model) # Add the logistic regression model"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"training-and-evaluating-the-model","dir":"Articles","previous_headings":"","what":"Training and Evaluating the Model","title":"AMR with tidymodels","text":"train model, split data training testing sets. , fit workflow training set evaluate performance. Explanation: initial_split() splits data training testing sets. fit() trains workflow training set. Notice fit(), antibiotic selector functions internally called . training, functions called since stored recipe. Next, evaluate model testing data. Explanation: predict() generates predictions testing set. metrics() computes evaluation metrics like accuracy kappa. appears can predict Gram based AMR results 0.995 accuracy based AMR results aminoglycosides beta-lactam antibiotics. ROC curve looks like :","code":"# Split data into training and testing sets set.seed(123) # For reproducibility data_split <- initial_split(data, prop = 0.8) # 80% training, 20% testing training_data <- training(data_split) # Training set testing_data <- testing(data_split)   # Testing set  # Fit the workflow to the training data fitted_workflow <- resistance_workflow %>%   fit(training_data) # Train the model #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For betalactams() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'TZP' #>   (piperacillin/tazobactam), 'CZO' (cefazolin), 'FEP' (cefepime), 'CXM' #>   (cefuroxime), 'FOX' (cefoxitin), 'CTX' (cefotaxime), 'CAZ' (ceftazidime), #>   'CRO' (ceftriaxone), 'IPM' (imipenem), and 'MEM' (meropenem) # Make predictions on the testing set predictions <- fitted_workflow %>%   predict(testing_data)                # Generate predictions probabilities <- fitted_workflow %>%   predict(testing_data, type = \"prob\") # Generate probabilities  predictions <- predictions %>%   bind_cols(probabilities) %>%   bind_cols(testing_data) # Combine with true labels  predictions #> # A tibble: 394 × 24 #>    .pred_class   `.pred_Gram-negative` `.pred_Gram-positive` mo        GEN   TOB #>    <fct>                         <dbl>                 <dbl> <fct>   <int> <int> #>  1 Gram-positive              1.07e- 1              8.93e- 1 Gram-p…     5     5 #>  2 Gram-positive              3.17e- 8              1.00e+ 0 Gram-p…     5     1 #>  3 Gram-negative              9.99e- 1              1.42e- 3 Gram-n…     5     5 #>  4 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     5     5 #>  5 Gram-negative              9.46e- 1              5.42e- 2 Gram-n…     5     5 #>  6 Gram-positive              1.07e- 1              8.93e- 1 Gram-p…     5     5 #>  7 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     1     5 #>  8 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     4     4 #>  9 Gram-negative              1   e+ 0              2.22e-16 Gram-n…     1     1 #> 10 Gram-positive              6.05e-11              1.00e+ 0 Gram-p…     4     4 #> # ℹ 384 more rows #> # ℹ 18 more variables: AMK <int>, KAN <int>, PEN <int>, OXA <int>, FLC <int>, #> #   AMX <int>, AMC <int>, AMP <int>, TZP <int>, CZO <int>, FEP <int>, #> #   CXM <int>, FOX <int>, CTX <int>, CAZ <int>, CRO <int>, IPM <int>, MEM <int>  # Evaluate model performance metrics <- predictions %>%   metrics(truth = mo, estimate = .pred_class) # Calculate performance metrics  metrics #> # A tibble: 2 × 3 #>   .metric  .estimator .estimate #>   <chr>    <chr>          <dbl> #> 1 accuracy binary         0.995 #> 2 kap      binary         0.989 predictions %>%   roc_curve(mo, `.pred_Gram-negative`) %>%   autoplot()"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"conclusion","dir":"Articles","previous_headings":"","what":"Conclusion","title":"AMR with tidymodels","text":"post, demonstrated build machine learning pipeline tidymodels framework AMR package. combining selector functions like aminoglycosides() betalactams() tidymodels, efficiently prepared data, trained model, evaluated performance. workflow extensible antibiotic classes resistance patterns, empowering users analyse AMR data systematically reproducibly.","code":""},{"path":"https://msberends.github.io/AMR/articles/EUCAST.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"How to apply EUCAST rules","text":"EUCAST rules? European Committee Antimicrobial Susceptibility Testing (EUCAST) states website: EUCAST expert rules tabulated collection expert knowledge intrinsic resistances, exceptional resistance phenotypes interpretive rules may applied antimicrobial susceptibility testing order reduce errors make appropriate recommendations reporting particular resistances. Europe, lot medical microbiological laboratories already apply rules (Brown et al., 2015). package features latest insights intrinsic resistance unusual phenotypes (v3.1, 2016). Moreover, eucast_rules() function use purpose can also apply additional rules, like forcing ampicillin = R isolates amoxicillin/clavulanic acid = R.","code":""},{"path":"https://msberends.github.io/AMR/articles/EUCAST.html","id":"examples","dir":"Articles","previous_headings":"","what":"Examples","title":"How to apply EUCAST rules","text":"rules can used discard impossible bug-drug combinations data. example, Klebsiella produces beta-lactamase prevents ampicillin (amoxicillin) working . words, practically every strain Klebsiella resistant ampicillin. Sometimes, laboratory data can still contain strains ampicillin susceptible ampicillin. antibiogram available identification available, antibiogram re-interpreted based identification (namely, Klebsiella). EUCAST expert rules solve , can applied using eucast_rules(): convenient function mo_is_intrinsic_resistant() uses guideline, allows check one specific microorganisms antibiotics: EUCAST rules can used correction, can also used filling known resistance susceptibility based results antimicrobials drugs. process called interpretive reading, basically form imputation, part eucast_rules() function well:","code":"oops <- data.frame(   mo = c(     \"Klebsiella\",     \"Escherichia\"   ),   ampicillin = \"S\" ) oops #>            mo ampicillin #> 1  Klebsiella          S #> 2 Escherichia          S  eucast_rules(oops, info = FALSE) #>            mo ampicillin #> 1  Klebsiella          R #> 2 Escherichia          S mo_is_intrinsic_resistant(   c(\"Klebsiella\", \"Escherichia\"),   \"ampicillin\" ) #> [1]  TRUE FALSE  mo_is_intrinsic_resistant(   \"Klebsiella\",   c(\"ampicillin\", \"kanamycin\") ) #> [1]  TRUE FALSE data <- data.frame(   mo = c(     \"Staphylococcus aureus\",     \"Enterococcus faecalis\",     \"Escherichia coli\",     \"Klebsiella pneumoniae\",     \"Pseudomonas aeruginosa\"   ),   VAN = \"-\", # Vancomycin   AMX = \"-\", # Amoxicillin   COL = \"-\", # Colistin   CAZ = \"-\", # Ceftazidime   CXM = \"-\", # Cefuroxime   PEN = \"S\", # Benzylenicillin   FOX = \"S\", # Cefoxitin   stringsAsFactors = FALSE ) data eucast_rules(data)"},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"type-of-input","dir":"Articles","previous_headings":"","what":"Type of input","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function takes data set input, regular data.frame. tries automatically determine right columns info isolates, name species columns results antimicrobial agents. See help page info set right settings data command ?mdro. WHONET data (data), settings automatically set correctly.","code":""},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"guidelines","dir":"Articles","previous_headings":"","what":"Guidelines","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function support multiple guidelines. can select guideline guideline parameter. Currently supported guidelines (case-insensitive): guideline = \"CMI2012\" (default) Magiorakos AP, Srinivasan et al. “Multidrug-resistant, extensively drug-resistant pandrug-resistant bacteria: international expert proposal interim standard definitions acquired resistance.” Clinical Microbiology Infection (2012) (link) guideline = \"EUCAST3.2\" (simply guideline = \"EUCAST\") European international guideline - EUCAST Expert Rules Version 3.2 “Intrinsic Resistance Unusual Phenotypes” (link) guideline = \"EUCAST3.1\" European international guideline - EUCAST Expert Rules Version 3.1 “Intrinsic Resistance Exceptional Phenotypes Tables” (link) guideline = \"TB\" international guideline multi-drug resistant tuberculosis - World Health Organization “Companion handbook guidelines programmatic management drug-resistant tuberculosis” (link) guideline = \"MRGN\" German national guideline - Mueller et al. (2015) Antimicrobial Resistance Infection Control 4:7. DOI: 10.1186/s13756-015-0047-6 guideline = \"BRMO\" Dutch national guideline - Rijksinstituut voor Volksgezondheid en Milieu “WIP-richtlijn BRMO (Bijzonder Resistente Micro-Organismen) (ZKH)” (link) Please suggest (country-specific) guidelines letting us know: https://github.com/msberends/AMR/issues/new.","code":""},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"custom-guidelines","dir":"Articles","previous_headings":"Guidelines","what":"Custom Guidelines","title":"How to determine multi-drug resistance (MDR)","text":"can also use custom guideline. Custom guidelines can set custom_mdro_guideline() function. great importance custom rules determine MDROs hospital, e.g., rules dependent ward, state contact isolation variables data. familiar case_when() dplyr package, recognise input method set rules. Rules must set using R considers ‘formula notation’: row/isolate matches first rule, value first ~ (case ‘Elderly Type ’) set MDRO value. Otherwise, second rule tried . maximum number rules unlimited. can print rules set console overview. Colours help reading console supports colours. outcome function can used guideline argument mdro() function: rules set (custom object case) exported shared file location using saveRDS() collaborate multiple users. custom rules set imported using readRDS().","code":"custom <- custom_mdro_guideline(   CIP == \"R\" & age > 60 ~ \"Elderly Type A\",   ERY == \"R\" & age > 60 ~ \"Elderly Type B\" ) custom #> A set of custom MDRO rules: #>   1. If CIP is  R  and age is higher than 60 then: Elderly Type A #>   2. If ERY is  R  and age is higher than 60 then: Elderly Type B #>   3. Otherwise: Negative #>  #> Unmatched rows will return NA. #> Results will be of class 'factor', with ordered levels: Negative < Elderly Type A < Elderly Type B x <- mdro(example_isolates, guideline = custom) table(x) #> x #>       Negative Elderly Type A Elderly Type B  #>           1070            198            732"},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"examples","dir":"Articles","previous_headings":"","what":"Examples","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function always returns ordered factor predefined guidelines. example, output default guideline Magiorakos et al. returns factor levels ‘Negative’, ‘MDR’, ‘XDR’ ‘PDR’ order. next example uses example_isolates data set. data set included package contains full antibiograms 2,000 microbial isolates. reflects reality can used practise AMR data analysis. test MDR/XDR/PDR guideline data set, get: Frequency table Class: factor > ordered (numeric) Length: 2,000 Levels: 4: Negative < Multi-drug-resistant (MDR) < Extensively drug-resistant … Available: 1,745 (87.25%, NA: 255 = 12.75%) Unique: 2 another example, create data set determine multi-drug resistant TB: column names automatically verified valid drug names codes, worked exactly way: data set now looks like : can now add interpretation MDR-TB data set. can use: shortcut mdr_tb(): Create frequency table results: Frequency table Class: factor > ordered (numeric) Length: 5,000 Levels: 5: Negative < Mono-resistant < Poly-resistant < Multi-drug-resistant <… Available: 5,000 (100%, NA: 0 = 0%) Unique: 5","code":"library(dplyr) # to support pipes: %>% library(cleaner) # to create frequency tables example_isolates %>%   mdro() %>%   freq() # show frequency table of the result #> Warning: in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) # random_sir() is a helper function to generate # a random vector with values S, I and R my_TB_data <- data.frame(   rifampicin = random_sir(5000),   isoniazid = random_sir(5000),   gatifloxacin = random_sir(5000),   ethambutol = random_sir(5000),   pyrazinamide = random_sir(5000),   moxifloxacin = random_sir(5000),   kanamycin = random_sir(5000) ) my_TB_data <- data.frame(   RIF = random_sir(5000),   INH = random_sir(5000),   GAT = random_sir(5000),   ETH = random_sir(5000),   PZA = random_sir(5000),   MFX = random_sir(5000),   KAN = random_sir(5000) ) head(my_TB_data) #>   rifampicin isoniazid gatifloxacin ethambutol pyrazinamide moxifloxacin #> 1          I         R            S          S            S            S #> 2          S         S            I          R            R            S #> 3          R         I            I          I            R            I #> 4          I         S            S          S            S            S #> 5          I         I            I          S            I            S #> 6          R         S            R          S            I            I #>   kanamycin #> 1         R #> 2         I #> 3         S #> 4         I #> 5         I #> 6         I mdro(my_TB_data, guideline = \"TB\") my_TB_data$mdr <- mdr_tb(my_TB_data) #> ℹ No column found as input for col_mo, assuming all rows contain #>   Mycobacterium tuberculosis. freq(my_TB_data$mdr)"},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"transforming","dir":"Articles","previous_headings":"","what":"Transforming","title":"How to conduct principal component analysis (PCA) for AMR","text":"PCA, need transform AMR data first. example_isolates data set package looks like: Now transform data set resistance percentages per taxonomic order genus:","code":"library(AMR) library(dplyr) glimpse(example_isolates) #> Rows: 2,000 #> Columns: 46 #> $ date    <date> 2002-01-02, 2002-01-03, 2002-01-07, 2002-01-07, 2002-01-13, 2… #> $ patient <chr> \"A77334\", \"A77334\", \"067927\", \"067927\", \"067927\", \"067927\", \"4… #> $ age     <dbl> 65, 65, 45, 45, 45, 45, 78, 78, 45, 79, 67, 67, 71, 71, 75, 50… #> $ gender  <chr> \"F\", \"F\", \"F\", \"F\", \"F\", \"F\", \"M\", \"M\", \"F\", \"F\", \"M\", \"M\", \"M… #> $ ward    <chr> \"Clinical\", \"Clinical\", \"ICU\", \"ICU\", \"ICU\", \"ICU\", \"Clinical\"… #> $ mo      <mo> \"B_ESCHR_COLI\", \"B_ESCHR_COLI\", \"B_STPHY_EPDR\", \"B_STPHY_EPDR\",… #> $ PEN     <sir> R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, S,… #> $ OXA     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ FLC     <sir> NA, NA, R, R, R, R, S, S, R, S, S, S, NA, NA, NA, NA, NA, R, R… #> $ AMX     <sir> NA, NA, NA, NA, NA, NA, R, R, NA, NA, NA, NA, NA, NA, R, NA, N… #> $ AMC     <sir> I, I, NA, NA, NA, NA, S, S, NA, NA, S, S, I, I, R, I, I, NA, N… #> $ AMP     <sir> NA, NA, NA, NA, NA, NA, R, R, NA, NA, NA, NA, NA, NA, R, NA, N… #> $ TZP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CZO     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, NA,… #> $ FEP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CXM     <sir> I, I, R, R, R, R, S, S, R, S, S, S, S, S, NA, S, S, R, R, S, S… #> $ FOX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, NA,… #> $ CTX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ CAZ     <sir> NA, NA, R, R, R, R, R, R, R, R, R, R, NA, NA, NA, S, S, R, R, … #> $ CRO     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ GEN     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ TOB     <sir> NA, NA, NA, NA, NA, NA, S, S, NA, NA, NA, NA, S, S, NA, NA, NA… #> $ AMK     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ KAN     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ TMP     <sir> R, R, S, S, R, R, R, R, S, S, NA, NA, S, S, S, S, S, R, R, R, … #> $ SXT     <sir> R, R, S, S, NA, NA, NA, NA, S, S, NA, NA, S, S, S, S, S, NA, N… #> $ NIT     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R,… #> $ FOS     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ LNZ     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… #> $ CIP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, NA, NA, NA, NA, S, S… #> $ MFX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ VAN     <sir> R, R, S, S, S, S, S, S, S, S, NA, NA, R, R, R, R, R, S, S, S, … #> $ TEC     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… #> $ TCY     <sir> R, R, S, S, S, S, S, S, S, I, S, S, NA, NA, I, R, R, S, I, R, … #> $ TGC     <sir> NA, NA, S, S, S, S, S, S, S, NA, S, S, NA, NA, NA, R, R, S, NA… #> $ DOX     <sir> NA, NA, S, S, S, S, S, S, S, NA, S, S, NA, NA, NA, R, R, S, NA… #> $ ERY     <sir> R, R, R, R, R, R, S, S, R, S, S, S, R, R, R, R, R, R, R, R, S,… #> $ CLI     <sir> R, R, NA, NA, NA, R, NA, NA, NA, NA, NA, NA, R, R, R, R, R, NA… #> $ AZM     <sir> R, R, R, R, R, R, S, S, R, S, S, S, R, R, R, R, R, R, R, R, S,… #> $ IPM     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ MEM     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ MTR     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CHL     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ COL     <sir> NA, NA, R, R, R, R, R, R, R, R, R, R, NA, NA, NA, R, R, R, R, … #> $ MUP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ RIF     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… resistance_data <- example_isolates %>%   group_by(     order = mo_order(mo), # group on anything, like order     genus = mo_genus(mo)   ) %>% #  and genus as we do here   summarise_if(is.sir, resistance) %>% # then get resistance of all drugs   select(     order, genus, AMC, CXM, CTX,     CAZ, GEN, TOB, TMP, SXT   ) # and select only relevant columns  head(resistance_data) #> # A tibble: 6 × 10 #> # Groups:   order [5] #>   order             genus          AMC   CXM   CTX   CAZ   GEN   TOB   TMP   SXT #>   <chr>             <chr>        <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 (unknown order)   (unknown ge…    NA    NA    NA    NA    NA    NA    NA    NA #> 2 Actinomycetales   Schaalia        NA    NA    NA    NA    NA    NA    NA    NA #> 3 Bacteroidales     Bacteroides     NA    NA    NA    NA    NA    NA    NA    NA #> 4 Campylobacterales Campylobact…    NA    NA    NA    NA    NA    NA    NA    NA #> 5 Caryophanales     Gemella         NA    NA    NA    NA    NA    NA    NA    NA #> 6 Caryophanales     Listeria        NA    NA    NA    NA    NA    NA    NA    NA"},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"perform-principal-component-analysis","dir":"Articles","previous_headings":"","what":"Perform principal component analysis","title":"How to conduct principal component analysis (PCA) for AMR","text":"new pca() function automatically filter rows contain numeric values selected variables, now need : result can reviewed good old summary() function: Good news. first two components explain total 93.3% variance (see PC1 PC2 values Proportion Variance. can create -called biplot base R biplot() function, see antimicrobial resistance per drug explain difference per microorganism.","code":"pca_result <- pca(resistance_data) #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. summary(pca_result) #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\" #> Importance of components: #>                           PC1    PC2    PC3     PC4     PC5     PC6       PC7 #> Standard deviation     2.1539 1.6807 0.6138 0.33879 0.20808 0.03140 1.232e-16 #> Proportion of Variance 0.5799 0.3531 0.0471 0.01435 0.00541 0.00012 0.000e+00 #> Cumulative Proportion  0.5799 0.9330 0.9801 0.99446 0.99988 1.00000 1.000e+00 #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\""},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"plotting-the-results","dir":"Articles","previous_headings":"","what":"Plotting the results","title":"How to conduct principal component analysis (PCA) for AMR","text":"can’t see explanation points. Perhaps works better new ggplot_pca() function, automatically adds right labels even groups:  can also print ellipse per group, edit appearance:","code":"biplot(pca_result) ggplot_pca(pca_result) ggplot_pca(pca_result, ellipse = TRUE) +   ggplot2::labs(title = \"An AMR/PCA biplot!\")"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"import-of-data","dir":"Articles","previous_headings":"","what":"Import of data","title":"How to work with WHONET data","text":"tutorial assumes already imported WHONET data e.g. readxl package. RStudio, can done using menu button ‘Import Dataset’ tab ‘Environment’. Choose option ‘Excel’ select exported file. Make sure date fields imported correctly. example syntax look like : package comes example data set WHONET. use analysis.","code":"library(readxl) data <- read_excel(path = \"path/to/your/file.xlsx\")"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"preparation","dir":"Articles","previous_headings":"","what":"Preparation","title":"How to work with WHONET data","text":"First, load relevant packages yet . use tidyverse analyses. . don’t know yet, suggest read website: https://www.tidyverse.org/. transform variables simplify automate analysis: Microorganisms transformed microorganism codes (called mo) using Catalogue Life reference data set, contains ~70,000 microorganisms taxonomic kingdoms Bacteria, Fungi Protozoa. tranformation .mo(). function also recognises almost WHONET abbreviations microorganisms. Antimicrobial results interpretations clean valid. words, contain values \"S\", \"\" \"R\". exactly .sir() function . errors warnings, values transformed succesfully. also created package dedicated data cleaning checking, called cleaner package. freq() function can used create frequency tables. let’s check data, couple frequency tables: Frequency table Class: character Length: 500 Available: 500 (100%, NA: 0 = 0%) Unique: 38 Shortest: 11 Longest: 40 (omitted 28 entries, n = 57 [11.4%]) Frequency table Class: factor > ordered > sir (numeric) Length: 500 Levels: 5: S < SDD < < R < NI Available: 481 (96.2%, NA: 19 = 3.8%) Unique: 3 Drug: Amoxicillin/clavulanic acid (AMC, J01CR02) Drug group: Beta-lactams/penicillins %SI: 78.59%","code":"library(dplyr) # part of tidyverse library(ggplot2) # part of tidyverse library(AMR) # this package library(cleaner) # to create frequency tables # transform variables data <- WHONET %>%   # get microbial ID based on given organism   mutate(mo = as.mo(Organism)) %>%   # transform everything from \"AMP_ND10\" to \"CIP_EE\" to the new `sir` class   mutate_at(vars(AMP_ND10:CIP_EE), as.sir) # our newly created `mo` variable, put in the mo_name() function data %>% freq(mo_name(mo), nmax = 10) # our transformed antibiotic columns # amoxicillin/clavulanic acid (J01CR02) as an example data %>% freq(AMC_ND2)"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"a-first-glimpse-at-results","dir":"Articles","previous_headings":"","what":"A first glimpse at results","title":"How to work with WHONET data","text":"easy ggplot already give lot information, using included ggplot_sir() function:","code":"data %>%   group_by(Country) %>%   select(Country, AMP_ND2, AMC_ED20, CAZ_ED10, CIP_ED5) %>%   ggplot_sir(translate_ab = \"ab\", facet = \"Country\", datalabels = FALSE)"},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-full-microbial-taxonomy","dir":"Articles","previous_headings":"","what":"microorganisms: Full Microbial Taxonomy","title":"Data sets for download / own use","text":"data set 78 678 rows 26 columns, containing following column names:mo, fullname, status, kingdom, phylum, class, order, family, genus, species, subspecies, rank, ref, oxygen_tolerance, source, lpsn, lpsn_parent, lpsn_renamed_to, mycobank, mycobank_parent, mycobank_renamed_to, gbif, gbif_parent, gbif_renamed_to, prevalence, snomed. data set R available microorganisms, load AMR package. last updated 4 October 2024 13:28:44 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (1.8 MB) Download tab-separated text file (17.7 MB) Download Microsoft Excel workbook (8.7 MB) Download Apache Feather file (8.3 MB) Download Apache Parquet file (3.8 MB) Download IBM SPSS Statistics data file (29 MB) Download Stata DTA file (92.5 MB) NOTE: exported files SPSS Stata contain first 50 SNOMED codes per record, file size otherwise exceed 100 MB; file size limit GitHub. file structures compression techniques inefficient. Advice? Use R instead. ’s free much better many ways. tab-separated text file Microsoft Excel workbook contain SNOMED codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source","dir":"Articles","previous_headings":"microorganisms: Full Microbial Taxonomy","what":"Source","title":"Data sets for download / own use","text":"data set contains full microbial taxonomy six kingdoms List Prokaryotic names Standing Nomenclature (LPSN), MycoBank, Global Biodiversity Information Facility (GBIF): Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name ‘Microorganism’, OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example-content","dir":"Articles","previous_headings":"microorganisms: Full Microbial Taxonomy","what":"Example content","title":"Data sets for download / own use","text":"Included (sub)species per taxonomic kingdom: Example rows filtering genus Escherichia:","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"antibiotics-antibiotic-antifungal-drugs","dir":"Articles","previous_headings":"","what":"antibiotics: Antibiotic (+Antifungal) Drugs","title":"Data sets for download / own use","text":"data set 487 rows 14 columns, containing following column names:ab, cid, name, group, atc, atc_group1, atc_group2, abbreviations, synonyms, oral_ddd, oral_units, iv_ddd, iv_units, loinc. data set R available antibiotics, load AMR package. last updated 7 February 2025 17:01:22 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (42 kB) Download tab-separated text file (0.1 MB) Download Microsoft Excel workbook (74 kB) Download Apache Feather file (0.1 MB) Download Apache Parquet file (0.1 MB) Download IBM SPSS Statistics data file (0.4 MB) Download Stata DTA file (0.4 MB) tab-separated text, Microsoft Excel, SPSS, Stata files contain ATC codes, common abbreviations, trade names LOINC codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-1","dir":"Articles","previous_headings":"antibiotics: Antibiotic (+Antifungal) Drugs","what":"Source","title":"Data sets for download / own use","text":"data set contains EARS-Net ATC codes gathered WHONET, compound IDs PubChem. also contains brand names (synonyms) found PubChem Defined Daily Doses (DDDs) oral parenteral administration. ATC/DDD index Collaborating Centre Drug Statistics Methodology (note: may used commercial purposes, freely available CC website personal use) PubChem US National Library Medicine WHONET software 2019 LOINC (Logical Observation Identifiers Names Codes)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"antivirals-antiviral-drugs","dir":"Articles","previous_headings":"","what":"antivirals: Antiviral Drugs","title":"Data sets for download / own use","text":"data set 120 rows 11 columns, containing following column names:av, name, atc, cid, atc_group, synonyms, oral_ddd, oral_units, iv_ddd, iv_units, loinc. data set R available antivirals, load AMR package. last updated 20 October 2023 12:51:48 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (6 kB) Download tab-separated text file (17 kB) Download Microsoft Excel workbook (16 kB) Download Apache Feather file (16 kB) Download Apache Parquet file (13 kB) Download IBM SPSS Statistics data file (32 kB) Download Stata DTA file (78 kB) tab-separated text, Microsoft Excel, SPSS, Stata files contain trade names LOINC codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-2","dir":"Articles","previous_headings":"antivirals: Antiviral Drugs","what":"Source","title":"Data sets for download / own use","text":"data set contains ATC codes gathered compound IDs PubChem. also contains brand names (synonyms) found PubChem Defined Daily Doses (DDDs) oral parenteral administration. ATC/DDD index Collaborating Centre Drug Statistics Methodology (note: may used commercial purposes, freely available CC website personal use) PubChem US National Library Medicine LOINC (Logical Observation Identifiers Names Codes)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"clinical_breakpoints-interpretation-from-mic-values-disk-diameters-to-sir","dir":"Articles","previous_headings":"","what":"clinical_breakpoints: Interpretation from MIC values & disk diameters to SIR","title":"Data sets for download / own use","text":"data set 34 063 rows 14 columns, containing following column names:guideline, type, host, method, site, mo, rank_index, ab, ref_tbl, disk_dose, breakpoint_S, breakpoint_R, uti, is_SDD. data set R available clinical_breakpoints, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (70 kB) Download tab-separated text file (3.1 MB) Download Microsoft Excel workbook (2 MB) Download Apache Feather file (1.5 MB) Download Apache Parquet file (0.1 MB) Download IBM SPSS Statistics data file (5.6 MB) Download Stata DTA file (9.3 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-3","dir":"Articles","previous_headings":"clinical_breakpoints: Interpretation from MIC values & disk diameters to SIR","what":"Source","title":"Data sets for download / own use","text":"data set contains interpretation rules MIC values disk diffusion diameters. Included guidelines CLSI (2011-2024) EUCAST (2011-2024). Clinical breakpoints package validated imported WHONET, free desktop Windows application developed supported Collaborating Centre Surveillance Antimicrobial Resistance. can read website. developers WHONET AMR package contact sharing work. highly appreciate development WHONET software. CEO CLSI chairman EUCAST endorsed work public use AMR package (consequently use breakpoints) June 2023, future development distributing clinical breakpoints discussed meeting CLSI, EUCAST, , developers WHONET AMR package. NOTE: AMR package (WHONET software well) contains internal methods apply guidelines, rather complex. example, breakpoints must applied certain species groups (case package available microorganisms.groups data set). important considered using breakpoints use.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"intrinsic_resistant-intrinsic-bacterial-resistance","dir":"Articles","previous_headings":"","what":"intrinsic_resistant: Intrinsic Bacterial Resistance","title":"Data sets for download / own use","text":"data set 301 583 rows 2 columns, containing following column names:mo ab. data set R available intrinsic_resistant, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (0.1 MB) Download tab-separated text file (10.9 MB) Download Microsoft Excel workbook (3 MB) Download Apache Feather file (2.5 MB) Download Apache Parquet file (0.3 MB) Download IBM SPSS Statistics data file (16.2 MB) Download Stata DTA file (25 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-4","dir":"Articles","previous_headings":"intrinsic_resistant: Intrinsic Bacterial Resistance","what":"Source","title":"Data sets for download / own use","text":"data set contains defined intrinsic resistance EUCAST bug-drug combinations, based ‘EUCAST Expert Rules’ ‘EUCAST Intrinsic Resistance Unusual Phenotypes’ v3.3 (2021).","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example-content-4","dir":"Articles","previous_headings":"intrinsic_resistant: Intrinsic Bacterial Resistance","what":"Example content","title":"Data sets for download / own use","text":"Example rows filtering Enterobacter cloacae:","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"dosage-dosage-guidelines-from-eucast","dir":"Articles","previous_headings":"","what":"dosage: Dosage Guidelines from EUCAST","title":"Data sets for download / own use","text":"data set 503 rows 9 columns, containing following column names:ab, name, type, dose, dose_times, administration, notes, original_txt, eucast_version. data set R available dosage, load AMR package. last updated 22 June 2023 13:10:59 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (3 kB) Download tab-separated text file (43 kB) Download Microsoft Excel workbook (25 kB) Download Apache Feather file (21 kB) Download Apache Parquet file (9 kB) Download IBM SPSS Statistics data file (64 kB) Download Stata DTA file (0.1 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-5","dir":"Articles","previous_headings":"dosage: Dosage Guidelines from EUCAST","what":"Source","title":"Data sets for download / own use","text":"EUCAST breakpoints used package based dosages data set. Currently included dosages data set meant :  (), ‘EUCAST Clinical Breakpoint Tables’ v11.0 (2021), ‘EUCAST Clinical Breakpoint Tables’ v12.0 (2022).","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example_isolates-example-data-for-practice","dir":"Articles","previous_headings":"","what":"example_isolates: Example Data for Practice","title":"Data sets for download / own use","text":"data set 2 000 rows 46 columns, containing following column names:date, patient, age, gender, ward, mo, PEN, OXA, FLC, AMX, AMC, AMP, TZP, CZO, FEP, CXM, FOX, CTX, CAZ, CRO, GEN, TOB, AMK, KAN, TMP, SXT, NIT, FOS, LNZ, CIP, MFX, VAN, TEC, TCY, TGC, DOX, ERY, CLI, AZM, IPM, MEM, MTR, CHL, COL, MUP, RIF. data set R available example_isolates, load AMR package. last updated 15 June 2024 13:33:49 UTC. Find info structure data set .","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-6","dir":"Articles","previous_headings":"example_isolates: Example Data for Practice","what":"Source","title":"Data sets for download / own use","text":"data set contains randomised fictitious data, reflects reality can used practise AMR data analysis.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example_isolates_unclean-example-data-for-practice","dir":"Articles","previous_headings":"","what":"example_isolates_unclean: Example Data for Practice","title":"Data sets for download / own use","text":"data set 3 000 rows 8 columns, containing following column names:patient_id, hospital, date, bacteria, AMX, AMC, CIP, GEN. data set R available example_isolates_unclean, load AMR package. last updated 27 August 2022 18:49:37 UTC. Find info structure data set .","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-7","dir":"Articles","previous_headings":"example_isolates_unclean: Example Data for Practice","what":"Source","title":"Data sets for download / own use","text":"data set contains randomised fictitious data, reflects reality can used practise AMR data analysis.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-groups-species-groups-and-microbiological-complexes","dir":"Articles","previous_headings":"","what":"microorganisms.groups: Species Groups and Microbiological Complexes","title":"Data sets for download / own use","text":"data set 521 rows 4 columns, containing following column names:mo_group, mo, mo_group_name, mo_name. data set R available microorganisms.groups, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (5 kB) Download tab-separated text file (49 kB) Download Microsoft Excel workbook (19 kB) Download Apache Feather file (19 kB) Download Apache Parquet file (13 kB) Download IBM SPSS Statistics data file (63 kB) Download Stata DTA file (81 kB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-8","dir":"Articles","previous_headings":"microorganisms.groups: Species Groups and Microbiological Complexes","what":"Source","title":"Data sets for download / own use","text":"data set contains species groups microbiological complexes, used clinical_breakpoints data set.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-codes-common-laboratory-codes","dir":"Articles","previous_headings":"","what":"microorganisms.codes: Common Laboratory Codes","title":"Data sets for download / own use","text":"data set 4 971 rows 2 columns, containing following column names:code mo. data set R available microorganisms.codes, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (22 kB) Download tab-separated text file (0.1 MB) Download Microsoft Excel workbook (82 kB) Download Apache Feather file (85 kB) Download Apache Parquet file (56 kB) Download IBM SPSS Statistics data file (0.1 MB) Download Stata DTA file (0.1 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-9","dir":"Articles","previous_headings":"microorganisms.codes: Common Laboratory Codes","what":"Source","title":"Data sets for download / own use","text":"data set contains commonly used codes microorganisms, laboratory systems WHONET.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"needed-r-packages","dir":"Articles","previous_headings":"","what":"Needed R packages","title":"How to predict antimicrobial resistance","text":"many uses R, need additional packages AMR data analysis. package works closely together tidyverse packages dplyr ggplot2. tidyverse tremendously improves way conduct data science - allows natural way writing syntaxes creating beautiful plots R. AMR package depends packages even extends use functions.","code":"library(dplyr) library(ggplot2) library(AMR)  # (if not yet installed, install with:) # install.packages(c(\"tidyverse\", \"AMR\"))"},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"prediction-analysis","dir":"Articles","previous_headings":"","what":"Prediction analysis","title":"How to predict antimicrobial resistance","text":"package contains function resistance_predict(), takes input functions AMR data analysis. Based date column, calculates cases per year uses regression model predict antimicrobial resistance. basically easy : function look date column col_date set. running commands, summary regression model printed unless using resistance_predict(..., info = FALSE). text printed summary - actual result (output) function data.frame containing year: number observations, actual observed resistance, estimated resistance standard error estimation: function plot available base R, can extended packages depend output based type input. extended function cope resistance predictions:  fastest way plot result. automatically adds right axes, error bars, titles, number available observations type model. also support ggplot2 package custom function ggplot_sir_predict() create appealing plots:","code":"# resistance prediction of piperacillin/tazobactam (TZP): resistance_predict(tbl = example_isolates, col_date = \"date\", col_ab = \"TZP\", model = \"binomial\")  # or: example_isolates %>%   resistance_predict(     col_ab = \"TZP\",     model = \"binomial\"   )  # to bind it to object 'predict_TZP' for example: predict_TZP <- example_isolates %>%   resistance_predict(     col_ab = \"TZP\",     model = \"binomial\"   ) predict_TZP #> # A tibble: 34 × 7 #>     year  value se_min se_max observations observed estimated #>  * <dbl>  <dbl>  <dbl>  <dbl>        <int>    <dbl>     <dbl> #>  1  2002 0.2        NA     NA           15   0.2       0.0562 #>  2  2003 0.0625     NA     NA           32   0.0625    0.0616 #>  3  2004 0.0854     NA     NA           82   0.0854    0.0676 #>  4  2005 0.05       NA     NA           60   0.05      0.0741 #>  5  2006 0.0508     NA     NA           59   0.0508    0.0812 #>  6  2007 0.121      NA     NA           66   0.121     0.0889 #>  7  2008 0.0417     NA     NA           72   0.0417    0.0972 #>  8  2009 0.0164     NA     NA           61   0.0164    0.106  #>  9  2010 0.0566     NA     NA           53   0.0566    0.116  #> 10  2011 0.183      NA     NA           93   0.183     0.127  #> # ℹ 24 more rows plot(predict_TZP) ggplot_sir_predict(predict_TZP) # choose for error bars instead of a ribbon ggplot_sir_predict(predict_TZP, ribbon = FALSE)"},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"choosing-the-right-model","dir":"Articles","previous_headings":"Prediction analysis","what":"Choosing the right model","title":"How to predict antimicrobial resistance","text":"Resistance easily predicted; look vancomycin resistance Gram-positive bacteria, spread (.e. standard error) enormous:  Vancomycin resistance 100% ten years, might remain low. can define model model parameter. model chosen generalised linear regression model using binomial distribution, assuming period zero resistance followed period increasing resistance leading slowly resistance. Valid values : vancomycin resistance Gram-positive bacteria, linear model might appropriate:  model also available object, attribute:","code":"example_isolates %>%   filter(mo_gramstain(mo, language = NULL) == \"Gram-positive\") %>%   resistance_predict(col_ab = \"VAN\", year_min = 2010, info = FALSE, model = \"binomial\") %>%   ggplot_sir_predict() example_isolates %>%   filter(mo_gramstain(mo, language = NULL) == \"Gram-positive\") %>%   resistance_predict(col_ab = \"VAN\", year_min = 2010, info = FALSE, model = \"linear\") %>%   ggplot_sir_predict() model <- attributes(predict_TZP)$model  summary(model)$family #>  #> Family: binomial  #> Link function: logit  summary(model)$coefficients #>                  Estimate  Std. Error   z value     Pr(>|z|) #> (Intercept) -200.67944891 46.17315349 -4.346237 1.384932e-05 #> year           0.09883005  0.02295317  4.305725 1.664395e-05"},{"path":"https://msberends.github.io/AMR/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Matthijs S. Berends. Author, maintainer. Dennis Souverein. Author, contributor. Erwin E. . Hassing. Author, contributor. Andrew P. Norgan. Contributor. Anita Williams. Contributor. Annick Lenglet. Contributor. Anthony Underwood. Contributor. Anton Mymrikov. Contributor. Bart C. Meijer. Contributor. Christian F. Luz. Contributor. Dmytro Mykhailenko. Contributor. Eric H. L. C. M. Hazenberg. Contributor. Gwen Knight. Contributor. Jason Stull. Contributor. Javier Sanchez. Contributor. Jonas Salm. Contributor. Judith M. Fonville. Contributor. Larisse Bolton. Contributor. Matthew Saab. Contributor. Peter Dutey-Magni. Contributor. Rogier P. Schade. Contributor. Sofia Ny. Contributor. Alex W. Friedrich. Thesis advisor. Bhanu N. M. Sinha. Thesis advisor. Casper J. Albers. Thesis advisor. Corinna Glasner. Thesis advisor.","code":""},{"path":"https://msberends.github.io/AMR/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Berends MS, Luz CF, Friedrich AW, Sinha BNM, Albers CJ, Glasner C (2022). “AMR: R Package Working Antimicrobial Resistance Data.” Journal Statistical Software, 104(3), 1–31. doi:10.18637/jss.v104.i03.","code":"@Article{,   title = {{AMR}: An {R} Package for Working with Antimicrobial Resistance Data},   author = {Matthijs S. Berends and Christian F. Luz and Alexander W. Friedrich and Bhanu N. M. Sinha and Casper J. Albers and Corinna Glasner},   journal = {Journal of Statistical Software},   year = {2022},   volume = {104},   number = {3},   pages = {1--31},   doi = {10.18637/jss.v104.i03}, }"},{"path":"https://msberends.github.io/AMR/index.html","id":"the-amr-package-for-r-","dir":"","previous_headings":"","what":"Antimicrobial Resistance Data Analysis","title":"Antimicrobial Resistance Data Analysis","text":"Provides --one solution antimicrobial resistance (AMR) data analysis One Health approach Used 175 countries, available 20 languages Generates antibiograms - traditional, combined, syndromic, even WISCA Provides full microbiological taxonomy extensive info antimicrobial drugs Applies recent CLSI EUCAST clinical veterinary breakpoints MICs, disk zones ECOFFs Corrects duplicate isolates, calculates predicts AMR per antimicrobial class Integrates WHONET, ATC, EARS-Net, PubChem, LOINC, SNOMED CT, NCBI 100% free costs dependencies, highly suitable places limited resources Now available Python ! Click read . https://msberends.github.io/AMR https://doi.org/10.18637/jss.v104.i03","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"introduction","dir":"","previous_headings":"","what":"Introduction","title":"Antimicrobial Resistance Data Analysis","text":"AMR package free open-source R package zero dependencies simplify analysis prediction Antimicrobial Resistance (AMR) work microbial antimicrobial data properties, using evidence-based methods. aim provide standard clean reproducible AMR data analysis, can therefore empower epidemiological analyses continuously enable surveillance treatment evaluation setting. Many different researchers around globe continually helping us make successful durable project! work published Journal Statistical Software (Volume 104(3); DOI 10.18637/jss.v104.i03) formed basis two PhD theses (DOI 10.33612/diss.177417131 DOI 10.33612/diss.192486375). installing package, R knows ~52,000 distinct microbial species (updated December 2022) ~600 antibiotic, antimycotic antiviral drugs name code (including ATC, EARS-Net, ASIARS-Net, PubChem, LOINC SNOMED CT), knows valid SIR MIC values. integral clinical breakpoint guidelines CLSI EUCAST included, even epidemiological cut-(ECOFF) values. supports can read data format, including WHONET data. package works Windows, macOS Linux versions R since R-3.0 (April 2013). designed work setting, including limited resources. created routine data analysis academic research Faculty Medical Sciences University Groningen, collaboration non-profit organisations Certe Medical Diagnostics Advice Foundation University Medical Center Groningen.","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"used-in-over-175-countries-available-in-20-languages","dir":"","previous_headings":"Introduction","what":"Used in over 175 countries, available in 20 languages","title":"Antimicrobial Resistance Data Analysis","text":"Since first public release early 2018, R package used almost countries world. Click map enlarge see country names. help contributors corners world, AMR package available  English,  Czech,  Chinese,  Danish,  Dutch,  Finnish,  French,  German,  Greek,  Italian,  Japanese,  Norwegian,  Polish,  Portuguese,  Romanian,  Russian,  Spanish,  Swedish,  Turkish,  Ukrainian. Antimicrobial drug (group) names colloquial microorganism names provided languages.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/index.html","id":"filtering-and-selecting-data","dir":"","previous_headings":"Practical examples","what":"Filtering and selecting data","title":"Antimicrobial Resistance Data Analysis","text":"One powerful functions package, aside calculating plotting AMR, selecting filtering based antimicrobial columns. can done using -called antimicrobial selectors, work base R, dplyr data.table. defined row filter Gram-negative bacteria intrinsic resistance cefotaxime (mo_is_gram_negative() mo_is_intrinsic_resistant()) column selection two antibiotic groups (aminoglycosides() carbapenems()), reference data microorganisms antibiotics AMR package make sure get meant:","code":"# AMR works great with dplyr, but it's not required or neccesary library(AMR) library(dplyr)  example_isolates %>%   mutate(bacteria = mo_fullname()) %>%   # filtering functions for microorganisms:   filter(mo_is_gram_negative(),          mo_is_intrinsic_resistant(ab = \"cefotax\")) %>%   # antimicrobial selectors:   select(bacteria,          aminoglycosides(),          carbapenems())"},{"path":"https://msberends.github.io/AMR/index.html","id":"generating-antibiograms","dir":"","previous_headings":"Practical examples","what":"Generating antibiograms","title":"Antimicrobial Resistance Data Analysis","text":"AMR package supports generating traditional, combined, syndromic, even weighted-incidence syndromic combination antibiograms (WISCA). used inside R Markdown Quarto, table printed right output format automatically (markdown, LaTeX, HTML, etc.). combination antibiograms, clear combined antibiotics yield higher empiric coverage: Like many functions package, antibiogram() comes support 20 languages often detected automatically based system language:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems())) antibiogram(example_isolates,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),             mo_transform = \"gramstain\") antibiogram(example_isolates,             antibiotics = c(\"cipro\", \"tobra\", \"genta\"), # any arbitrary name or code will work             mo_transform = \"gramstain\",             ab_transform = \"name\",             language = \"uk\") # Ukrainian"},{"path":"https://msberends.github.io/AMR/index.html","id":"interpreting-and-plotting-mic-and-sir-values","dir":"","previous_headings":"Practical examples","what":"Interpreting and plotting MIC and SIR values","title":"Antimicrobial Resistance Data Analysis","text":"AMR package allows interpretation MIC disk diffusion values based CLSI EUCAST. Moreover, ggplot2 package extended new scale functions, allow plotting log2-distributed MIC values SIR values.","code":"library(ggplot2) library(AMR)  # generate some random values some_mic_values <- random_mic(size = 100) some_groups <- sample(LETTERS[1:5], 20, replace = TRUE) interpretation <- as.sir(some_mic_values,                          guideline = \"EUCAST 2024\",                          mo = \"E. coli\", # or any code or name resembling a known species                          ab = \"Cipro\")   # or any code or name resembling an antibiotic  # create the plot ggplot(data.frame(mic = some_mic_values,                   group = some_groups,                   sir = interpretation),        aes(x = group, y = mic, colour = sir)) +   theme_minimal() +   geom_boxplot(fill = NA, colour = \"grey\") +   geom_jitter(width = 0.25) +      # NEW scale function: plot MIC values to x, y, colour or fill   scale_y_mic() +      # NEW scale function: write out S/I/R in any of the 20 supported languages   #                     and set colourblind-friendly colours   scale_colour_sir()"},{"path":"https://msberends.github.io/AMR/index.html","id":"calculating-resistance-per-group","dir":"","previous_headings":"Practical examples","what":"Calculating resistance per group","title":"Antimicrobial Resistance Data Analysis","text":"manual approach, can use resistance susceptibility() function: use antimicrobial selectors select series antibiotic columns:","code":"example_isolates %>%   # group by ward:   group_by(ward) %>%   # calculate AMR using resistance() for gentamicin and tobramycin   # and get their 95% confidence intervals using sir_confidence_interval():   summarise(across(c(GEN, TOB),                    list(total_R = resistance,                         conf_int = function(x) sir_confidence_interval(x, collapse = \"-\")))) library(AMR) library(dplyr)  out <- example_isolates %>%   # group by ward:   group_by(ward) %>%   # calculate AMR using resistance(), over all aminoglycosides and polymyxins:   summarise(across(c(aminoglycosides(), polymyxins()),             resistance)) out # transform the antibiotic columns to names: out %>% set_ab_names() # transform the antibiotic column to ATC codes: out %>% set_ab_names(property = \"atc\")"},{"path":"https://msberends.github.io/AMR/index.html","id":"what-else-can-you-do-with-this-package","dir":"","previous_headings":"","what":"What else can you do with this package?","title":"Antimicrobial Resistance Data Analysis","text":"package intended comprehensive toolbox integrated AMR data analysis. package can used : Reference taxonomy microorganisms, since package contains microbial (sub)species List Prokaryotic names Standing Nomenclature (LPSN) Global Biodiversity Information Facility (GBIF) (manual) Interpreting raw MIC disk diffusion values, based CLSI EUCAST guideline (manual) Retrieving antimicrobial drug names, doses forms administration clinical health care records (manual) Determining first isolates used AMR data analysis (manual) Calculating antimicrobial resistance (tutorial) Determining multi-drug resistance (MDR) / multi-drug resistant organisms (MDRO) (tutorial) Calculating (empirical) susceptibility mono therapy combination therapies (tutorial) Predicting future antimicrobial resistance using regression models (tutorial) Getting properties microorganism (like Gram stain, species, genus family) (manual) Getting properties antimicrobial (like name, code EARS-Net/ATC/LOINC/PubChem, defined daily dose trade name) (manual) Plotting antimicrobial resistance (tutorial) Applying EUCAST expert rules (manual) Getting SNOMED codes microorganism, getting properties microorganism based SNOMED code (manual) Getting LOINC codes antibiotic, getting properties antibiotic based LOINC code (manual) Machine reading EUCAST CLSI guidelines 2011-2021 translate MIC values disk diffusion diameters SIR (link) Principal component analysis AMR (tutorial)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/index.html","id":"latest-official-version","dir":"","previous_headings":"Get this package","what":"Latest official version","title":"Antimicrobial Resistance Data Analysis","text":"package available official R network (CRAN). Install package R CRAN using command: downloaded installed automatically. RStudio, click menu Tools > Install Packages… type “AMR” press Install. Note: functions website may available latest release. use functions data sets mentioned website, install latest development version.","code":"install.packages(\"AMR\")"},{"path":"https://msberends.github.io/AMR/index.html","id":"latest-development-version","dir":"","previous_headings":"Get this package","what":"Latest development version","title":"Antimicrobial Resistance Data Analysis","text":"Please read Developer Guideline . latest unpublished development version can installed GitHub two ways: Manually, using: Automatically, using rOpenSci R-universe platform, adding R-universe address list repositories (‘repos’): , can install update AMR package like official release (e.g., using install.packages(\"AMR\") RStudio via Tools > Check Package Updates…).","code":"install.packages(\"remotes\") # if you haven't already remotes::install_github(\"msberends/AMR\") options(repos = c(getOption(\"repos\"),                   msberends = \"https://msberends.r-universe.dev\"))"},{"path":"https://msberends.github.io/AMR/index.html","id":"get-started","dir":"","previous_headings":"","what":"Get started","title":"Antimicrobial Resistance Data Analysis","text":"find conduct AMR data analysis, please continue reading get started click link ‘’ menu.","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"partners","dir":"","previous_headings":"","what":"Partners","title":"Antimicrobial Resistance Data Analysis","text":"development package part , related , made possible following non-profit organisations initiatives:","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"copyright","dir":"","previous_headings":"","what":"Copyright","title":"Antimicrobial Resistance Data Analysis","text":"R package free, open-source software licensed GNU General Public License v2.0 (GPL-2). nutshell, means package: May used commercial purposes May used private purposes May used patent purposes May modified, although: Modifications must released license distributing package Changes made code must documented May distributed, although: Source code must made available package distributed copy license copyright notice must included package. Comes LIMITATION liability Comes warranty","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions — AMR-deprecated","title":"Deprecated Functions — AMR-deprecated","text":"functions -called 'Deprecated'. removed future version package. Using functions give warning name function replaced (one).","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions — AMR-deprecated","text":"","code":"ab_class(...)  ab_selector(...)"},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":null,"dir":"Reference","previous_headings":"","what":"Options for the AMR package — AMR-options","title":"Options for the AMR package — AMR-options","text":"overview package-specific options() can set AMR package.","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"options","dir":"Reference","previous_headings":"","what":"Options","title":"Options for the AMR package — AMR-options","text":"AMR_antibiogram_formatting_type  numeric (1-12) use antibiogram(), indicate formatting type use. AMR_breakpoint_type  character use .sir(), indicate breakpoint type use. must either \"ECOFF\", \"animal\", \"human\". AMR_cleaning_regex  regular expression (case-insensitive) use .mo() mo_* functions, clean user input. default outcome mo_cleaning_regex(), removes texts brackets texts \"species\" \"serovar\". AMR_custom_ab  file location RDS file, use custom antimicrobial drugs package. explained add_custom_antimicrobials(). AMR_custom_mo  file location RDS file, use custom microorganisms package. explained add_custom_microorganisms(). AMR_eucastrules  character set default types rules eucast_rules() function, must one : \"breakpoints\", \"expert\", \"\", \"custom\", \"\", defaults c(\"breakpoints\", \"expert\"). AMR_guideline  character set default guideline interpreting MIC values disk diffusion diameters .sir(). Can guideline name (e.g., \"CLSI\") name year (e.g. \"CLSI 2019\"). default latest implemented EUCAST guideline, currently \"EUCAST 2024\". Supported guideline currently EUCAST (2011-2024) CLSI (2011-2024). AMR_ignore_pattern  regular expression ignore (.e., make NA) match given .mo() mo_* functions. AMR_include_PKPD  logical use .sir(), indicate PK/PD clinical breakpoints must applied last resort - default TRUE. AMR_include_screening  logical use .sir(), indicate clinical breakpoints screening allowed - default FALSE. AMR_keep_synonyms  logical use .mo() mo_* functions, indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE. AMR_locale  character set language AMR package, can one supported language names ISO-639-1 codes: English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). default current system language (supported, English otherwise). AMR_mo_source  file location manual code list used .mo() mo_* functions. explained set_mo_source().","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"saving-settings-between-sessions","dir":"Reference","previous_headings":"","what":"Saving Settings Between Sessions","title":"Options for the AMR package — AMR-options","text":"Settings R saved globally thus lost R exited. can save options .Rprofile file, user-specific file. can edit using:   file, can set options ...   ...add Portuguese language support antibiotics, allow PK/PD rules interpreting MIC values .sir().","code":"utils::file.edit(\"~/.Rprofile\") options(AMR_locale = \"pt\")  options(AMR_include_PKPD = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"share-options-within-team","dir":"Reference","previous_headings":"","what":"Share Options Within Team","title":"Options for the AMR package — AMR-options","text":"global approach, e.g. within (data) team, save options file remote file location, shared network drive, user read file automatically start-. work way: Save plain text file e.g. \"X:/team_folder/R_options.R\" fill preferred settings. user, open .Rprofile file using utils::file.edit(\"~/.Rprofile\") put :   Reload R/RStudio check settings getOption(), e.g. getOption(\"AMR_locale\") set value. Now team settings configured one place, can maintained .","code":"source(\"X:/team_folder/R_options.R\")"},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":null,"dir":"Reference","previous_headings":"","what":"The AMR Package — AMR","title":"The AMR Package — AMR","text":"Welcome AMR package. AMR package free open-source R package zero dependencies simplify analysis prediction Antimicrobial Resistance (AMR) work microbial antimicrobial data properties, using evidence-based methods. aim provide standard clean reproducible AMR data analysis, can therefore empower epidemiological analyses continuously enable surveillance treatment evaluation setting. Many different researchers around globe continually helping us make successful durable project! work published Journal Statistical Software (Volume 104(3); doi:10.18637/jss.v104.i03 ) formed basis two PhD theses (doi:10.33612/diss.177417131  doi:10.33612/diss.192486375 ). installing package, R knows ~79 000 microorganisms (updated June 2024) ~610 antibiotic, antimycotic antiviral drugs name code (including ATC, EARS-Net, ASIARS-Net, PubChem, LOINC SNOMED CT), knows valid SIR MIC values. integral clinical breakpoint guidelines CLSI EUCAST included, even epidemiological cut-(ECOFF) values. supports can read data format, including WHONET data. package works Windows, macOS Linux versions R since R-3.0 (April 2013). designed work setting, including limited resources. created routine data analysis academic research Faculty Medical Sciences public University Groningen, collaboration non-profit organisations Certe Medical Diagnostics Advice Foundation University Medical Center Groningen. AMR package available English, Chinese, Czech, Danish, Dutch, Finnish, French, German, Greek, Italian, Japanese, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Turkish, Ukrainian. Antimicrobial drug (group) names colloquial microorganism names provided languages.","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"The AMR Package — AMR","text":"cite AMR publications use: Berends MS, Luz CF, Friedrich AW, Sinha BNM, Albers CJ, Glasner C (2022). \"AMR: R Package Working Antimicrobial Resistance Data.\" Journal Statistical Software, 104(3), 1-31. doi:10.18637/jss.v104.i03 BibTeX entry LaTeX users :","code":"@Article{,   title = {{AMR}: An {R} Package for Working with Antimicrobial Resistance Data},   author = {Matthijs S. Berends and Christian F. Luz and Alexander W. Friedrich and Bhanu N. M. Sinha and Casper J. Albers and Corinna Glasner},   journal = {Journal of Statistical Software},   year = {2022},   volume = {104},   number = {3},   pages = {1--31},   doi = {10.18637/jss.v104.i03}, }"},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"The AMR Package — AMR","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"The AMR Package — AMR","text":"Maintainer: Matthijs S. Berends m.s.berends@umcg.nl (ORCID) Authors: Dennis Souverein (ORCID) [contributor] Erwin E. . Hassing [contributor] contributors: Andrew P. Norgan (ORCID) [contributor] Anita Williams (ORCID) [contributor] Annick Lenglet (ORCID) [contributor] Anthony Underwood (ORCID) [contributor] Anton Mymrikov [contributor] Bart C. Meijer [contributor] Christian F. Luz (ORCID) [contributor] Dmytro Mykhailenko [contributor] Eric H. L. C. M. Hazenberg [contributor] Gwen Knight (ORCID) [contributor] Jason Stull (ORCID) [contributor] Javier Sanchez (ORCID) [contributor] Jonas Salm [contributor] Judith M. Fonville [contributor] Larisse Bolton (ORCID) [contributor] Matthew Saab [contributor] Peter Dutey-Magni (ORCID) [contributor] Rogier P. Schade [contributor] Sofia Ny (ORCID) [contributor] Alex W. Friedrich (ORCID) [thesis advisor] Bhanu N. M. Sinha (ORCID) [thesis advisor] Casper J. Albers (ORCID) [thesis advisor] Corinna Glasner (ORCID) [thesis advisor]","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":null,"dir":"Reference","previous_headings":"","what":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"antimicrobial drugs official names, ATC codes, ATC groups defined daily dose (DDD) included package, using Collaborating Centre Drug Statistics Methodology.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"","code":"as.ab(\"meropenem\") #> Class 'ab' #> [1] MEM ab_name(\"J01DH02\") #> [1] \"Meropenem\"  ab_tradenames(\"flucloxacillin\") #>  [1] \"culpen\"               \"floxacillin\"          \"floxacillinsodium\"    #>  [4] \"floxapen\"             \"floxapensodiumsalt\"   \"fluclox\"              #>  [7] \"flucloxacilina\"       \"flucloxacilline\"      \"flucloxacillinum\"     #> [10] \"fluorochloroxacillin\" \"staphylex\""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 500 Isolates - WHONET Example — WHONET","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"example data set exact structure export file WHONET. files can used package, example data set shows. antibiotic results example_isolates data set. patient names created using online surname generators place practice purposes.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"","code":"WHONET"},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"tibble 500 observations 53 variables: Identification number ID sample Specimen number ID specimen Organism Name microorganism. analysis, transform valid microbial class, using .mo(). Country Country origin Laboratory Name laboratory Last name Fictitious last name patient First name Fictitious initial patient Sex Fictitious gender patient Age Fictitious age patient Age category Age group, can also looked using age_groups() Date admissionDate hospital admission Specimen dateDate specimen received laboratory Specimen type Specimen type group Specimen type (Numeric) Translation \"Specimen type\" Reason Reason request Differential Diagnosis Isolate number ID isolate Organism type Type microorganism, can also looked using mo_type() Serotype Serotype microorganism Beta-lactamase Microorganism produces beta-lactamase? ESBL Microorganism produces extended spectrum beta-lactamase? Carbapenemase Microorganism produces carbapenemase? MRSA screening test Microorganism possible MRSA? Inducible clindamycin resistance Clindamycin can induced? Comment comments Date data entryDate data entered WHONET AMP_ND10:CIP_EE 28 different antibiotics. can lookup abbreviations antibiotics data set, use e.g. ab_name(\"AMP\") get official name immediately. analysis, transform valid antibiotic class, using .sir().","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"","code":"WHONET #> # A tibble: 500 × 53 #>    `Identification number` `Specimen number` Organism Country         Laboratory #>    <chr>                               <int> <chr>    <chr>           <chr>      #>  1 fe41d7bafa                           1748 SPN      Belgium         National … #>  2 91f175ec37                           1767 eco      The Netherlands National … #>  3 cc4015056e                           1343 eco      The Netherlands National … #>  4 e864b692f5                           1894 MAP      Denmark         National … #>  5 3d051fe345                           1739 PVU      Belgium         National … #>  6 c80762a08d                           1846 103      The Netherlands National … #>  7 8022d3727c                           1628 103      Denmark         National … #>  8 f3dc5f553d                           1493 eco      The Netherlands National … #>  9 15add38f6c                           1847 eco      France          National … #> 10 fd41248def                           1458 eco      Germany         National … #> # ℹ 490 more rows #> # ℹ 48 more variables: `Last name` <chr>, `First name` <chr>, Sex <chr>, #> #   Age <dbl>, `Age category` <chr>, `Date of admission` <date>, #> #   `Specimen date` <date>, `Specimen type` <chr>, #> #   `Specimen type (Numeric)` <dbl>, Reason <chr>, `Isolate number` <int>, #> #   `Organism type` <chr>, Serotype <chr>, `Beta-lactamase` <lgl>, ESBL <lgl>, #> #   Carbapenemase <lgl>, `MRSA screening test` <lgl>, …"},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":null,"dir":"Reference","previous_headings":"","what":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"Use function e.g. clinical texts health care records. returns list antimicrobial drugs, doses forms administration found texts.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"","code":"ab_from_text(text, type = c(\"drug\", \"dose\", \"administration\"),   collapse = NULL, translate_ab = FALSE, thorough_search = NULL,   info = interactive(), ...)"},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"text text analyse type type property search , either \"drug\", \"dose\" \"administration\", see Examples collapse character pass paste(, collapse = ...) return one character per element text, see Examples translate_ab type = \"drug\": column name antibiotics data set translate antibiotic abbreviations , using ab_property(). default FALSE. Using TRUE equal using \"name\". thorough_search logical indicate whether input must extensively searched misspelling faulty input values. Setting TRUE take considerably time using FALSE. default, turn TRUE input elements contain maximum three words. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed .ab()","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"list,  character collapse NULL","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"function also internally used .ab(), although searches first drug name throw note drug names returned. Note: .ab() function may use long regular expression match brand names antimicrobial drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"argument-type","dir":"Reference","previous_headings":"","what":"Argument type","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"default, function search antimicrobial drug names. text elements searched official names, ATC codes brand names. uses .ab() internally, correct misspelling. type = \"dose\" (similar, like \"dosing\", \"doses\"), text elements searched numeric values higher 100 resemble years. output numeric. supports unit (g, mg, IE, etc.) multiple values one clinical text, see Examples. type = \"administration\" (abbreviations, like \"admin\", \"adm\"), text elements searched form drug administration. supports following forms (including common abbreviations): buccal, implant, inhalation, instillation, intravenous, nasal, oral, parenteral, rectal, sublingual, transdermal vaginal. Abbreviations oral ('po', 'per os') become \"oral\", values intravenous ('iv', 'intraven') become \"iv\". supports multiple values one clinical text, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"argument-collapse","dir":"Reference","previous_headings":"","what":"Argument collapse","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"Without using collapse, function return list. can convenient use e.g. inside mutate()):df %>% mutate(abx = ab_from_text(clinical_text)) returned AB codes can transformed official names, groups, etc. ab_* functions ab_name() ab_group(), using translate_ab argument. using collapse, function return character:df %>% mutate(abx = ab_from_text(clinical_text, collapse = \"|\"))","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"","code":"# mind the bad spelling of amoxicillin in this line, # straight from a true health care record: ab_from_text(\"28/03/2020 regular amoxicilliin 500mg po tid\") #> [[1]] #> Class 'ab' #> [1] AMX #>   ab_from_text(\"500 mg amoxi po and 400mg cipro iv\") #> [[1]] #> Class 'ab' #> [1] AMX CIP #>  ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", type = \"dose\") #> [[1]] #> [1] 500 400 #>  ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", type = \"admin\") #> [[1]] #> [1] \"oral\" \"iv\"   #>   ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", collapse = \", \") #> [1] \"AMX, CIP\" # \\donttest{ # if you want to know which antibiotic groups were administered, do e.g.: abx <- ab_from_text(\"500 mg amoxi po and 400mg cipro iv\") ab_group(abx[[1]]) #> [1] \"Beta-lactams/penicillins\" \"Quinolones\"                if (require(\"dplyr\")) {   tibble(clinical_text = c(     \"given 400mg cipro and 500 mg amox\",     \"started on doxy iv today\"   )) %>%     mutate(       abx_codes = ab_from_text(clinical_text),       abx_doses = ab_from_text(clinical_text, type = \"doses\"),       abx_admin = ab_from_text(clinical_text, type = \"admin\"),       abx_coll = ab_from_text(clinical_text, collapse = \"|\"),       abx_coll_names = ab_from_text(clinical_text,         collapse = \"|\",         translate_ab = \"name\"       ),       abx_coll_doses = ab_from_text(clinical_text,         type = \"doses\",         collapse = \"|\"       ),       abx_coll_admin = ab_from_text(clinical_text,         type = \"admin\",         collapse = \"|\"       )     ) } #> Loading required package: dplyr #>  #> Attaching package: ‘dplyr’ #> The following objects are masked from ‘package:stats’: #>  #>     filter, lag #> The following objects are masked from ‘package:base’: #>  #>     intersect, setdiff, setequal, union #> # A tibble: 2 × 8 #>   clinical_text            abx_codes abx_doses abx_admin abx_coll abx_coll_names #>   <chr>                    <list>    <list>    <list>    <chr>    <chr>          #> 1 given 400mg cipro and 5… <ab [2]>  <dbl [2]> <chr [1]> CIP|AMX  Ciprofloxacin… #> 2 started on doxy iv today <ab [1]>  <dbl [1]> <chr [1]> DOX      Doxycycline    #> # ℹ 2 more variables: abx_coll_doses <chr>, abx_coll_admin <chr> # }"},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of an Antibiotic — ab_property","title":"Get Properties of an Antibiotic — ab_property","text":"Use functions return specific property antibiotic antibiotics data set. input values evaluated internally .ab().","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of an Antibiotic — ab_property","text":"","code":"ab_name(x, language = get_AMR_locale(), tolower = FALSE, ...)  ab_cid(x, ...)  ab_synonyms(x, ...)  ab_tradenames(x, ...)  ab_group(x, language = get_AMR_locale(), ...)  ab_atc(x, only_first = FALSE, ...)  ab_atc_group1(x, language = get_AMR_locale(), ...)  ab_atc_group2(x, language = get_AMR_locale(), ...)  ab_loinc(x, ...)  ab_ddd(x, administration = \"oral\", ...)  ab_ddd_units(x, administration = \"oral\", ...)  ab_info(x, language = get_AMR_locale(), ...)  ab_url(x, open = FALSE, ...)  ab_property(x, property = \"name\", language = get_AMR_locale(), ...)  set_ab_names(data, ..., property = \"name\", language = get_AMR_locale(),   snake_case = NULL)"},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of an Antibiotic — ab_property","text":"x (vector ) text can coerced valid antibiotic drug code .ab() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. tolower logical indicate whether first character every output transformed lower case character. lead e.g. \"polymyxin B\" \"polymyxin b\". ... case set_ab_names() data data.frame: columns select (supports tidy selection column1:column4), otherwise arguments passed .ab() only_first logical indicate whether first ATC code must returned, giving preference J0-codes (.e., antimicrobial drug group) administration way administration, either \"oral\" \"iv\" open browse URL using utils::browseURL() property one column names one antibiotics data set: vector_or(colnames(antibiotics), sort = FALSE). data data.frame columns need renamed, character vector column names snake_case logical indicate whether names -called snake case: lower case spaces/slashes replaced underscore (_)","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of an Antibiotic — ab_property","text":"integer case ab_cid() named list case ab_info() multiple ab_atc()/ab_synonyms()/ab_tradenames() double case ab_ddd() data.frame case set_ab_names() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of an Antibiotic — ab_property","text":"output translated possible. function ab_url() return direct URL official website. warning returned required ATC code available. function set_ab_names() special column renaming function data.frames. renames columns names resemble antimicrobial drugs. always makes sure new column names unique. property = \"atc\" set, preference given ATC codes J-group.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of an Antibiotic — ab_property","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of an Antibiotic — ab_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of an Antibiotic — ab_property","text":"","code":"# all properties: ab_name(\"AMX\") #> [1] \"Amoxicillin\" ab_atc(\"AMX\") #> [1] \"J01CA04\" ab_cid(\"AMX\") #> [1] 33613 ab_synonyms(\"AMX\") #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           ab_tradenames(\"AMX\") #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           ab_group(\"AMX\") #> [1] \"Beta-lactams/penicillins\" ab_atc_group1(\"AMX\") #> [1] \"Beta-lactam antibacterials, penicillins\" ab_atc_group2(\"AMX\") #> [1] \"Penicillins with extended spectrum\" ab_url(\"AMX\") #>                                                             Amoxicillin  #> \"https://atcddd.fhi.no/atc_ddd_index//?code=J01CA04&showdescription=no\"   # smart lowercase transformation ab_name(x = c(\"AMC\", \"PLB\")) #> [1] \"Amoxicillin/clavulanic acid\" \"Polymyxin B\"                 ab_name(x = c(\"AMC\", \"PLB\"), tolower = TRUE) #> [1] \"amoxicillin/clavulanic acid\" \"polymyxin B\"                  # defined daily doses (DDD) ab_ddd(\"AMX\", \"oral\") #> [1] 1.5 ab_ddd_units(\"AMX\", \"oral\") #> [1] \"g\" ab_ddd(\"AMX\", \"iv\") #> [1] 3 ab_ddd_units(\"AMX\", \"iv\") #> [1] \"g\"  ab_info(\"AMX\") # all properties as a list #> $ab #> [1] \"AMX\" #>  #> $cid #> [1] 33613 #>  #> $name #> [1] \"Amoxicillin\" #>  #> $group #> [1] \"Beta-lactams/penicillins\" #>  #> $atc #> [1] \"J01CA04\" #>  #> $atc_group1 #> [1] \"Beta-lactam antibacterials, penicillins\" #>  #> $atc_group2 #> [1] \"Penicillins with extended spectrum\" #>  #> $tradenames #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           #>  #> $loinc #>  [1] \"101498-4\" \"15-8\"     \"16-6\"     \"16365-9\"  \"17-4\"     \"18-2\"     #>  [7] \"18861-5\"  \"18862-3\"  \"19-0\"     \"20-8\"     \"21-6\"     \"22-4\"     #> [13] \"25274-2\"  \"25310-4\"  \"3344-9\"   \"55614-2\"  \"55615-9\"  \"55616-7\"  #> [19] \"6976-5\"   \"6977-3\"   \"80133-2\"  #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] 1.5 #>  #> $ddd$oral$units #> [1] \"g\" #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] 3 #>  #> $ddd$iv$units #> [1] \"g\" #>  #>  #>   # all ab_* functions use as.ab() internally, so you can go from 'any' to 'any': ab_atc(\"AMP\") #> [1] \"J01CA01\" \"S01AA19\" ab_group(\"J01CA01\") #> [1] \"Beta-lactams/penicillins\" ab_loinc(\"ampicillin\") #>  [1] \"101477-8\" \"101478-6\" \"18864-9\"  \"18865-6\"  \"20374-5\"  \"21066-6\"  #>  [7] \"23618-2\"  \"27-3\"     \"28-1\"     \"29-9\"     \"30-7\"     \"31-5\"     #> [13] \"32-3\"     \"33-1\"     \"3355-5\"   \"33562-0\"  \"33919-2\"  \"34-9\"     #> [19] \"43883-8\"  \"43884-6\"  \"6979-9\"   \"6980-7\"   \"87604-5\"  ab_name(\"21066-6\") #> [1] \"Ampicillin\" ab_name(6249) #> [1] \"Ampicillin\" ab_name(\"J01CA01\") #> [1] \"Ampicillin\"  # spelling from different languages and dyslexia are no problem ab_atc(\"ceftriaxon\") #> [1] \"J01DD04\" ab_atc(\"cephtriaxone\") #> [1] \"J01DD04\" ab_atc(\"cephthriaxone\") #> [1] \"J01DD04\" ab_atc(\"seephthriaaksone\") #> [1] \"J01DD04\"  # use set_ab_names() for renaming columns colnames(example_isolates) #>  [1] \"date\"    \"patient\" \"age\"     \"gender\"  \"ward\"    \"mo\"      \"PEN\"     #>  [8] \"OXA\"     \"FLC\"     \"AMX\"     \"AMC\"     \"AMP\"     \"TZP\"     \"CZO\"     #> [15] \"FEP\"     \"CXM\"     \"FOX\"     \"CTX\"     \"CAZ\"     \"CRO\"     \"GEN\"     #> [22] \"TOB\"     \"AMK\"     \"KAN\"     \"TMP\"     \"SXT\"     \"NIT\"     \"FOS\"     #> [29] \"LNZ\"     \"CIP\"     \"MFX\"     \"VAN\"     \"TEC\"     \"TCY\"     \"TGC\"     #> [36] \"DOX\"     \"ERY\"     \"CLI\"     \"AZM\"     \"IPM\"     \"MEM\"     \"MTR\"     #> [43] \"CHL\"     \"COL\"     \"MUP\"     \"RIF\"     colnames(set_ab_names(example_isolates)) #>  [1] \"date\"                          \"patient\"                       #>  [3] \"age\"                           \"gender\"                        #>  [5] \"ward\"                          \"mo\"                            #>  [7] \"benzylpenicillin\"              \"oxacillin\"                     #>  [9] \"flucloxacillin\"                \"amoxicillin\"                   #> [11] \"amoxicillin_clavulanic_acid\"   \"ampicillin\"                    #> [13] \"piperacillin_tazobactam\"       \"cefazolin\"                     #> [15] \"cefepime\"                      \"cefuroxime\"                    #> [17] \"cefoxitin\"                     \"cefotaxime\"                    #> [19] \"ceftazidime\"                   \"ceftriaxone\"                   #> [21] \"gentamicin\"                    \"tobramycin\"                    #> [23] \"amikacin\"                      \"kanamycin\"                     #> [25] \"trimethoprim\"                  \"trimethoprim_sulfamethoxazole\" #> [27] \"nitrofurantoin\"                \"fosfomycin\"                    #> [29] \"linezolid\"                     \"ciprofloxacin\"                 #> [31] \"moxifloxacin\"                  \"vancomycin\"                    #> [33] \"teicoplanin\"                   \"tetracycline\"                  #> [35] \"tigecycline\"                   \"doxycycline\"                   #> [37] \"erythromycin\"                  \"clindamycin\"                   #> [39] \"azithromycin\"                  \"imipenem\"                      #> [41] \"meropenem\"                     \"metronidazole\"                 #> [43] \"chloramphenicol\"               \"colistin\"                      #> [45] \"mupirocin\"                     \"rifampicin\"                    colnames(set_ab_names(example_isolates, NIT:VAN)) #>  [1] \"date\"           \"patient\"        \"age\"            \"gender\"         #>  [5] \"ward\"           \"mo\"             \"PEN\"            \"OXA\"            #>  [9] \"FLC\"            \"AMX\"            \"AMC\"            \"AMP\"            #> [13] \"TZP\"            \"CZO\"            \"FEP\"            \"CXM\"            #> [17] \"FOX\"            \"CTX\"            \"CAZ\"            \"CRO\"            #> [21] \"GEN\"            \"TOB\"            \"AMK\"            \"KAN\"            #> [25] \"TMP\"            \"SXT\"            \"nitrofurantoin\" \"fosfomycin\"     #> [29] \"linezolid\"      \"ciprofloxacin\"  \"moxifloxacin\"   \"vancomycin\"     #> [33] \"TEC\"            \"TCY\"            \"TGC\"            \"DOX\"            #> [37] \"ERY\"            \"CLI\"            \"AZM\"            \"IPM\"            #> [41] \"MEM\"            \"MTR\"            \"CHL\"            \"COL\"            #> [45] \"MUP\"            \"RIF\"            # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     set_ab_names()    # this does the same:   example_isolates %>%     rename_with(set_ab_names)    # set_ab_names() works with any AB property:   example_isolates %>%     set_ab_names(property = \"atc\")    example_isolates %>%     set_ab_names(where(is.sir)) %>%     colnames()    example_isolates %>%     set_ab_names(NIT:VAN) %>%     colnames() } #>  [1] \"date\"           \"patient\"        \"age\"            \"gender\"         #>  [5] \"ward\"           \"mo\"             \"PEN\"            \"OXA\"            #>  [9] \"FLC\"            \"AMX\"            \"AMC\"            \"AMP\"            #> [13] \"TZP\"            \"CZO\"            \"FEP\"            \"CXM\"            #> [17] \"FOX\"            \"CTX\"            \"CAZ\"            \"CRO\"            #> [21] \"GEN\"            \"TOB\"            \"AMK\"            \"KAN\"            #> [25] \"TMP\"            \"SXT\"            \"nitrofurantoin\" \"fosfomycin\"     #> [29] \"linezolid\"      \"ciprofloxacin\"  \"moxifloxacin\"   \"vancomycin\"     #> [33] \"TEC\"            \"TCY\"            \"TGC\"            \"DOX\"            #> [37] \"ERY\"            \"CLI\"            \"AZM\"            \"IPM\"            #> [41] \"MEM\"            \"MTR\"            \"CHL\"            \"COL\"            #> [45] \"MUP\"            \"RIF\"            # }"},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Custom Antimicrobials — add_custom_antimicrobials","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"add_custom_antimicrobials() can add custom antimicrobial drug names codes.","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"","code":"add_custom_antimicrobials(x)  clear_custom_antimicrobials()"},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"x data.frame resembling antibiotics data set, least containing columns \"ab\" \"name\"","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"Important: Due R works, add_custom_antimicrobials() function run every R session - added antimicrobials stored sessions thus lost R exited. two ways circumvent automate process adding antimicrobials: Method 1: Using package option AMR_custom_ab, preferred method. use method: Create data set structure antibiotics data set (containing least columns \"ab\" \"name\") save saveRDS() location choice, e.g. \"~/my_custom_ab.rds\", remote location. Set file location package option AMR_custom_ab: options(AMR_custom_ab = \"~/my_custom_ab.rds\"). can even remote file location, https URL. Since options saved R sessions, best save option .Rprofile file loaded start-R. , open .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"), add text save file:   Upon package load, file loaded run add_custom_antimicrobials() function. Method 2: Loading antimicrobial additions directly .Rprofile file. Note definitions stored user-specific R file, suboptimal workflow. use method: Edit .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"). Add text like save file:   Use clear_custom_antimicrobials() clear previously added antimicrobials.","code":"# Add custom antimicrobial codes: options(AMR_custom_ab = \"~/my_custom_ab.rds\") # Add custom antibiotic drug codes:  AMR::add_custom_antimicrobials(    data.frame(ab = \"TESTAB\",               name = \"Test Antibiotic\",               group = \"Test Group\")  )"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"","code":"# \\donttest{  # returns NA and throws a warning (which is suppressed here): suppressWarnings(   as.ab(\"testab\") ) #> Class 'ab' #> [1] <NA>  # now add a custom entry - it will be considered by as.ab() and # all ab_*() functions add_custom_antimicrobials(   data.frame(     ab = \"TESTAB\",     name = \"Test Antibiotic\",     # you can add any property present in the     # 'antibiotics' data set, such as 'group':     group = \"Test Group\"   ) ) #> ℹ Added one record to the internal antibiotics data set.  # \"testab\" is now a new antibiotic: as.ab(\"testab\") #> Class 'ab' #> [1] TESTAB ab_name(\"testab\") #> [1] \"Test Antibiotic\" ab_group(\"testab\") #> [1] \"Test Group\"  ab_info(\"testab\") #> $ab #> [1] \"TESTAB\" #>  #> $cid #> [1] NA #>  #> $name #> [1] \"Test Antibiotic\" #>  #> $group #> [1] \"Test Group\" #>  #> $atc #> [1] NA #>  #> $atc_group1 #> [1] NA #>  #> $atc_group2 #> [1] NA #>  #> $tradenames #> [1] NA #>  #> $loinc #> [1] NA #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] NA #>  #> $ddd$oral$units #> [1] NA #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] NA #>  #> $ddd$iv$units #> [1] NA #>  #>  #>    # Add Co-fluampicil, which is one of the many J01CR50 codes, see # https://atcddd.fhi.no/ddd/list_of_ddds_combined_products/ add_custom_antimicrobials(   data.frame(     ab = \"COFLU\",     name = \"Co-fluampicil\",     atc = \"J01CR50\",     group = \"Beta-lactams/penicillins\"   ) ) #> ℹ Added one record to the internal antibiotics data set. ab_atc(\"Co-fluampicil\") #> [1] \"J01CR50\" ab_name(\"J01CR50\") #> [1] \"Co-fluampicil\"  # even antimicrobial selectors work x <- data.frame(   random_column = \"some value\",   coflu = as.sir(\"S\"),   ampicillin = as.sir(\"R\") ) x #>   random_column coflu ampicillin #> 1    some value     S          R x[, betalactams()] #> ℹ For betalactams() using columns 'coflu' (co-fluampicil) and #>   'ampicillin' #>   coflu ampicillin #> 1     S          R # }"},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Custom Microorganisms — add_custom_microorganisms","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"add_custom_microorganisms() can add custom microorganisms, non-taxonomic outcome laboratory analysis.","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"","code":"add_custom_microorganisms(x)  clear_custom_microorganisms()"},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"x data.frame resembling microorganisms data set, least containing column \"genus\" (case-insensitive)","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"function fill missing taxonomy , specific taxonomic columns missing, see Examples. Important: Due R works, add_custom_microorganisms() function run every R session - added microorganisms stored sessions thus lost R exited. two ways circumvent automate process adding microorganisms: Method 1: Using package option AMR_custom_mo, preferred method. use method: Create data set structure microorganisms data set (containing least column \"genus\") save saveRDS() location choice, e.g. \"~/my_custom_mo.rds\", remote location. Set file location package option AMR_custom_mo: options(AMR_custom_mo = \"~/my_custom_mo.rds\"). can even remote file location, https URL. Since options saved R sessions, best save option .Rprofile file loaded start-R. , open .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"), add text save file:   Upon package load, file loaded run add_custom_microorganisms() function. Method 2: Loading microorganism directly .Rprofile file. Note definitions stored user-specific R file, suboptimal workflow. use method: Edit .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"). Add text like save file:   Use clear_custom_microorganisms() clear previously added microorganisms.","code":"# Add custom microorganism codes: options(AMR_custom_mo = \"~/my_custom_mo.rds\") # Add custom antibiotic drug codes:  AMR::add_custom_microorganisms(    data.frame(genus = \"Enterobacter\",               species = \"asburiae/cloacae\")  )"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"","code":"# \\donttest{ # a combination of species is not formal taxonomy, so # this will result in \"Enterobacter cloacae cloacae\", # since it resembles the input best: mo_name(\"Enterobacter asburiae/cloacae\") #> [1] \"Enterobacter asburiae\"  # now add a custom entry - it will be considered by as.mo() and # all mo_*() functions add_custom_microorganisms(   data.frame(     genus = \"Enterobacter\",     species = \"asburiae/cloacae\"   ) ) #> ℹ Added Enterobacter asburiae/cloacae to the internal microorganisms data #>   set.  # E. asburiae/cloacae is now a new microorganism: mo_name(\"Enterobacter asburiae/cloacae\") #> [1] \"Enterobacter asburiae/cloacae\"  # its code: as.mo(\"Enterobacter asburiae/cloacae\") #> Class 'mo' #> [1] CUSTOM1_ENTRB_ASB/  # all internal algorithms will work as well: mo_name(\"Ent asburia cloacae\") #> [1] \"Enterobacter asburiae/cloacae\"  # and even the taxonomy was added based on the genus! mo_family(\"E. asburiae/cloacae\") #> [1] \"Enterobacteriaceae\" mo_gramstain(\"Enterobacter asburiae/cloacae\") #> [1] \"Gram-negative\"  mo_info(\"Enterobacter asburiae/cloacae\") #> $mo #> [1] \"CUSTOM1_ENTRB_ASB/\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Enterobacter\" #>  #> $species #> [1] \"asburiae/cloacae\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] NA #>  #> $url #> [1] \"\" #>  #> $ref #> [1] \"Self-added, 2025\" #>  #> $snomed #> [1] NA #>  #> $lpsn #> [1] NA #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] NA #>  #> $group_members #> character(0) #>    # the function tries to be forgiving: add_custom_microorganisms(   data.frame(     GENUS = \"BACTEROIDES / PARABACTEROIDES SLASHLINE\",     SPECIES = \"SPECIES\"   ) ) #> ℹ Added Bacteroides/Parabacteroides to the internal microorganisms data #>   set. mo_name(\"BACTEROIDES / PARABACTEROIDES\") #> [1] \"Bacteroides/Parabacteroides\" mo_rank(\"BACTEROIDES / PARABACTEROIDES\") #> [1] \"genus\"  # taxonomy still works, even though a slashline genus was given as input: mo_family(\"Bacteroides/Parabacteroides\") #> [1] \"Bacteroidaceae\"   # for groups and complexes, set them as species or subspecies: add_custom_microorganisms(   data.frame(     genus = \"Citrobacter\",     species = c(\"freundii\", \"braakii complex\"),     subspecies = c(\"complex\", \"\")   ) ) #> ℹ Added Citrobacter braakii complex and Citrobacter freundii complex to the #>   internal microorganisms data set. mo_name(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"Citrobacter freundii complex\" \"Citrobacter braakii complex\"  mo_species(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"freundii complex\" \"braakii complex\"  mo_gramstain(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"Gram-negative\" \"Gram-negative\" # }"},{"path":"https://msberends.github.io/AMR/reference/age.html","id":null,"dir":"Reference","previous_headings":"","what":"Age in Years of Individuals — age","title":"Age in Years of Individuals — age","text":"Calculates age years based reference date, system date default.","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Age in Years of Individuals — age","text":"","code":"age(x, reference = Sys.Date(), exact = FALSE, na.rm = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Age in Years of Individuals — age","text":"x date(s), character (vectors) coerced .POSIXlt() reference reference date(s) (default today), character (vectors) coerced .POSIXlt() exact logical indicate whether age calculation exact, .e. decimals. divides number days year--date (YTD) x number days year reference (either 365 366). na.rm logical indicate whether missing values removed ... arguments passed .POSIXlt(), origin","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Age in Years of Individuals — age","text":"integer (decimals) exact = FALSE, double (decimals) otherwise","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Age in Years of Individuals — age","text":"Ages 0 returned NA warning. Ages 120 give warning. function vectorises x reference, meaning either can length 1 argument larger length.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Age in Years of Individuals — age","text":"","code":"# 10 random pre-Y2K birth dates df <- data.frame(birth_date = as.Date(\"2000-01-01\") - runif(10) * 25000)  # add ages df$age <- age(df$birth_date)  # add exact ages df$age_exact <- age(df$birth_date, exact = TRUE)  # add age at millenium switch df$age_at_y2k <- age(df$birth_date, \"2000-01-01\")  df #>    birth_date age age_exact age_at_y2k #> 1  1965-12-05  59  59.21644         34 #> 2  1980-03-01  44  44.98082         19 #> 3  1949-11-01  75  75.30959         50 #> 4  1947-02-14  78  78.02192         52 #> 5  1940-02-19  85  85.00822         59 #> 6  1988-01-10  37  37.11781         11 #> 7  1997-08-27  27  27.49041          2 #> 8  1978-01-26  47  47.07397         21 #> 9  1972-06-17  52  52.68493         27 #> 10 1986-08-10  38  38.53699         13"},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":null,"dir":"Reference","previous_headings":"","what":"Split Ages into Age Groups — age_groups","title":"Split Ages into Age Groups — age_groups","text":"Split ages age groups defined split argument. allows easier demographic (antimicrobial resistance) analysis.","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Split Ages into Age Groups — age_groups","text":"","code":"age_groups(x, split_at = c(12, 25, 55, 75), na.rm = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Split Ages into Age Groups — age_groups","text":"x age, e.g. calculated age() split_at values split x - default age groups 0-11, 12-24, 25-54, 55-74 75+. See Details. na.rm logical indicate whether missing values removed","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Split Ages into Age Groups — age_groups","text":"Ordered factor","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Split Ages into Age Groups — age_groups","text":"split ages, input split_at argument can : numeric vector. value e.g. c(10, 20) split x 0-9, 10-19 20+. value 50 split x 0-49 50+. default split young children (0-11), youth (12-24), young adults (25-54), middle-aged adults (55-74) elderly (75+). character: \"children\" \"kids\", equivalent : c(0, 1, 2, 4, 6, 13, 18). split 0, 1, 2-3, 4-5, 6-12, 13-17 18+. \"elderly\" \"seniors\", equivalent : c(65, 75, 85). split 0-64, 65-74, 75-84, 85+. \"fives\", equivalent : 1:20 * 5. split 0-4, 5-9, ..., 95-99, 100+. \"tens\", equivalent : 1:10 * 10. split 0-9, 10-19, ..., 90-99, 100+.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Split Ages into Age Groups — age_groups","text":"","code":"ages <- c(3, 8, 16, 54, 31, 76, 101, 43, 21)  # split into 0-49 and 50+ age_groups(ages, 50) #> [1] 0-49 0-49 0-49 50+  0-49 50+  50+  0-49 0-49 #> Levels: 0-49 < 50+  # split into 0-19, 20-49 and 50+ age_groups(ages, c(20, 50)) #> [1] 0-19  0-19  0-19  50+   20-49 50+   50+   20-49 20-49 #> Levels: 0-19 < 20-49 < 50+  # split into groups of ten years age_groups(ages, 1:10 * 10) #> [1] 0-9   0-9   10-19 50-59 30-39 70-79 100+  40-49 20-29 #> 11 Levels: 0-9 < 10-19 < 20-29 < 30-39 < 40-49 < 50-59 < 60-69 < ... < 100+ age_groups(ages, split_at = \"tens\") #> [1] 0-9   0-9   10-19 50-59 30-39 70-79 100+  40-49 20-29 #> 11 Levels: 0-9 < 10-19 < 20-29 < 30-39 < 40-49 < 50-59 < 60-69 < ... < 100+  # split into groups of five years age_groups(ages, 1:20 * 5) #> [1] 0-4   5-9   15-19 50-54 30-34 75-79 100+  40-44 20-24 #> 21 Levels: 0-4 < 5-9 < 10-14 < 15-19 < 20-24 < 25-29 < 30-34 < ... < 100+ age_groups(ages, split_at = \"fives\") #> [1] 0-4   5-9   15-19 50-54 30-34 75-79 100+  40-44 20-24 #> 21 Levels: 0-4 < 5-9 < 10-14 < 15-19 < 20-24 < 25-29 < 30-34 < ... < 100+  # split specifically for children age_groups(ages, c(1, 2, 4, 6, 13, 18)) #> [1] 2-3   6-12  13-17 18+   18+   18+   18+   18+   18+   #> Levels: 0 < 1 < 2-3 < 4-5 < 6-12 < 13-17 < 18+ age_groups(ages, \"children\") #> [1] 2-3   6-12  13-17 18+   18+   18+   18+   18+   18+   #> Levels: 0 < 1 < 2-3 < 4-5 < 6-12 < 13-17 < 18+  # \\donttest{ # resistance of ciprofloxacin per age group if (require(\"dplyr\") && require(\"ggplot2\")) {   example_isolates %>%     filter_first_isolate() %>%     filter(mo == as.mo(\"Escherichia coli\")) %>%     group_by(age_group = age_groups(age)) %>%     select(age_group, CIP) %>%     ggplot_sir(       x = \"age_group\",       minimum = 0,       x.title = \"Age Group\",       title = \"Ciprofloxacin resistance per age group\"     ) } #> Loading required package: ggplot2  # }"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Create detailed antibiograms options traditional, combination, syndromic, Bayesian WISCA methods. Adhering previously described approaches (see Source) especially Bayesian WISCA model (Weighted-Incidence Syndromic Combination Antibiogram) Bielicki et al., functions provide flexible output formats including plots tables, ideal integration R Markdown Quarto reports.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"","code":"antibiogram(x, antibiotics = where(is.sir), mo_transform = \"shortname\",   ab_transform = \"name\", syndromic_group = NULL, add_total_n = FALSE,   only_all_tested = FALSE, digits = ifelse(wisca, 1, 0),   formatting_type = getOption(\"AMR_antibiogram_formatting_type\", 14),   col_mo = NULL, language = get_AMR_locale(), minimum = 30,   combine_SI = TRUE, sep = \" + \", wisca = FALSE, simulations = 1000,   conf_interval = 0.95, interval_side = \"two-tailed\",   info = interactive())  wisca(x, antibiotics = where(is.sir), ab_transform = \"name\",   syndromic_group = NULL, add_total_n = FALSE, only_all_tested = FALSE,   digits = 1,   formatting_type = getOption(\"AMR_antibiogram_formatting_type\", 14),   col_mo = NULL, language = get_AMR_locale(), minimum = 30,   combine_SI = TRUE, sep = \" + \", simulations = 1000,   conf_interval = 0.95, interval_side = \"two-tailed\",   info = interactive())  retrieve_wisca_parameters(wisca_model, ...)  # S3 method for class 'antibiogram' plot(x, ...)  # S3 method for class 'antibiogram' autoplot(object, ...)  # S3 method for class 'antibiogram' knit_print(x, italicise = TRUE,   na = getOption(\"knitr.kable.NA\", default = \"\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Bielicki JA et al. (2016). Selecting appropriate empirical antibiotic regimens paediatric bloodstream infections: application Bayesian decision model local pooled antimicrobial resistance surveillance data Journal Antimicrobial Chemotherapy 71(3); doi:10.1093/jac/dkv397 Bielicki JA et al. (2020). Evaluation coverage 3 antibiotic regimens neonatal sepsis hospital setting across Asian countries JAMA Netw Open. 3(2):e1921124; doi:10.1001.jamanetworkopen.2019.21124 Klinker KP et al. (2021). Antimicrobial stewardship antibiograms: importance moving beyond traditional antibiograms. Therapeutic Advances Infectious Disease, May 5;8:20499361211011373; doi:10.1177/20499361211011373 Barbieri E et al. (2021). Development Weighted-Incidence Syndromic Combination Antibiogram (WISCA) guide choice empiric antibiotic treatment urinary tract infection paediatric patients: Bayesian approach Antimicrobial Resistance & Infection Control May 1;10(1):74; doi:10.1186/s13756-021-00939-2 M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"x data.frame containing least column microorganisms columns antimicrobial results (class 'sir', see .sir()) antibiotics vector antimicrobial name code (evaluated .ab(), column name x, (combinations ) antimicrobial selectors aminoglycosides() carbapenems(). combination antibiograms, can also set values separated \"+\", \"TZP+TOB\" \"cipro + genta\", given columns resembling antimicrobials exist x. See Examples. mo_transform character transform microorganism input - must \"name\", \"shortname\" (default), \"gramstain\", one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". Can also NULL transform input NA consider microorganisms 'unknown'. ab_transform character transform antimicrobial input - must one column names antibiotics data set (defaults \"name\"): \"ab\", \"cid\", \"name\", \"group\", \"atc\", \"atc_group1\", \"atc_group2\", \"abbreviations\", \"synonyms\", \"oral_ddd\", \"oral_units\", \"iv_ddd\", \"iv_units\", \"loinc\". Can also NULL transform input. syndromic_group column name x, values calculated split rows x, e.g. using ifelse() case_when(). See Examples. add_total_n logical indicate whether n_tested available numbers per pathogen added table (default TRUE). add lowest highest number available isolates per antimicrobial (e.g, E. coli 200 isolates available ciprofloxacin 150 amoxicillin, returned number \"150-200\"). option unavailable wisca = TRUE; case, use retrieve_wisca_parameters() get parameters used WISCA. only_all_tested (combination antibiograms): logical indicate isolates must tested antimicrobials, see Details digits number digits use rounding antimicrobial coverage, defaults 1 WISCA 0 otherwise formatting_type numeric value (1–22 WISCA, 1-12 non-WISCA) indicating 'cells' antibiogram table formatted. See Details > Formatting Type list options. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). language language translate text, defaults system language (see get_AMR_locale()) minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether susceptibility determined results either S, SDD, , instead S (default TRUE) sep separating character antimicrobial columns combination antibiograms wisca logical indicate whether Weighted-Incidence Syndromic Combination Antibiogram (WISCA) must generated (default FALSE). use Bayesian decision model estimate regimen coverage probabilities using Monte Carlo simulations. Set simulations, conf_interval, interval_side adjust. simulations (WISCA) numerical value set number Monte Carlo simulations conf_interval (WISCA) numerical value set confidence interval (default 0.95) interval_side (WISCA) side confidence interval, either \"two-tailed\" (default), \"left\" \"right\" info logical indicate info printed - default TRUE interactive mode wisca_model outcome wisca() antibiogram(..., wisca = TRUE) ... used R Markdown Quarto: arguments passed knitr::kable() (otherwise, use) object antibiogram() object italicise logical indicate whether microorganism names knitr table made italic, using italicise_taxonomy(). na character use showing NA values","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"functions return table values 0 100 susceptibility, resistance. Remember filter data let contain first isolates! needed exclude duplicates reduce selection bias. Use first_isolate() determine one four available algorithms: isolate-based, patient-based, episode-based, phenotype-based. estimating antimicrobial coverage, especially creating WISCA, outcome might become reliable including top n species encountered data. can filter top n using top_n_microorganisms(). example, use top_n_microorganisms(your_data, n = 10) pre-processing step include top 10 species data. numeric values antibiogram stored long format attribute long_numeric. can retrieve using attributes(x)$long_numeric, x outcome antibiogram() wisca(). ideal e.g. advanced plotting.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"formatting-type","dir":"Reference","previous_headings":"","what":"Formatting Type","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"formatting 'cells' table can set argument formatting_type. examples, 5 antimicrobial coverage (4-6 indicates confidence level), 15 number susceptible isolates, 300 number tested (.e., available) isolates: 5 15 300 15/300 5 (300) 5% (300) 5 (N=300) 5% (N=300) 5 (15/300) 5% (15/300) 5 (N=15/300) 5% (N=15/300) 5 (4-6) 5% (4-6%) - default 5 (4-6,300) 5% (4-6%,300) 5 (4-6,N=300) 5% (4-6%,N=300) 5 (4-6,15/300) 5% (4-6%,15/300) 5 (4-6,N=15/300) 5% (4-6%,N=15/300) default 14, can set globally package option AMR_antibiogram_formatting_type, e.g. options(AMR_antibiogram_formatting_type = 5). note WISCA, total numbers tested susceptible isolates less useful report, since included Bayesian model apparent susceptibility confidence level. Set digits (defaults 0) alter rounding susceptibility percentages.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"antibiogram-types","dir":"Reference","previous_headings":"","what":"Antibiogram Types","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"various antibiogram types, summarised Klinker et al. (2021, doi:10.1177/20499361211011373 ), supported antibiogram(). clinical coverage estimations, use WISCA whenever possible, since provides precise coverage estimates accounting pathogen incidence antimicrobial susceptibility, shown Bielicki et al. (2020, doi:10.1001.jamanetworkopen.2019.21124 ). See section Explaining WISCA page. note WISCA pathogen-agnostic, meaning outcome stratied pathogen, rather syndrome. Traditional Antibiogram Case example: Susceptibility Pseudomonas aeruginosa piperacillin/tazobactam (TZP) Code example:   Combination Antibiogram Case example: Additional susceptibility Pseudomonas aeruginosa TZP + tobramycin versus TZP alone Code example:   Syndromic Antibiogram Case example: Susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) Code example:   Weighted-Incidence Syndromic Combination Antibiogram (WISCA) WISCA can applied antibiogram, see section Explaining WISCA page information. Code example:   WISCA uses sophisticated Bayesian decision model combine local pooled antimicrobial resistance data. approach evaluates local patterns can also draw multi-centre datasets improve regimen accuracy, even low-incidence infections like paediatric bloodstream infections (BSIs).","code":"antibiogram(your_data,             antibiotics = \"TZP\") antibiogram(your_data,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\")) antibiogram(your_data,             antibiotics = penicillins(),             syndromic_group = \"ward\") antibiogram(your_data,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),             wisca = TRUE)  # this is equal to: wisca(your_data,       antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"grouped-tibbles","dir":"Reference","previous_headings":"","what":"Grouped tibbles","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"type antibiogram, grouped tibbles can also used calculate susceptibilities various groups. Code example:","code":"library(dplyr) your_data %>%   group_by(has_sepsis, is_neonate, sex) %>%   wisca(antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"stepped-approach-for-clinical-insight","dir":"Reference","previous_headings":"","what":"Stepped Approach for Clinical Insight","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"clinical practice, antimicrobial coverage decisions evolve microbiological data becomes available. theoretical stepped approach ensures empirical coverage can continuously assessed improve patient outcomes: Initial Empirical Therapy (Admission / Pre-Culture Data) admission, pathogen information available. Action: broad-spectrum coverage based local resistance patterns syndromic antibiograms. Using pathogen-agnostic yet incidence-weighted WISCA preferred. Code example:   Refinement Gram Stain Results blood culture becomes positive, Gram stain provides initial crucial first stratification (Gram-positive vs. Gram-negative). Action: narrow coverage based Gram stain-specific resistance patterns. Code example:   Definitive Therapy Based Species Identification cultivation pathogen, full pathogen identification allows precise targeting therapy. Action: adjust treatment pathogen-specific antibiograms, minimizing resistance risks. Code example:   structuring antibiograms around stepped approach, clinicians can make data-driven adjustments stage, ensuring optimal empirical targeted therapy reducing unnecessary broad-spectrum antimicrobial use.","code":"antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = NA) # all pathogens set to `NA`  # preferred: use WISCA wisca(your_data,       antibiotics = selected_regimens) antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = \"gramstain\") # all pathogens set to Gram-pos/Gram-neg antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = \"shortname\") # all pathogens set to 'G. species', e.g., E. coli"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"inclusion-in-combination-antibiograms","dir":"Reference","previous_headings":"","what":"Inclusion in Combination Antibiograms","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Note combination antibiograms, important realise susceptibility can calculated two ways, can set only_all_tested argument (default FALSE). See example two antimicrobials, Drug Drug B, antibiogram() works calculate %SI:","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - --------------------------------------------------------------------"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"plotting","dir":"Reference","previous_headings":"","what":"Plotting","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"types antibiograms listed can plotted (using ggplot2::autoplot() base R's plot() barplot()). mentioned , numeric values antibiogram stored long format attribute long_numeric. can retrieve using attributes(x)$long_numeric, x outcome antibiogram() wisca(). outcome antibiogram() can also used directly R Markdown / Quarto (.e., knitr) reports. case, knitr::kable() applied automatically microorganism names even printed italics default (see argument italicise). can also use functions specific 'table reporting' packages transform output antibiogram() needs, e.g. flextable::as_flextable() gt::gt().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"explaining-wisca","dir":"Reference","previous_headings":"","what":"Explaining WISCA","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"WISCA, outlined Bielicki et al. (doi:10.1093/jac/dkv397 ), stands Weighted-Incidence Syndromic Combination Antibiogram, estimates probability adequate empirical antimicrobial regimen coverage specific infection syndromes. method leverages Bayesian decision model random effects pathogen incidence susceptibility, enabling robust estimates presence sparse data. Bayesian model assumes conjugate priors parameter estimation. example, coverage probability \\(\\theta\\) given antimicrobial regimen modelled using Beta distribution prior: $$\\theta \\sim \\text{Beta}(\\alpha_0, \\beta_0)$$ \\(\\alpha_0\\) \\(\\beta_0\\) represent prior successes failures, respectively, informed expert knowledge weakly informative priors (e.g., \\(\\alpha_0 = 1, \\beta_0 = 1\\)). likelihood function constructed based observed data, number covered cases regimen follows binomial distribution: $$y \\sim \\text{Binomial}(n, \\theta)$$ Posterior parameter estimates obtained combining prior likelihood using Bayes' theorem. posterior distribution \\(\\theta\\) also Beta distribution: $$\\theta | y \\sim \\text{Beta}(\\alpha_0 + y, \\beta_0 + n - y)$$ Pathogen incidence, representing proportion infections caused different pathogens, modelled using Dirichlet distribution, natural conjugate prior multinomial outcomes. Dirichlet distribution parameterised vector concentration parameters \\(\\alpha\\), \\(\\alpha_i\\) corresponds specific pathogen. prior typically chosen uniform (\\(\\alpha_i = 1\\)), reflecting assumption equal prior probability across pathogens. posterior distribution pathogen incidence given : $$\\text{Dirichlet}(\\alpha_1 + n_1, \\alpha_2 + n_2, \\dots, \\alpha_K + n_K)$$ \\(n_i\\) number infections caused pathogen \\(\\) observed data. practical implementation, pathogen incidences sampled posterior using normalised Gamma-distributed random variables: $$x_i \\sim \\text{Gamma}(\\alpha_i + n_i, 1)$$ $$p_i = \\frac{x_i}{\\sum_{j=1}^K x_j}$$ \\(x_i\\) represents unnormalised pathogen counts, \\(p_i\\) normalised proportion pathogen \\(\\). hierarchical modelling, pathogen-level effects (e.g., differences resistance patterns) regimen-level effects modelled using Gaussian priors log-odds. hierarchical structure ensures partial pooling estimates across groups, improving stability strata small sample sizes. model implemented using Hamiltonian Monte Carlo (HMC) sampling. Stratified results can provided based covariates age, sex, clinical complexity (e.g., prior antimicrobial treatments renal/urological comorbidities) using dplyr's group_by() pre-processing step running wisca(). Posterior odds ratios (ORs) derived quantify effect covariates coverage probabilities: $$\\text{}_{\\text{covariate}} = \\frac{\\exp(\\beta_{\\text{covariate}})}{\\exp(\\beta_0)}$$ combining empirical data prior knowledge, WISCA overcomes limitations traditional combination antibiograms, offering disease-specific, patient-stratified estimates robust uncertainty quantification. tool invaluable antimicrobial stewardship programs empirical treatment guideline refinement. Note: WISCA never gives output pathogen/species level, incidences susceptibilities already weighted species.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Implementation: Dr. Larisse Bolton Dr. Matthijs Berends","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # \\donttest{ # Traditional antibiogram ----------------------------------------------  antibiogram(example_isolates,   antibiotics = c(aminoglycosides(), carbapenems()) ) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 10 × 7 #>    Pathogen       Amikacin    Gentamicin Imipenem Kanamycin Meropenem Tobramycin #>  * <chr>          <chr>       <chr>      <chr>    <chr>     <chr>     <chr>      #>  1 CoNS           0% (0-8%)   86% (82-9… 52% (37… 0% (0-8%) 52% (37-… 22% (12-3… #>  2 E. coli        100% (98-1… 98% (96-9… 100% (9… NA        100% (99… 97% (96-9… #>  3 E. faecalis    0% (0-9%)   0% (0-9%)  100% (9… 0% (0-9%) NA        0% (0-9%)  #>  4 K. pneumoniae  NA          90% (79-9… 100% (9… NA        100% (93… 90% (79-9… #>  5 P. aeruginosa  NA          100% (88-… NA       0% (0-12… NA        100% (88-… #>  6 P. mirabilis   NA          94% (80-9… 94% (79… NA        NA        94% (80-9… #>  7 S. aureus      NA          99% (97-1… NA       NA        NA        98% (92-1… #>  8 S. epidermidis 0% (0-8%)   79% (71-8… NA       0% (0-8%) NA        51% (40-6… #>  9 S. hominis     NA          92% (84-9… NA       NA        NA        85% (74-9… #> 10 S. pneumoniae  0% (0-3%)   0% (0-3%)  NA       0% (0-3%) NA        0% (0-3%)  #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  antibiogram(example_isolates,   antibiotics = aminoglycosides(),   ab_transform = \"atc\",   mo_transform = \"gramstain\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # An Antibiogram (non-WISCA / 95% CI): 2 × 5 #>   Pathogen      J01GB01      J01GB03      J01GB04    J01GB06      #> * <chr>         <chr>        <chr>        <chr>      <chr>        #> 1 Gram-negative 96% (94-97%) 96% (95-98%) 0% (0-10%) 98% (96-99%) #> 2 Gram-positive 34% (31-38%) 63% (60-66%) 0% (0-1%)  0% (0-1%)    #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  antibiogram(example_isolates,   antibiotics = carbapenems(),   ab_transform = \"name\",   mo_transform = \"name\") #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 5 × 3 #>   Pathogen                                 Imipenem       Meropenem      #> * <chr>                                    <chr>          <chr>          #> 1 Coagulase-negative Staphylococcus (CoNS) 52% (37-67%)   52% (37-67%)   #> 2 Enterococcus faecalis                    100% (91-100%) NA             #> 3 Escherichia coli                         100% (99-100%) 100% (99-100%) #> 4 Klebsiella pneumoniae                    100% (93-100%) 100% (93-100%) #> 5 Proteus mirabilis                        94% (79-99%)   NA             #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Combined antibiogram -------------------------------------------------  # combined antibiotics yield higher empiric coverage antibiogram(example_isolates,   antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),   mo_transform = \"gramstain\") #> # An Antibiogram (non-WISCA / 95% CI): 2 × 4 #>   Pathogen  Piperacillin/tazobac…¹ Piperacillin/tazobac…² Piperacillin/tazobac…³ #> * <chr>     <chr>                  <chr>                  <chr>                  #> 1 Gram-neg… 88% (85-91%)           99% (97-99%)           98% (97-99%)           #> 2 Gram-pos… 86% (82-89%)           98% (96-98%)           95% (93-97%)           #> # ℹ abbreviated names: ¹​`Piperacillin/tazobactam`, #> #   ²​`Piperacillin/tazobactam + Gentamicin`, #> #   ³​`Piperacillin/tazobactam + Tobramycin` #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  # names of antibiotics do not need to resemble columns exactly: antibiogram(example_isolates,   antibiotics = c(\"Cipro\", \"cipro + genta\"),   mo_transform = \"gramstain\",   ab_transform = \"name\",   sep = \" & \") #> # An Antibiogram (non-WISCA / 95% CI): 2 × 3 #>   Pathogen      Ciprofloxacin `Ciprofloxacin & Gentamicin` #> * <chr>         <chr>         <chr>                        #> 1 Gram-negative 91% (88-93%)  99% (97-99%)                 #> 2 Gram-positive 77% (74-80%)  93% (91-94%)                 #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Syndromic antibiogram ------------------------------------------------  # the data set could contain a filter for e.g. respiratory specimens antibiogram(example_isolates,   antibiotics = c(aminoglycosides(), carbapenems()),   syndromic_group = \"ward\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 14 × 8 #>    `Syndromic Group` Pathogen   Amikacin Gentamicin Imipenem Kanamycin Meropenem #>  * <chr>             <chr>      <chr>    <chr>      <chr>    <chr>     <chr>     #>  1 Clinical          CoNS       NA       89% (84-9… 57% (39… NA        57% (39-… #>  2 ICU               CoNS       NA       79% (68-8… NA       NA        NA        #>  3 Outpatient        CoNS       NA       84% (66-9… NA       NA        NA        #>  4 Clinical          E. coli    100% (9… 98% (96-9… 100% (9… NA        100% (99… #>  5 ICU               E. coli    100% (9… 99% (95-1… 100% (9… NA        100% (97… #>  6 Clinical          K. pneumo… NA       92% (81-9… 100% (9… NA        100% (92… #>  7 Clinical          P. mirabi… NA       100% (88-… NA       NA        NA        #>  8 Clinical          S. aureus  NA       99% (95-1… NA       NA        NA        #>  9 ICU               S. aureus  NA       100% (95-… NA       NA        NA        #> 10 Clinical          S. epider… NA       82% (72-9… NA       NA        NA        #> 11 ICU               S. epider… NA       72% (60-8… NA       NA        NA        #> 12 Clinical          S. hominis NA       96% (85-9… NA       NA        NA        #> 13 Clinical          S. pneumo… 0% (0-5… 0% (0-5%)  NA       0% (0-5%) NA        #> 14 ICU               S. pneumo… 0% (0-1… 0% (0-12%) NA       0% (0-12… NA        #> # ℹ 1 more variable: Tobramycin <chr> #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  # now define a data set with only E. coli ex1 <- example_isolates[which(mo_genus() == \"Escherichia\"), ] #> ℹ Using column 'mo' as input for mo_genus()  # with a custom language, though this will be determined automatically # (i.e., this table will be in Spanish on Spanish systems) antibiogram(ex1,   antibiotics = aminoglycosides(),   ab_transform = \"name\",   syndromic_group = ifelse(ex1$ward == \"ICU\",     \"UCI\", \"No UCI\"   ),   language = \"es\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # An Antibiogram (non-WISCA / 95% CI): 2 × 5 #>   `Grupo sindrómico` Patógeno Amikacina      Gentamicina   Tobramicina  #> * <chr>              <chr>    <chr>          <chr>         <chr>        #> 1 No UCI             E. coli  100% (97-100%) 98% (96-99%)  98% (96-99%) #> 2 UCI                E. coli  100% (93-100%) 99% (95-100%) 96% (92-99%) #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # WISCA antibiogram ----------------------------------------------------  # WISCA are not stratified by species, but rather on syndromes antibiogram(example_isolates,   antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),   syndromic_group = \"ward\",   wisca = TRUE) #> Warning: WISCA must be based on the species level as WISCA parameters are based on #> this. For that reason, mo_transform will be ignored. #> # An Antibiogram (WISCA / 95% CI): 3 × 4 #>   `Syndromic Group` `Piperacillin/tazobactam` Piperacillin/tazobactam + Gentam…¹ #> * <chr>             <chr>                     <chr>                              #> 1 Clinical          78% (70-85.4%)            98.3% (94.9-99.7%)                 #> 2 ICU               90.6% (84.4-94.9%)        97.6% (92.7-99.7%)                 #> 3 Outpatient        92.2% (79.7-99%)          94.7% (82.7-99.5%)                 #> # ℹ abbreviated name: ¹​`Piperacillin/tazobactam + Gentamicin` #> # ℹ 1 more variable: `Piperacillin/tazobactam + Tobramycin` <chr> #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Print the output for R Markdown / Quarto -----------------------------  ureido <- antibiogram(example_isolates,   antibiotics = ureidopenicillins(),   syndromic_group = \"ward\",   wisca = TRUE) #> Warning: WISCA must be based on the species level as WISCA parameters are based on #> this. For that reason, mo_transform will be ignored. #> ℹ For ureidopenicillins() using column 'TZP' (piperacillin/tazobactam)  # in an Rmd file, you would just need to return `ureido` in a chunk, # but to be explicit here: if (requireNamespace(\"knitr\")) {   cat(knitr::knit_print(ureido)) } #>  #>  #> |Syndromic Group |Piperacillin/tazobactam | #> |:---------------|:-----------------------| #> |Clinical        |78% (70.1-85.7%)        | #> |ICU             |90.4% (83.6-95%)        | #> |Outpatient      |91.8% (78.5-98.9%)      |   # Generate plots with ggplot2 or base R --------------------------------  ab1 <- antibiogram(example_isolates,   antibiotics = c(\"AMC\", \"CIP\", \"TZP\", \"TZP+TOB\"),   mo_transform = \"gramstain\") ab2 <- antibiogram(example_isolates,   antibiotics = c(\"AMC\", \"CIP\", \"TZP\", \"TZP+TOB\"),   mo_transform = \"gramstain\",   syndromic_group = \"ward\")  if (requireNamespace(\"ggplot2\")) {   ggplot2::autoplot(ab1) }  if (requireNamespace(\"ggplot2\")) {   ggplot2::autoplot(ab2) }   plot(ab1)  plot(ab2)  # }"},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Sets with 607 Antimicrobial Drugs — antibiotics","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Two data sets containing antibiotics/antimycotics antivirals. Use .ab() one ab_* functions retrieve values antibiotics data set. Three identifiers included data set: antibiotic ID (ab, primarily used package) defined WHONET/EARS-Net, ATC code (atc) defined , Compound ID (cid) found PubChem. properties data set derived one codes. Note drugs multiple ATC codes.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"","code":"antibiotics  antivirals"},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"object class tbl_df (inherits tbl, data.frame) 120 rows 11 columns.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"for-the-antibiotics-data-set-a-tibble-with-observations-and-variables-","dir":"Reference","previous_headings":"","what":"For the antibiotics data set: a tibble with 487 observations and 14 variables:","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"ab Antibiotic ID used package (AMC), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available. unique identifier. cid Compound ID found PubChem. unique identifier. name Official name used WHONET/EARS-Net . unique identifier. group short concise group name, based WHONET WHOCC definitions atc ATC codes (Anatomical Therapeutic Chemical) defined WHOCC, like J01CR02 atc_group1 Official pharmacological subgroup (3rd level ATC code) defined WHOCC, like \"Macrolides, lincosamides streptogramins\" atc_group2 Official chemical subgroup (4th level ATC code) defined WHOCC, like \"Macrolides\" abbr List abbreviations used many countries, also antibiotic susceptibility testing (AST) synonyms Synonyms (often trade names) drug, found PubChem based compound ID oral_ddd Defined Daily Dose (DDD), oral treatment, currently available 179 drugs oral_units Units oral_ddd iv_ddd Defined Daily Dose (DDD), parenteral (intravenous) treatment, currently available 153 drugs iv_units Units iv_ddd loinc codes associated name antimicrobial drug Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Use ab_loinc() retrieve quickly, see ab_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"for-the-antivirals-data-set-a-tibble-with-observations-and-variables-","dir":"Reference","previous_headings":"","what":"For the antivirals data set: a tibble with 120 observations and 11 variables:","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"av Antiviral ID used package (ACI), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available. unique identifier. Combinations codes contain + indicate , ATA+COBI atazanavir/cobicistat. name Official name used WHONET/EARS-Net . unique identifier. atc ATC codes (Anatomical Therapeutic Chemical) defined WHOCC cid Compound ID found PubChem. unique identifier. atc_group Official pharmacological subgroup (3rd level ATC code) defined WHOCC synonyms Synonyms (often trade names) drug, found PubChem based compound ID oral_ddd Defined Daily Dose (DDD), oral treatment oral_units Units oral_ddd iv_ddd Defined Daily Dose (DDD), parenteral treatment iv_units Units iv_ddd loinc codes associated name antiviral drug Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Use av_loinc() retrieve quickly, see av_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology (WHOCC): https://atcddd.fhi./atc_ddd_index/ Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Accessed https://loinc.org October 19th, 2023. European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Properties based ATC code available ATC available. properties : atc_group1, atc_group2, oral_ddd, oral_units, iv_ddd iv_units. Synonyms (.e. trade names) derived PubChem Compound ID (column cid) consequently available CID available.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Like data sets package, data sets publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"","code":"antibiotics #> # A tibble: 487 × 14 #>    ab        cid name   group atc   atc_group1 atc_group2 abbreviations synonyms #>    <ab>    <dbl> <chr>  <chr> <lis> <chr>      <chr>      <list>        <named > #>  1 AMA      4649 4-ami… Anti… <chr> Drugs for… Aminosali… <chr [1]>     <chr>    #>  2 ACM   6450012 Acety… Macr… <chr> NA         NA         <chr [1]>     <chr>    #>  3 ASP  49787020 Acety… Macr… <chr> NA         NA         <chr [1]>     <chr>    #>  4 ALS      8954 Aldes… Othe… <chr> Drugs for… Drugs for… <chr [1]>     <chr>    #>  5 AMK     37768 Amika… Amin… <chr> Aminoglyc… Other ami… <chr [5]>     <chr>    #>  6 AKF        NA Amika… Amin… <chr> NA         NA         <chr [1]>     <chr>    #>  7 AMO     54260 Amoro… Anti… <chr> Antifunga… Other ant… <chr [1]>     <chr>    #>  8 AMX     33613 Amoxi… Beta… <chr> Beta-lact… Penicilli… <chr [3]>     <chr>    #>  9 AMC  23665637 Amoxi… Beta… <chr> Beta-lact… Combinati… <chr [5]>     <chr>    #> 10 AXS    465441 Amoxi… Beta… <chr> NA         NA         <chr [1]>     <chr>    #> # ℹ 477 more rows #> # ℹ 5 more variables: oral_ddd <dbl>, oral_units <chr>, iv_ddd <dbl>, #> #   iv_units <chr>, loinc <list> antivirals #> # A tibble: 120 × 11 #>    av       name      atc      cid atc_group synonyms oral_ddd oral_units iv_ddd #>    <av>     <chr>     <chr>  <dbl> <chr>     <list>      <dbl> <chr>       <dbl> #>  1 ABA      Abacavir  J05A… 4.41e5 Nucleosi… <chr>         0.6 g              NA #>  2 ACI      Aciclovir J05A… 1.35e8 Nucleosi… <chr>         4   g               4 #>  3 ADD      Adefovir… J05A… 6.09e4 Nucleosi… <chr>        10   mg             NA #>  4 AME      Amenamev… J05A… 1.14e7 Other an… <chr>         0.4 g              NA #>  5 AMP      Amprenav… J05A… 6.50e4 Protease… <chr>         1.2 g              NA #>  6 ASU      Asunapre… J05A… 1.61e7 Antivira… <chr>         0.2 g              NA #>  7 ATA      Atazanav… J05A… 1.48e5 Protease… <chr>         0.3 g              NA #>  8 ATA+COBI Atazanav… J05A… 8.66e7 Antivira… <chr>        NA   NA             NA #>  9 ATA+RIT  Atazanav… J05A… 2.51e7 Antivira… <chr>         0.3 g              NA #> 10 BAM      Baloxavi… J05A… 1.24e8 Other an… <chr>        40   mg             NA #> # ℹ 110 more rows #> # ℹ 2 more variables: iv_units <chr>, loinc <list>"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":null,"dir":"Reference","previous_headings":"","what":"Antimicrobial Selectors — antimicrobial_selectors","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"functions allow filtering rows selecting columns based antimicrobial test results specific antimicrobial class group, without need define columns antimicrobial abbreviations. short, column name resembles antimicrobial drug, picked functions matches pharmaceutical class: \"cefazolin\", \"kefzol\", \"CZO\" \"J01DB04\" picked using:","code":"library(dplyr) my_data_with_all_these_columns %>%   select(cephalosporins())"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"","code":"amr_class(amr_class, only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  amr_selector(filter, only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  aminoglycosides(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  aminopenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  antifungals(only_sir_columns = FALSE, return_all = TRUE, ...)  antimycobacterials(only_sir_columns = FALSE, return_all = TRUE, ...)  betalactams(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  betalactams_with_inhibitor(only_sir_columns = FALSE, return_all = TRUE,   ...)  carbapenems(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  cephalosporins(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_1st(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_2nd(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_3rd(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_4th(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_5th(only_sir_columns = FALSE, return_all = TRUE, ...)  fluoroquinolones(only_sir_columns = FALSE, return_all = TRUE, ...)  glycopeptides(only_sir_columns = FALSE, return_all = TRUE, ...)  isoxazolylpenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  lincosamides(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  lipoglycopeptides(only_sir_columns = FALSE, return_all = TRUE, ...)  macrolides(only_sir_columns = FALSE, return_all = TRUE, ...)  monobactams(only_sir_columns = FALSE, return_all = TRUE, ...)  nitrofurans(only_sir_columns = FALSE, return_all = TRUE, ...)  oxazolidinones(only_sir_columns = FALSE, return_all = TRUE, ...)  penicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  phenicols(only_sir_columns = FALSE, return_all = TRUE, ...)  polymyxins(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  quinolones(only_sir_columns = FALSE, return_all = TRUE, ...)  rifamycins(only_sir_columns = FALSE, return_all = TRUE, ...)  streptogramins(only_sir_columns = FALSE, return_all = TRUE, ...)  tetracyclines(only_sir_columns = FALSE, return_all = TRUE, ...)  trimethoprims(only_sir_columns = FALSE, return_all = TRUE, ...)  ureidopenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  administrable_per_os(only_sir_columns = FALSE, return_all = TRUE, ...)  administrable_iv(only_sir_columns = FALSE, return_all = TRUE, ...)  not_intrinsic_resistant(only_sir_columns = FALSE, col_mo = NULL,   version_expertrules = 3.3, ...)"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"amr_class antimicrobial class part , \"carba\" \"carbapenems\". columns group, atc_group1 atc_group2 antibiotics data set searched (case-insensitive) value. only_sir_columns logical indicate whether columns class sir must selected (default FALSE), see .sir() only_treatable logical indicate whether antimicrobial drugs excluded laboratory tests (default TRUE), gentamicin-high (GEH) imipenem/EDTA (IPE) return_all logical indicate whether matched columns must returned (default TRUE). FALSE, first unique antimicrobial returned, e.g. columns \"genta\" \"gentamicin\" exist data, first hit gentamicin returned. ... ignored, place allow future extensions filter expression evaluated antibiotics data set, name %like% \"trim\" col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). version_expertrules version number use EUCAST Expert Rules Intrinsic Resistance guideline. Can \"3.3\", \"3.2\", \"3.1\".","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"used inside selecting filtering, returns character vector column names, additional class \"amr_selector\". used individually, returns 'ab' vector possible antimicrobials function able select filter.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"functions can used data set calls selecting columns filtering rows. work base R, Tidyverse, data.table. heavily inspired Tidyverse selection helpers everything(), limited dplyr verbs. Nonetheless, convenient use dplyr functions select(), filter() summarise(), see Examples. selectors can also used tidymodels packages recipe parsnip. See info tutorial using AMR functions predictive modelling. columns data functions called searched known antimicrobial names, abbreviations, brand names, codes (ATC, EARS-Net, , etc.) according antibiotics data set. means selector aminoglycosides() pick column names like 'gen', 'genta', 'J01GB03', 'tobra', 'Tobracin', etc. amr_class() function can used filter/select manually defined antimicrobial class. searches results antibiotics data set within columns group, atc_group1 atc_group2. amr_selector() function can used internally filter antibiotics data set results, see Examples. allows filtering (part ) certain name, /group name even minimum DDDs oral treatment. function yields highest flexibility, also least user-friendly, since requires hard-coded filter set. administrable_per_os() administrable_iv() functions also rely antibiotics data set - antimicrobials matched DDD (defined daily dose) resp. oral IV treatment available antibiotics data set. not_intrinsic_resistant() function can used select antimicrobials pose intrinsic resistance microorganisms data set. example, data set contains microorganism codes names E. coli K. pneumoniae contains column \"vancomycin\", column removed (rather, unselected) using function. currently applies 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021) determine intrinsic resistance, using eucast_rules() function internally. determination, function quite slow terms performance.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"full-list-of-supported-antimicrobial-classes","dir":"Reference","previous_headings":"","what":"Full list of supported (antimicrobial) classes","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"aminoglycosides() can select:  amikacin (AMK), amikacin/fosfomycin (AKF), apramycin (APR), arbekacin (ARB), astromicin (AST), bekanamycin (BEK), dibekacin (DKB), framycetin (FRM), gentamicin (GEN), gentamicin-high (GEH), habekacin (HAB), hygromycin (HYG), isepamicin (ISE), kanamycin (KAN), kanamycin-high (KAH), kanamycin/cephalexin (KAC), micronomicin (MCR), neomycin (NEO), netilmicin (NET), pentisomicin (PIM), plazomicin (PLZ), propikacin (PKA), ribostamycin (RST), sisomicin (SIS), streptoduocin (STR), streptomycin (STR1), streptomycin-high (STH), tobramycin (TOB), tobramycin-high (TOH) aminopenicillins() can select:  amoxicillin (AMX) ampicillin (AMP) antifungals() can select:  amorolfine (AMO), amphotericin B (AMB), amphotericin B-high (AMH), anidulafungin (ANI), butoconazole (), caspofungin (CAS), ciclopirox (CIX), clotrimazole (CTR), econazole (ECO), fluconazole (FLU), flucytosine (FCT), fosfluconazole (FFL), griseofulvin (GRI), hachimycin (HCH), ibrexafungerp (IBX), isavuconazole (ISV), isoconazole (ISO), itraconazole (ITR), ketoconazole (KET), manogepix (MGX), micafungin (MIF), miconazole (MCZ), nystatin (NYS), oteseconazole (OTE), pimaricin (PMR), posaconazole (POS), rezafungin (RZF), ribociclib (RBC), sulconazole (SUC), terbinafine (TRB), terconazole (TRC), voriconazole (VOR) antimycobacterials() can select:  4-aminosalicylic acid (AMA), calcium aminosalicylate (CLA), capreomycin (CAP), clofazimine (CLF), delamanid (DLM), enviomycin (ENV), ethambutol (ETH), ethambutol/isoniazid (ETI), ethionamide (ETI1), isoniazid (INH), isoniazid/sulfamethoxazole/trimethoprim/pyridoxine (IST), morinamide (MRN), p-aminosalicylic acid (PAS), pretomanid (PMD), protionamide (PTH), pyrazinamide (PZA), rifabutin (RIB), rifampicin (RIF), rifampicin/ethambutol/isoniazid (REI), rifampicin/isoniazid (RFI), rifampicin/pyrazinamide/ethambutol/isoniazid (RPEI), rifampicin/pyrazinamide/isoniazid (RPI), rifamycin (RFM), rifapentine (RFP), simvastatin/fenofibrate (SMF), sodium aminosalicylate (SDA), streptomycin/isoniazid (STI), terizidone (TRZ), thioacetazone (TAT), thioacetazone/isoniazid (THI1), tiocarlide (TCR), viomycin (VIO) betalactams() can select:  amoxicillin (AMX), amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin (AMP), ampicillin/sulbactam (SAM), apalcillin (APL), aspoxicillin (APX), azidocillin (AZD), azlocillin (AZL), aztreonam (ATM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), bacampicillin (BAM), benzathine benzylpenicillin (BNB), benzathine phenoxymethylpenicillin (BNP), benzylpenicillin (PEN), biapenem (BIA), carbenicillin (CRB), carindacillin (CRN), carumonam (CAR), cefacetrile (CAC), cefaclor (CEC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefamandole (MAN), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetamet (CAT), cefetamet pivoxil (CPI), cefetecol (CCL), cefetrizole (CZL), cefiderocol (FDC), cefixime (CFM), cefmenoxime (CMX), cefmetazole (CMZ), cefodizime (DIZ), cefonicid (CID), cefoperazone (CFP), cefoperazone/sulbactam (CSL), ceforanide (CND), cefoselis (CSE), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotetan (CTT), cefotiam (CTF), cefotiam hexetil (CHE), cefovecin (FOV), cefoxitin (FOX), cefoxitin screening (FOX1), cefozopran (ZOP), cefpimizole (CFZ), cefpiramide (CPM), cefpirome (CPO), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefprozil (CPR), cefquinome (CEQ), cefroxadine (CRD), cefsulodin (CFS), cefsumide (CSU), ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftezole (CTL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), cefuroxime (CXM), cefuroxime axetil (CXA), cephradine (CED), ciclacillin (CIC), clometocillin (CLM), cloxacillin (CLO), dicloxacillin (DIC), doripenem (DOR), epicillin (EPC), ertapenem (ETP), flucloxacillin (FLC), hetacillin (HET), imipenem (IPM), imipenem/EDTA (IPE), imipenem/relebactam (IMR), latamoxef (LTM), lenampicillin (LEN), loracarbef (LOR), mecillinam (MEC), meropenem (MEM), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), metampicillin (MTM), meticillin (MET), mezlocillin (MEZ), mezlocillin/sulbactam (MSU), nafcillin (NAF), oxacillin (OXA), panipenem (PAN), penamecillin (PNM), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), pheneticillin (PHE), phenoxymethylpenicillin (PHN), piperacillin (PIP), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), piridicillin (PRC), pivampicillin (PVM), pivmecillinam (PME), procaine benzylpenicillin (PRB), propicillin (PRP), razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), sarmoxicillin (SRX), sulbenicillin (SBC), sultamicillin (SLT6), talampicillin (TAL), tebipenem (TBP), temocillin (TEM), ticarcillin (TIC), ticarcillin/clavulanic acid (TCC), tigemonam (TMN) betalactams_with_inhibitor() can select:  amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin/sulbactam (SAM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefoperazone/sulbactam (CSL), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefpodoxime/clavulanic acid (CDC), ceftaroline/avibactam (CPA), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), ceftolozane/tazobactam (CZT), ceftriaxone/beta-lactamase inhibitor (CEB), imipenem/relebactam (IMR), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), mezlocillin/sulbactam (MSU), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), ticarcillin/clavulanic acid (TCC) carbapenems() can select:  biapenem (BIA), doripenem (DOR), ertapenem (ETP), imipenem (IPM), imipenem/EDTA (IPE), imipenem/relebactam (IMR), meropenem (MEM), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), panipenem (PAN), razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), tebipenem (TBP) cephalosporins() can select:  cefacetrile (CAC), cefaclor (CEC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefamandole (MAN), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetamet (CAT), cefetamet pivoxil (CPI), cefetecol (CCL), cefetrizole (CZL), cefiderocol (FDC), cefixime (CFM), cefmenoxime (CMX), cefmetazole (CMZ), cefodizime (DIZ), cefonicid (CID), cefoperazone (CFP), cefoperazone/sulbactam (CSL), ceforanide (CND), cefoselis (CSE), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotetan (CTT), cefotiam (CTF), cefotiam hexetil (CHE), cefovecin (FOV), cefoxitin (FOX), cefoxitin screening (FOX1), cefozopran (ZOP), cefpimizole (CFZ), cefpiramide (CPM), cefpirome (CPO), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefprozil (CPR), cefquinome (CEQ), cefroxadine (CRD), cefsulodin (CFS), cefsumide (CSU), ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftezole (CTL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), cefuroxime (CXM), cefuroxime axetil (CXA), cephradine (CED), latamoxef (LTM), loracarbef (LOR) cephalosporins_1st() can select:  cefacetrile (CAC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefroxadine (CRD), ceftezole (CTL), cephradine (CED) cephalosporins_2nd() can select:  cefaclor (CEC), cefamandole (MAN), cefmetazole (CMZ), cefonicid (CID), ceforanide (CND), cefotetan (CTT), cefotiam (CTF), cefoxitin (FOX), cefoxitin screening (FOX1), cefprozil (CPR), cefuroxime (CXM), cefuroxime axetil (CXA), loracarbef (LOR) cephalosporins_3rd() can select:  cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefetamet (CAT), cefetamet pivoxil (CPI), cefixime (CFM), cefmenoxime (CMX), cefodizime (DIZ), cefoperazone (CFP), cefoperazone/sulbactam (CSL), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotiam hexetil (CHE), cefovecin (FOV), cefpimizole (CFZ), cefpiramide (CPM), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefsulodin (CFS), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), latamoxef (LTM) cephalosporins_4th() can select:  cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetecol (CCL), cefoselis (CSE), cefozopran (ZOP), cefpirome (CPO), cefquinome (CEQ) cephalosporins_5th() can select:  ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT) fluoroquinolones() can select:  besifloxacin (BES), ciprofloxacin (CIP), clinafloxacin (CLX), danofloxacin (DAN), delafloxacin (DFX), difloxacin (DIF), enoxacin (ENX), enrofloxacin (ENR), finafloxacin (FIN), fleroxacin (FLE), garenoxacin (GRN), gatifloxacin (GAT), gemifloxacin (GEM), grepafloxacin (GRX), lascufloxacin (LSC), levofloxacin (LVX), levonadifloxacin (LND), lomefloxacin (LOM), marbofloxacin (MAR), metioxate (MXT), miloxacin (MIL), moxifloxacin (MFX), nadifloxacin (NAD), nifuroquine (NIF), norfloxacin (), ofloxacin (OFX), orbifloxacin (ORB), pazufloxacin (PAZ), pefloxacin (PEF), pradofloxacin (PRA), premafloxacin (PRX), prulifloxacin (PRU), rufloxacin (RFL), sarafloxacin (SAR), sitafloxacin (SIT), sparfloxacin (SPX), temafloxacin (TMX), tilbroquinol (TBQ), tioxacin (TXC), tosufloxacin (TFX), trovafloxacin (TVA) glycopeptides() can select:  avoparcin (AVO), dalbavancin (DAL), norvancomycin (NVA), oritavancin (ORI), ramoplanin (RAM), teicoplanin (TEC), teicoplanin-macromethod (TCM), telavancin (TLV), vancomycin (VAN), vancomycin-macromethod (VAM) isoxazolylpenicillins() can select:  cloxacillin (CLO), dicloxacillin (DIC), flucloxacillin (FLC), meticillin (MET), oxacillin (OXA) lincosamides() can select:  clindamycin (CLI), lincomycin (LIN), pirlimycin (PRL) lipoglycopeptides() can select:  dalbavancin (DAL), oritavancin (ORI), telavancin (TLV) macrolides() can select:  acetylmidecamycin (ACM), acetylspiramycin (ASP), azithromycin (AZM), clarithromycin (CLR), dirithromycin (DIR), erythromycin (ERY), flurithromycin (FLR1), gamithromycin (GAM), josamycin (JOS), kitasamycin (KIT), meleumycin (MEL), midecamycin (MID), miocamycin (MCM), nafithromycin (ZWK), oleandomycin (OLE), rokitamycin (ROK), roxithromycin (RXT), solithromycin (SOL), spiramycin (SPI), telithromycin (TLT), tildipirosin (TIP), tilmicosin (TIL), troleandomycin (TRL), tulathromycin (TUL), tylosin (TYL), tylvalosin (TYL1) monobactams() can select:  aztreonam (ATM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), carumonam (CAR), tigemonam (TMN) nitrofurans() can select:  furazidin (FUR), furazolidone (FRZ), nifurtoinol (NFR), nitrofurantoin (NIT), nitrofurazone (NIZ) oxazolidinones() can select:  cadazolid (CDZ), cycloserine (CYC), linezolid (LNZ), tedizolid (TZD), thiacetazone (THA) penicillins() can select:  amoxicillin (AMX), amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin (AMP), ampicillin/sulbactam (SAM), apalcillin (APL), aspoxicillin (APX), azidocillin (AZD), azlocillin (AZL), bacampicillin (BAM), benzathine benzylpenicillin (BNB), benzathine phenoxymethylpenicillin (BNP), benzylpenicillin (PEN), carbenicillin (CRB), carindacillin (CRN), ciclacillin (CIC), clometocillin (CLM), cloxacillin (CLO), dicloxacillin (DIC), epicillin (EPC), flucloxacillin (FLC), hetacillin (HET), lenampicillin (LEN), mecillinam (MEC), metampicillin (MTM), meticillin (MET), mezlocillin (MEZ), mezlocillin/sulbactam (MSU), nafcillin (NAF), oxacillin (OXA), penamecillin (PNM), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), pheneticillin (PHE), phenoxymethylpenicillin (PHN), piperacillin (PIP), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), piridicillin (PRC), pivampicillin (PVM), pivmecillinam (PME), procaine benzylpenicillin (PRB), propicillin (PRP), sarmoxicillin (SRX), sulbenicillin (SBC), sultamicillin (SLT6), talampicillin (TAL), temocillin (TEM), ticarcillin (TIC), ticarcillin/clavulanic acid (TCC) phenicols() can select:  chloramphenicol (CHL), florfenicol (FLR), thiamphenicol (THI) polymyxins() can select:  colistin (COL), polymyxin B (PLB), polymyxin B/polysorbate 80 (POP) quinolones() can select:  besifloxacin (BES), cinoxacin (CIN), ciprofloxacin (CIP), ciprofloxacin/metronidazole (CIM), ciprofloxacin/ornidazole (CIO), ciprofloxacin/tinidazole (CIT), clinafloxacin (CLX), danofloxacin (DAN), delafloxacin (DFX), difloxacin (DIF), enoxacin (ENX), enrofloxacin (ENR), finafloxacin (FIN), fleroxacin (FLE), flumequine (FLM), garenoxacin (GRN), gatifloxacin (GAT), gemifloxacin (GEM), grepafloxacin (GRX), lascufloxacin (LSC), levofloxacin (LVX), levonadifloxacin (LND), lomefloxacin (LOM), marbofloxacin (MAR), metioxate (MXT), miloxacin (MIL), moxifloxacin (MFX), nadifloxacin (NAD), nalidixic acid (NAL), nemonoxacin (NEM), nifuroquine (NIF), nitroxoline (NTR), norfloxacin (), ofloxacin (OFX), orbifloxacin (ORB), oxolinic acid (OXO), pazufloxacin (PAZ), pefloxacin (PEF), pipemidic acid (PPA), piromidic acid (PIR), pradofloxacin (PRA), premafloxacin (PRX), prulifloxacin (PRU), rosoxacin (ROS), rufloxacin (RFL), sarafloxacin (SAR), sitafloxacin (SIT), sparfloxacin (SPX), temafloxacin (TMX), tilbroquinol (TBQ), tioxacin (TXC), tosufloxacin (TFX), trovafloxacin (TVA) rifamycins() can select:  rifabutin (RIB), rifampicin (RIF), rifampicin/ethambutol/isoniazid (REI), rifampicin/isoniazid (RFI), rifampicin/pyrazinamide/ethambutol/isoniazid (RPEI), rifampicin/pyrazinamide/isoniazid (RPI), rifamycin (RFM), rifapentine (RFP) streptogramins() can select:  pristinamycin (PRI) quinupristin/dalfopristin (QDA) tetracyclines() can select:  cetocycline (CTO), chlortetracycline (CTE), clomocycline (CLM1), demeclocycline (DEM), doxycycline (DOX), eravacycline (ERV), lymecycline (LYM), metacycline (MTC), minocycline (MNO), omadacycline (OMC), oxytetracycline (OXY), penimepicycline (PNM1), rolitetracycline (RLT), sarecycline (SRC), tetracycline (TCY), tigecycline (TGC) trimethoprims() can select:  brodimoprim (BDP), sulfadiazine (SDI), sulfadiazine/tetroxoprim (SLT), sulfadiazine/trimethoprim (SLT1), sulfadimethoxine (SUD), sulfadimidine (SDM), sulfadimidine/trimethoprim (SLT2), sulfafurazole (SLF), sulfaisodimidine (SLF1), sulfalene (SLF2), sulfamazone (SZO), sulfamerazine (SLF3), sulfamerazine/trimethoprim (SLT3), sulfamethizole (SLF4), sulfamethoxazole (SMX), sulfamethoxypyridazine (SLF5), sulfametomidine (SLF6), sulfametoxydiazine (SLF7), sulfametrole/trimethoprim (SLT4), sulfamoxole (SLF8), sulfamoxole/trimethoprim (SLT5), sulfanilamide (SLF9), sulfaperin (SLF10), sulfaphenazole (SLF11), sulfapyridine (SLF12), sulfathiazole (SUT), sulfathiourea (SLF13), trimethoprim (TMP), trimethoprim/sulfamethoxazole (SXT) ureidopenicillins() can select:  azlocillin (AZL), mezlocillin (MEZ), piperacillin (PIP), piperacillin/tazobactam (TZP)","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …   # you can use the selectors separately to retrieve all possible antimicrobials: carbapenems() #> ℹ The function carbapenems() should be used inside a dplyr verb or #>   data.frame call, e.g.: #>   • your_data %>% select(carbapenems()) #>   • your_data %>% select(column_a, column_b, carbapenems()) #>   • your_data %>% filter(any(carbapenems() == \"R\")) #>   • your_data[, carbapenems()] #>   • your_data[, c(\"column_a\", \"column_b\", carbapenems())] #>  #> Now returning a vector of all possible antimicrobials that carbapenems() #> can select. #> Class 'ab' #>  [1] BIA DOR ETP IMR IPE IPM MEM MEV MNC PAN RIA RIT RZM TBP   # Though they are primarily intended to use for selections and filters. # Examples sections below are split into 'dplyr', 'base R', and 'data.table':  # \\donttest{ if (FALSE) { # \\dontrun{ # dplyr -------------------------------------------------------------------  library(dplyr, warn.conflicts = FALSE)  example_isolates %>% select(carbapenems())  # select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB' example_isolates %>% select(mo, aminoglycosides())  # you can combine selectors like you are used with tidyverse # e.g., for betalactams, but not the ones with an enzyme inhibitor: example_isolates %>% select(betalactams(), -betalactams_with_inhibitor())  # select only antimicrobials with DDDs for oral treatment example_isolates %>% select(administrable_per_os())  # get AMR for all aminoglycosides e.g., per ward: example_isolates %>%   group_by(ward) %>%   summarise(across(aminoglycosides(),                    resistance))  # You can combine selectors with '&' to be more specific: example_isolates %>%   select(penicillins() & administrable_per_os())  # get AMR for only drugs that matter - no intrinsic resistance: example_isolates %>%   filter(mo_genus() %in% c(\"Escherichia\", \"Klebsiella\")) %>%   group_by(ward) %>%   summarise_at(not_intrinsic_resistant(),                resistance)  # get susceptibility for antimicrobials whose name contains \"trim\": example_isolates %>%   filter(first_isolate()) %>%   group_by(ward) %>%   summarise(across(amr_selector(name %like% \"trim\"), susceptibility))  # this will select columns 'IPM' (imipenem) and 'MEM' (meropenem): example_isolates %>%   select(carbapenems())  # this will select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB': example_isolates %>%   select(mo, aminoglycosides())  # any() and all() work in dplyr's filter() too: example_isolates %>%   filter(     any(aminoglycosides() == \"R\"),     all(cephalosporins_2nd() == \"R\")   )  # also works with c(): example_isolates %>%   filter(any(c(carbapenems(), aminoglycosides()) == \"R\"))  # not setting any/all will automatically apply all(): example_isolates %>%   filter(aminoglycosides() == \"R\")  # this will select columns 'mo' and all antimycobacterial drugs ('RIF'): example_isolates %>%   select(mo, amr_class(\"mycobact\"))  # get bug/drug combinations for only glycopeptides in Gram-positives: example_isolates %>%   filter(mo_is_gram_positive()) %>%   select(mo, glycopeptides()) %>%   bug_drug_combinations() %>%   format()  data.frame(   some_column = \"some_value\",   J01CA01 = \"S\" ) %>% # ATC code of ampicillin   select(penicillins()) # only the 'J01CA01' column will be selected  # with recent versions of dplyr, this is all equal: x <- example_isolates[carbapenems() == \"R\", ] y <- example_isolates %>% filter(carbapenems() == \"R\") z <- example_isolates %>% filter(if_all(carbapenems(), ~ .x == \"R\")) identical(x, y) && identical(y, z)  } # } # base R ------------------------------------------------------------------  # select columns 'IPM' (imipenem) and 'MEM' (meropenem) example_isolates[, carbapenems()] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 2,000 × 2 #>    IPM   MEM   #>    <sir> <sir> #>  1   NA    NA  #>  2   NA    NA  #>  3   NA    NA  #>  4   NA    NA  #>  5   NA    NA  #>  6   NA    NA  #>  7   NA    NA  #>  8   NA    NA  #>  9   NA    NA  #> 10   NA    NA  #> # ℹ 1,990 more rows  # select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB' example_isolates[, c(\"mo\", aminoglycosides())] #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 2,000 × 5 #>    mo           GEN   TOB   AMK   KAN   #>    <mo>         <sir> <sir> <sir> <sir> #>  1 B_ESCHR_COLI   NA    NA    NA    NA  #>  2 B_ESCHR_COLI   NA    NA    NA    NA  #>  3 B_STPHY_EPDR   NA    NA    NA    NA  #>  4 B_STPHY_EPDR   NA    NA    NA    NA  #>  5 B_STPHY_EPDR   NA    NA    NA    NA  #>  6 B_STPHY_EPDR   NA    NA    NA    NA  #>  7 B_STPHY_AURS   NA    S     NA    NA  #>  8 B_STPHY_AURS   NA    S     NA    NA  #>  9 B_STPHY_EPDR   NA    NA    NA    NA  #> 10 B_STPHY_EPDR   NA    NA    NA    NA  #> # ℹ 1,990 more rows  # select only antimicrobials with DDDs for oral treatment example_isolates[, administrable_per_os()] #> ℹ For administrable_per_os() using columns 'OXA' (oxacillin), 'FLC' #>   (flucloxacillin), 'AMX' (amoxicillin), 'AMC' (amoxicillin/clavulanic acid), #>   'AMP' (ampicillin), 'CXM' (cefuroxime), 'KAN' (kanamycin), 'TMP' #>   (trimethoprim), 'NIT' (nitrofurantoin), 'FOS' (fosfomycin), 'LNZ' #>   (linezolid), 'CIP' (ciprofloxacin), 'MFX' (moxifloxacin), 'VAN' #>   (vancomycin), 'TCY' (tetracycline), 'DOX' (doxycycline), 'ERY' #>   (erythromycin), 'CLI' (clindamycin), 'AZM' (azithromycin), 'MTR' #>   (metronidazole), 'CHL' (chloramphenicol), 'COL' (colistin), and 'RIF' #>   (rifampicin) #> # A tibble: 2,000 × 23 #>    OXA   FLC   AMX   AMC   AMP   CXM   KAN   TMP   NIT   FOS   LNZ   CIP   MFX   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA    I     NA    R     NA    NA    R     NA    NA  #>  2   NA    NA    NA    I     NA    I     NA    R     NA    NA    R     NA    NA  #>  3   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    NA    NA  #>  4   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    NA    NA  #>  5   NA    R     NA    NA    NA    R     NA    R     NA    NA    NA    NA    NA  #>  6   NA    R     NA    NA    NA    R     NA    R     NA    NA    NA    NA    NA  #>  7   NA    S     R     S     R     S     NA    R     NA    NA    NA    NA    NA  #>  8   NA    S     R     S     R     S     NA    R     NA    NA    NA    NA    NA  #>  9   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    S     NA  #> 10   NA    S     NA    NA    NA    S     NA    S     NA    NA    NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 10 more variables: VAN <sir>, TCY <sir>, DOX <sir>, ERY <sir>, CLI <sir>, #> #   AZM <sir>, MTR <sir>, CHL <sir>, COL <sir>, RIF <sir>  # filter using any() or all() example_isolates[any(carbapenems() == \"R\"), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 55 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  4 2005-04-21 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  5 2005-04-22 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  8 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  9 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> 10 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> # ℹ 45 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … subset(example_isolates, any(carbapenems() == \"R\")) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 55 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  4 2005-04-21 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  5 2005-04-22 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  8 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  9 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> 10 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> # ℹ 45 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter on any or all results in the carbapenem columns (i.e., IPM, MEM): example_isolates[any(carbapenems()), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Filtering any of columns 'IPM' and 'MEM' to contain value \"S\", \"I\" or \"R\" #> # A tibble: 962 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  4 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  5 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  6 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  7 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  8 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #>  9 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #> 10 2002-03-08 4FC193     69 M      Clinical B_ESCHR_COLI   R     NA    NA    R   #> # ℹ 952 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … example_isolates[all(carbapenems()), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Filtering all of columns 'IPM' and 'MEM' to contain value \"S\", \"I\" or \"R\" #> # A tibble: 756 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-04-14 F30196     73 M      Outpat… B_STRPT_GRPB    S     NA    S     S   #>  2 2003-04-08 114570     74 M      ICU     B_STRPT_PYGN    S     NA    S     S   #>  3 2003-04-08 114570     74 M      ICU     B_STRPT_GRPA    S     NA    S     S   #>  4 2003-04-08 114570     74 M      ICU     B_STRPT_GRPA    S     NA    S     S   #>  5 2003-08-14 F71508      0 F      Clinic… B_STRPT_GRPB    S     NA    S     S   #>  6 2003-10-16 650870     63 F      ICU     B_ESCHR_COLI    R     NA    NA    R   #>  7 2003-10-20 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-10-20 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  9 2003-11-04 2FC253     87 F      ICU     B_ESCHR_COLI    R     NA    NA    NA  #> 10 2003-11-04 2FC253     87 F      ICU     B_ESCHR_COLI    R     NA    NA    NA  #> # ℹ 746 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter with multiple antimicrobial selectors using c() example_isolates[all(c(carbapenems(), aminoglycosides()) == \"R\"), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 26 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  2 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  4 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  5 2008-12-06 501361     43 F      Clinical B_STNTR_MLTP   R     NA    NA    R   #>  6 2011-05-09 207325     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2012-03-12 582258     80 M      ICU      B_STPHY_CONS   R     R     R     R   #>  8 2012-05-19 C25552     89 F      Outpati… B_STPHY_CONS   R     R     R     R   #>  9 2012-07-17 F05015     83 M      ICU      B_STPHY_CONS   R     R     R     R   #> 10 2012-07-20 404299     66 F      Clinical B_STPHY_CONS   R     R     R     R   #> # ℹ 16 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter + select in one go: get penicillins in carbapenem-resistant strains example_isolates[any(carbapenems() == \"R\"), penicillins()] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #> # A tibble: 55 × 7 #>    PEN   OXA   FLC   AMX   AMC   AMP   TZP   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    NA    NA    NA    NA  #>  2   NA    NA    NA    NA    NA    NA    NA  #>  3   R     NA    NA    R     R     R     R   #>  4   NA    NA    NA    NA    NA    NA    R   #>  5   NA    NA    NA    NA    NA    NA    R   #>  6   NA    NA    NA    NA    NA    NA    R   #>  7   NA    NA    NA    NA    NA    NA    R   #>  8   NA    NA    NA    NA    NA    NA    R   #>  9   R     NA    NA    NA    S     NA    S   #> 10   R     NA    NA    NA    S     NA    S   #> # ℹ 45 more rows  # You can combine selectors with '&' to be more specific. For example, # penicillins() would select benzylpenicillin ('peni G') and # administrable_per_os() would select erythromycin. Yet, when combined these # drugs are both omitted since benzylpenicillin is not administrable per os # and erythromycin is not a penicillin: example_isolates[, penicillins() & administrable_per_os()] #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #> ℹ For administrable_per_os() using columns 'OXA' (oxacillin), 'FLC' #>   (flucloxacillin), 'AMX' (amoxicillin), 'AMC' (amoxicillin/clavulanic acid), #>   'AMP' (ampicillin), 'CXM' (cefuroxime), 'KAN' (kanamycin), 'TMP' #>   (trimethoprim), 'NIT' (nitrofurantoin), 'FOS' (fosfomycin), 'LNZ' #>   (linezolid), 'CIP' (ciprofloxacin), 'MFX' (moxifloxacin), 'VAN' #>   (vancomycin), 'TCY' (tetracycline), 'DOX' (doxycycline), 'ERY' #>   (erythromycin), 'CLI' (clindamycin), 'AZM' (azithromycin), 'MTR' #>   (metronidazole), 'CHL' (chloramphenicol), 'COL' (colistin), and 'RIF' #>   (rifampicin) #> # A tibble: 2,000 × 5 #>    OXA   FLC   AMX   AMC   AMP   #>    <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA  #>  2   NA    NA    NA    I     NA  #>  3   NA    R     NA    NA    NA  #>  4   NA    R     NA    NA    NA  #>  5   NA    R     NA    NA    NA  #>  6   NA    R     NA    NA    NA  #>  7   NA    S     R     S     R   #>  8   NA    S     R     S     R   #>  9   NA    R     NA    NA    NA  #> 10   NA    S     NA    NA    NA  #> # ℹ 1,990 more rows  # amr_selector() applies a filter in the `antibiotics` data set and is thus # very flexible. For instance, to select antimicrobials with an oral DDD # of at least 1 gram: example_isolates[, amr_selector(oral_ddd > 1 & oral_units == \"g\")] #> ℹ For amr_selector(oral_ddd > 1 & oral_units == \"g\") using columns 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'KAN' (kanamycin), 'FOS' #>   (fosfomycin), 'LNZ' (linezolid), 'VAN' (vancomycin), 'ERY' (erythromycin), #>   'CLI' (clindamycin), 'MTR' (metronidazole), and 'CHL' (chloramphenicol) #> # A tibble: 2,000 × 13 #>    OXA   FLC   AMX   AMC   AMP   KAN   FOS   LNZ   VAN   ERY   CLI   MTR   CHL   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA    NA    NA    R     R     R     R     NA    NA  #>  2   NA    NA    NA    I     NA    NA    NA    R     R     R     R     NA    NA  #>  3   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  4   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  5   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  6   NA    R     NA    NA    NA    NA    NA    NA    S     R     R     NA    NA  #>  7   NA    S     R     S     R     NA    NA    NA    S     S     NA    NA    NA  #>  8   NA    S     R     S     R     NA    NA    NA    S     S     NA    NA    NA  #>  9   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #> 10   NA    S     NA    NA    NA    NA    NA    NA    S     S     NA    NA    NA  #> # ℹ 1,990 more rows   # data.table --------------------------------------------------------------  # data.table is supported as well, just use it in the same way as with # base R, but add `with = FALSE` if using a single AB selector.  if (require(\"data.table\")) {   dt <- as.data.table(example_isolates)    # this does not work, it returns column *names*   dt[, carbapenems()] } #> Loading required package: data.table #>  #> Attaching package: ‘data.table’ #> The following objects are masked from ‘package:dplyr’: #>  #>     between, first, last #> The following objects are masked from ‘package:AMR’: #>  #>     %like%, like #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> Warning: It should never be needed to print an antimicrobial selector class. Are you #> using data.table? Then add the argument with = FALSE, see our examples at #> ?amr_selector. #> Class 'amr_selector' #> [1] IPM MEM if (require(\"data.table\")) {   # so `with = FALSE` is required   dt[, carbapenems(), with = FALSE] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #>         IPM   MEM #>       <sir> <sir> #>    1:  <NA>  <NA> #>    2:  <NA>  <NA> #>    3:  <NA>  <NA> #>    4:  <NA>  <NA> #>    5:  <NA>  <NA> #>   ---             #> 1996:  <NA>  <NA> #> 1997:     S     S #> 1998:     S     S #> 1999:     S     S #> 2000:     S     S  # for multiple selections or AB selectors, `with = FALSE` is not needed: if (require(\"data.table\")) {   dt[, c(\"mo\", aminoglycosides())] } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #>                 mo   GEN   TOB   AMK   KAN #>               <mo> <sir> <sir> <sir> <sir> #>    1: B_ESCHR_COLI  <NA>  <NA>  <NA>  <NA> #>    2: B_ESCHR_COLI  <NA>  <NA>  <NA>  <NA> #>    3: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>    4: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>    5: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>   ---                                      #> 1996: B_STRPT_PNMN     R     R     R     R #> 1997: B_ESCHR_COLI     S     S     S  <NA> #> 1998: B_STPHY_CONS     S  <NA>  <NA>  <NA> #> 1999: B_ESCHR_COLI     S     S  <NA>  <NA> #> 2000: B_KLBSL_PNMN     S     S  <NA>  <NA> if (require(\"data.table\")) {   dt[, c(carbapenems(), aminoglycosides())] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #>         IPM   MEM   GEN   TOB   AMK   KAN #>       <sir> <sir> <sir> <sir> <sir> <sir> #>    1:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    2:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    3:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    4:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    5:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>   ---                                     #> 1996:  <NA>  <NA>     R     R     R     R #> 1997:     S     S     S     S     S  <NA> #> 1998:     S     S     S  <NA>  <NA>  <NA> #> 1999:     S     S     S     S  <NA>  <NA> #> 2000:     S     S     S     S  <NA>  <NA>  # row filters are also supported: if (require(\"data.table\")) {   dt[any(carbapenems() == \"S\"), ] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #>            date patient   age gender       ward           mo   PEN   OXA   FLC #>          <Date>  <char> <num> <char>     <char>         <mo> <sir> <sir> <sir> #>   1: 2002-01-19  738003    71      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #>   2: 2002-01-19  738003    71      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #>   3: 2002-01-22  F35553    50      M        ICU B_PROTS_MRBL     R  <NA>  <NA> #>   4: 2002-01-22  F35553    50      M        ICU B_PROTS_MRBL     R  <NA>  <NA> #>   5: 2002-02-05  067927    45      F        ICU B_SERRT_MRCS     R  <NA>  <NA> #>  ---                                                                           #> 905: 2005-04-12  D71461    70      M        ICU B_ESCHR_COLI     R  <NA>  <NA> #> 906: 2009-11-12  650870    69      F Outpatient B_ESCHR_COLI     R  <NA>  <NA> #> 907: 2012-06-14  8CBCF2    41      F   Clinical B_STPHY_CONS     R     S     S #> 908: 2012-10-11  175532    78      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #> 909: 2013-11-23  A97263    77      M   Clinical B_KLBSL_PNMN     R  <NA>  <NA> #>        AMX   AMC   AMP   TZP   CZO   FEP   CXM   FOX   CTX   CAZ   CRO   GEN #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S  <NA>     S  <NA> #>   2:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S  <NA>     S  <NA> #>   3:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S     S     S  <NA> #>   4:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S     S     S  <NA> #>   5:     R     R     R  <NA>     R  <NA>     R     R  <NA>  <NA>  <NA>  <NA> #>  ---                                                                         #> 905:     S     S     S     S  <NA>     S     S     S     S     S     S     S #> 906:     S     S     S     S     S     S     S     S     S     S     S     S #> 907:  <NA>     S  <NA>  <NA>     S     S     S     S     S     R     S     S #> 908:     R     S     R     S  <NA>     S     R     R     S     S     S     S #> 909:     R     S     R     S  <NA>     S     S     S     S     S     S     S #>        TOB   AMK   KAN   TMP   SXT   NIT   FOS   LNZ   CIP   MFX   VAN   TEC #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:     S  <NA>  <NA>     S     S  <NA>  <NA>     R  <NA>  <NA>     R     R #>   2:     S  <NA>  <NA>     S     S  <NA>  <NA>     R  <NA>  <NA>     R     R #>   3:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>   4:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>   5:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>  ---                                                                         #> 905:     S     S  <NA>  <NA>     S     S  <NA>     R     S  <NA>     R     R #> 906:     S     S  <NA>     S     S     S  <NA>     R     S  <NA>     R     R #> 907:  <NA>  <NA>  <NA>     S     S  <NA>  <NA>  <NA>     S  <NA>     S  <NA> #> 908:     S  <NA>  <NA>     R     R     R  <NA>     R     R     R     R     R #> 909:     S  <NA>  <NA>     S     S     S  <NA>     R     S  <NA>     R     R #>        TCY   TGC   DOX   ERY   CLI   AZM   IPM   MEM   MTR   CHL   COL   MUP #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:  <NA>  <NA>  <NA>     R     R     R     S  <NA>  <NA>  <NA>  <NA>  <NA> #>   2:  <NA>  <NA>  <NA>     R     R     R     S  <NA>  <NA>  <NA>  <NA>  <NA> #>   3:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>   4:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>   5:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>  ---                                                                         #> 905:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>  <NA>  <NA> #> 906:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>  <NA>  <NA> #> 907:  <NA>  <NA>  <NA>     S     S     S     S     S  <NA>  <NA>     R  <NA> #> 908:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>     S  <NA> #> 909:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>     S  <NA> #>        RIF #>      <sir> #>   1:     R #>   2:     R #>   3:     R #>   4:     R #>   5:     R #>  ---       #> 905:     R #> 906:     R #> 907:  <NA> #> 908:     R #> 909:     R if (require(\"data.table\")) {   dt[any(carbapenems() == \"S\"), penicillins(), with = FALSE] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #>        PEN   OXA   FLC   AMX   AMC   AMP   TZP #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   2:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   3:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   4:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   5:     R  <NA>  <NA>     R     R     R  <NA> #>  ---                                           #> 905:     R  <NA>  <NA>     S     S     S     S #> 906:     R  <NA>  <NA>     S     S     S     S #> 907:     R     S     S  <NA>     S  <NA>  <NA> #> 908:     R  <NA>  <NA>     R     S     R     S #> 909:     R  <NA>  <NA>     R     S     R     S # }"},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to an Antibiotic ID — as.ab","title":"Transform Input to an Antibiotic ID — as.ab","text":"Use function determine antibiotic drug code one antibiotics. data set antibiotics searched abbreviations, official names synonyms (brand names).","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to an Antibiotic ID — as.ab","text":"","code":"as.ab(x, flag_multiple_results = TRUE, info = interactive(), ...)  is.ab(x)"},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to an Antibiotic ID — as.ab","text":"x character vector determine antibiotic ID flag_multiple_results logical indicate whether note printed console probably one antibiotic drug code name can retrieved single input value. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed internal functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to an Antibiotic ID — as.ab","text":"character vector additional class ab","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to an Antibiotic ID — as.ab","text":"entries antibiotics data set three different identifiers: human readable EARS-Net code (column ab, used ECDC WHONET), ATC code (column atc, used ), CID code (column cid, Compound ID, used PubChem). data set contains 5,000 official brand names many different countries, found PubChem. drugs contain multiple ATC codes. properties searched user input. .ab() can correct different forms misspelling: Wrong spelling drug names (\"tobramicin\" \"gentamycin\"), corrects audible similarities f/ph, x/ks, c/z/s, t/th, etc. many vowels consonants Switching two characters (\"mreopenem\", often case clinical data, doctors typed fast) Digitalised paper records, leaving artefacts like 0/o/O (zero O's), B/8, n/r, etc. Use ab_* functions get properties based returned antibiotic ID, see Examples. Note: .ab() ab_* functions may use long regular expression match brand names antimicrobial drugs. may fail systems. can add manual codes considered .ab() ab_* functions, see add_custom_antimicrobials().","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Input to an Antibiotic ID — as.ab","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Transform Input to an Antibiotic ID — as.ab","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Input to an Antibiotic ID — as.ab","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to an Antibiotic ID — as.ab","text":"","code":"# these examples all return \"ERY\", the ID of erythromycin: as.ab(\"J01FA01\") #> Class 'ab' #> [1] ERY as.ab(\"J 01 FA 01\") #> Class 'ab' #> [1] ERY as.ab(\"Erythromycin\") #> Class 'ab' #> [1] ERY as.ab(\"eryt\") #> Class 'ab' #> [1] ERY as.ab(\"   eryt 123\") #> Class 'ab' #> [1] ERY as.ab(\"ERYT\") #> Class 'ab' #> [1] ERY as.ab(\"ERY\") #> Class 'ab' #> [1] ERY as.ab(\"eritromicine\") # spelled wrong, yet works #> Class 'ab' #> [1] ERY as.ab(\"Erythrocin\") # trade name #> Class 'ab' #> [1] ERY as.ab(\"Romycin\") # trade name #> Class 'ab' #> [1] ERY  # spelling from different languages and dyslexia are no problem ab_atc(\"ceftriaxon\") #> [1] \"J01DD04\" ab_atc(\"cephtriaxone\") # small spelling error #> [1] \"J01DD04\" ab_atc(\"cephthriaxone\") # or a bit more severe #> [1] \"J01DD04\" ab_atc(\"seephthriaaksone\") # and even this works #> [1] \"J01DD04\"  # use ab_* functions to get a specific properties (see ?ab_property); # they use as.ab() internally: ab_name(\"J01FA01\") #> [1] \"Erythromycin\" ab_name(\"eryt\") #> [1] \"Erythromycin\"  # \\donttest{ if (require(\"dplyr\")) {   # you can quickly rename 'sir' columns using set_ab_names() with dplyr:   example_isolates %>%     set_ab_names(where(is.sir), property = \"atc\") } #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           J01CE01 J01CF04 J01CF05 #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir>   <sir>   <sir>   #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R       NA      NA    #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R       NA      NA    #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R       NA      S     #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R       NA      S     #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R       NA      S     #> # ℹ 1,990 more rows #> # ℹ 37 more variables: J01CA04 <sir>, J01CR02 <sir>, J01CA01 <sir>, #> #   J01CR05 <sir>, J01DB04 <sir>, J01DE01 <sir>, J01DC02 <sir>, J01DC01 <sir>, #> #   J01DD01 <sir>, J01DD02 <sir>, J01DD04 <sir>, J01GB03 <sir>, J01GB01 <sir>, #> #   J01GB06 <sir>, J01GB04 <sir>, J01EA01 <sir>, J01EE01 <sir>, J01XE01 <sir>, #> #   J01XX01 <sir>, J01XX08 <sir>, J01MA02 <sir>, J01MA14 <sir>, J01XA01 <sir>, #> #   J01XA02 <sir>, J01AA07 <sir>, J01AA12 <sir>, J01AA02 <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to an Antiviral Drug ID — as.av","title":"Transform Input to an Antiviral Drug ID — as.av","text":"Use function determine antiviral drug code one antiviral drugs. data set antivirals searched abbreviations, official names synonyms (brand names).","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to an Antiviral Drug ID — as.av","text":"","code":"as.av(x, flag_multiple_results = TRUE, info = interactive(), ...)  is.av(x)"},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to an Antiviral Drug ID — as.av","text":"x character vector determine antiviral drug ID flag_multiple_results logical indicate whether note printed console probably one antiviral drug code name can retrieved single input value. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed internal functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to an Antiviral Drug ID — as.av","text":"character vector additional class ab","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to an Antiviral Drug ID — as.av","text":"entries antivirals data set three different identifiers: human readable EARS-Net code (column ab, used ECDC WHONET), ATC code (column atc, used ), CID code (column cid, Compound ID, used PubChem). data set contains 5,000 official brand names many different countries, found PubChem. drugs contain multiple ATC codes. properties searched user input. .av() can correct different forms misspelling: Wrong spelling drug names (\"acyclovir\"), corrects audible similarities f/ph, x/ks, c/z/s, t/th, etc. many vowels consonants Switching two characters (\"aycclovir\", often case clinical data, doctors typed fast) Digitalised paper records, leaving artefacts like 0/o/O (zero O's), B/8, n/r, etc. Use av_* functions get properties based returned antiviral drug ID, see Examples. Note: .av() av_* functions may use long regular expression match brand names antimicrobial drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Input to an Antiviral Drug ID — as.av","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Transform Input to an Antiviral Drug ID — as.av","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Input to an Antiviral Drug ID — as.av","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to an Antiviral Drug ID — as.av","text":"","code":"# these examples all return \"ACI\", the ID of aciclovir: as.av(\"J05AB01\") #> Class 'av' #> [1] ACI as.av(\"J 05 AB 01\") #> Class 'av' #> [1] ACI as.av(\"Aciclovir\") #> Class 'av' #> [1] ACI as.av(\"aciclo\") #> Class 'av' #> [1] ACI as.av(\"   aciclo 123\") #> Class 'av' #> [1] ACI as.av(\"ACICL\") #> Class 'av' #> [1] ACI as.av(\"ACI\") #> Class 'av' #> [1] ACI as.av(\"Virorax\") # trade name #> Class 'av' #> [1] ACI as.av(\"Zovirax\") # trade name #> Class 'av' #> [1] ACI  as.av(\"acyklofir\") # severe spelling error, yet works #> Class 'av' #> [1] ACI  # use av_* functions to get a specific properties (see ?av_property); # they use as.av() internally: av_name(\"J05AB01\") #> [1] \"Aciclovir\" av_name(\"acicl\") #> [1] \"Aciclovir\""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to Disk Diffusion Diameters — as.disk","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"transforms vector new class disk, disk diffusion growth zone size (around antibiotic disk) millimetres 0 50.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"","code":"as.disk(x, na.rm = FALSE)  NA_disk_  is.disk(x)"},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"object class disk (inherits integer) length 1.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"x vector na.rm logical indicating whether missing values removed","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"integer additional class disk","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"Interpret disk values SIR values .sir(). supports guidelines EUCAST CLSI. Disk diffusion growth zone sizes must 0 50 millimetres. Values higher 50 lower 100 maximised 50. others input values outside 0-50 range return NA. NA_disk_ missing value new disk class.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"","code":"# transform existing disk zones to the `disk` class (using base R) df <- data.frame(   microorganism = \"Escherichia coli\",   AMP = 20,   CIP = 14,   GEN = 18,   TOB = 16 ) df[, 2:5] <- lapply(df[, 2:5], as.disk) str(df) #> 'data.frame':\t1 obs. of  5 variables: #>  $ microorganism: chr \"Escherichia coli\" #>  $ AMP          : 'disk' int 20 #>  $ CIP          : 'disk' int 14 #>  $ GEN          : 'disk' int 18 #>  $ TOB          : 'disk' int 16  # \\donttest{ # transforming is easier with dplyr: if (require(\"dplyr\")) {   df %>% mutate(across(AMP:TOB, as.disk)) } #>      microorganism AMP CIP GEN TOB #> 1 Escherichia coli  20  14  18  16 # }  # interpret disk values, see ?as.sir as.sir(   x = as.disk(18),   mo = \"Strep pneu\", # `mo` will be coerced with as.mo()   ab = \"ampicillin\", # and `ab` with as.ab()   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: 'ampicillin' (AMP), EUCAST 2024... #>  OK  #> Class 'sir' #> [1] R  # interpret whole data set, pretend to be all from urinary tract infections: as.sir(df, uti = TRUE) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'AMP' (ampicillin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'CIP' (ciprofloxacin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'GEN' (gentamicin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'TOB' (tobramycin), EUCAST #> 2024... #>  OK  #>      microorganism AMP  CIP GEN TOB #> 1 Escherichia coli   S <NA>   S   S"},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"transforms vectors new class mic, treats input decimal numbers, maintaining operators (\">=\") allowing valid MIC values known field (medical) microbiology.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"","code":"as.mic(x, na.rm = FALSE, keep_operators = \"all\")  is.mic(x)  NA_mic_  rescale_mic(x, mic_range, keep_operators = \"edges\", as.mic = TRUE)  # S3 method for class 'mic' droplevels(x, as.mic = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"x character numeric vector na.rm logical indicating whether missing values removed keep_operators character specifying handle operators (> <=) input. Accepts one three values: \"\" (TRUE) keep operators, \"none\" (FALSE) remove operators, \"edges\" keep operators ends range. mic_range manual range rescale MIC values, e.g., mic_range = c(0.001, 32). Use NA prevent rescaling one side, e.g., mic_range = c(NA, 32). .mic logical indicate whether mic class kept - default TRUE rescale_mic() FALSE droplevels(). setting FALSE rescale_mic(), output factor levels acknowledge mic_range. ... arguments passed methods","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"Ordered factor additional class mic, mathematical operations acts numeric vector. Bear mind outcome mathematical operation MICs return numeric value.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"interpret MIC values SIR values, use .sir() MIC values. supports guidelines EUCAST (2011-2024) CLSI (2011-2024). class MIC values quite special data type: formally ordered factor valid MIC values factor levels (make sure valid MIC values retained), mathematical operation acts decimal numbers:   makes possible maintain operators often come MIC values, \">=\" \"<=\", even filtering using numeric values data analysis, e.g.:   -called group generic functions implemented MIC class (!, !=, <, >=, exp(), log2()). functions stats package also implemented (quantile(), median(), fivenum()). Since sd() var() non-generic functions, extended. Use mad() alternative, use e.g. sd(.numeric(x)) x vector MIC values. Using .double() .numeric() MIC values remove operators return numeric vector. use .integer() MIC values R convention factors, return index factor levels (often useless regular users). Use droplevels() drop unused levels. default, return plain factor. Use droplevels(..., .mic = TRUE) maintain mic class. rescale_mic(), existing MIC ranges can limited defined range MIC values. can useful better compare MIC distributions. ggplot2, use one scale_*_mic() functions plot MIC values. allows custom MIC ranges plot intermediate log2 levels missing MIC values. NA_mic_ missing value new mic class, analogous e.g. base R's NA_character_.","code":"x <- random_mic(10) x #> Class 'mic' #>  [1] 16     1      8      8      64     >=128  0.0625 32     32     16  is.factor(x) #> [1] TRUE  x[1] * 2 #> [1] 32  median(x) #> [1] 26 x[x > 4] #> Class 'mic' #> [1] 16    8     8     64    >=128 32    32    16  df <- data.frame(x, hospital = \"A\") subset(df, x > 4) # or with dplyr: df %>% filter(x > 4) #>        x hospital #> 1     16        A #> 5     64        A #> 6  >=128        A #> 8     32        A #> 9     32        A #> 10    16        A"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"","code":"mic_data <- as.mic(c(\">=32\", \"1.0\", \"1\", \"1.00\", 8, \"<=0.128\", \"8\", \"16\", \"16\")) mic_data #> Class 'mic' #> [1] >=32    1       1       1       8       <=0.128 8       16      16      is.mic(mic_data) #> [1] TRUE  # this can also coerce combined MIC/SIR values: as.mic(\"<=0.002; S\") #> Class 'mic' #> [1] <=0.002  # mathematical processing treats MICs as numeric values fivenum(mic_data) #> [1]  0.128  1.000  8.000 16.000 32.000 quantile(mic_data) #>     0%    25%    50%    75%   100%  #>  0.128  1.000  8.000 16.000 32.000  all(mic_data < 512) #> [1] TRUE  # rescale MICs using rescale_mic() rescale_mic(mic_data, mic_range = c(4, 16)) #> Class 'mic' #> [1] >=16 <=4  <=4  <=4  8    <=4  8    >=16 >=16  # interpret MIC values as.sir(   x = as.mic(2),   mo = as.mo(\"Streptococcus pneumoniae\"),   ab = \"AMX\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMX' (amoxicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for amoxicillin (AMX) in Streptococcus pneumoniae - assuming body site 'Meningitis'. #> Class 'sir' #> [1] R as.sir(   x = as.mic(c(0.01, 2, 4, 8)),   mo = as.mo(\"Streptococcus pneumoniae\"),   ab = \"AMX\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMX' (amoxicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for amoxicillin (AMX) in Streptococcus pneumoniae - assuming body site 'Meningitis'. #> Class 'sir' #> [1] S R R R  # plot MIC values, see ?plot plot(mic_data)  plot(mic_data, mo = \"E. coli\", ab = \"cipro\")   if (require(\"ggplot2\")) {   autoplot(mic_data, mo = \"E. coli\", ab = \"cipro\") }  if (require(\"ggplot2\")) {   autoplot(mic_data, mo = \"E. coli\", ab = \"cipro\", language = \"nl\") # Dutch }"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Use function get valid microorganism code (mo) based arbitrary user input. Determination done using intelligent rules complete taxonomic tree kingdoms Animalia, Archaea, Bacteria, Chromista, Protozoa, microbial species kingdom Fungi (see Source). input can almost anything: full name (like \"Staphylococcus aureus\"), abbreviated name (\"S. aureus\"), abbreviation known field (\"MRSA\"), just genus. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"","code":"as.mo(x, Becker = FALSE, Lancefield = FALSE,   minimum_matching_score = NULL,   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE),   reference_df = get_mo_source(),   ignore_pattern = getOption(\"AMR_ignore_pattern\", NULL),   cleaning_regex = getOption(\"AMR_cleaning_regex\", mo_cleaning_regex()),   only_fungi = getOption(\"AMR_only_fungi\", FALSE),   language = get_AMR_locale(), info = interactive(), ...)  is.mo(x)  mo_uncertainties()  mo_renamed()  mo_failures()  mo_reset_session()  mo_cleaning_regex()"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"x character vector data.frame one two columns Becker logical indicate whether staphylococci categorised coagulase-negative staphylococci (\"CoNS\") coagulase-positive staphylococci (\"CoPS\") instead species, according Karsten Becker et al. (see Source). Please see Details full list staphylococcal species converted. excludes Staphylococcus aureus default, use Becker = \"\" also categorise S. aureus \"CoPS\". Lancefield logical indicate whether beta-haemolytic Streptococcus categorised Lancefield groups instead species, according Rebecca C. Lancefield (see Source). streptococci categorised first group, e.g. Streptococcus dysgalactiae group C, although officially also categorised groups G L. . Please see Details full list streptococcal species converted. excludes enterococci default (group D), use Lancefield = \"\" also categorise enterococci group D. minimum_matching_score numeric value set lower limit MO matching score. left blank, determined automatically based character length x, taxonomic kingdom human pathogenicity. keep_synonyms logical indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE, return note old taxonomic names processed. default can set package option AMR_keep_synonyms, .e. options(AMR_keep_synonyms = TRUE) options(AMR_keep_synonyms = FALSE). reference_df data.frame used extra reference translating x valid mo. See set_mo_source() get_mo_source() automate usage codes (e.g. used analysis organisation). ignore_pattern Perl-compatible regular expression (case-insensitive) matches x must return NA. can convenient exclude known non-relevant input can also set package option AMR_ignore_pattern, e.g. options(AMR_ignore_pattern = \"(reported|contaminated flora)\"). cleaning_regex Perl-compatible regular expression (case-insensitive) clean input x. Every matched part x removed. default, outcome mo_cleaning_regex(), removes texts brackets texts \"species\" \"serovar\". default can set package option AMR_cleaning_regex. only_fungi logical indicate fungi must found, making sure e.g. misspellings always return records kingdom Fungi. can set globally microorganism functions package option AMR_only_fungi, .e. options(AMR_only_fungi = TRUE). language language translate text like \"growth\", defaults system language (see get_AMR_locale()) info logical indicate info must printed, e.g. progress bar 25 items coerced, list old taxonomic names. default TRUE interactive mode. ... arguments passed functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"character vector additional class mo","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"microorganism (MO) code package (class: mo) human-readable typically looks like examples:   Values coerced considered 'unknown' return MO code UNKNOWN warning. Use mo_* functions get properties based returned code, see Examples. .mo() function uses novel scientifically validated (doi:10.18637/jss.v104.i03 ) matching score algorithm (see Matching Score Microorganisms ) match input available microbial taxonomy package. implicates e.g. \"E. coli\" (microorganism highly prevalent humans) return microbial ID Escherichia coli Entamoeba coli (microorganism less prevalent humans), although latter alphabetically come first.","code":"Code               Full name   ---------------    --------------------------------------   B_KLBSL            Klebsiella   B_KLBSL_PNMN       Klebsiella pneumoniae   B_KLBSL_PNMN_RHNS  Klebsiella pneumoniae rhinoscleromatis   |   |    |    |   |   |    |    |   |   |    |    \\---> subspecies, a 3-5 letter acronym   |   |    \\----> species, a 3-6 letter acronym   |   \\----> genus, a 4-8 letter acronym   \\----> kingdom: A (Archaea), AN (Animalia), B (Bacteria),                   C (Chromista), F (Fungi), PL (Plantae),                   P (Protozoa)"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"coping-with-uncertain-results","dir":"Reference","previous_headings":"","what":"Coping with Uncertain Results","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Results non-exact taxonomic input based matching score. lowest allowed score can set minimum_matching_score argument. default determined based character length input, taxonomic kingdom, human pathogenicity taxonomic outcome. values matched uncertainty, message shown suggest user inspect results mo_uncertainties(), returns data.frame specifications. increase quality matching, cleaning_regex argument used clean input. must regular expression matches parts input removed input matched available microbial taxonomy. matched Perl-compatible case-insensitive. default value cleaning_regex outcome helper function mo_cleaning_regex(). three helper functions can run using .mo() function: Use mo_uncertainties() get data.frame prints pretty format taxonomic names guessed. output contains matching score matches (see Matching Score Microorganisms ). Use mo_failures() get character vector values coerced valid value. Use mo_renamed() get data.frame values coerced based old, previously accepted taxonomic names.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"for-mycologists","dir":"Reference","previous_headings":"","what":"For Mycologists","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"matching score algorithm gives precedence bacteria fungi. analysing fungi, sure use only_fungi = TRUE, better yet, add code run every session:   make sure bacteria 'non-fungi' returned .mo(), mo_* functions.","code":"options(AMR_only_fungi = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"coagulase-negative-and-coagulase-positive-staphylococci","dir":"Reference","previous_headings":"","what":"Coagulase-negative and Coagulase-positive Staphylococci","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Becker = TRUE, following staphylococci converted corresponding coagulase group: Coagulase-negative: S. americanisciuri, S. argensis, S. arlettae, S. auricularis, S. borealis, S. brunensis, S. caeli, S. caledonicus, S. canis, S. capitis, S. capitis capitis, S. capitis urealyticus, S. capitis ureolyticus, S. caprae, S. carnosus, S. carnosus carnosus, S. carnosus utilis, S. casei, S. caseolyticus, S. chromogenes, S. cohnii, S. cohnii cohnii, S. cohnii urealyticum, S. cohnii urealyticus, S. condimenti, S. croceilyticus, S. debuckii, S. devriesei, S. durrellii, S. edaphicus, S. epidermidis, S. equorum, S. equorum equorum, S. equorum linens, S. felis, S. fleurettii, S. gallinarum, S. haemolyticus, S. hominis, S. hominis hominis, S. hominis novobiosepticus, S. jettensis, S. kloosii, S. lentus, S. lloydii, S. lugdunensis, S. marylandisciuri, S. massiliensis, S. microti, S. muscae, S. nepalensis, S. pasteuri, S. petrasii, S. petrasii croceilyticus, S. petrasii jettensis, S. petrasii petrasii, S. petrasii pragensis, S. pettenkoferi, S. piscifermentans, S. pragensis, S. pseudoxylosus, S. pulvereri, S. ratti, S. rostri, S. saccharolyticus, S. saprophyticus, S. saprophyticus bovis, S. saprophyticus saprophyticus, S. schleiferi, S. schleiferi schleiferi, S. sciuri, S. sciuri carnaticus, S. sciuri lentus, S. sciuri rodentium, S. sciuri sciuri, S. shinii, S. simulans, S. stepanovicii, S. succinus, S. succinus casei, S. succinus succinus, S. taiwanensis, S. urealyticus, S. ureilyticus, S. veratri, S. vitulinus, S. vitulus, S. warneri, S. xylosus Coagulase-positive: S. agnetis, S. argenteus, S. coagulans, S. cornubiensis, S. delphini, S. hyicus, S. hyicus chromogenes, S. hyicus hyicus, S. intermedius, S. lutrae, S. pseudintermedius, S. roterodami, S. schleiferi coagulans, S. schweitzeri, S. simiae, S. singaporensis based : Becker K et al. (2014). Coagulase-Negative Staphylococci. Clin Microbiol Rev. 27(4): 870-926; doi:10.1128/CMR.00109-13 Becker K et al. (2019). Implications identifying recently defined members S. aureus complex, S. argenteus S. schweitzeri: position paper members ESCMID Study Group staphylococci Staphylococcal Diseases (ESGS). Clin Microbiol Infect; doi:10.1016/j.cmi.2019.02.028 Becker K et al. (2020). Emergence coagulase-negative staphylococci. Expert Rev Anti Infect Ther. 18(4):349-366; doi:10.1080/14787210.2020.1730813 newly named staphylococcal species, S. brunensis (2024) S. shinii (2023), looked scientific reference make sure species considered correct coagulase group.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"lancefield-groups-in-streptococci","dir":"Reference","previous_headings":"","what":"Lancefield Groups in Streptococci","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Lancefield = TRUE, following streptococci converted corresponding Lancefield group: Streptococcus Group : S. pyogenes Streptococcus Group B: S. agalactiae Streptococcus Group C: S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis, S. equi, S. equi equi, S. equi ruminatorum, S. equi zooepidemicus Streptococcus Group F: S. anginosus, S. anginosus anginosus, S. anginosus whileyi, S. constellatus, S. constellatus constellatus, S. constellatus pharyngis, S. constellatus viborgensis, S. intermedius Streptococcus Group G: S. canis, S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis Streptococcus Group H: S. sanguinis Streptococcus Group K: S. salivarius, S. salivarius salivarius, S. salivarius thermophilus Streptococcus Group L: S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis based : Lancefield RC (1933). serological differentiation human groups hemolytic streptococci. J Exp Med. 57(4): 571-95; doi:10.1084/jem.57.4.571","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/ Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"ambiguous user input .mo() mo_* functions, returned results chosen based matching score using mo_matching_score(). matching score \\(m\\), calculated : $$m_{(x, n)} = \\frac{l_{n} - 0.5 \\cdot \\min \\begin{cases}l_{n} \\\\ \\textrm{lev}(x, n)\\end{cases}}{l_{n} \\cdot p_{n} \\cdot k_{n}}$$ : \\(x\\) user input; \\(n\\) taxonomic name (genus, species, subspecies); \\(l_n\\) length \\(n\\); \\(lev\\) Levenshtein distance function (counting insertion 1, deletion substitution 2) needed change \\(x\\) \\(n\\); \\(p_n\\) human pathogenic prevalence group \\(n\\), described ; \\(k_n\\) taxonomic kingdom \\(n\\), set Bacteria = 1, Fungi = 1.25, Protozoa = 1.5, Chromista = 1.75, Archaea = 2, others = 3. grouping human pathogenic prevalence \\(p\\) based recent work Bartlett et al. (2022, doi:10.1099/mic.0.001269 ) extensively studied medical-scientific literature categorise bacterial species groups: Established, taxonomic species infected least three persons three references. records prevalence = 1.15 microorganisms data set; Putative, taxonomic species fewer three known cases. records prevalence = 1.25 microorganisms data set. Furthermore, Genera World Health Organization's () Priority Pathogen List prevalence = 1.0 microorganisms data set; genus present established list also prevalence = 1.15 microorganisms data set; genus present putative list prevalence = 1.25 microorganisms data set; species subspecies genus present two aforementioned groups, prevalence = 1.5 microorganisms data set; non-bacterial genus, species subspecies genus present following list, prevalence = 1.25 microorganisms data set: Absidia, Acanthamoeba, Acremonium, Actinomucor, Aedes, Alternaria, Amoeba, Ancylostoma, Angiostrongylus, Anisakis, Anopheles, Apophysomyces, Arthroderma, Aspergillus, Aureobasidium, Basidiobolus, Beauveria, Bipolaris, Blastobotrys, Blastocystis, Blastomyces, Candida, Capillaria, Chaetomium, Chilomastix, Chrysonilia, Chrysosporium, Cladophialophora, Cladosporium, Clavispora, Coccidioides, Cokeromyces, Conidiobolus, Coniochaeta, Contracaecum, Cordylobia, Cryptococcus, Cryptosporidium, Cunninghamella, Curvularia, Cyberlindnera, Debaryozyma, Demodex, Dermatobia, Dientamoeba, Diphyllobothrium, Dirofilaria, Echinostoma, Entamoeba, Enterobius, Epidermophyton, Exidia, Exophiala, Exserohilum, Fasciola, Fonsecaea, Fusarium, Geotrichum, Giardia, Graphium, Haloarcula, Halobacterium, Halococcus, Hansenula, Hendersonula, Heterophyes, Histomonas, Histoplasma, Hortaea, Hymenolepis, Hypomyces, Hysterothylacium, Kloeckera, Kluyveromyces, Kodamaea, Lacazia, Leishmania, Lichtheimia, Lodderomyces, Lomentospora, Madurella, Malassezia, Malbranchea, Metagonimus, Meyerozyma, Microsporidium, Microsporum, Millerozyma, Mortierella, Mucor, Mycocentrospora, Nannizzia, Necator, Nectria, Ochroconis, Oesophagostomum, Oidiodendron, Opisthorchis, Paecilomyces, Paracoccidioides, Pediculus, Penicillium, Phaeoacremonium, Phaeomoniella, Phialophora, Phlebotomus, Phoma, Pichia, Piedraia, Pithomyces, Pityrosporum, Pneumocystis, Pseudallescheria, Pseudoscopulariopsis, Pseudoterranova, Pulex, Purpureocillium, Quambalaria, Rhinocladiella, Rhizomucor, Rhizopus, Rhodotorula, Saccharomyces, Saksenaea, Saprochaete, Sarcoptes, Scedosporium, Schistosoma, Schizosaccharomyces, Scolecobasidium, Scopulariopsis, Scytalidium, Spirometra, Sporobolomyces, Sporopachydermia, Sporothrix, Sporotrichum, Stachybotrys, Strongyloides, Syncephalastrum, Syngamus, Taenia, Talaromyces, Teleomorph, Toxocara, Trichinella, Trichobilharzia, Trichoderma, Trichomonas, Trichophyton, Trichosporon, Trichostrongylus, Trichuris, Tritirachium, Trombicula, Trypanosoma, Tunga, Ulocladium, Ustilago, Verticillium, Wallemia, Wangiella, Wickerhamomyces, Wuchereria, Yarrowia, Zygosaccharomyces; records prevalence = 2.0 microorganisms data set. calculating matching score, characters \\(x\\) \\(n\\) ignored -Z, -z, 0-9, spaces parentheses. matches sorted descending matching score user input values, top match returned. lead effect e.g., \"E. coli\" return microbial ID Escherichia coli (\\(m = 0.688\\), highly prevalent microorganism found humans) Entamoeba coli (\\(m = 0.381\\), less prevalent microorganism humans), although latter alphabetically come first.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"","code":"# \\donttest{ # These examples all return \"B_STPHY_AURS\", the ID of S. aureus: as.mo(c(   \"sau\", # WHONET code   \"stau\",   \"STAU\",   \"staaur\",   \"S. aureus\",   \"S aureus\",   \"Sthafilokkockus aureus\", # handles incorrect spelling   \"Staphylococcus aureus (MRSA)\",   \"MRSA\", # Methicillin Resistant S. aureus   \"VISA\", # Vancomycin Intermediate S. aureus   \"VRSA\", # Vancomycin Resistant S. aureus   115329001 # SNOMED CT code )) #> Class 'mo' #>  [1] B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS #>  [6] B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS #> [11] B_STPHY_AURS B_STPHY_AURS  # Dyslexia is no problem - these all work: as.mo(c(   \"Ureaplasma urealyticum\",   \"Ureaplasma urealyticus\",   \"Ureaplasmium urealytica\",   \"Ureaplazma urealitycium\" )) #> Class 'mo' #> [1] B_URPLS_URLY B_URPLS_URLY B_URPLS_URLY B_URPLS_URLY  # input will get cleaned up with the input given in the `cleaning_regex` argument, # which defaults to `mo_cleaning_regex()`: cat(mo_cleaning_regex(), \"\\n\") #> ([^A-Za-z- \\(\\)\\[\\]{}]+|([({]|\\[).+([})]|\\])|(^| )( ?[a-z-]+[-](resistant|susceptible) ?|e?spp([^a-z]+|$)|e?ssp([^a-z]+|$)|serogr.?up[a-z]*|e?ss([^a-z]+|$)|e?sp([^a-z]+|$)|var([^a-z]+|$)|serovar[a-z]*|sube?species|biovar[a-z]*|e?species|Ig[ADEGM]|e?subsp|biotype|titer|dummy))   as.mo(\"Streptococcus group A\") #> Class 'mo' #> [1] B_STRPT_GRPA  as.mo(\"S. epidermidis\") # will remain species: B_STPHY_EPDR #> Class 'mo' #> [1] B_STPHY_EPDR as.mo(\"S. epidermidis\", Becker = TRUE) # will not remain species: B_STPHY_CONS #> Class 'mo' #> [1] B_STPHY_CONS  as.mo(\"S. pyogenes\") # will remain species: B_STRPT_PYGN #> Class 'mo' #> [1] B_STRPT_PYGN as.mo(\"S. pyogenes\", Lancefield = TRUE) # will not remain species: B_STRPT_GRPA #> Class 'mo' #> [1] B_STRPT_GRPA  # All mo_* functions use as.mo() internally too (see ?mo_property): mo_genus(\"E. coli\") #> [1] \"Escherichia\" mo_gramstain(\"ESCO\") #> [1] \"Gram-negative\" mo_is_intrinsic_resistant(\"ESCCOL\", ab = \"vanco\") #> ℹ Determining intrinsic resistance based on 'EUCAST Expert Rules' and #>   'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3 (2021). This note #>   will be shown once per session. #> [1] TRUE # }"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":null,"dir":"Reference","previous_headings":"","what":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Clean existing SIR values, interpret minimum inhibitory concentration (MIC) values disk diffusion diameters according EUCAST CLSI. .sir() transforms input new class sir, ordered factor containing levels S, SDD, , R, NI. breakpoints currently implemented: clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024; veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024; ECOFFs (Epidemiological Cut-Values): EUCAST 2020-2024 CLSI 2022-2024. breakpoints used interpretation available clinical_breakpoints data set.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"","code":"as.sir(x, ...)  NA_sir_  is.sir(x)  is_sir_eligible(x, threshold = 0.05)  # Default S3 method as.sir(x, S = \"^(S|U)+$\", I = \"^(I)+$\", R = \"^(R)+$\",   NI = \"^(N|NI|V)+$\", SDD = \"^(SDD|D|H)+$\", ...)  # S3 method for class 'mic' as.sir(x, mo = NULL, ab = deparse(substitute(x)),   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   conserve_capped_values = FALSE, add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE, ...)  # S3 method for class 'disk' as.sir(x, mo = NULL, ab = deparse(substitute(x)),   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE, ...)  # S3 method for class 'data.frame' as.sir(x, ..., col_mo = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   conserve_capped_values = FALSE, add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE)  sir_interpretation_history(clean = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"interpretations minimum inhibitory concentration (MIC) values disk diffusion diameters: CLSI M39: Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 2011-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/. CLSI M100: Performance Standard Antimicrobial Susceptibility Testing, 2011-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m100/. CLSI VET01: Performance Standards Antimicrobial Disk Dilution Susceptibility Tests Bacteria Isolated Animals, 2019-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/veterinary-medicine/documents/vet01/. CLSI VET09: Understanding Susceptibility Test Data Component Antimicrobial Stewardship Veterinary Settings, 2019-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/veterinary-medicine/documents/vet09/. EUCAST Breakpoint tables interpretation MICs zone diameters, 2011-2024, European Committee Antimicrobial Susceptibility Testing (EUCAST). https://www.eucast.org/clinical_breakpoints. WHONET source machine-reading clinical breakpoints (read ), 1989-2024, Collaborating Centre Surveillance Antimicrobial Resistance. https://whonet.org/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"x vector values (class mic: MIC values mg/L, class disk: disk diffusion radius millimetres) ... using data.frame: names columns apply .sir() (supports tidy selection column1:column4). Otherwise: arguments passed methods. threshold maximum fraction invalid antimicrobial interpretations x, see Examples S, , R, NI, SDD case-independent regular expression translate input result. regular expression run non-letters whitespaces removed input. mo vector (column name) characters can coerced valid microorganism codes .mo(), can left empty determine automatically ab vector (column name) characters can coerced valid antimicrobial drug code .ab() guideline defaults EUCAST 2024 (latest implemented EUCAST guideline clinical_breakpoints data set), can set package option AMR_guideline. Currently supports EUCAST (2011-2024) CLSI (2011-2024), see Details. uti (Urinary Tract Infection) vector (column name) logicals (TRUE FALSE) specify whether UTI specific interpretation guideline chosen. using .sir() data.frame, can also column containing logicals left blank, data set searched column 'specimen', rows within column containing 'urin' ('urine', 'urina') regarded isolates UTI. See Examples. conserve_capped_values logical indicate MIC values starting \">\" (\">=\") must always return \"R\" , MIC values starting \"<\" (\"<=\") must always return \"S\" add_intrinsic_resistance (useful using EUCAST guideline) logical indicate whether intrinsic antibiotic resistance must also considered applicable bug-drug combinations, meaning e.g. ampicillin always return \"R\" Klebsiella species. Determination based intrinsic_resistant data set, based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021). reference_data data.frame used interpretation, defaults clinical_breakpoints data set. Changing argument allows using interpretation guidelines. argument must contain data set equal structure clinical_breakpoints data set (column names column types). Please note guideline argument ignored reference_data manually set. include_screening logical indicate clinical breakpoints screening allowed - default FALSE. Can also set package option AMR_include_screening. include_PKPD logical indicate PK/PD clinical breakpoints must applied last resort - default TRUE. Can also set package option AMR_include_PKPD. breakpoint_type type breakpoints use, either \"ECOFF\", \"animal\", \"human\". ECOFF stands Epidemiological Cut-values. default \"human\", can also set package option AMR_breakpoint_type. host set values veterinary species, automatically set \"animal\". host vector (column name) characters indicate host. useful veterinary breakpoints, requires breakpoint_type = \"animal\". values can text resembling animal species, even 20 supported languages package. foreign languages, sure set language set_AMR_locale() (though automatically guessed based system language). verbose logical indicate notes printed interpretation MIC values disk diffusion values. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). clean logical indicate whether previously stored results forgotten returning 'logbook' results","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Ordered factor new class sir","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Note: clinical breakpoints package validated , imported , WHONET. public use AMR package endorsed CLSI EUCAST. See clinical_breakpoints information. NA_sir_ missing value new sir class, analogous e.g. base R's NA_character_.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"how-it-works","dir":"Reference","previous_headings":"","what":"How it Works","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":".sir() function can work four ways: cleaning raw / untransformed data. data cleaned contain valid values, namely: S susceptible, intermediate 'susceptible, increased exposure', R resistant, NI non-interpretable, SDD susceptible dose-dependent. can set using regular expression. Furthermore, .sir() try best clean intelligence. example, mixed values SIR interpretations MIC values \"<0.25; S\" coerced \"S\". Combined interpretations multiple test methods (seen laboratory records) \"S; S\" coerced \"S\", value like \"S; \" return NA warning input invalid. interpreting minimum inhibitory concentration (MIC) values according EUCAST CLSI. must clean MIC values first using .mic(), also gives columns new data class mic. Also, sure column microorganism names codes. found automatically, can set manually using mo argument. Using dplyr, SIR interpretation can done easily either:   Operators like \"<=\" stripped interpretation. using conserve_capped_values = TRUE, MIC value e.g. \">2\" always return \"R\", even breakpoint according chosen guideline \">=4\". prevent capped values raw laboratory data treated conservatively. default behaviour (conserve_capped_values = FALSE) considers \">2\" lower \">=4\" might case return \"S\" \"\". interpreting disk diffusion diameters according EUCAST CLSI. must clean disk zones first using .disk(), also gives columns new data class disk. Also, sure column microorganism names codes. found automatically, can set manually using mo argument. Using dplyr, SIR interpretation can done easily either:   interpreting complete data set, automatic determination MIC values, disk diffusion diameters, microorganism names codes, antimicrobial test results. done simply running .sir(your_data). points 2, 3 4: Use sir_interpretation_history() retrieve data.frame (tibble tibble package installed) results last .sir() call.","code":"your_data %>% mutate_if(is.mic, as.sir) your_data %>% mutate(across(where(is.mic), as.sir)) your_data %>% mutate_if(is.mic, as.sir, ab = \"column_with_antibiotics\", mo = \"column_with_microorganisms\") your_data %>% mutate_if(is.mic, as.sir, ab = c(\"cipro\", \"ampicillin\", ...), mo = c(\"E. coli\", \"K. pneumoniae\", ...))  # for veterinary breakpoints, also set `host`: your_data %>% mutate_if(is.mic, as.sir, host = \"column_with_animal_species\", guideline = \"CLSI\") your_data %>% mutate_if(is.disk, as.sir) your_data %>% mutate(across(where(is.disk), as.sir)) your_data %>% mutate_if(is.disk, as.sir, ab = \"column_with_antibiotics\", mo = \"column_with_microorganisms\") your_data %>% mutate_if(is.disk, as.sir, ab = c(\"cipro\", \"ampicillin\", ...), mo = c(\"E. coli\", \"K. pneumoniae\", ...))  # for veterinary breakpoints, also set `host`: your_data %>% mutate_if(is.disk, as.sir, host = \"column_with_animal_species\", guideline = \"CLSI\")"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"supported-guidelines","dir":"Reference","previous_headings":"","what":"Supported Guidelines","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"interpreting MIC values well disk diffusion diameters, currently implemented guidelines clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024, veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024. Thus, guideline argument must set e.g., \"EUCAST 2024\" \"CLSI 2024\". simply using \"EUCAST\" (default) \"CLSI\" input, latest included version guideline automatically selected. can set data set using reference_data argument. guideline argument ignored. can set default guideline package option AMR_guideline (e.g. .Rprofile file), :   veterinary guidelines, might best options:   applying veterinary breakpoints (setting host setting breakpoint_type = \"animal\"), CLSI VET09 guideline applied cope missing animal species-specific breakpoints.","code":"options(AMR_guideline = \"CLSI\")   options(AMR_guideline = \"CLSI 2018\")   options(AMR_guideline = \"EUCAST 2020\")   # or to reset:   options(AMR_guideline = NULL) options(AMR_guideline = \"CLSI\")   options(AMR_breakpoint_type = \"animal\")"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"after-interpretation","dir":"Reference","previous_headings":"","what":"After Interpretation","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"using .sir(), can use eucast_rules() defined EUCAST (1) apply inferred susceptibility resistance based results antimicrobials (2) apply intrinsic resistance based taxonomic properties microorganism. determine isolates multi-drug resistant, sure run mdro() (applies MDR/PDR/XDR guideline 2012 default) data set contains S//R values. Read interpreting multidrug-resistant organisms .","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"machine-readable-clinical-breakpoints","dir":"Reference","previous_headings":"","what":"Machine-Readable Clinical Breakpoints","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"repository package contains machine-readable version guidelines. CSV file consisting 34 063 rows 14 columns. file machine-readable, since contains one row every unique combination test method (MIC disk diffusion), antimicrobial drug microorganism. allows easy implementation rules laboratory information systems (LIS). Note contains interpretation guidelines humans - interpretation guidelines CLSI animals removed.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"other","dir":"Reference","previous_headings":"","what":"Other","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"function .sir() detects input contains class sir. input data.frame, iterates columns returns logical vector. base R function .double() can used retrieve quantitative values sir object: \"S\" = 1, \"\"/\"SDD\" = 2, \"R\" = 3. values rendered NA . Note: use .integer(), since (R works internally) return factor level indices, aforementioned quantitative values. function is_sir_eligible() returns TRUE column contains 5% invalid antimicrobial interpretations (S //R /NI /SDD), FALSE otherwise. threshold 5% can set threshold argument. input data.frame, iterates columns returns logical vector.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … summary(example_isolates) # see all SIR results at a glance #>       date              patient               age           gender          #>  Min.   :2002-01-02   Length:2000        Min.   : 0.00   Length:2000        #>  1st Qu.:2005-07-31   Class :character   1st Qu.:63.00   Class :character   #>  Median :2009-07-31   Mode  :character   Median :74.00   Mode  :character   #>  Mean   :2009-11-20                      Mean   :70.69                      #>  3rd Qu.:2014-05-30                      3rd Qu.:82.00                      #>  Max.   :2017-12-28                      Max.   :97.00                      #>      ward                mo                  PEN                 #>  Length:2000        Class :mo             Class:sir              #>  Class :character   <NA>  :0              %S   :25.6% (n=417)    #>  Mode  :character   Unique:90             %SDD : 0.0% (n=0)      #>                     #1    :B_ESCHR_COLI   %I   : 0.7% (n=11)     #>                     #2    :B_STPHY_CONS   %R   :73.7% (n=1201)   #>                     #3    :B_STPHY_AURS   %NI  : 0.0% (n=0)      #>     OXA                   FLC                   AMX                #>  Class:sir             Class:sir             Class:sir             #>  %S   :68.8% (n=251)   %S   :70.5% (n=665)   %S   :40.2% (n=543)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)     %I   : 0.2% (n=3)     #>  %R   :31.2% (n=114)   %R   :29.5% (n=278)   %R   :59.6% (n=804)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     AMC                    AMP                   TZP                #>  Class:sir              Class:sir             Class:sir             #>  %S   :71.4% (n=1342)   %S   :40.2% (n=543)   %S   :86.1% (n=862)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 4.8% (n=91)     %I   : 0.2% (n=3)     %I   : 1.3% (n=13)    #>  %R   :23.7% (n=446)    %R   :59.6% (n=804)   %R   :12.6% (n=126)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     CZO                   FEP                   CXM                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :54.9% (n=245)   %S   :85.6% (n=620)   %S   :72.5% (n=1297)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 0.4% (n=2)     %I   : 0.1% (n=1)     %I   : 1.2% (n=22)     #>  %R   :44.6% (n=199)   %R   :14.2% (n=103)   %R   :26.3% (n=470)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     FOX                   CTX                   CAZ                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :71.6% (n=586)   %S   :84.4% (n=796)   %S   :33.5% (n=607)    #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 1.0% (n=8)     %I   : 0.1% (n=1)     %I   : 0.0% (n=0)      #>  %R   :27.4% (n=224)   %R   :15.5% (n=146)   %R   :66.5% (n=1204)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     CRO                   GEN                    TOB                #>  Class:sir             Class:sir              Class:sir             #>  %S   :84.4% (n=796)   %S   :74.0% (n=1372)   %S   :65.1% (n=879)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.1% (n=1)     %I   : 1.5% (n=27)     %I   : 0.5% (n=7)     #>  %R   :15.5% (n=146)   %R   :24.6% (n=456)    %R   :34.4% (n=465)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     AMK                   KAN                    TMP                #>  Class:sir             Class:sir              Class:sir             #>  %S   :36.3% (n=251)   %S   : 0.0% (n=0)      %S   :61.2% (n=918)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)      %I   : 0.7% (n=10)    #>  %R   :63.7% (n=441)   %R   :100.0% (n=471)   %R   :38.1% (n=571)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     SXT                    NIT                   FOS                #>  Class:sir              Class:sir             Class:sir             #>  %S   :79.1% (n=1392)   %S   :76.0% (n=565)   %S   :57.8% (n=203)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.3% (n=6)      %I   : 6.9% (n=51)    %I   : 0.0% (n=0)     #>  %R   :20.5% (n=361)    %R   :17.1% (n=127)   %R   :42.2% (n=148)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     LNZ                   CIP                    MFX                #>  Class:sir             Class:sir              Class:sir             #>  %S   :30.7% (n=314)   %S   :78.9% (n=1112)   %S   :64.5% (n=136)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 4.9% (n=69)     %I   : 1.9% (n=4)     #>  %R   :69.3% (n=709)   %R   :16.2% (n=228)    %R   :33.6% (n=71)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     VAN                    TEC                   TCY                #>  Class:sir              Class:sir             Class:sir             #>  %S   :61.7% (n=1149)   %S   :24.3% (n=237)   %S   :68.3% (n=820)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)      %I   : 0.0% (n=0)     %I   : 1.9% (n=23)    #>  %R   :38.3% (n=712)    %R   :75.7% (n=739)   %R   :29.8% (n=357)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     TGC                   DOX                   ERY                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :87.3% (n=697)   %S   :71.7% (n=814)   %S   :42.3% (n=801)    #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 0.0% (n=0)     %I   : 0.6% (n=7)     %I   : 0.5% (n=9)      #>  %R   :12.7% (n=101)   %R   :27.7% (n=315)   %R   :57.2% (n=1084)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     CLI                   AZM                    IPM                #>  Class:sir             Class:sir              Class:sir             #>  %S   :38.6% (n=586)   %S   :42.3% (n=801)    %S   :92.7% (n=824)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.3% (n=4)     %I   : 0.5% (n=9)      %I   : 1.1% (n=10)    #>  %R   :61.2% (n=930)   %R   :57.2% (n=1084)   %R   : 6.2% (n=55)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     MEM                   MTR                  CHL                #>  Class:sir             Class:sir            Class:sir             #>  %S   :94.1% (n=780)   %S   :85.3% (n=29)   %S   :78.6% (n=121)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)    %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)    %I   : 0.0% (n=0)     #>  %R   : 5.9% (n=49)    %R   :14.7% (n=5)    %R   :21.4% (n=33)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)    %NI  : 0.0% (n=0)     #>     COL                    MUP                   RIF                #>  Class:sir              Class:sir             Class:sir             #>  %S   :18.8% (n=309)    %S   :93.0% (n=251)   %S   :30.2% (n=303)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)      %I   : 1.1% (n=3)     %I   : 0.2% (n=2)     #>  %R   :81.2% (n=1331)   %R   : 5.9% (n=16)    %R   :69.6% (n=698)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      # For INTERPRETING disk diffusion and MIC values -----------------------  # example data sets, with combined MIC values and disk zones df_wide <- data.frame(   microorganism = \"Escherichia coli\",   amoxicillin = as.mic(8),   cipro = as.mic(0.256),   tobra = as.disk(16),   genta = as.disk(18),   ERY = \"R\" ) df_long <- data.frame(   bacteria = rep(\"Escherichia coli\", 4),   antibiotic = c(\"amoxicillin\", \"cipro\", \"tobra\", \"genta\"),   mics = as.mic(c(0.01, 1, 4, 8)),   disks = as.disk(c(6, 10, 14, 18)) )  # \\donttest{ ## Using dplyr ------------------------------------------------- if (require(\"dplyr\")) {   # approaches that all work without additional arguments:   df_wide %>% mutate_if(is.mic, as.sir)   df_wide %>% mutate_if(function(x) is.mic(x) | is.disk(x), as.sir)   df_wide %>% mutate(across(where(is.mic), as.sir))   df_wide %>% mutate_at(vars(amoxicillin:tobra), as.sir)   df_wide %>% mutate(across(amoxicillin:tobra, as.sir))      # approaches that all work with additional arguments:   df_long %>%     # given a certain data type, e.g. MIC values     mutate_if(is.mic, as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               guideline = \"CLSI\")   df_long %>%     mutate(across(where(is.mic),                   function(x) as.sir(x,                                      mo = \"bacteria\",                                      ab = \"antibiotic\",                                      guideline = \"CLSI\")))   df_wide %>%     # given certain columns, e.g. from 'cipro' to 'genta'     mutate_at(vars(cipro:genta), as.sir,               mo = \"bacteria\",               guideline = \"CLSI\")   df_wide %>%     mutate(across(cipro:genta,                        function(x) as.sir(x,                                           mo = \"bacteria\",                                           guideline = \"CLSI\")))                           # for veterinary breakpoints, add 'host':   df_long$animal_species <- c(\"cats\", \"dogs\", \"horses\", \"cattle\")   df_long %>%     # given a certain data type, e.g. MIC values     mutate_if(is.mic, as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               host = \"animal_species\",               guideline = \"CLSI\")   df_long %>%     mutate(across(where(is.mic),                   function(x) as.sir(x,                                      mo = \"bacteria\",                                      ab = \"antibiotic\",                                      host = \"animal_species\",                                      guideline = \"CLSI\")))   df_wide %>%     mutate_at(vars(cipro:genta), as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               host = \"animal_species\",               guideline = \"CLSI\")   df_wide %>%     mutate(across(cipro:genta,                        function(x) as.sir(x,                                           mo = \"bacteria\",                                           host = \"animal_species\",                                           guideline = \"CLSI\")))      # to include information about urinary tract infections (UTI)   data.frame(mo = \"E. coli\",              nitrofuratoin = c(\"<= 2\", 32),              from_the_bladder = c(TRUE, FALSE)) %>%     as.sir(uti = \"from_the_bladder\")    data.frame(mo = \"E. coli\",              nitrofuratoin = c(\"<= 2\", 32),              specimen = c(\"urine\", \"blood\")) %>%     as.sir() # automatically determines urine isolates    df_wide %>%     mutate_at(vars(cipro:genta), as.sir, mo = \"E. coli\", uti = TRUE) } #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for gentamicin (GEN) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'antibiotic' (TESTAB, test Antibiotic), CLSI #> 2024... #>  WARNING  #>   • No CLSI 2024 MIC breakpoints available for test Antibiotic (TESTAB). #> Interpreting disk diffusion zones: 'antibiotic' (TESTAB, test Antibiotic), #> CLSI 2024... #>  WARNING  #>   • No CLSI 2024 DISK breakpoints available for test Antibiotic (TESTAB). #> Interpreting disk diffusion zones: 'antibiotic' (TESTAB, test Antibiotic), #> CLSI 2024... #>  WARNING  #>   • No CLSI 2024 DISK breakpoints available for test Antibiotic (TESTAB). #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `cipro = (function (x, ...) ...`. #> Caused by warning: #> ! The following animal host(s) could not be coerced: \"animal_species\" #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `across(...)`. #> Caused by warning: #> ! The following animal host(s) could not be coerced: \"animal_species\" #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'nitrofuratoin' (NIT, #> nitrofurantoin), EUCAST 2024... #>  OK  #> ℹ Assuming value \"urine\" in column 'specimen' reflects a urinary tract #>   infection. #>   Use as.sir(uti = FALSE) to prevent this. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'nitrofuratoin' (NIT, #> nitrofurantoin), EUCAST 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), EUCAST 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), EUCAST #> 2024... #>  OK  #>      microorganism amoxicillin cipro tobra genta ERY #> 1 Escherichia coli           8  <NA>     S     S   R   ## Using base R ------------------------------------------------  as.sir(df_wide) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: column 'amoxicillin' (AMX), EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: column 'cipro' (CIP, ciprofloxacin), EUCAST #> 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: column 'tobra' (TOB, tobramycin), EUCAST #> 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting disk diffusion zones: column 'genta' (GEN, gentamicin), EUCAST #> 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for gentamicin (GEN) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Assigning class 'sir' to already clean column 'ERY' (erythromycin)... #>  OK  #>      microorganism amoxicillin cipro tobra genta ERY #> 1 Escherichia coli           S     I     S     S   R  # return a 'logbook' about the results: sir_interpretation_history() #> # A tibble: 57 × 16 #>    datetime            index ab_given mo_given host_given ab   mo               #>  * <dttm>              <int> <chr>    <chr>    <chr>      <ab> <mo>             #>  1 2025-02-22 20:31:37     4 AMX      B_STRPT… human      AMX  B_STRPT_PNMN     #>  2 2025-02-22 20:31:43     4 genta    Escheri… human      GEN  B_[ORD]_ENTRBCTR #>  3 2025-02-22 20:31:44     4 genta    Escheri… human      GEN  B_[ORD]_ENTRBCTR #>  4 2025-02-22 20:31:44     4 genta    Escheri… cattle     GEN  B_ESCHR_COLI     #>  5 2025-02-22 20:31:45     4 genta    Escheri… cattle     GEN  B_ESCHR_COLI     #>  6 2025-02-22 20:31:37     3 AMX      B_STRPT… human      AMX  B_STRPT_PNMN     #>  7 2025-02-22 20:31:43     3 tobra    Escheri… human      TOB  B_[ORD]_ENTRBCTR #>  8 2025-02-22 20:31:44     3 tobra    Escheri… human      TOB  B_[ORD]_ENTRBCTR #>  9 2025-02-22 20:31:44     3 tobra    Escheri… horses     TOB  B_ESCHR_COLI     #> 10 2025-02-22 20:31:45     3 tobra    Escheri… horses     TOB  B_ESCHR_COLI     #> # ℹ 47 more rows #> # ℹ 9 more variables: host <chr>, method <chr>, input <chr>, outcome <sir>, #> #   notes <chr>, guideline <chr>, ref_table <chr>, uti <lgl>, #> #   breakpoint_S_R <chr>  # for single values as.sir(   x = as.mic(2),   mo = as.mo(\"S. pneumoniae\"),   ab = \"AMP\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMP' (ampicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ampicillin (AMP) in Streptococcus pneumoniae - assuming body site 'Non-meningitis'. #> Class 'sir' #> [1] R  as.sir(   x = as.disk(18),   mo = \"Strep pneu\", # `mo` will be coerced with as.mo()   ab = \"ampicillin\", # and `ab` with as.ab()   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: 'ampicillin' (AMP), EUCAST 2024... #>  OK  #> Class 'sir' #> [1] R   # For CLEANING existing SIR values ------------------------------------  as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", \"A\", \"B\", \"C\")) #> Warning: in as.sir(): 3 results in column '20' truncated (38%) that were invalid #> antimicrobial interpretations: \"A\", \"B\", and \"C\" as.sir(\"<= 0.002; S\") # will return \"S\" sir_data <- as.sir(c(rep(\"S\", 474), rep(\"I\", 36), rep(\"R\", 370))) is.sir(sir_data) #> [1] TRUE plot(sir_data) # for percentages  barplot(sir_data) # for frequencies   # as common in R, you can use as.integer() to return factor indices: as.integer(as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", NA))) #> [1]  1  2  3  4  5 NA # but for computational use, as.double() will return 1 for S, 2 for I/SDD, and 3 for R: as.double(as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", NA))) #> [1]  1  2  2  3 NA NA  # the dplyr way if (require(\"dplyr\")) {   example_isolates %>%     mutate_at(vars(PEN:RIF), as.sir)   # same:   example_isolates %>%     as.sir(PEN:RIF)    # fastest way to transform all columns with already valid AMR results to class `sir`:   example_isolates %>%     mutate_if(is_sir_eligible, as.sir)    # since dplyr 1.0.0, this can also be:   # example_isolates %>%   #   mutate(across(where(is_sir_eligible), as.sir)) } #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":null,"dir":"Reference","previous_headings":"","what":"Get ATC Properties from WHOCC Website — atc_online_property","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"Gets data WHOCC website determine properties Anatomical Therapeutic Chemical (ATC) (e.g. antibiotic), name, defined daily dose (DDD) standard unit.","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"","code":"atc_online_property(atc_code, property, administration = \"O\",   url = \"https://atcddd.fhi.no/atc_ddd_index/?code=%s&showdescription=no\",   url_vet = \"https://atcddd.fhi.no/atcvet/atcvet_index/?code=%s&showdescription=no\")  atc_online_groups(atc_code, ...)  atc_online_ddd(atc_code, ...)  atc_online_ddd_units(atc_code, ...)"},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"https://atcddd.fhi./atc_ddd_alterations__cumulative/ddd_alterations/abbrevations/","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"atc_code character (vector) ATC code(s) antibiotics, coerced .ab() ab_atc() internally valid ATC code property property ATC code. Valid values \"ATC\", \"Name\", \"DDD\", \"U\" (\"unit\"), \"Adm.R\", \"Note\" groups. last option, hierarchical groups ATC code returned, see Examples. administration type administration using property = \"Adm.R\", see Details url url website WHOCC. sign %s can used placeholder ATC codes. url_vet url website WHOCC veterinary medicine. sign %s can used placeholder ATC_vet codes (start \"Q\"). ... arguments pass atc_property","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"Options argument administration: \"Implant\" = Implant \"Inhal\" = Inhalation \"Instill\" = Instillation \"N\" = nasal \"O\" = oral \"P\" = parenteral \"R\" = rectal \"SL\" = sublingual/buccal \"TD\" = transdermal \"V\" = vaginal Abbreviations return values using property = \"U\" (unit): \"g\" = gram \"mg\" = milligram \"mcg\" = microgram \"U\" = unit \"TU\" = thousand units \"MU\" = million units \"mmol\" = millimole \"ml\" = millilitre (e.g. eyedrops) N.B. function requires internet connection works following packages installed: curl, rvest, xml2.","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"","code":"# \\donttest{ if (requireNamespace(\"curl\") && requireNamespace(\"rvest\") && requireNamespace(\"xml2\")) {   # oral DDD (Defined Daily Dose) of amoxicillin   atc_online_property(\"J01CA04\", \"DDD\", \"O\")   atc_online_ddd(ab_atc(\"amox\"))    # parenteral DDD (Defined Daily Dose) of amoxicillin   atc_online_property(\"J01CA04\", \"DDD\", \"P\")    atc_online_property(\"J01CA04\", property = \"groups\") # search hierarchical groups of amoxicillin } #> Loading required namespace: rvest #> [1] \"ANTIINFECTIVES FOR SYSTEMIC USE\"         #> [2] \"ANTIBACTERIALS FOR SYSTEMIC USE\"         #> [3] \"BETA-LACTAM ANTIBACTERIALS, PENICILLINS\" #> [4] \"Penicillins with extended spectrum\"      # }"},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":null,"dir":"Reference","previous_headings":"","what":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"Use function e.g. clinical texts health care records. returns list antiviral drugs, doses forms administration found texts.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"","code":"av_from_text(text, type = c(\"drug\", \"dose\", \"administration\"),   collapse = NULL, translate_av = FALSE, thorough_search = NULL,   info = interactive(), ...)"},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"text text analyse type type property search , either \"drug\", \"dose\" \"administration\", see Examples collapse character pass paste(, collapse = ...) return one character per element text, see Examples translate_av type = \"drug\": column name antivirals data set translate antibiotic abbreviations , using av_property(). default FALSE. Using TRUE equal using \"name\". thorough_search logical indicate whether input must extensively searched misspelling faulty input values. Setting TRUE take considerably time using FALSE. default, turn TRUE input elements contain maximum three words. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed .av()","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"list,  character collapse NULL","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"function also internally used .av(), although searches first drug name throw note drug names returned. Note: .av() function may use long regular expression match brand names antiviral drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"argument-type","dir":"Reference","previous_headings":"","what":"Argument type","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"default, function search antiviral drug names. text elements searched official names, ATC codes brand names. uses .av() internally, correct misspelling. type = \"dose\" (similar, like \"dosing\", \"doses\"), text elements searched numeric values higher 100 resemble years. output numeric. supports unit (g, mg, IE, etc.) multiple values one clinical text, see Examples. type = \"administration\" (abbreviations, like \"admin\", \"adm\"), text elements searched form drug administration. supports following forms (including common abbreviations): buccal, implant, inhalation, instillation, intravenous, nasal, oral, parenteral, rectal, sublingual, transdermal vaginal. Abbreviations oral ('po', 'per os') become \"oral\", values intravenous ('iv', 'intraven') become \"iv\". supports multiple values one clinical text, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"argument-collapse","dir":"Reference","previous_headings":"","what":"Argument collapse","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"Without using collapse, function return list. can convenient use e.g. inside mutate()):df %>% mutate(avx = av_from_text(clinical_text)) returned AV codes can transformed official names, groups, etc. av_* functions av_name() av_group(), using translate_av argument. using collapse, function return character:df %>% mutate(avx = av_from_text(clinical_text, collapse = \"|\"))","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"","code":"av_from_text(\"28/03/2020 valaciclovir po tid\") #> [[1]] #> Class 'av' #> [1] VALA #>  av_from_text(\"28/03/2020 valaciclovir po tid\", type = \"admin\") #> [[1]] #> [1] \"oral\" #>"},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of an Antiviral Drug — av_property","title":"Get Properties of an Antiviral Drug — av_property","text":"Use functions return specific property antiviral drug antivirals data set. input values evaluated internally .av().","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of an Antiviral Drug — av_property","text":"","code":"av_name(x, language = get_AMR_locale(), tolower = FALSE, ...)  av_cid(x, ...)  av_synonyms(x, ...)  av_tradenames(x, ...)  av_group(x, language = get_AMR_locale(), ...)  av_atc(x, ...)  av_loinc(x, ...)  av_ddd(x, administration = \"oral\", ...)  av_ddd_units(x, administration = \"oral\", ...)  av_info(x, language = get_AMR_locale(), ...)  av_url(x, open = FALSE, ...)  av_property(x, property = \"name\", language = get_AMR_locale(), ...)"},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of an Antiviral Drug — av_property","text":"x (vector ) text can coerced valid antiviral drug code .av() language language returned text - default system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. tolower logical indicate whether first character every output transformed lower case character. ... arguments passed .av() administration way administration, either \"oral\" \"iv\" open browse URL using utils::browseURL() property one column names one antivirals data set: vector_or(colnames(antivirals), sort = FALSE).","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of an Antiviral Drug — av_property","text":"integer case av_cid() named list case av_info() multiple av_atc()/av_synonyms()/av_tradenames() double case av_ddd() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of an Antiviral Drug — av_property","text":"output translated possible. function av_url() return direct URL official website. warning returned required ATC code available.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of an Antiviral Drug — av_property","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of an Antiviral Drug — av_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of an Antiviral Drug — av_property","text":"","code":"# all properties: av_name(\"ACI\") #> [1] \"Aciclovir\" av_atc(\"ACI\") #> [1] \"J05AB01\" av_cid(\"ACI\") #> [1] 135398513 av_synonyms(\"ACI\") #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            av_tradenames(\"ACI\") #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            av_group(\"ACI\") #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" av_url(\"ACI\") #>                                                              Aciclovir  #> \"https://atcddd.fhi.no/atc_ddd_index/?code=J05AB01&showdescription=no\"   # lowercase transformation av_name(x = c(\"ACI\", \"VALA\")) #> [1] \"Aciclovir\"    \"Valaciclovir\" av_name(x = c(\"ACI\", \"VALA\"), tolower = TRUE) #> [1] \"aciclovir\"    \"valaciclovir\"  # defined daily doses (DDD) av_ddd(\"ACI\", \"oral\") #> [1] 4 av_ddd_units(\"ACI\", \"oral\") #> [1] \"g\" av_ddd(\"ACI\", \"iv\") #> [1] 4 av_ddd_units(\"ACI\", \"iv\") #> [1] \"g\"  av_info(\"ACI\") # all properties as a list #> $av #> [1] \"ACI\" #>  #> $cid #> [1] 135398513 #>  #> $name #> [1] \"Aciclovir\" #>  #> $group #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" #>  #> $atc #> [1] \"J05AB01\" #>  #> $tradenames #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            #>  #> $loinc #> [1] \"\" #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] 4 #>  #> $ddd$oral$units #> [1] \"g\" #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] 4 #>  #> $ddd$iv$units #> [1] \"g\" #>  #>  #>   # all av_* functions use as.av() internally, so you can go from 'any' to 'any': av_atc(\"ACI\") #> [1] \"J05AB01\" av_group(\"J05AB01\") #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" av_loinc(\"abacavir\") #> [1] \"29113-8\" \"30273-7\" \"30287-7\" \"30303-2\" \"78772-1\" \"78773-9\" \"79134-3\" #> [8] \"80118-3\" av_name(\"29113-8\") #> [1] \"Abacavir\" av_name(135398513) #> [1] \"Aciclovir\" av_name(\"J05AB01\") #> [1] \"Aciclovir\""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":null,"dir":"Reference","previous_headings":"","what":"Check Availability of Columns — availability","title":"Check Availability of Columns — availability","text":"Easy check data availability columns data set. makes easy get idea antimicrobial combinations can used calculation e.g. susceptibility() resistance().","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Check Availability of Columns — availability","text":"","code":"availability(tbl, width = NULL)"},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Check Availability of Columns — availability","text":"tbl data.frame list width number characters present visual availability - default filling width console","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Check Availability of Columns — availability","text":"data.frame column names tbl row names","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Check Availability of Columns — availability","text":"function returns data.frame columns \"resistant\" \"visual_resistance\". values columns calculated resistance().","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Check Availability of Columns — availability","text":"","code":"availability(example_isolates) #>         count available     visual_availabilty resistant      visual_resistance #> date     2000    100.0% |####################|                                  #> patient  2000    100.0% |####################|                                  #> age      2000    100.0% |####################|                                  #> gender   2000    100.0% |####################|                                  #> ward     2000    100.0% |####################|                                  #> mo       2000    100.0% |####################|                                  #> PEN      1629     81.5% |################----|     73.7% |##############------| #> OXA       365     18.3% |###-----------------|     31.2% |######--------------| #> FLC       943     47.2% |#########-----------|     29.5% |#####---------------| #> AMX      1350     67.5% |#############-------|     59.6% |###########---------| #> AMC      1879     94.0% |##################--|     23.7% |####----------------| #> AMP      1350     67.5% |#############-------|     59.6% |###########---------| #> TZP      1001     50.0% |##########----------|     12.6% |##------------------| #> CZO       446     22.3% |####----------------|     44.6% |########------------| #> FEP       724     36.2% |#######-------------|     14.2% |##------------------| #> CXM      1789     89.5% |#################---|     26.3% |#####---------------| #> FOX       818     40.9% |########------------|     27.4% |#####---------------| #> CTX       943     47.2% |#########-----------|     15.5% |###-----------------| #> CAZ      1811     90.6% |##################--|     66.5% |#############-------| #> CRO       943     47.2% |#########-----------|     15.5% |###-----------------| #> GEN      1855     92.8% |##################--|     24.6% |####----------------| #> TOB      1351     67.6% |#############-------|     34.4% |######--------------| #> AMK       692     34.6% |######--------------|     63.7% |############--------| #> KAN       471     23.6% |####----------------|    100.0% |####################| #> TMP      1499     75.0% |###############-----|     38.1% |#######-------------| #> SXT      1759     88.0% |#################---|     20.5% |####----------------| #> NIT       743     37.2% |#######-------------|     17.1% |###-----------------| #> FOS       351     17.6% |###-----------------|     42.2% |########------------| #> LNZ      1023     51.2% |##########----------|     69.3% |#############-------| #> CIP      1409     70.5% |#############-------|     16.2% |###-----------------| #> MFX       211     10.6% |##------------------|     33.6% |######--------------| #> VAN      1861     93.1% |##################--|     38.3% |#######-------------| #> TEC       976     48.8% |#########-----------|     75.7% |###############-----| #> TCY      1200     60.0% |###########---------|     29.8% |#####---------------| #> TGC       798     39.9% |########------------|     12.7% |##------------------| #> DOX      1136     56.8% |###########---------|     27.7% |#####---------------| #> ERY      1894     94.7% |##################--|     57.2% |###########---------| #> CLI      1520     76.0% |###############-----|     61.2% |############--------| #> AZM      1894     94.7% |##################--|     57.2% |###########---------| #> IPM       889     44.5% |########------------|      6.2% |#-------------------| #> MEM       829     41.5% |########------------|      5.9% |#-------------------| #> MTR        34      1.7% |--------------------|     14.7% |##------------------| #> CHL       154      7.7% |#-------------------|     21.4% |####----------------| #> COL      1640     82.0% |################----|     81.2% |################----| #> MUP       270     13.5% |##------------------|      5.9% |#-------------------| #> RIF      1003     50.2% |##########----------|     69.6% |#############-------| # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     filter(mo == as.mo(\"Escherichia coli\")) %>%     select_if(is.sir) %>%     availability() } #>     count available       visual_availabilty resistant        visual_resistance #> PEN   467    100.0% |######################|    100.0% |######################| #> OXA     0      0.0% |----------------------|                                    #> FLC     0      0.0% |----------------------|                                    #> AMX   392     83.9% |##################----|     50.0% |###########-----------| #> AMC   467    100.0% |######################|     13.1% |##--------------------| #> AMP   392     83.9% |##################----|     50.0% |###########-----------| #> TZP   416     89.1% |###################---|      5.5% |#---------------------| #> CZO    82     17.6% |###-------------------|      2.4% |----------------------| #> FEP   317     67.9% |##############--------|      2.8% |----------------------| #> CXM   465     99.6% |######################|      5.4% |#---------------------| #> FOX   377     80.7% |#################-----|      6.9% |#---------------------| #> CTX   459     98.3% |#####################-|      2.4% |----------------------| #> CAZ   460     98.5% |#####################-|      2.4% |----------------------| #> CRO   459     98.3% |#####################-|      2.4% |----------------------| #> GEN   460     98.5% |#####################-|      2.0% |----------------------| #> TOB   462     98.9% |#####################-|      2.6% |----------------------| #> AMK   171     36.6% |########--------------|      0.0% |----------------------| #> KAN     0      0.0% |----------------------|                                    #> TMP   396     84.8% |##################----|     39.1% |########--------------| #> SXT   465     99.6% |######################|     31.6% |######----------------| #> NIT   458     98.1% |#####################-|      2.8% |----------------------| #> FOS    61     13.1% |##--------------------|      0.0% |----------------------| #> LNZ   467    100.0% |######################|    100.0% |######################| #> CIP   456     97.6% |#####################-|     12.5% |##--------------------| #> MFX    57     12.2% |##--------------------|    100.0% |######################| #> VAN   467    100.0% |######################|    100.0% |######################| #> TEC   467    100.0% |######################|    100.0% |######################| #> TCY     3      0.6% |----------------------|     66.7% |##############--------| #> TGC    68     14.6% |###-------------------|      0.0% |----------------------| #> DOX     0      0.0% |----------------------|                                    #> ERY   467    100.0% |######################|    100.0% |######################| #> CLI   467    100.0% |######################|    100.0% |######################| #> AZM   467    100.0% |######################|    100.0% |######################| #> IPM   422     90.4% |###################---|      0.0% |----------------------| #> MEM   418     89.5% |###################---|      0.0% |----------------------| #> MTR     2      0.4% |----------------------|      0.0% |----------------------| #> CHL     0      0.0% |----------------------|                                    #> COL   240     51.4% |###########-----------|      0.0% |----------------------| #> MUP     0      0.0% |----------------------|                                    #> RIF   467    100.0% |######################|    100.0% |######################| # }"},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Bug-Drug Combinations — bug_drug_combinations","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"Determine antimicrobial resistance (AMR) bug-drug combinations data set least 30 (default) isolates available per species. Use format() result prettify publishable/printable format, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"","code":"bug_drug_combinations(x, col_mo = NULL, FUN = mo_shortname,   include_n_rows = FALSE, ...)  # S3 method for class 'bug_drug_combinations' format(x, translate_ab = \"name (ab, atc)\",   language = get_AMR_locale(), minimum = 30, combine_SI = TRUE,   add_ab_group = TRUE, remove_intrinsic_resistant = FALSE,   decimal.mark = getOption(\"OutDec\"), big.mark = ifelse(decimal.mark ==   \",\", \".\", \",\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"x data set antibiotic columns, amox, AMX AMC col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). FUN function call mo column transform microorganism codes - default mo_shortname() include_n_rows logical indicate total number rows must included output ... arguments passed FUN translate_ab character length 1 containing column names antibiotics data set language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether values S, SDD, summed, resistance based R - default TRUE add_ab_group logical indicate group antimicrobials must included first column remove_intrinsic_resistant logical indicate rows columns 100% resistance tested antimicrobials must removed table decimal.mark character used indicate numeric     decimal point. big.mark character; empty used mark every     big.interval decimals (hence big)     decimal point.","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"function bug_drug_combinations() returns data.frame columns \"mo\", \"ab\", \"S\", \"SDD\", \"\", \"R\", \"total\".","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"function format() calculates resistance per bug-drug combination returns table ready reporting/publishing. Use combine_SI = TRUE (default) test R vs. S+combine_SI = FALSE test R+vs. S. table can also directly used R Markdown / Quarto without need e.g. knitr::kable().","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # \\donttest{ x <- bug_drug_combinations(example_isolates) head(x) #> # A tibble: 6 × 8 #>   mo                ab        S   SDD     I     R    NI total #>   <chr>             <chr> <int> <int> <int> <int> <int> <int> #> 1 (unknown species) AMC      15     0     0     0     0    15 #> 2 (unknown species) AMK       0     0     0     0     0     0 #> 3 (unknown species) AMP      15     0     0     1     0    16 #> 4 (unknown species) AMX      15     0     0     1     0    16 #> 5 (unknown species) AZM       3     0     0     3     0     6 #> 6 (unknown species) CAZ       0     0     0     0     0     0 #> Use 'format()' on this result to get a publishable/printable format. format(x, translate_ab = \"name (atc)\") #> # A tibble: 39 × 12 #>    Group     Drug  CoNS  `E. coli` `E. faecalis` `K. pneumoniae` `P. aeruginosa` #>    <chr>     <chr> <chr> <chr>     <chr>         <chr>           <chr>           #>  1 \"Aminogl… Amik… \"100… \"  0.0% … \"100.0% (39/… \"\"              \"\"              #>  2 \"\"        Gent… \" 13… \"  2.0% … \"100.0% (39/… \" 10.3% (6/58)\" \"  0.0% (0/30)\" #>  3 \"\"        Kana… \"100… \"\"        \"100.0% (39/… \"\"              \"100.0% (30/30… #>  4 \"\"        Tobr… \" 78… \"  2.6% … \"100.0% (39/… \" 10.3% (6/58)\" \"  0.0% (0/30)\" #>  5 \"Antimyc… Rifa… \"\"    \"100.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  6 \"Beta-la… Amox… \" 93… \" 50.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  7 \"\"        Amox… \" 42… \" 13.1% … \"\"            \" 10.3% (6/58)\" \"100.0% (30/30… #>  8 \"\"        Ampi… \" 93… \" 50.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  9 \"\"        Benz… \" 77… \"100.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #> 10 \"\"        Fluc… \" 42… \"\"        \"\"            \"\"              \"\"              #> # ℹ 29 more rows #> # ℹ 5 more variables: `P. mirabilis` <chr>, `S. aureus` <chr>, #> #   `S. epidermidis` <chr>, `S. hominis` <chr>, `S. pneumoniae` <chr>  # Use FUN to change to transformation of microorganism codes bug_drug_combinations(example_isolates,   FUN = mo_gramstain ) #> # A tibble: 80 × 8 #>    mo            ab        S   SDD     I     R    NI total #>  * <chr>         <chr> <int> <int> <int> <int> <int> <int> #>  1 Gram-negative AMC     463     0    89   174     0   726 #>  2 Gram-negative AMK     251     0     0     5     0   256 #>  3 Gram-negative AMP     226     0     0   405     0   631 #>  4 Gram-negative AMX     226     0     0   405     0   631 #>  5 Gram-negative AZM       1     0     2   696     0   699 #>  6 Gram-negative CAZ     607     0     0    27     0   634 #>  7 Gram-negative CHL       1     0     0    30     0    31 #>  8 Gram-negative CIP     610     0    11    63     0   684 #>  9 Gram-negative CLI      18     0     1   709     0   728 #> 10 Gram-negative COL     309     0     0    78     0   387 #> # ℹ 70 more rows #> Use 'format()' on this result to get a publishable/printable format.  bug_drug_combinations(example_isolates,   FUN = function(x) {     ifelse(x == as.mo(\"Escherichia coli\"),       \"E. coli\",       \"Others\"     )   } ) #> # A tibble: 80 × 8 #>    mo      ab        S   SDD     I     R    NI total #>  * <chr>   <chr> <int> <int> <int> <int> <int> <int> #>  1 E. coli AMC     332     0    74    61     0   467 #>  2 E. coli AMK     171     0     0     0     0   171 #>  3 E. coli AMP     196     0     0   196     0   392 #>  4 E. coli AMX     196     0     0   196     0   392 #>  5 E. coli AZM       0     0     0   467     0   467 #>  6 E. coli CAZ     449     0     0    11     0   460 #>  7 E. coli CHL       0     0     0     0     0     0 #>  8 E. coli CIP     398     0     1    57     0   456 #>  9 E. coli CLI       0     0     0   467     0   467 #> 10 E. coli COL     240     0     0     0     0   240 #> # ℹ 70 more rows #> Use 'format()' on this result to get a publishable/printable format. # }"},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Data set containing clinical breakpoints interpret MIC disk diffusion SIR values, according international guidelines. dataset contain breakpoints humans, 7 different animal groups, ECOFFs. breakpoints currently implemented: clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024; veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024; ECOFFs (Epidemiological Cut-Values): EUCAST 2020-2024 CLSI 2022-2024. Use .sir() transform MICs disks measurements SIR values.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"","code":"clinical_breakpoints"},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"tibble 34 063 observations 14 variables: guideline Name guideline type Breakpoint type, either \"ECOFF\", \"animal\", \"human\" host Host infectious agent. mostly useful veterinary breakpoints either \"ECOFF\", \"aquatic\", \"cats\", \"cattle\", \"dogs\", \"horse\", \"human\", \"poultry\", \"swine\" method Testing method, either \"DISK\" \"MIC\" site Body site breakpoint must applied, e.g. \"Oral\" \"Respiratory\" mo Microbial ID, see .mo() rank_index Taxonomic rank index mo 1 (subspecies/infraspecies) 5 (unknown microorganism) ab Antibiotic code used package, EARS-Net WHONET, see .ab() ref_tbl Info guideline rule can found disk_dose Dose used disk diffusion method breakpoint_S Lowest MIC value highest number millimetres leads \"S\" breakpoint_R Highest MIC value lowest number millimetres leads \"R\" uti logical value (TRUE/FALSE) indicate whether rule applies urinary tract infection (UTI) is_SDD logical value (TRUE/FALSE) indicate whether intermediate range \"S\" \"R\" interpreted \"SDD\", instead \"\". currently applies 24 breakpoints.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"different-types-of-breakpoints","dir":"Reference","previous_headings":"","what":"Different types of breakpoints","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Supported types breakpoints ECOFF, animal, human. ECOFF (Epidemiological cut-) values used antimicrobial susceptibility testing differentiate wild-type non-wild-type strains bacteria fungi. default \"human\", can also set package option AMR_breakpoint_type. Use .sir(..., breakpoint_type = ...) interpret raw data using specific breakpoint type, e.g. .sir(..., breakpoint_type = \"ECOFF\") use ECOFFs.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"imported-from-whonet","dir":"Reference","previous_headings":"","what":"Imported from WHONET","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Clinical breakpoints package validated imported WHONET, free desktop Windows application developed supported Collaborating Centre Surveillance Antimicrobial Resistance. can read website. developers WHONET AMR package contact sharing work. highly appreciate great development WHONET software.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"response-from-clsi-and-eucast","dir":"Reference","previous_headings":"","what":"Response from CLSI and EUCAST","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"CEO CLSI chairman EUCAST endorsed work public use AMR package (consequently use breakpoints) June 2023, future development distributing clinical breakpoints discussed meeting CLSI, EUCAST, , developers WHONET software, developers AMR package.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"download","dir":"Reference","previous_headings":"","what":"Download","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository. allow machine reading EUCAST CLSI guidelines, almost impossible MS Excel PDF files distributed EUCAST CLSI, though initiatives started overcome burdens. NOTE: AMR package (WHONET software well) contains rather complex internal methods apply guidelines. example, breakpoints must applied certain species groups (case package available microorganisms.groups data set). important considered using breakpoints use.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"","code":"clinical_breakpoints #> # A tibble: 34,063 × 14 #>    guideline   type  host  method site    mo            rank_index ab   ref_tbl  #>    <chr>       <chr> <chr> <chr>  <chr>   <mo>               <dbl> <ab> <chr>    #>  1 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 MEM  A. xylo… #>  2 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 MEM  A. xylo… #>  3 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 SXT  A. xylo… #>  4 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 SXT  A. xylo… #>  5 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 TZP  A. xylo… #>  6 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 TZP  A. xylo… #>  7 EUCAST 2024 human human DISK   NA      B_ACNTB                3 AMK  Acineto… #>  8 EUCAST 2024 human human DISK   Uncomp… B_ACNTB                3 AMK  Acineto… #>  9 EUCAST 2024 human human MIC    NA      B_ACNTB                3 AMK  Acineto… #> 10 EUCAST 2024 human human MIC    Uncomp… B_ACNTB                3 AMK  Acineto… #> # ℹ 34,053 more rows #> # ℹ 5 more variables: disk_dose <chr>, breakpoint_S <dbl>, breakpoint_R <dbl>, #> #   uti <lgl>, is_SDD <lgl>"},{"path":"https://msberends.github.io/AMR/reference/count.html","id":null,"dir":"Reference","previous_headings":"","what":"Count Available Isolates — count","title":"Count Available Isolates — count","text":"functions can used count resistant/susceptible microbial isolates. functions support quasiquotation pipes, can used summarise() dplyr package also support grouped variables, see Examples. count_resistant() used count resistant isolates, count_susceptible() used count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Count Available Isolates — count","text":"","code":"count_resistant(..., only_all_tested = FALSE)  count_susceptible(..., only_all_tested = FALSE)  count_S(..., only_all_tested = FALSE)  count_SI(..., only_all_tested = FALSE)  count_I(..., only_all_tested = FALSE)  count_IR(..., only_all_tested = FALSE)  count_R(..., only_all_tested = FALSE)  count_all(..., only_all_tested = FALSE)  n_sir(..., only_all_tested = FALSE)  count_df(data, translate_ab = \"name\", language = get_AMR_locale(),   combine_SI = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Count Available Isolates — count","text":"... one vectors (columns) antibiotic interpretations. transformed internally .sir() needed. only_all_tested (combination therapies, .e. using one variable ...): logical indicate isolates must tested antibiotics, see section Combination Therapy data data.frame containing columns class sir (see .sir()) translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Count Available Isolates — count","text":"integer","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Count Available Isolates — count","text":"functions meant count isolates. Use resistance()/susceptibility() functions calculate microbial resistance/susceptibility. function count_resistant() equal function count_R(). function count_susceptible() equal function count_SI(). function n_sir() alias count_all(). can used count available isolates, .e. input antibiotics available result (S, R). use equal n_distinct(). function equal count_susceptible(...) + count_resistant(...). function count_df() takes variable data sir class (created .sir()) counts number S's, 's R's. also supports grouped variables. function sir_df() works exactly like count_df(), adds percentage S, R.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Count Available Isolates — count","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"combination-therapy","dir":"Reference","previous_headings":"","what":"Combination Therapy","title":"Count Available Isolates — count","text":"using one variable ... (= combination therapy), use only_all_tested count isolates tested antibiotics/variables test . See example two antibiotics, Drug Drug B, susceptibility() works calculate %SI:   Please note , combination therapies, only_all_tested = TRUE applies :   , combination therapies, only_all_tested = FALSE applies :   Using only_all_tested impact using one antibiotic input.","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - -------------------------------------------------------------------- count_S()    +   count_I()    +   count_R()    = count_all()   proportion_S() + proportion_I() + proportion_R() = 1 count_S()    +   count_I()    +   count_R()    >= count_all()   proportion_S() + proportion_I() + proportion_R() >= 1"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Count Available Isolates — count","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info.  # base R ------------------------------------------------------------ count_resistant(example_isolates$AMX) # counts \"R\" #> [1] 804 count_susceptible(example_isolates$AMX) # counts \"S\" and \"I\" #> [1] 546 count_all(example_isolates$AMX) # counts \"S\", \"I\" and \"R\" #> [1] 1350  # be more specific count_S(example_isolates$AMX) #> Using count_S() is discouraged; use count_susceptible() instead to also #> consider \"I\" and \"SDD\" being susceptible. This note will be shown once for #> this session. #> [1] 543 count_SI(example_isolates$AMX) #> Note that count_SI() will also count dose-dependent susceptibility, #> 'SDD'. This note will be shown once for this session. #> [1] 546 count_I(example_isolates$AMX) #> Note that count_I() will also count dose-dependent susceptibility, 'SDD'. #> This note will be shown once for this session. #> [1] 3 count_IR(example_isolates$AMX) #> Using count_IR() is discouraged; use count_resistant() instead to not #> consider \"I\" and \"SDD\" being resistant. This note will be shown once for #> this session. #> [1] 807 count_R(example_isolates$AMX) #> [1] 804  # Count all available isolates count_all(example_isolates$AMX) #> [1] 1350 n_sir(example_isolates$AMX) #> [1] 1350  # n_sir() is an alias of count_all(). # Since it counts all available isolates, you can # calculate back to count e.g. susceptible isolates. # These results are the same: count_susceptible(example_isolates$AMX) #> [1] 546 susceptibility(example_isolates$AMX) * n_sir(example_isolates$AMX) #> [1] 546  # dplyr ------------------------------------------------------------- # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       R = count_R(CIP),       I = count_I(CIP),       S = count_S(CIP),       n1 = count_all(CIP), # the actual total; sum of all three       n2 = n_sir(CIP), # same - analogous to n_distinct       total = n()     ) # NOT the number of tested isolates!    # Number of available isolates for a whole antibiotic class   # (i.e., in this data set columns GEN, TOB, AMK, KAN)   example_isolates %>%     group_by(ward) %>%     summarise(across(aminoglycosides(), n_sir))    # Count co-resistance between amoxicillin/clav acid and gentamicin,   # so we can see that combination therapy does a lot more than mono therapy.   # Please mind that `susceptibility()` calculates percentages right away instead.   example_isolates %>% count_susceptible(AMC) # 1433   example_isolates %>% count_all(AMC) # 1879    example_isolates %>% count_susceptible(GEN) # 1399   example_isolates %>% count_all(GEN) # 1855    example_isolates %>% count_susceptible(AMC, GEN) # 1764   example_isolates %>% count_all(AMC, GEN) # 1936    # Get number of S+I vs. R immediately of selected columns   example_isolates %>%     select(AMX, CIP) %>%     count_df(translate = FALSE)    # It also supports grouping variables   example_isolates %>%     select(ward, AMX, CIP) %>%     group_by(ward) %>%     count_df(translate = FALSE) } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 12 × 4 #>    ward       antibiotic interpretation value #>  * <chr>      <chr>      <ord>          <int> #>  1 Clinical   AMX        SI               357 #>  2 Clinical   AMX        R                487 #>  3 Clinical   CIP        SI               741 #>  4 Clinical   CIP        R                128 #>  5 ICU        AMX        SI               158 #>  6 ICU        AMX        R                270 #>  7 ICU        CIP        SI               362 #>  8 ICU        CIP        R                 85 #>  9 Outpatient AMX        SI                31 #> 10 Outpatient AMX        R                 47 #> 11 Outpatient CIP        SI                78 #> 12 Outpatient CIP        R                 15 # }"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":null,"dir":"Reference","previous_headings":"","what":"Define Custom EUCAST Rules — custom_eucast_rules","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"Define custom EUCAST rules organisation specific analysis use output function eucast_rules().","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"","code":"custom_eucast_rules(...)"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"... rules formula notation, see instructions, Examples","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"list containing custom rules","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"organisations adoption EUCAST rules. function can used define custom EUCAST rules used eucast_rules() function.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"basics","dir":"Reference","previous_headings":"","what":"Basics","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"familiar case_when() function dplyr package, recognise input method set rules. Rules must set using R considers 'formula notation'. rule written tilde (~) consequence rule written tilde:   two custom EUCAST rules: TZP (piperacillin/tazobactam) \"S\", aminopenicillins (ampicillin amoxicillin) must made \"S\", TZP \"R\", aminopenicillins must made \"R\". rules can also printed console, immediately clear work:   rules (part tilde, example TZP == \"S\" TZP == \"R\") must evaluable data set: able run filter data set without errors. means example column TZP must exist. create sample data set test rules set:","code":"x <- custom_eucast_rules(TZP == \"S\" ~ aminopenicillins == \"S\",                          TZP == \"R\" ~ aminopenicillins == \"R\") x #> A set of custom EUCAST rules: #> #>   1. If TZP is \"S\" then set to  S : #>      amoxicillin (AMX), ampicillin (AMP) #> #>   2. If TZP is \"R\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) df <- data.frame(mo = c(\"Escherichia coli\", \"Klebsiella pneumoniae\"),                  TZP = as.sir(\"R\"),                  ampi = as.sir(\"S\"),                  cipro = as.sir(\"S\")) df #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    S     S #> 2 Klebsiella pneumoniae   R    S     S  eucast_rules(df, rules = \"custom\", custom_rules = x, info = FALSE) #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    R     S #> 2 Klebsiella pneumoniae   R    R     S"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"using-taxonomic-properties-in-rules","dir":"Reference","previous_headings":"","what":"Using taxonomic properties in rules","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"one exception columns used rules: column names microorganisms data set can also used, exist data set. column names : \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". Thus, next example work well, despite fact df data set contain column genus:","code":"y <- custom_eucast_rules(TZP == \"S\" & genus == \"Klebsiella\" ~ aminopenicillins == \"S\",                          TZP == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\")  eucast_rules(df, rules = \"custom\", custom_rules = y, info = FALSE) #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    S     S #> 2 Klebsiella pneumoniae   R    R     S"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"usage-of-multiple-antibiotics-and-antibiotic-group-names","dir":"Reference","previous_headings":"","what":"Usage of multiple antibiotics and antibiotic group names","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"can define antibiotic groups instead single antibiotics rule consequence, part tilde (~). examples , antibiotic group aminopenicillins includes ampicillin amoxicillin. Rules can also applied multiple antibiotics antibiotic groups simultaneously. Use c() function combine multiple antibiotics. instance, following example sets aminopenicillins ureidopenicillins \"R\" column TZP (piperacillin/tazobactam) \"R\":   34 antibiotic groups allowed rules (case-insensitive) can used combination: aminoglycosides(amikacin, amikacin/fosfomycin, apramycin, arbekacin, astromicin, bekanamycin, dibekacin, framycetin, gentamicin, gentamicin-high, habekacin, hygromycin, isepamicin, kanamycin, kanamycin-high, kanamycin/cephalexin, micronomicin, neomycin, netilmicin, pentisomicin, plazomicin, propikacin, ribostamycin, sisomicin, streptoduocin, streptomycin, streptomycin-high, tobramycin, tobramycin-high) aminopenicillins(amoxicillin ampicillin) antifungals(amorolfine, amphotericin B, amphotericin B-high, anidulafungin, butoconazole, caspofungin, ciclopirox, clotrimazole, econazole, fluconazole, flucytosine, fosfluconazole, griseofulvin, hachimycin, ibrexafungerp, isavuconazole, isoconazole, itraconazole, ketoconazole, manogepix, micafungin, miconazole, nystatin, oteseconazole, pimaricin, posaconazole, rezafungin, ribociclib, sulconazole, terbinafine, terconazole, voriconazole) antimycobacterials(4-aminosalicylic acid, calcium aminosalicylate, capreomycin, clofazimine, delamanid, enviomycin, ethambutol, ethambutol/isoniazid, ethionamide, isoniazid, isoniazid/sulfamethoxazole/trimethoprim/pyridoxine, morinamide, p-aminosalicylic acid, pretomanid, protionamide, pyrazinamide, rifabutin, rifampicin, rifampicin/ethambutol/isoniazid, rifampicin/isoniazid, rifampicin/pyrazinamide/ethambutol/isoniazid, rifampicin/pyrazinamide/isoniazid, rifamycin, rifapentine, simvastatin/fenofibrate, sodium aminosalicylate, streptomycin/isoniazid, terizidone, thioacetazone, thioacetazone/isoniazid, tiocarlide, viomycin) betalactams(amoxicillin, amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin, ampicillin/sulbactam, apalcillin, aspoxicillin, azidocillin, azlocillin, aztreonam, aztreonam/avibactam, aztreonam/nacubactam, bacampicillin, benzathine benzylpenicillin, benzathine phenoxymethylpenicillin, benzylpenicillin, biapenem, carbenicillin, carindacillin, carumonam, cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, ciclacillin, clometocillin, cloxacillin, dicloxacillin, doripenem, epicillin, ertapenem, flucloxacillin, hetacillin, imipenem, imipenem/EDTA, imipenem/relebactam, latamoxef, lenampicillin, loracarbef, mecillinam, meropenem, meropenem/nacubactam, meropenem/vaborbactam, metampicillin, meticillin, mezlocillin, mezlocillin/sulbactam, nafcillin, oxacillin, panipenem, penamecillin, penicillin/novobiocin, penicillin/sulbactam, pheneticillin, phenoxymethylpenicillin, piperacillin, piperacillin/sulbactam, piperacillin/tazobactam, piridicillin, pivampicillin, pivmecillinam, procaine benzylpenicillin, propicillin, razupenem, ritipenem, ritipenem acoxil, sarmoxicillin, sulbenicillin, sultamicillin, talampicillin, tebipenem, temocillin, ticarcillin, ticarcillin/clavulanic acid, tigemonam) betalactams_with_inhibitor(amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin/sulbactam, aztreonam/avibactam, aztreonam/nacubactam, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefoperazone/sulbactam, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefpodoxime/clavulanic acid, ceftaroline/avibactam, ceftazidime/avibactam, ceftazidime/clavulanic acid, ceftolozane/tazobactam, ceftriaxone/beta-lactamase inhibitor, imipenem/relebactam, meropenem/nacubactam, meropenem/vaborbactam, mezlocillin/sulbactam, penicillin/novobiocin, penicillin/sulbactam, piperacillin/sulbactam, piperacillin/tazobactam, ticarcillin/clavulanic acid) carbapenems(biapenem, doripenem, ertapenem, imipenem, imipenem/EDTA, imipenem/relebactam, meropenem, meropenem/nacubactam, meropenem/vaborbactam, panipenem, razupenem, ritipenem, ritipenem acoxil, tebipenem) cephalosporins(cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, latamoxef, loracarbef) cephalosporins_1st(cefacetrile, cefadroxil, cefalexin, cefaloridine, cefalotin, cefapirin, cefatrizine, cefazedone, cefazolin, cefroxadine, ceftezole, cephradine) cephalosporins_2nd(cefaclor, cefamandole, cefmetazole, cefonicid, ceforanide, cefotetan, cefotiam, cefoxitin, cefoxitin screening, cefprozil, cefuroxime, cefuroxime axetil, loracarbef) cephalosporins_3rd(cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefetamet, cefetamet pivoxil, cefixime, cefmenoxime, cefodizime, cefoperazone, cefoperazone/sulbactam, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotiam hexetil, cefovecin, cefpimizole, cefpiramide, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefsulodin, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, latamoxef) cephalosporins_4th(cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetecol, cefoselis, cefozopran, cefpirome, cefquinome) cephalosporins_5th(ceftaroline, ceftaroline/avibactam, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam) cephalosporins_except_caz(cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, latamoxef, loracarbef) fluoroquinolones(besifloxacin, ciprofloxacin, clinafloxacin, danofloxacin, delafloxacin, difloxacin, enoxacin, enrofloxacin, finafloxacin, fleroxacin, garenoxacin, gatifloxacin, gemifloxacin, grepafloxacin, lascufloxacin, levofloxacin, levonadifloxacin, lomefloxacin, marbofloxacin, metioxate, miloxacin, moxifloxacin, nadifloxacin, nifuroquine, norfloxacin, ofloxacin, orbifloxacin, pazufloxacin, pefloxacin, pradofloxacin, premafloxacin, prulifloxacin, rufloxacin, sarafloxacin, sitafloxacin, sparfloxacin, temafloxacin, tilbroquinol, tioxacin, tosufloxacin, trovafloxacin) glycopeptides(avoparcin, dalbavancin, norvancomycin, oritavancin, ramoplanin, teicoplanin, teicoplanin-macromethod, telavancin, vancomycin, vancomycin-macromethod) glycopeptides_except_lipo(avoparcin, norvancomycin, ramoplanin, teicoplanin, teicoplanin-macromethod, vancomycin, vancomycin-macromethod) isoxazolylpenicillins(cloxacillin, dicloxacillin, flucloxacillin, meticillin, oxacillin) lincosamides(clindamycin, lincomycin, pirlimycin) lipoglycopeptides(dalbavancin, oritavancin, telavancin) macrolides(acetylmidecamycin, acetylspiramycin, azithromycin, clarithromycin, dirithromycin, erythromycin, flurithromycin, gamithromycin, josamycin, kitasamycin, meleumycin, midecamycin, miocamycin, nafithromycin, oleandomycin, rokitamycin, roxithromycin, solithromycin, spiramycin, telithromycin, tildipirosin, tilmicosin, troleandomycin, tulathromycin, tylosin, tylvalosin) monobactams(aztreonam, aztreonam/avibactam, aztreonam/nacubactam, carumonam, tigemonam) nitrofurans(furazidin, furazolidone, nifurtoinol, nitrofurantoin, nitrofurazone) oxazolidinones(cadazolid, cycloserine, linezolid, tedizolid, thiacetazone) penicillins(amoxicillin, amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin, ampicillin/sulbactam, apalcillin, aspoxicillin, azidocillin, azlocillin, bacampicillin, benzathine benzylpenicillin, benzathine phenoxymethylpenicillin, benzylpenicillin, carbenicillin, carindacillin, ciclacillin, clometocillin, cloxacillin, dicloxacillin, epicillin, flucloxacillin, hetacillin, lenampicillin, mecillinam, metampicillin, meticillin, mezlocillin, mezlocillin/sulbactam, nafcillin, oxacillin, penamecillin, penicillin/novobiocin, penicillin/sulbactam, pheneticillin, phenoxymethylpenicillin, piperacillin, piperacillin/sulbactam, piperacillin/tazobactam, piridicillin, pivampicillin, pivmecillinam, procaine benzylpenicillin, propicillin, sarmoxicillin, sulbenicillin, sultamicillin, talampicillin, temocillin, ticarcillin, ticarcillin/clavulanic acid) phenicols(chloramphenicol, florfenicol, thiamphenicol) polymyxins(colistin, polymyxin B, polymyxin B/polysorbate 80) quinolones(besifloxacin, cinoxacin, ciprofloxacin, ciprofloxacin/metronidazole, ciprofloxacin/ornidazole, ciprofloxacin/tinidazole, clinafloxacin, danofloxacin, delafloxacin, difloxacin, enoxacin, enrofloxacin, finafloxacin, fleroxacin, flumequine, garenoxacin, gatifloxacin, gemifloxacin, grepafloxacin, lascufloxacin, levofloxacin, levonadifloxacin, lomefloxacin, marbofloxacin, metioxate, miloxacin, moxifloxacin, nadifloxacin, nalidixic acid, nemonoxacin, nifuroquine, nitroxoline, norfloxacin, ofloxacin, orbifloxacin, oxolinic acid, pazufloxacin, pefloxacin, pipemidic acid, piromidic acid, pradofloxacin, premafloxacin, prulifloxacin, rosoxacin, rufloxacin, sarafloxacin, sitafloxacin, sparfloxacin, temafloxacin, tilbroquinol, tioxacin, tosufloxacin, trovafloxacin) rifamycins(rifabutin, rifampicin, rifampicin/ethambutol/isoniazid, rifampicin/isoniazid, rifampicin/pyrazinamide/ethambutol/isoniazid, rifampicin/pyrazinamide/isoniazid, rifamycin, rifapentine) streptogramins(pristinamycin quinupristin/dalfopristin) tetracyclines(cetocycline, chlortetracycline, clomocycline, demeclocycline, doxycycline, eravacycline, lymecycline, metacycline, minocycline, omadacycline, oxytetracycline, penimepicycline, rolitetracycline, sarecycline, tetracycline, tigecycline) tetracyclines_except_tgc(cetocycline, chlortetracycline, clomocycline, demeclocycline, doxycycline, eravacycline, lymecycline, metacycline, minocycline, omadacycline, oxytetracycline, penimepicycline, rolitetracycline, sarecycline, tetracycline) trimethoprims(brodimoprim, sulfadiazine, sulfadiazine/tetroxoprim, sulfadiazine/trimethoprim, sulfadimethoxine, sulfadimidine, sulfadimidine/trimethoprim, sulfafurazole, sulfaisodimidine, sulfalene, sulfamazone, sulfamerazine, sulfamerazine/trimethoprim, sulfamethizole, sulfamethoxazole, sulfamethoxypyridazine, sulfametomidine, sulfametoxydiazine, sulfametrole/trimethoprim, sulfamoxole, sulfamoxole/trimethoprim, sulfanilamide, sulfaperin, sulfaphenazole, sulfapyridine, sulfathiazole, sulfathiourea, trimethoprim, trimethoprim/sulfamethoxazole) ureidopenicillins(azlocillin, mezlocillin, piperacillin, piperacillin/tazobactam)","code":"x <- custom_eucast_rules(TZP == \"R\" ~ c(aminopenicillins, ureidopenicillins) == \"R\") x #> A set of custom EUCAST rules: #> #>   1. If TZP is \"R\" then set to \"R\": #>      amoxicillin (AMX), ampicillin (AMP), azlocillin (AZL), mezlocillin (MEZ), piperacillin (PIP), piperacillin/tazobactam (TZP)"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"","code":"x <- custom_eucast_rules(   AMC == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\",   AMC == \"I\" & genus == \"Klebsiella\" ~ aminopenicillins == \"I\" ) x #> A set of custom EUCAST rules: #>  #>   1. If AMC is  R  and genus is \"Klebsiella\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   2. If AMC is  I  and genus is \"Klebsiella\" then set to  I : #>      amoxicillin (AMX), ampicillin (AMP)  # run the custom rule set (verbose = TRUE will return a logbook instead of the data set): eucast_rules(example_isolates,   rules = \"custom\",   custom_rules = x,   info = FALSE,   verbose = TRUE ) #> # A tibble: 8 × 9 #>     row col   mo_fullname     old   new   rule  rule_group rule_name rule_source #>   <int> <chr> <chr>           <ord> <chr> <chr> <chr>      <chr>     <chr>       #> 1    33 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 2    33 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 3    34 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 4    34 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 5   531 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 6   531 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 7  1485 AMP   Klebsiella oxy… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 8  1485 AMX   Klebsiella oxy… R     I     \"rep… Custom EU… Custom E… Object 'x'…  # combine rule sets x2 <- c(   x,   custom_eucast_rules(TZP == \"R\" ~ carbapenems == \"R\") ) x2 #> A set of custom EUCAST rules: #>  #>   1. If AMC is  R  and genus is \"Klebsiella\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   2. If AMC is  I  and genus is \"Klebsiella\" then set to  I : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   3. If TZP is  R  then set to  R : #>      biapenem (BIA), doripenem (DOR), ertapenem (ETP), imipenem (IPM), #>      imipenem/EDTA (IPE), imipenem/relebactam (IMR), meropenem (MEM), #>      meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), panipenem (PAN), #>      razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), tebipenem (TBP)"},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"EUCAST breakpoints used package based dosages data set. can retrieved eucast_dosage().","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"","code":"dosage"},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"tibble 503 observations 9 variables: ab Antibiotic ID used package (AMC), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available name Official name antimicrobial drug used WHONET/EARS-Net type Type dosage, either \"high_dosage\", \"standard_dosage\", \"uncomplicated_uti\" dose Dose, \"2 g\" \"25 mg/kg\" dose_times Number times dose must administered administration Route administration, either \"im\", \"iv\", \"oral\" notes Additional dosage notes original_txt Original text PDF file EUCAST eucast_version Version number EUCAST Clinical Breakpoints guideline dosages apply, either 13, 12, 11","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"","code":"dosage #> # A tibble: 503 × 9 #>    ab   name            type  dose  dose_times administration notes original_txt #>    <ab> <chr>           <chr> <chr>      <int> <chr>          <chr> <chr>        #>  1 AMK  Amikacin        stan… 25-3…          1 iv             \"\"    25-30 mg/kg… #>  2 AMX  Amoxicillin     high… 2 g            6 iv             \"\"    2 g x 6 iv   #>  3 AMX  Amoxicillin     stan… 1 g            3 iv             \"\"    1 g x 3-4 iv #>  4 AMX  Amoxicillin     high… 0.75…          3 oral           \"\"    0.75-1 g x … #>  5 AMX  Amoxicillin     stan… 0.5 g          3 oral           \"\"    0.5 g x 3 o… #>  6 AMX  Amoxicillin     unco… 0.5 g          3 oral           \"\"    0.5 g x 3 o… #>  7 AMC  Amoxicillin/cl… high… 2 g …          3 iv             \"\"    (2 g amoxic… #>  8 AMC  Amoxicillin/cl… stan… 1 g …          3 iv             \"\"    (1 g amoxic… #>  9 AMC  Amoxicillin/cl… high… 0.87…          3 oral           \"\"    (0.875 g am… #> 10 AMC  Amoxicillin/cl… stan… 0.5 …          3 oral           \"\"    (0.5 g amox… #> # ℹ 493 more rows #> # ℹ 1 more variable: eucast_version <dbl>"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply EUCAST Rules — eucast_rules","title":"Apply EUCAST Rules — eucast_rules","text":"Apply rules clinical breakpoints notes expected resistant phenotypes defined European Committee Antimicrobial Susceptibility Testing (EUCAST, https://www.eucast.org), see Source. Use eucast_dosage() get data.frame advised dosages certain bug-drug combination, based dosage data set. improve interpretation antibiogram EUCAST rules applied, non-EUCAST rules can applied default, see Details.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply EUCAST Rules — eucast_rules","text":"","code":"eucast_rules(x, col_mo = NULL, info = interactive(),   rules = getOption(\"AMR_eucastrules\", default = c(\"breakpoints\", \"expert\")),   verbose = FALSE, version_breakpoints = 12, version_expertrules = 3.3,   ampc_cephalosporin_resistance = NA, only_sir_columns = FALSE,   custom_rules = NULL, overwrite = TRUE, ...)  eucast_dosage(ab, administration = \"iv\", version_breakpoints = 12)"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Apply EUCAST Rules — eucast_rules","text":"EUCAST Expert Rules. Version 2.0, 2012. Leclercq et al. EUCAST expert rules antimicrobial susceptibility testing. Clin Microbiol Infect. 2013;19(2):141-60; doi:10.1111/j.1469-0691.2011.03703.x EUCAST Expert Rules, Intrinsic Resistance Exceptional Phenotypes Tables. Version 3.1, 2016. (link) EUCAST Intrinsic Resistance Unusual Phenotypes. Version 3.2, 2020. (link) EUCAST Intrinsic Resistance Unusual Phenotypes. Version 3.3, 2021. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 9.0, 2019. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 10.0, 2020. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 11.0, 2021. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 12.0, 2022. (link)","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply EUCAST Rules — eucast_rules","text":"x data set antibiotic columns, amox, AMX AMC col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). info logical indicate whether progress printed console - default print interactive sessions rules character vector specifies rules applied. Must one \"breakpoints\", \"expert\", \"\", \"custom\", \"\", defaults c(\"breakpoints\", \"expert\"). default value can set another value using package option AMR_eucastrules: options(AMR_eucastrules = \"\"). using \"custom\", sure fill argument custom_rules . Custom rules can created custom_eucast_rules(). verbose logical turn Verbose mode (default ). Verbose mode, function apply rules data, instead returns data set logbook form extensive info rows columns effected way. Using Verbose mode takes lot time. version_breakpoints version number use EUCAST Clinical Breakpoints guideline. Can \"12.0\", \"11.0\", \"10.0\". version_expertrules version number use EUCAST Expert Rules Intrinsic Resistance guideline. Can \"3.3\", \"3.2\", \"3.1\". ampc_cephalosporin_resistance character value applied cefotaxime, ceftriaxone ceftazidime AmpC de-repressed cephalosporin-resistant mutants - default NA. Currently works version_expertrules 3.2 higher; version 'EUCAST Expert Rules Enterobacterales' state results cefotaxime, ceftriaxone ceftazidime reported note, results suppressed (emptied) three drugs. value NA (default) argument remove results three drugs, e.g. value \"R\" make results drugs resistant. Use NULL FALSE alter results three drugs AmpC de-repressed cephalosporin-resistant mutants. Using TRUE equal using \"R\".  EUCAST Expert Rules v3.2, rule applies : Citrobacter braakii, Citrobacter freundii, Citrobacter gillenii, Citrobacter murliniae, Citrobacter rodenticum, Citrobacter sedlakii, Citrobacter werkmanii, Citrobacter youngae, Enterobacter, Hafnia alvei, Klebsiella aerogenes, Morganella morganii, Providencia, Serratia. only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE) custom_rules custom rules apply, created custom_eucast_rules() overwrite logical indicate whether non-NA values must overwritten (defaults TRUE). FALSE, NA values changed. ... column name antibiotic, see section Antibiotics ab (vector ) text can coerced valid antibiotic drug code .ab() administration route administration, either \"im\", \"iv\", \"oral\"","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply EUCAST Rules — eucast_rules","text":"input x, possibly edited values antibiotics. , verbose = TRUE, data.frame original new values affected bug-drug combinations.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Apply EUCAST Rules — eucast_rules","text":"Note: function translate MIC values SIR values. Use .sir() . Note: ampicillin (AMP, J01CA01) available amoxicillin (AMX, J01CA04) , latter used rules dependency ampicillin. drugs interchangeable comes expression antimicrobial resistance. file containing EUCAST rules located : https://github.com/msberends/AMR/blob/main/data-raw/eucast_rules.tsv.  Note: Old taxonomic names replaced current taxonomy applicable. example, Ochrobactrum anthropi renamed Brucella anthropi 2020; original EUCAST rules v3.1 v3.2 yet contain new taxonomic name. AMR package contains full microbial taxonomy updated June 24th, 2024, see microorganisms.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"custom-rules","dir":"Reference","previous_headings":"","what":"Custom Rules","title":"Apply EUCAST Rules — eucast_rules","text":"Custom rules can created using custom_eucast_rules(), e.g.:","code":"x <- custom_eucast_rules(AMC == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\",                          AMC == \"I\" & genus == \"Klebsiella\" ~ aminopenicillins == \"I\")  eucast_rules(example_isolates, rules = \"custom\", custom_rules = x)"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"-other-rules","dir":"Reference","previous_headings":"","what":"'Other' Rules","title":"Apply EUCAST Rules — eucast_rules","text":"processing, two non-EUCAST rules drug combinations can applied improve efficacy EUCAST rules, reliability data (analysis). rules : drug enzyme inhibitor set S drug without enzyme inhibitor S drug without enzyme inhibitor set R drug enzyme inhibitor R Important examples include amoxicillin amoxicillin/clavulanic acid, trimethoprim trimethoprim/sulfamethoxazole. Needless say, rules work, drugs must available data set. Since rules officially approved EUCAST, applied default. use rules, include \"\" rules argument, use eucast_rules(..., rules = \"\"). can also set package option AMR_eucastrules, .e. run options(AMR_eucastrules = \"\").","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"antibiotics","dir":"Reference","previous_headings":"","what":"Antibiotics","title":"Apply EUCAST Rules — eucast_rules","text":"define antibiotics column names, leave determine automatically guess_ab_col() input text (case-insensitive), use NULL skip column (e.g. TIC = NULL skip ticarcillin). Manually defined non-existing columns skipped warning. following antibiotics eligible functions eucast_rules() mdro(). shown format 'name (antimicrobial ID, ATC code)', sorted alphabetically: Amikacin (AMK, J01GB06), amoxicillin (AMX, J01CA04), amoxicillin/clavulanic acid (AMC, J01CR02), ampicillin (AMP, J01CA01), ampicillin/sulbactam (SAM, J01CR01), apramycin (APR, QA07AA92), arbekacin (ARB, J01GB12), aspoxicillin (APX, J01CA19), azidocillin (AZD, J01CE04), azithromycin (AZM, J01FA10), azlocillin (AZL, J01CA09), aztreonam (ATM, J01DF01), bacampicillin (BAM, J01CA06), bekanamycin (BEK, J01GB13), benzathine benzylpenicillin (BNB, J01CE08), benzathine phenoxymethylpenicillin (BNP, J01CE10), benzylpenicillin (PEN, J01CE01), besifloxacin (BES, S01AE08), biapenem (BIA, J01DH05), carbenicillin (CRB, J01CA03), carindacillin (CRN, J01CA05), carumonam (CAR, J01DF02), cefacetrile (CAC, J01DB10), cefaclor (CEC, J01DC04), cefadroxil (CFR, J01DB05), cefalexin (LEX, J01DB01), cefaloridine (RID, J01DB02), cefalotin (CEP, J01DB03), cefamandole (MAN, J01DC03), cefapirin (HAP, J01DB08), cefatrizine (CTZ, J01DB07), cefazedone (CZD, J01DB06), cefazolin (CZO, J01DB04), cefcapene (CCP, J01DD17), cefdinir (CDR, J01DD15), cefditoren (DIT, J01DD16), cefepime (FEP, J01DE01), cefepime/amikacin (CFA, J01RA06), cefetamet (CAT, J01DD10), cefiderocol (FDC, J01DI04), cefixime (CFM, J01DD08), cefmenoxime (CMX, J01DD05), cefmetazole (CMZ, J01DC09), cefodizime (DIZ, J01DD09), cefonicid (CID, J01DC06), cefoperazone (CFP, J01DD12), cefoperazone/sulbactam (CSL, J01DD62), ceforanide (CND, J01DC11), cefotaxime (CTX, J01DD01), cefotaxime/clavulanic acid (CTC, J01DD51), cefotetan (CTT, J01DC05), cefotiam (CTF, J01DC07), cefovecin (FOV, QJ01DD91), cefoxitin (FOX, J01DC01), cefozopran (ZOP, J01DE03), cefpiramide (CPM, J01DD11), cefpirome (CPO, J01DE02), cefpodoxime (CPD, J01DD13), cefprozil (CPR, J01DC10), cefquinome (CEQ, QG51AA07), cefroxadine (CRD, J01DB11), cefsulodin (CFS, J01DD03), ceftaroline (CPT, J01DI02), ceftazidime (CAZ, J01DD02), ceftazidime/clavulanic acid (CCV, J01DD52), cefteram (CEM, J01DD18), ceftezole (CTL, J01DB12), ceftibuten (CTB, J01DD14), ceftiofur (TIO, QJ01DD90), ceftizoxime (CZX, J01DD07), ceftobiprole medocaril (CFM1, J01DI01), ceftolozane/tazobactam (CZT, J01DI54), ceftriaxone (CRO, J01DD04), ceftriaxone/beta-lactamase inhibitor (CEB, J01DD63), cefuroxime (CXM, J01DC02), cephradine (CED, J01DB09), chloramphenicol (CHL, J01BA01), ciprofloxacin (CIP, J01MA02), clarithromycin (CLR, J01FA09), clindamycin (CLI, J01FF01), clometocillin (CLM, J01CE07), cloxacillin (CLO, J01CF02), colistin (COL, J01XB01), cycloserine (CYC, J04AB01), dalbavancin (DAL, J01XA04), danofloxacin (DAN, QJ01MA92), daptomycin (DAP, J01XX09), delafloxacin (DFX, J01MA23), dibekacin (DKB, J01GB09), dicloxacillin (DIC, J01CF01), difloxacin (DIF, QJ01MA94), dirithromycin (DIR, J01FA13), doripenem (DOR, J01DH04), doxycycline (DOX, J01AA02), enoxacin (ENX, J01MA04), enrofloxacin (ENR, QJ01MA90), epicillin (EPC, J01CA07), ertapenem (ETP, J01DH03), erythromycin (ERY, J01FA01), fleroxacin (FLE, J01MA08), flucloxacillin (FLC, J01CF05), flurithromycin (FLR1, J01FA14), fosfomycin (FOS, J01XX01), framycetin (FRM, D09AA01), fusidic acid (FUS, J01XC01), gamithromycin (GAM, QJ01FA95), garenoxacin (GRN, J01MA19), gatifloxacin (GAT, J01MA16), gemifloxacin (GEM, J01MA15), gentamicin (GEN, J01GB03), grepafloxacin (GRX, J01MA11), hetacillin (HET, J01CA18), imipenem (IPM, J01DH51), imipenem/relebactam (IMR, J01DH56), isepamicin (ISE, J01GB11), josamycin (JOS, J01FA07), kanamycin (KAN, J01GB04), kitasamycin (KIT, QJ01FA93), lascufloxacin (LSC, J01MA25), latamoxef (LTM, J01DD06), levofloxacin (LVX, J01MA12), levonadifloxacin (LND, J01MA24), lincomycin (LIN, J01FF02), linezolid (LNZ, J01XX08), lomefloxacin (LOM, J01MA07), loracarbef (LOR, J01DC08), marbofloxacin (MAR, QJ01MA93), mecillinam (MEC, J01CA11), meropenem (MEM, J01DH02), meropenem/vaborbactam (MEV, J01DH52), metampicillin (MTM, J01CA14), meticillin (MET, J01CF03), mezlocillin (MEZ, J01CA10), micronomicin (MCR, S01AA22), midecamycin (MID, J01FA03), minocycline (MNO, J01AA08), miocamycin (MCM, J01FA11), moxifloxacin (MFX, J01MA14), nadifloxacin (NAD, D10AF05), nafcillin (NAF, J01CF06), nalidixic acid (NAL, J01MB02), neomycin (NEO, J01GB05), netilmicin (NET, J01GB07), nitrofurantoin (NIT, J01XE01), norfloxacin (, J01MA06), novobiocin (NOV, QJ01XX95), ofloxacin (OFX, J01MA01), oleandomycin (OLE, J01FA05), orbifloxacin (ORB, QJ01MA95), oritavancin (ORI, J01XA05), oxacillin (OXA, J01CF04), panipenem (PAN, J01DH55), pazufloxacin (PAZ, J01MA18), pefloxacin (PEF, J01MA03), penamecillin (PNM, J01CE06), pheneticillin (PHE, J01CE05), phenoxymethylpenicillin (PHN, J01CE02), piperacillin (PIP, J01CA12), piperacillin/tazobactam (TZP, J01CR05), pirlimycin (PRL, QJ51FF90), pivampicillin (PVM, J01CA02), pivmecillinam (PME, J01CA08), plazomicin (PLZ, J01GB14), polymyxin B (PLB, J01XB02), pradofloxacin (PRA, QJ01MA97), pristinamycin (PRI, J01FG01), procaine benzylpenicillin (PRB, J01CE09), propicillin (PRP, J01CE03), prulifloxacin (PRU, J01MA17), quinupristin/dalfopristin (QDA, QJ01FG02), ribostamycin (RST, J01GB10), rifampicin (RIF, J04AB02), rokitamycin (ROK, J01FA12), roxithromycin (RXT, J01FA06), rufloxacin (RFL, J01MA10), sarafloxacin (SAR, QJ01MA98), sisomicin (SIS, J01GB08), sitafloxacin (SIT, J01MA21), solithromycin (SOL, J01FA16), sparfloxacin (SPX, J01MA09), spiramycin (SPI, J01FA02), streptoduocin (STR, J01GA02), streptomycin (STR1, J01GA01), sulbenicillin (SBC, J01CA16), sulfadiazine (SDI, J01EC02), sulfadiazine/trimethoprim (SLT1, J01EE02), sulfadimethoxine (SUD, J01ED01), sulfadimidine (SDM, J01EB03), sulfadimidine/trimethoprim (SLT2, J01EE05), sulfafurazole (SLF, J01EB05), sulfaisodimidine (SLF1, J01EB01), sulfalene (SLF2, J01ED02), sulfamazone (SZO, J01ED09), sulfamerazine (SLF3, J01ED07), sulfamerazine/trimethoprim (SLT3, J01EE07), sulfamethizole (SLF4, J01EB02), sulfamethoxazole (SMX, J01EC01), sulfamethoxypyridazine (SLF5, J01ED05), sulfametomidine (SLF6, J01ED03), sulfametoxydiazine (SLF7, J01ED04), sulfametrole/trimethoprim (SLT4, J01EE03), sulfamoxole (SLF8, J01EC03), sulfamoxole/trimethoprim (SLT5, J01EE04), sulfanilamide (SLF9, J01EB06), sulfaperin (SLF10, J01ED06), sulfaphenazole (SLF11, J01ED08), sulfapyridine (SLF12, J01EB04), sulfathiazole (SUT, J01EB07), sulfathiourea (SLF13, J01EB08), sultamicillin (SLT6, J01CR04), talampicillin (TAL, J01CA15), tebipenem (TBP, J01DH06), tedizolid (TZD, J01XX11), teicoplanin (TEC, J01XA02), telavancin (TLV, J01XA03), telithromycin (TLT, J01FA15), temafloxacin (TMX, J01MA05), temocillin (TEM, J01CA17), tetracycline (TCY, J01AA07), ticarcillin (TIC, J01CA13), ticarcillin/clavulanic acid (TCC, J01CR03), tigecycline (TGC, J01AA12), tilbroquinol (TBQ, P01AA05), tildipirosin (TIP, QJ01FA96), tilmicosin (TIL, QJ01FA91), tobramycin (TOB, J01GB01), tosufloxacin (TFX, J01MA22), trimethoprim (TMP, J01EA01), trimethoprim/sulfamethoxazole (SXT, J01EE01), troleandomycin (TRL, J01FA08), trovafloxacin (TVA, J01MA13), tulathromycin (TUL, QJ01FA94), tylosin (TYL, QJ01FA90), tylvalosin (TYL1, QJ01FA92), vancomycin (VAN, J01XA01)","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Apply EUCAST Rules — eucast_rules","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Apply EUCAST Rules — eucast_rules","text":"","code":"# \\donttest{ a <- data.frame(   mo = c(     \"Staphylococcus aureus\",     \"Enterococcus faecalis\",     \"Escherichia coli\",     \"Klebsiella pneumoniae\",     \"Pseudomonas aeruginosa\"   ),   VAN = \"-\", # Vancomycin   AMX = \"-\", # Amoxicillin   COL = \"-\", # Colistin   CAZ = \"-\", # Ceftazidime   CXM = \"-\", # Cefuroxime   PEN = \"S\", # Benzylpenicillin   FOX = \"S\", # Cefoxitin   stringsAsFactors = FALSE )  head(a) #>                       mo VAN AMX COL CAZ CXM PEN FOX #> 1  Staphylococcus aureus   -   -   -   -   -   S   S #> 2  Enterococcus faecalis   -   -   -   -   -   S   S #> 3       Escherichia coli   -   -   -   -   -   S   S #> 4  Klebsiella pneumoniae   -   -   -   -   -   S   S #> 5 Pseudomonas aeruginosa   -   -   -   -   -   S   S   # apply EUCAST rules: some results wil be changed b <- eucast_rules(a) #> Warning: in eucast_rules(): not all columns with antimicrobial results are of #> class 'sir'. Transform them on beforehand, with e.g.: #>   - a %>% as.sir(PEN:FOX) #>   - a %>% mutate_if(is_sir_eligible, as.sir) #>   - a %>% mutate(across(where(is_sir_eligible), as.sir))  head(b) #>                       mo VAN AMX COL CAZ CXM PEN FOX #> 1  Staphylococcus aureus   -   -   R   R   S   S   S #> 2  Enterococcus faecalis   -   -   R   R   R   S   R #> 3       Escherichia coli   R   -   -   -   -   R   S #> 4  Klebsiella pneumoniae   R   R   -   -   -   R   S #> 5 Pseudomonas aeruginosa   R   R   -   -   R   R   R   # do not apply EUCAST rules, but rather get a data.frame # containing all details about the transformations: c <- eucast_rules(a, verbose = TRUE) #> Warning: in eucast_rules(): not all columns with antimicrobial results are of #> class 'sir'. Transform them on beforehand, with e.g.: #>   - a %>% as.sir(PEN:FOX) #>   - a %>% mutate_if(is_sir_eligible, as.sir) #>   - a %>% mutate(across(where(is_sir_eligible), as.sir)) head(c) #>   row col           mo_fullname old new rule   rule_group #> 1   1 CXM Staphylococcus aureus   -   S       Breakpoints #> 2   1 CAZ Staphylococcus aureus   -   R      Expert Rules #> 3   1 COL Staphylococcus aureus   -   R      Expert Rules #> 4   2 CAZ Enterococcus faecalis   -   R      Expert Rules #> 5   2 COL Enterococcus faecalis   -   R      Expert Rules #> 6   2 CXM Enterococcus faecalis   -   R      Expert Rules #>                                                 rule_name #> 1                                          Staphylococcus #> 2 Table 4: Intrinsic resistance in gram-positive bacteria #> 3 Table 4: Intrinsic resistance in gram-positive bacteria #> 4 Table 4: Intrinsic resistance in gram-positive bacteria #> 5 Table 4: Intrinsic resistance in gram-positive bacteria #> 6 Table 4: Intrinsic resistance in gram-positive bacteria #>                                                                                 rule_source #> 1                                           'EUCAST Clinical Breakpoint Tables' v12.0, 2022 #> 2 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 3 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 4 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 5 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 6 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 # }  # Dosage guidelines:  eucast_dosage(c(\"tobra\", \"genta\", \"cipro\"), \"iv\") #> ℹ Dosages for antimicrobial drugs, as meant for 'EUCAST Clinical Breakpoint #>   Tables' v12.0 (2022). This note will be shown once per session. #> # A tibble: 3 × 4 #>   ab   name          standard_dosage  high_dosage  #>   <ab> <chr>         <chr>            <chr>        #> 1 TOB  Tobramycin    6-7 mg/kg x 1 iv NA           #> 2 GEN  Gentamicin    6-7 mg/kg x 1 iv NA           #> 3 CIP  Ciprofloxacin 0.4 g x 2 iv     0.4 g x 3 iv  eucast_dosage(c(\"tobra\", \"genta\", \"cipro\"), \"iv\", version_breakpoints = 10) #> ℹ Dosages for antimicrobial drugs, as meant for 'EUCAST Clinical Breakpoint #>   Tables' v10.0 (2020). This note will be shown once per session. #> # A tibble: 3 × 4 #>   ab   name          standard_dosage  high_dosage  #>   <ab> <chr>         <chr>            <chr>        #> 1 TOB  Tobramycin    6-7 mg/kg x 1 iv NA           #> 2 GEN  Gentamicin    6-7 mg/kg x 1 iv NA           #> 3 CIP  Ciprofloxacin 0.4 g x 2 iv     0.4 g x 3 iv"},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 2 000 Example Isolates — example_isolates","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"data set containing 2 000 microbial isolates full antibiograms. data set contains randomised fictitious data, reflects reality can used practise AMR data analysis. examples, please read tutorial website.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"","code":"example_isolates"},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"tibble 2 000 observations 46 variables: date Date receipt laboratory patient ID patient age Age patient gender Gender patient, either \"F\" \"M\" ward Ward type patient admitted, either \"Clinical\", \"ICU\", \"Outpatient\" mo ID microorganism created .mo(), see also microorganisms data set PEN:RIF 40 different antibiotics class sir (see .sir()); column names occur antibiotics data set can translated set_ab_names() ab_name()","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Unclean Data — example_isolates_unclean","title":"Data Set with Unclean Data — example_isolates_unclean","text":"data set containing 3 000 microbial isolates cleaned consequently ready AMR data analysis. data set can used practice.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Unclean Data — example_isolates_unclean","text":"","code":"example_isolates_unclean"},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Unclean Data — example_isolates_unclean","text":"tibble 3 000 observations 8 variables: patient_id ID patient date date receipt laboratory hospital ID hospital, C bacteria info microorganism can transformed .mo(), see also microorganisms AMX:GEN 4 different antibiotics transformed .sir()","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Unclean Data — example_isolates_unclean","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Unclean Data — example_isolates_unclean","text":"","code":"example_isolates_unclean #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria      AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <chr>         <chr> <chr> <chr> <chr> #>  1 J3         A        2012-11-21 E. coli       R     I     S     S     #>  2 R7         A        2018-04-03 K. pneumoniae R     I     S     S     #>  3 P3         A        2014-09-19 E. coli       R     S     S     S     #>  4 P10        A        2015-12-10 E. coli       S     I     S     S     #>  5 B7         A        2015-03-02 E. coli       S     S     S     S     #>  6 W3         A        2018-03-31 S. aureus     R     S     R     S     #>  7 J8         A        2016-06-14 E. coli       R     S     S     S     #>  8 M3         A        2015-10-25 E. coli       R     S     S     S     #>  9 J3         A        2019-06-19 E. coli       S     S     S     S     #> 10 G6         A        2015-04-27 S. aureus     S     S     S     S     #> # ℹ 2,990 more rows"},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":null,"dir":"Reference","previous_headings":"","what":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"Export Data Set NCBI BioSample Antibiogram","code":""},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"","code":"export_ncbi_biosample(x, filename = paste0(\"biosample_\", format(Sys.time(),   \"%Y-%m-%d-%H%M%S\"), \".xlsx\"), type = \"pathogen MIC\",   columns = where(is.mic), save_as_xlsx = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"x data set filename character string specifying file name type character string specifying type data set, either \"pathogen MIC\" \"beta-lactamase MIC\", see https://www.ncbi.nlm.nih.gov/biosample/docs/","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine First Isolates — first_isolate","title":"Determine First Isolates — first_isolate","text":"Determine first isolates microorganisms every patient per episode (needed) per specimen type. functions support four methods summarised Hindler et al. 2007 (doi:10.1086/511864 ). determine patient episodes necessarily based microorganisms, use is_new_episode() also supports grouping dplyr package.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine First Isolates — first_isolate","text":"","code":"first_isolate(x = NULL, col_date = NULL, col_patient_id = NULL,   col_mo = NULL, col_testcode = NULL, col_specimen = NULL,   col_icu = NULL, col_keyantimicrobials = NULL, episode_days = 365,   testcodes_exclude = NULL, icu_exclude = FALSE, specimen_group = NULL,   type = \"points\", method = c(\"phenotype-based\", \"episode-based\",   \"patient-based\", \"isolate-based\"), ignore_I = TRUE, points_threshold = 2,   info = interactive(), include_unknown = FALSE,   include_untested_sir = TRUE, ...)  filter_first_isolate(x = NULL, col_date = NULL, col_patient_id = NULL,   col_mo = NULL, episode_days = 365, method = c(\"phenotype-based\",   \"episode-based\", \"patient-based\", \"isolate-based\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Determine First Isolates — first_isolate","text":"Methodology functions strictly based : M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/. Hindler JF Stelling J (2007). Analysis Presentation Cumulative Antibiograms: New Consensus Guideline Clinical Laboratory Standards Institute. Clinical Infectious Diseases, 44(6), 867-873. doi:10.1086/511864","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine First Isolates — first_isolate","text":"x data.frame containing isolates. Can left blank automatic determination, see Examples. col_date column name result date (date received lab) - default first column date class col_patient_id column name unique IDs patients - default first column starts 'patient' 'patid' (case insensitive) col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). col_testcode column name test codes. Use col_testcode = NULL exclude certain test codes (test codes screening). case testcodes_exclude ignored. col_specimen column name specimen type group col_icu column name logicals (TRUE/FALSE) whether ward department Intensive Care Unit (ICU). can also logical vector length rows x. col_keyantimicrobials (useful method = \"phenotype-based\") column name key antimicrobials determine first isolates, see key_antimicrobials(). default first column starts 'key' followed 'ab' 'antibiotics' 'antimicrobials' (case insensitive). Use col_keyantimicrobials = FALSE prevent . Can also output key_antimicrobials(). episode_days episode days genus/species combination determined 'first isolate' . default 365 days based guideline CLSI, see Source. testcodes_exclude character vector test codes excluded (case-insensitive) icu_exclude logical indicate whether ICU isolates excluded (rows value TRUE column set col_icu) specimen_group value column set col_specimen filter type type determine weighed isolates; can \"keyantimicrobials\" \"points\", see Details method method apply, either \"phenotype-based\", \"episode-based\", \"patient-based\" \"isolate-based\" (can abbreviated), see Details. default \"phenotype-based\" antimicrobial test results present data, \"episode-based\" otherwise. ignore_I logical indicate whether antibiotic interpretations \"\" ignored type = \"keyantimicrobials\", see Details points_threshold minimum number points require differences antibiogram lead inclusion isolate type = \"points\", see Details info logical indicate info printed - default TRUE interactive mode include_unknown logical indicate whether 'unknown' microorganisms included , .e. microbial code \"UNKNOWN\", defaults FALSE. WHONET users, means records organism code \"con\" (contamination) excluded default. Isolates microbial ID NA always excluded first isolate. include_untested_sir logical indicate whether also rows without antibiotic results still eligible becoming first isolate. Use include_untested_sir = FALSE always return FALSE rows. checks data set columns class sir consequently requires transforming columns antibiotic results using .sir() first. ... arguments passed first_isolate() using filter_first_isolate(), otherwise arguments passed key_antimicrobials() (universal, gram_negative, gram_positive)","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine First Isolates — first_isolate","text":"logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine First Isolates — first_isolate","text":"methodology implemented functions strictly based recommendations outlined CLSI Guideline M39 research overview Hindler et al. (2007, doi:10.1086/511864 ). conduct epidemiological analyses antimicrobial resistance data, -called first isolates included prevent overestimation underestimation antimicrobial resistance. Different methods can used , see . functions context-aware. means x argument can left blank used inside data.frame call, see Examples. first_isolate() function wrapper around is_new_episode() function, efficient data sets containing microorganism codes names. isolates microbial ID NA excluded first isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"different-methods","dir":"Reference","previous_headings":"","what":"Different methods","title":"Determine First Isolates — first_isolate","text":"According previously-mentioned sources, different methods (algorithms) select first isolates increasing reliability: isolate-based, patient-based, episode-based phenotype-based. methods select combination taxonomic genus species (subspecies). mentioned methods covered first_isolate() function:","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"isolate-based","dir":"Reference","previous_headings":"","what":"Isolate-based","title":"Determine First Isolates — first_isolate","text":"method require selection, isolates included. , however, respect arguments set first_isolate() function. example, default setting include_unknown (FALSE) omit selection rows without microbial ID.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"patient-based","dir":"Reference","previous_headings":"","what":"Patient-based","title":"Determine First Isolates — first_isolate","text":"include every genus-species combination per patient , set episode_days Inf. method makes sure duplicate isolates selected patient. method preferred e.g. identify first MRSA finding patient determine incidence. Conversely, large longitudinal data set, mean isolates excluded found years initial isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"episode-based","dir":"Reference","previous_headings":"","what":"Episode-based","title":"Determine First Isolates — first_isolate","text":"include every genus-species combination per patient episode , set episode_days sensible number days. Depending type analysis, 14, 30, 60 365. Short episodes common analysing specific hospital ward data ICU cases, long episodes common analysing regional national data. common method correct duplicate isolates. Patients categorised episodes based ID dates (e.g., date specimen receipt laboratory result). common method, take account antimicrobial test results. means e.g. methicillin-resistant Staphylococcus aureus (MRSA) isolate differentiated wildtype Staphylococcus aureus isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"phenotype-based","dir":"Reference","previous_headings":"","what":"Phenotype-based","title":"Determine First Isolates — first_isolate","text":"reliable method, since also weighs antibiogram (antimicrobial test results) yielding -called 'first weighted isolates'. two different methods weigh antibiogram: Using type = \"points\" argument points_threshold (default) method weighs antimicrobial drugs available data set. difference S R (vice versa) counts 0.5 points, difference S R (vice versa) counts 1 point. sum points exceeds points_threshold, defaults 2, isolate selected first weighted isolate. antimicrobials internally selected using all_antimicrobials() function. output function need passed first_isolate() function. Using type = \"keyantimicrobials\" argument ignore_I method weighs specific antimicrobial drugs, called key antimicrobials. difference S R (vice versa) key antimicrobials select isolate first weighted isolate. ignore_I = FALSE, also differences S R (vice versa) lead . Key antimicrobials internally selected using key_antimicrobials() function, can also added manually variable data set col_keyantimicrobials argument. Another option pass output key_antimicrobials() function directly col_keyantimicrobials argument. default method phenotype-based (using type = \"points\") episode-based (using episode_days = 365). makes sure every genus-species combination selected per patient per year, taking account antimicrobial test results. antimicrobial test results available data set, episode-based method applied default.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine First Isolates — first_isolate","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  example_isolates[first_isolate(info = TRUE), ] #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Using column 'date' as input for col_date. #> ℹ Using column 'patient' as input for col_patient_id. #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> ℹ Excluding 16 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 1,387 'phenotype-based' first isolates (69.4% of total where a #>    microbial ID was available) #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # \\donttest{ # get all first Gram-negatives example_isolates[which(first_isolate(info = FALSE) & mo_is_gram_negative()), ] #> ℹ Using column 'mo' as input for mo_is_gram_negative() #> # A tibble: 441 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  4 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  5 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  6 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #>  7 2002-03-08 4FC193     69 M      Clinical B_ESCHR_COLI   R     NA    NA    R   #>  8 2002-03-16 4FC193     69 M      Clinical B_PSDMN_AERG   R     NA    NA    R   #>  9 2002-04-01 496896     46 F      ICU      B_ESCHR_COLI   R     NA    NA    NA  #> 10 2002-04-23 EE2510     69 F      ICU      B_ESCHR_COLI   R     NA    NA    NA  #> # ℹ 431 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  if (require(\"dplyr\")) {   # filter on first isolates using dplyr:   example_isolates %>%     filter(first_isolate(info = TRUE)) } #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> ℹ Excluding 16 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 1,387 'phenotype-based' first isolates (69.4% of total where a #>    microbial ID was available) #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … if (require(\"dplyr\")) {   # short-hand version:   example_isolates %>%     filter_first_isolate(info = FALSE) } #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … if (require(\"dplyr\")) {   # flag the first isolates per group:   example_isolates %>%     group_by(ward) %>%     mutate(first = first_isolate(info = TRUE)) %>%     select(ward, date, patient, mo, first) } #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #>  #> Group: ward = \"Clinical\" #> ℹ Excluding 9 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 865 'phenotype-based' first isolates (70.1% of total where a #>    microbial ID was available) #>  #> Group: ward = \"ICU\" #> ℹ Excluding 6 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 452 'phenotype-based' first isolates (70.0% of total where a #>    microbial ID was available) #>  #> Group: ward = \"Outpatient\" #> ℹ Excluding 1 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 99 'phenotype-based' first isolates (82.5% of total where a #>    microbial ID was available) #> # A tibble: 2,000 × 5 #> # Groups:   ward [3] #>    ward     date       patient mo           first #>    <chr>    <date>     <chr>   <mo>         <lgl> #>  1 Clinical 2002-01-02 A77334  B_ESCHR_COLI TRUE  #>  2 Clinical 2002-01-03 A77334  B_ESCHR_COLI FALSE #>  3 ICU      2002-01-07 067927  B_STPHY_EPDR TRUE  #>  4 ICU      2002-01-07 067927  B_STPHY_EPDR FALSE #>  5 ICU      2002-01-13 067927  B_STPHY_EPDR FALSE #>  6 ICU      2002-01-13 067927  B_STPHY_EPDR FALSE #>  7 Clinical 2002-01-14 462729  B_STPHY_AURS TRUE  #>  8 Clinical 2002-01-14 462729  B_STPHY_AURS FALSE #>  9 ICU      2002-01-16 067927  B_STPHY_EPDR TRUE  #> 10 ICU      2002-01-17 858515  B_STPHY_EPDR TRUE  #> # ℹ 1,990 more rows # }"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":null,"dir":"Reference","previous_headings":"","what":"G-test for Count Data — g.test","title":"G-test for Count Data — g.test","text":"g.test() performs chi-squared contingency table tests goodness--fit tests, just like chisq.test() reliable (1). G-test can used see whether number observations category fits theoretical expectation (called G-test goodness--fit), see whether proportions one variable different different values variable (called G-test independence).","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"G-test for Count Data — g.test","text":"","code":"g.test(x, y = NULL, p = rep(1/length(x), length(x)), rescale.p = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"G-test for Count Data — g.test","text":"code function identical chisq.test(), except : calculation statistic changed \\(2 * sum(x * log(x / E))\\) Yates' continuity correction removed apply G-test possibility simulate p values simulate.p.value removed","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"G-test for Count Data — g.test","text":"x numeric vector matrix. x y can also     factors. y numeric vector; ignored x matrix.      x factor, y factor length. p vector probabilities length x.     error given entry p negative. rescale.p logical scalar; TRUE p rescaled     (necessary) sum 1.  rescale.p FALSE,     p sum 1, error given.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"G-test for Count Data — g.test","text":"list class \"htest\" containing following   components: statistic value chi-squared test statistic. parameter degrees freedom approximate     chi-squared distribution test statistic, NA     p-value computed Monte Carlo simulation. p.value p-value test. method character string indicating type test     performed, whether Monte Carlo simulation continuity     correction used. data.name character string giving name(s) data. observed observed counts. expected expected counts null hypothesis. residuals Pearson residuals,     (observed - expected) / sqrt(expected). stdres standardized residuals,     (observed - expected) / sqrt(V), V     residual cell variance (Agresti, 2007, section 2.4.5     case x matrix, n * p * (1 - p) otherwise).","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"G-test for Count Data — g.test","text":"x matrix one row column, x vector y given, goodness--fit test performed (x treated one-dimensional contingency table). entries x must non-negative integers. case, hypothesis tested whether population probabilities equal p, equal p given. x matrix least two rows columns, taken two-dimensional contingency table: entries x must non-negative integers.  Otherwise, x y must vectors factors length; cases missing values removed, objects coerced factors, contingency table computed .  Pearson's chi-squared test performed null hypothesis joint distribution cell counts 2-dimensional contingency table product row column marginals. p-value computed asymptotic chi-squared distribution test statistic. contingency table case simulation done random sampling set contingency tables given marginals, works marginals strictly positive. Note usual sampling situation assumed chi-squared test (G-test) rather Fisher's exact test. goodness--fit case simulation done random sampling discrete distribution specified p, sample size n = sum(x). simulation done R may slow.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"g-test-of-goodness-of-fit-likelihood-ratio-test-","dir":"Reference","previous_headings":"","what":"G-test Of Goodness-of-Fit (Likelihood Ratio Test)","title":"G-test for Count Data — g.test","text":"Use G-test goodness--fit one nominal variable two values (male female, red, pink white flowers). compare observed counts numbers observations category expected counts, calculate using kind theoretical expectation (1:1 sex ratio 1:2:1 ratio genetic cross). expected number observations category small, G-test may give inaccurate results, use exact test instead (fisher.test()). G-test goodness--fit alternative chi-square test goodness--fit (chisq.test()); tests advantages disadvantages, results two tests usually similar.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"g-test-of-independence","dir":"Reference","previous_headings":"","what":"G-test of Independence","title":"G-test for Count Data — g.test","text":"Use G-test independence two nominal variables, two possible values. want know whether proportions one variable different among values variable. also possible G-test independence two nominal variables. example, Jackson et al. (2013) also data children 3, analysis old vs. young, thigh vs. arm, reaction vs. reaction, analyzed together. Fisher's exact test (fisher.test()) exact test, G-test still approximation. 2x2 table, Fisher's Exact test may slower still run seconds, even sum observations multiple millions. G-test independence alternative chi-square test independence (chisq.test()), give approximately results.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"how-the-test-works","dir":"Reference","previous_headings":"","what":"How the Test Works","title":"G-test for Count Data — g.test","text":"Unlike exact test goodness--fit (fisher.test()), G-test directly calculate probability obtaining observed results something extreme. Instead, like almost statistical tests, G-test intermediate step; uses data calculate test statistic measures far observed data null expectation. use mathematical relationship, case chi-square distribution, estimate probability obtaining value test statistic. G-test uses log ratio two likelihoods test statistic, also called likelihood ratio test log-likelihood ratio test. formula calculate G-statistic : \\(G = 2 * sum(x * log(x / E))\\) E expected values. Since chi-square distributed, p value can calculated R :   df degrees freedom. two categories want find ones significantly different null expectation, can use method testing category vs. sum categories, Bonferroni correction. use G-tests category, course.","code":"p <- stats::pchisq(G, df, lower.tail = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"G-test for Count Data — g.test","text":"McDonald, J.H. 2014. Handbook Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, Maryland. http://www.biostathandbook.com/gtestgof.html.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"G-test for Count Data — g.test","text":"","code":"# = EXAMPLE 1 = # Shivrain et al. (2006) crossed clearfield rice (which are resistant # to the herbicide imazethapyr) with red rice (which are susceptible to # imazethapyr). They then crossed the hybrid offspring and examined the # F2 generation, where they found 772 resistant plants, 1611 moderately # resistant plants, and 737 susceptible plants. If resistance is controlled # by a single gene with two co-dominant alleles, you would expect a 1:2:1 # ratio.  x <- c(772, 1611, 737) g.test(x, p = c(1, 2, 1) / 4) #>  #> \tG-test of goodness-of-fit (likelihood ratio test) #>  #> data:  x #> X-squared = 4.1471, p-value = 0.1257 #>   # There is no significant difference from a 1:2:1 ratio. # Meaning: resistance controlled by a single gene with two co-dominant # alleles, is plausible.   # = EXAMPLE 2 = # Red crossbills (Loxia curvirostra) have the tip of the upper bill either # right or left of the lower bill, which helps them extract seeds from pine # cones. Some have hypothesized that frequency-dependent selection would # keep the number of right and left-billed birds at a 1:1 ratio. Groth (1992) # observed 1752 right-billed and 1895 left-billed crossbills.  x <- c(1752, 1895) g.test(x) #>  #> \tG-test of goodness-of-fit (likelihood ratio test) #>  #> data:  x #> X-squared = 5.6085, p-value = 0.01787 #>   # There is a significant difference from a 1:1 ratio. # Meaning: there are significantly more left-billed birds."},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Clinical or Epidemic Episodes — get_episode","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"functions determine items vector can considered (start ) new episode. can used determine clinical episodes epidemiological analysis. get_episode() function returns index number episode per group, is_new_episode() function returns TRUE every new get_episode() index. absolute relative episode determination supported.","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"","code":"get_episode(x, episode_days = NULL, case_free_days = NULL, ...)  is_new_episode(x, episode_days = NULL, case_free_days = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"x vector dates (class Date POSIXt), sorted internally determine episodes episode_days episode length days specify time period new episode begins, can also less day Inf, see Details case_free_days (inter-epidemic) interval length days new episode start, can also less day Inf, see Details ... ignored, place allow future extensions","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"get_episode(): integer vector is_new_episode(): logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"Episodes can determined two ways: absolute relative. Absolute method uses episode_days define episode length days, new episode start. common use case AMR data analysis microbial epidemiology: episodes S. aureus bacteraemia ICU patients example. episode length 30 days, new S. aureus isolates ICU episode 30 days considered different (new) episode. Thus, method counts since start previous episode. Relative method uses case_free_days quantify duration case-free days (inter-epidemic interval), new episode start. common use case infectious disease epidemiology: episodes norovirus outbreaks hospital example. case-free period 14 days, new norovirus cases time considered different (new) episode. Thus, methods counts since last case previous episode. table: ** marks start new episode, 8 January 2023 7 days since start previous episode (1 January 2023).  *** marks start new episode, 21 January 2023 7 days since last case previous episode (8 January 2023). Either episode_days case_free_days must provided function.","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"difference-between-get-episode-and-is-new-episode-","dir":"Reference","previous_headings":"","what":"Difference between get_episode() and is_new_episode()","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"get_episode() function returns index number episode, cases/patients/isolates first episode number 1, cases/patients/isolates second episode number 2, etc. is_new_episode() function hand, returns TRUE every new get_episode() index. specify, setting episode_days = 365 (using method 1 explained ), two functions differ:","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"other","dir":"Reference","previous_headings":"","what":"Other","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"first_isolate() function wrapper around is_new_episode() function, efficient data sets containing microorganism codes names allows different isolate selection methods. dplyr package required functions work, episode functions support variable grouping work conveniently inside dplyr verbs filter(), mutate() summarise().","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"","code":"# difference between absolute and relative determination of episodes: x <- data.frame(dates = as.Date(c(   \"2021-01-01\",   \"2021-01-02\",   \"2021-01-05\",   \"2021-01-08\",   \"2021-02-21\",   \"2021-02-22\",   \"2021-02-23\",   \"2021-02-24\",   \"2021-03-01\",   \"2021-03-01\" ))) x$absolute <- get_episode(x$dates, episode_days = 7) x$relative <- get_episode(x$dates, case_free_days = 7) x #>         dates absolute relative #> 1  2021-01-01        1        1 #> 2  2021-01-02        1        1 #> 3  2021-01-05        1        1 #> 4  2021-01-08        2        1 #> 5  2021-02-21        3        2 #> 6  2021-02-22        3        2 #> 7  2021-02-23        3        2 #> 8  2021-02-24        3        2 #> 9  2021-03-01        4        2 #> 10 2021-03-01        4        2   # `example_isolates` is a data set available in the AMR package. # See ?example_isolates df <- example_isolates[sample(seq_len(2000), size = 100), ]  get_episode(df$date, episode_days = 60) # indices #>   [1] 47 28 47 47 48  1 48 13  7 38 28  5 14 49 13 22 31 18 28 21  6  8  6 39 13 #>  [26]  5 11 15  9 42 48 43 28 18  4 45 26  2 31 43 10  4 23 47 25 35  1 38 24 15 #>  [51] 19  4 16  5 16  1 44 39 41 25  4 11 30  1 17 30  3 48  8 39  6 38  5 20 37 #>  [76] 36 44 31 48 40 42 43 44 33 12  4 10 34 11  8 12 32 10 33 29 46 34 27 22 23 is_new_episode(df$date, episode_days = 60) # TRUE/FALSE #>   [1]  TRUE  TRUE FALSE FALSE  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE #>  [13]  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE #>  [25] FALSE FALSE  TRUE  TRUE  TRUE  TRUE FALSE  TRUE FALSE FALSE  TRUE  TRUE #>  [37]  TRUE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE  TRUE FALSE FALSE #>  [49]  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE #>  [61] FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE #>  [73] FALSE  TRUE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE #>  [85]  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE #>  [97] FALSE  TRUE FALSE FALSE  # filter on results from the third 60-day episode only, using base R df[which(get_episode(df$date, 60) == 3), ] #> # A tibble: 1 × 46 #>   date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>   <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #> 1 2003-01-31 151041     85 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, RIF <sir>  # the functions also work for less than a day, e.g. to include one per hour: get_episode(   c(     Sys.time(),     Sys.time() + 60 * 60   ),   episode_days = 1 / 24 ) #> [1] 1 2  # \\donttest{ if (require(\"dplyr\")) {   # is_new_episode() can also be used in dplyr verbs to determine patient   # episodes based on any (combination of) grouping variables:   df %>%     mutate(condition = sample(       x = c(\"A\", \"B\", \"C\"),       size = 100,       replace = TRUE     )) %>%     group_by(patient, condition) %>%     mutate(new_episode = is_new_episode(date, 365)) %>%     select(patient, date, condition, new_episode) %>%     arrange(patient, condition, date) } #> # A tibble: 100 × 4 #> # Groups:   patient, condition [99] #>    patient date       condition new_episode #>    <chr>   <date>     <chr>     <lgl>       #>  1 067927  2002-01-29 A         TRUE        #>  2 069276  2015-06-18 A         TRUE        #>  3 078381  2014-08-17 A         TRUE        #>  4 09B453  2010-03-21 C         TRUE        #>  5 0DBF93  2015-10-12 A         TRUE        #>  6 118928  2004-02-01 B         TRUE        #>  7 119392  2010-11-01 A         TRUE        #>  8 122506  2007-08-10 B         TRUE        #>  9 146120  2011-09-23 B         TRUE        #> 10 151041  2003-01-31 B         TRUE        #> # ℹ 90 more rows  if (require(\"dplyr\")) {   df %>%     group_by(ward, patient) %>%     transmute(date,       patient,       new_index = get_episode(date, 60),       new_logical = is_new_episode(date, 60)     ) %>%     arrange(patient, ward, date) } #> # A tibble: 100 × 5 #> # Groups:   ward, patient [94] #>    ward     date       patient new_index new_logical #>    <chr>    <date>     <chr>       <int> <lgl>       #>  1 ICU      2002-01-29 067927          1 TRUE        #>  2 Clinical 2015-06-18 069276          1 TRUE        #>  3 ICU      2014-08-17 078381          1 TRUE        #>  4 Clinical 2010-03-21 09B453          1 TRUE        #>  5 Clinical 2015-10-12 0DBF93          1 TRUE        #>  6 Clinical 2004-02-01 118928          1 TRUE        #>  7 Clinical 2010-11-01 119392          1 TRUE        #>  8 Clinical 2007-08-10 122506          1 TRUE        #>  9 Clinical 2011-09-23 146120          1 TRUE        #> 10 Clinical 2003-01-31 151041          1 TRUE        #> # ℹ 90 more rows  if (require(\"dplyr\")) {   df %>%     group_by(ward) %>%     summarise(       n_patients = n_distinct(patient),       n_episodes_365 = sum(is_new_episode(date, episode_days = 365)),       n_episodes_60 = sum(is_new_episode(date, episode_days = 60)),       n_episodes_30 = sum(is_new_episode(date, episode_days = 30))     ) } #> # A tibble: 3 × 5 #>   ward       n_patients n_episodes_365 n_episodes_60 n_episodes_30 #>   <chr>           <int>          <int>         <int>         <int> #> 1 Clinical           57             14            38            48 #> 2 ICU                31             12            25            28 #> 3 Outpatient          6              5             6             6  # grouping on patients and microorganisms leads to the same # results as first_isolate() when using 'episode-based': if (require(\"dplyr\")) {   x <- df %>%     filter_first_isolate(       include_unknown = TRUE,       method = \"episode-based\"     )    y <- df %>%     group_by(patient, mo) %>%     filter(is_new_episode(date, 365)) %>%     ungroup()    identical(x, y) } #> [1] TRUE  # but is_new_episode() has a lot more flexibility than first_isolate(), # since you can now group on anything that seems relevant: if (require(\"dplyr\")) {   df %>%     group_by(patient, mo, ward) %>%     mutate(flag_episode = is_new_episode(date, 365)) %>%     select(group_vars(.), flag_episode) } #> # A tibble: 100 × 4 #> # Groups:   patient, mo, ward [98] #>    patient mo           ward     flag_episode #>    <chr>   <mo>         <chr>    <lgl>        #>  1 F5F794  B_STPHY_AURS Clinical TRUE         #>  2 146120  B_STPHY_AURS Clinical TRUE         #>  3 501607  B_STRPT_AGLC ICU      TRUE         #>  4 402950  B_ENTRC_FCLS ICU      TRUE         #>  5 560323  B_ESCHR_COLI Clinical TRUE         #>  6 4FC193  B_ESCHR_COLI Clinical TRUE         #>  7 D80438  B_CRYNB_STRT Clinical TRUE         #>  8 F81262  B_STRPT_PNMN Clinical TRUE         #>  9 619705  B_STRPT_ANGN ICU      TRUE         #> 10 280834  B_STPHY_CONS Clinical TRUE         #> # ℹ 90 more rows # }"},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA Biplot with ggplot2 — ggplot_pca","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"Produces ggplot2 variant -called biplot PCA (principal component analysis), flexible appealing base R biplot() function.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"","code":"ggplot_pca(x, choices = 1:2, scale = 1, pc.biplot = TRUE,   labels = NULL, labels_textsize = 3, labels_text_placement = 1.5,   groups = NULL, ellipse = TRUE, ellipse_prob = 0.68,   ellipse_size = 0.5, ellipse_alpha = 0.5, points_size = 2,   points_alpha = 0.25, arrows = TRUE, arrows_colour = \"darkblue\",   arrows_size = 0.5, arrows_textsize = 3, arrows_textangled = TRUE,   arrows_alpha = 0.75, base_textsize = 10, ...)"},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"ggplot_pca() function based ggbiplot() function ggbiplot package Vince Vu, found GitHub: https://github.com/vqv/ggbiplot (retrieved: 2 March 2020, latest commit: 7325e88; 12 February 2015). per GPL-2 licence demands documentation code changes, changes made based source code : Rewritten code remove dependency packages plyr, scales grid Parametrised options, like arrow ellipse settings Hardened input possibilities defining exact type user input every argument Added total amount explained variance caption plot Cleaned syntax based lintr package, fixed grammatical errors added integrity checks Updated documentation","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"x object returned pca(), prcomp() princomp() choices length 2 vector specifying components plot. default     biplot strict sense. scale variables scaled lambda ^ scale     observations scaled lambda ^ (1-scale)     lambda singular values computed     princomp. Normally 0 <= scale <= 1, warning     issued specified scale outside range. pc.biplot true, use Gabriel (1971) refers \"principal component     biplot\", lambda = 1 observations scaled sqrt(n)     variables scaled sqrt(n).  inner products     variables approximate covariances distances observations     approximate Mahalanobis distance. labels optional vector labels observations. set, labels placed respective points. using pca() function input x, determined automatically based attribute non_numeric_cols, see pca(). labels_textsize size text used labels labels_text_placement adjustment factor placement variable names (>=1 means away arrow head) groups optional vector groups labels, length labels. set, points labels coloured according groups. using pca() function input x, determined automatically based attribute non_numeric_cols, see pca(). ellipse logical indicate whether normal data ellipse drawn group (set groups) ellipse_prob statistical size ellipse normal probability ellipse_size size ellipse line ellipse_alpha alpha (transparency) ellipse line points_size size points points_alpha alpha (transparency) points arrows logical indicate whether arrows drawn arrows_colour colour arrow text arrows_size size (thickness) arrow lines arrows_textsize size text end arrows arrows_textangled logical whether text end arrows angled arrows_alpha alpha (transparency) arrows text base_textsize text size plot elements except labels arrows ... arguments passed functions","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"colours labels points can changed adding another scale layer colour, scale_colour_viridis_d() scale_colour_brewer().","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # \\donttest{ if (require(\"dplyr\")) {   # calculate the resistance per group first   resistance_data <- example_isolates %>%     group_by(       order = mo_order(mo), # group on anything, like order       genus = mo_genus(mo)     ) %>% #   and genus as we do here;     filter(n() >= 30) %>% # filter on only 30 results per group     summarise_if(is.sir, resistance) # then get resistance of all drugs    # now conduct PCA for certain antimicrobial drugs   pca_result <- resistance_data %>%     pca(AMC, CXM, CTX, CAZ, GEN, TOB, TMP, SXT)    summary(pca_result)    # old base R plotting method:   biplot(pca_result, main = \"Base R biplot\")    # new ggplot2 plotting method using this package:   if (require(\"ggplot2\")) {     ggplot_pca(pca_result) +       labs(title = \"ggplot2 biplot\")   }   if (require(\"ggplot2\")) {     # still extendible with any ggplot2 function     ggplot_pca(pca_result) +       scale_colour_viridis_d() +       labs(title = \"ggplot2 biplot\")   } } #> Warning: There were 73 warnings in `summarise()`. #> The first warning was: #> ℹ In argument: `PEN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 5: `order = \"Lactobacillales\"` `genus = \"Enterococcus\"`. #> Caused by warning: #> ! Introducing NA: only 14 results available for PEN in group: order = #> \"Lactobacillales\", genus = \"Enterococcus\" (minimum = 30). #> ℹ Run `dplyr::last_dplyr_warnings()` to see the 72 remaining warnings. #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\"  #>    # }"},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":null,"dir":"Reference","previous_headings":"","what":"AMR Plots with ggplot2 — ggplot_sir","title":"AMR Plots with ggplot2 — ggplot_sir","text":"Use functions create bar plots AMR data analysis. functions rely ggplot2 functions.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"AMR Plots with ggplot2 — ggplot_sir","text":"","code":"ggplot_sir(data, position = NULL, x = \"antibiotic\",   fill = \"interpretation\", facet = NULL, breaks = seq(0, 1, 0.1),   limits = NULL, translate_ab = \"name\", combine_SI = TRUE,   minimum = 30, language = get_AMR_locale(), nrow = NULL, colours = c(S   = \"#3CAEA3\", SI = \"#3CAEA3\", I = \"#F6D55C\", IR = \"#ED553B\", R = \"#ED553B\"),   datalabels = TRUE, datalabels.size = 2.5, datalabels.colour = \"grey15\",   title = NULL, subtitle = NULL, caption = NULL,   x.title = \"Antimicrobial\", y.title = \"Proportion\", ...)  geom_sir(position = NULL, x = c(\"antibiotic\", \"interpretation\"),   fill = \"interpretation\", translate_ab = \"name\", minimum = 30,   language = get_AMR_locale(), combine_SI = TRUE, ...)"},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"AMR Plots with ggplot2 — ggplot_sir","text":"data data.frame column(s) class sir (see .sir()) position position adjustment bars, either \"fill\", \"stack\" \"dodge\" x variable show x axis, either \"antibiotic\" (default) \"interpretation\" grouping variable fill variable categorise using plots legend, either \"antibiotic\" (default) \"interpretation\" grouping variable facet variable split plots , either \"interpretation\" (default) \"antibiotic\" grouping variable breaks numeric vector positions limits numeric vector length two providing limits scale, use NA refer existing minimum maximum translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. nrow (using facet) number rows colours named vactor colour used filling. default colours colour-blind friendly. datalabels show datalabels using labels_sir_count() datalabels.size size datalabels datalabels.colour colour datalabels title text show title plot subtitle text show subtitle plot caption text show caption plot x.title text show x axis description y.title text show y axis description ... arguments passed geom_sir() , case scale_sir_colours(), named values set colours. default colours colour-blind friendly, maintaining convention e.g. 'susceptible' green 'resistant' red. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"AMR Plots with ggplot2 — ggplot_sir","text":"default, names antibiotics shown plots using ab_name(). can set translate_ab argument. See count_df(). geom_sir() take variable data sir class (created .sir()) using sir_df() plot bars percentage S, , R. default behaviour bars stacked different antibiotics x axis. Additional functions include: facet_sir() creates 2d plots (default based S//R) using ggplot2::facet_wrap(). scale_y_percent() transforms y axis 0 100% range using ggplot2::scale_y_continuous(). scale_sir_colours() sets colours bars (green S, yellow , red R). multilingual support. default colours colour-blind friendly, maintaining convention e.g. 'susceptible' green 'resistant' red. theme_sir() [ggplot2 theme][ggplot2::theme() minimal distraction. labels_sir_count() print datalabels bars percentage amount isolates using ggplot2::geom_text(). ggplot_sir() wrapper around functions uses data first input. makes possible use function pipe (%>%). See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"AMR Plots with ggplot2 — ggplot_sir","text":"","code":"# \\donttest{ if (require(\"ggplot2\") && require(\"dplyr\")) {   # get antimicrobial results for drugs against a UTI:   ggplot(example_isolates %>% select(AMX, NIT, FOS, TMP, CIP)) +     geom_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # prettify the plot using some additional functions:   df <- example_isolates %>% select(AMX, NIT, FOS, TMP, CIP)   ggplot(df) +     geom_sir() +     scale_y_percent() +     scale_sir_colours(aesthetics = \"fill\") +     labels_sir_count() +     theme_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # or better yet, simplify this using the wrapper function - a single command:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # get only proportions and no counts:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(datalabels = FALSE) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # add other ggplot2 arguments as you like:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(       width = 0.5,       colour = \"black\",       size = 1,       linetype = 2,       alpha = 0.25     ) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # you can alter the colours with colour names:   example_isolates %>%     select(AMX) %>%     ggplot_sir(colours = c(SI = \"yellow\")) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # but you can also use the built-in colour-blind friendly colours for   # your plots, where \"S\" is green, \"I\" is yellow and \"R\" is red:   data.frame(     x = c(\"Value1\", \"Value2\", \"Value3\"),     y = c(1, 2, 3),     z = c(\"Value4\", \"Value5\", \"Value6\")   ) %>%     ggplot() +     geom_col(aes(x = x, y = y, fill = z)) +     scale_sir_colours(aesthetics = \"fill\",                       Value4 = \"S\", Value5 = \"I\", Value6 = \"R\") }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # resistance of ciprofloxacine per age group   example_isolates %>%     mutate(first_isolate = first_isolate()) %>%     filter(       first_isolate == TRUE,       mo == as.mo(\"Escherichia coli\")     ) %>%     # age_groups() is also a function in this AMR package:     group_by(age_group = age_groups(age)) %>%     select(age_group, CIP) %>%     ggplot_sir(x = \"age_group\") } #> Warning: Removed 6 rows containing missing values or values outside the scale range #> (`geom_col()`). #> Warning: Removed 6 rows containing missing values or values outside the scale range #> (`geom_text()`).  if (require(\"ggplot2\") && require(\"dplyr\")) {   # a shorter version which also adjusts data label colours:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(colours = FALSE) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # it also supports groups (don't forget to use the group var on `x` or `facet`):   example_isolates %>%     filter(mo_is_gram_negative(), ward != \"Outpatient\") %>%     # select only UTI-specific drugs     select(ward, AMX, NIT, FOS, TMP, CIP) %>%     group_by(ward) %>%     ggplot_sir(       x = \"ward\",       facet = \"antibiotic\",       nrow = 1,       title = \"AMR of Anti-UTI Drugs Per Ward\",       x.title = \"Ward\",       datalabels = FALSE     ) } #> ℹ Using column 'mo' as input for mo_is_gram_negative()  # }"},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":null,"dir":"Reference","previous_headings":"","what":"Guess Antibiotic Column — guess_ab_col","title":"Guess Antibiotic Column — guess_ab_col","text":"tries find column name data set based information antibiotics data set. Also supports WHONET abbreviations.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Guess Antibiotic Column — guess_ab_col","text":"","code":"guess_ab_col(x = NULL, search_string = NULL, verbose = FALSE,   only_sir_columns = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Guess Antibiotic Column — guess_ab_col","text":"x data.frame search_string text search x , checked .ab() value column x verbose logical indicate whether additional info printed only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE)","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Guess Antibiotic Column — guess_ab_col","text":"column name x, NULL result found.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Guess Antibiotic Column — guess_ab_col","text":"can look antibiotic (trade) name abbreviation search x antibiotics data set column containing name code antibiotic.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Guess Antibiotic Column — guess_ab_col","text":"","code":"df <- data.frame(   amox = \"S\",   tetr = \"R\" )  guess_ab_col(df, \"amoxicillin\") #> [1] \"amox\" guess_ab_col(df, \"J01AA07\") # ATC code of tetracycline #> [1] \"tetr\"  guess_ab_col(df, \"J01AA07\", verbose = TRUE) #> Auto-guessing columns suitable for analysis #> ... #>  OK. #> ℹ Using column 'amox' as input for AMX (amoxicillin). #> ℹ Using column 'tetr' as input for TCY (tetracycline). #> ℹ Using column 'tetr' as input for J01AA07 (tetracycline). #> [1] \"tetr\" # NOTE: Using column 'tetr' as input for J01AA07 (tetracycline).  # WHONET codes df <- data.frame(   AMP_ND10 = \"R\",   AMC_ED20 = \"S\" ) guess_ab_col(df, \"ampicillin\") #> [1] \"AMP_ND10\" guess_ab_col(df, \"J01CR02\") #> [1] \"AMC_ED20\" guess_ab_col(df, as.ab(\"augmentin\")) #> [1] \"AMC_ED20\""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"Data set containing defined intrinsic resistance EUCAST bug-drug combinations.","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"","code":"intrinsic_resistant"},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"tibble 301 583 observations 2 variables: mo Microorganism ID ab Antibiotic ID","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"data set based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021).","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository. allow machine reading EUCAST CLSI guidelines, almost impossible MS Excel PDF files distributed EUCAST CLSI.","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"","code":"intrinsic_resistant #> # A tibble: 301,583 × 2 #>    mo          ab   #>    <mo>        <ab> #>  1 B_GRAMP     ATM  #>  2 B_GRAMP     COL  #>  3 B_GRAMP     NAL  #>  4 B_GRAMP     PLB  #>  5 B_GRAMP     TEM  #>  6 B_ANAER-POS ATM  #>  7 B_ANAER-POS COL  #>  8 B_ANAER-POS NAL  #>  9 B_ANAER-POS PLB  #> 10 B_ANAER-POS TEM  #> # ℹ 301,573 more rows"},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":null,"dir":"Reference","previous_headings":"","what":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"According binomial nomenclature, lowest four taxonomic levels (family, genus, species, subspecies) printed italics. function finds taxonomic names within strings makes italic.","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"","code":"italicise_taxonomy(string, type = c(\"markdown\", \"ansi\", \"html\"))  italicize_taxonomy(string, type = c(\"markdown\", \"ansi\", \"html\"))"},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"string character (vector) type type conversion taxonomic names, either \"markdown\", \"html\" \"ansi\", see Details","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"function finds taxonomic names makes italic based microorganisms data set. taxonomic names can italicised using markdown (default) adding * taxonomic names, <> <\/> using html. using 'ansi', ANSI colours added using \\033[3m \\033[23m taxonomic names. multiple ANSI colours available, conversion occur. function also supports abbreviation genus followed species, \"E. coli\" \"K. pneumoniae ozaenae\".","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"","code":"italicise_taxonomy(\"An overview of Staphylococcus aureus isolates\") #> [1] \"An overview of *Staphylococcus aureus* isolates\" italicise_taxonomy(\"An overview of S. aureus isolates\") #> [1] \"An overview of *S. aureus* isolates\"  cat(italicise_taxonomy(\"An overview of S. aureus isolates\", type = \"ansi\")) #> An overview of S. aureus isolates"},{"path":"https://msberends.github.io/AMR/reference/join.html","id":null,"dir":"Reference","previous_headings":"","what":"Join microorganisms to a Data Set — join","title":"Join microorganisms to a Data Set — join","text":"Join data set microorganisms easily existing data set character vector.","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Join microorganisms to a Data Set — join","text":"","code":"inner_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  left_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  right_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  full_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  semi_join_microorganisms(x, by = NULL, ...)  anti_join_microorganisms(x, by = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Join microorganisms to a Data Set — join","text":"x existing data set join, character vector. case character vector, resulting data.frame contain column 'x' values. variable join - left empty search column class mo (created .mo()) \"mo\" column name exists x, otherwise column name x values exist microorganisms$mo (= \"bacteria_id\"), another column microorganisms (named, like = c(\"bacteria_id\" = \"fullname\")) suffix non-joined duplicate variables x y, suffixes added output disambiguate . character vector length 2. ... ignored, place allow future extensions","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Join microorganisms to a Data Set — join","text":"data.frame","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Join microorganisms to a Data Set — join","text":"Note: opposed join() functions dplyr, character vectors supported default existing columns get suffix \"2\" newly joined columns get suffix. dplyr package installed, join functions used. Otherwise, much slower merge() interaction() functions base R used.","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Join microorganisms to a Data Set — join","text":"","code":"left_join_microorganisms(as.mo(\"K. pneumoniae\")) #> # A tibble: 1 × 26 #>   mo           fullname   status kingdom phylum class order family genus species #>   <mo>         <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #> 1 B_KLBSL_PNMN Klebsiell… accep… Bacter… Pseud… Gamm… Ente… Enter… Kleb… pneumo… #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list> left_join_microorganisms(\"B_KLBSL_PNMN\") #> # A tibble: 1 × 26 #>   mo           fullname   status kingdom phylum class order family genus species #>   <mo>         <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #> 1 B_KLBSL_PNMN Klebsiell… accep… Bacter… Pseud… Gamm… Ente… Enter… Kleb… pneumo… #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list>  df <- data.frame(   date = seq(     from = as.Date(\"2018-01-01\"),     to = as.Date(\"2018-01-07\"),     by = 1   ),   bacteria = as.mo(c(     \"S. aureus\", \"MRSA\", \"MSSA\", \"STAAUR\",     \"E. coli\", \"E. coli\", \"E. coli\"   )),   stringsAsFactors = FALSE ) colnames(df) #> [1] \"date\"     \"bacteria\"  df_joined <- left_join_microorganisms(df, \"bacteria\") colnames(df_joined) #>  [1] \"date\"                \"bacteria\"            \"fullname\"            #>  [4] \"status\"              \"kingdom\"             \"phylum\"              #>  [7] \"class\"               \"order\"               \"family\"              #> [10] \"genus\"               \"species\"             \"subspecies\"          #> [13] \"rank\"                \"ref\"                 \"oxygen_tolerance\"    #> [16] \"source\"              \"lpsn\"                \"lpsn_parent\"         #> [19] \"lpsn_renamed_to\"     \"mycobank\"            \"mycobank_parent\"     #> [22] \"mycobank_renamed_to\" \"gbif\"                \"gbif_parent\"         #> [25] \"gbif_renamed_to\"     \"prevalence\"          \"snomed\"               # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     left_join_microorganisms() %>%     colnames() } #> Joining, by = \"mo\" #>  [1] \"date\"                \"patient\"             \"age\"                 #>  [4] \"gender\"              \"ward\"                \"mo\"                  #>  [7] \"PEN\"                 \"OXA\"                 \"FLC\"                 #> [10] \"AMX\"                 \"AMC\"                 \"AMP\"                 #> [13] \"TZP\"                 \"CZO\"                 \"FEP\"                 #> [16] \"CXM\"                 \"FOX\"                 \"CTX\"                 #> [19] \"CAZ\"                 \"CRO\"                 \"GEN\"                 #> [22] \"TOB\"                 \"AMK\"                 \"KAN\"                 #> [25] \"TMP\"                 \"SXT\"                 \"NIT\"                 #> [28] \"FOS\"                 \"LNZ\"                 \"CIP\"                 #> [31] \"MFX\"                 \"VAN\"                 \"TEC\"                 #> [34] \"TCY\"                 \"TGC\"                 \"DOX\"                 #> [37] \"ERY\"                 \"CLI\"                 \"AZM\"                 #> [40] \"IPM\"                 \"MEM\"                 \"MTR\"                 #> [43] \"CHL\"                 \"COL\"                 \"MUP\"                 #> [46] \"RIF\"                 \"fullname\"            \"status\"              #> [49] \"kingdom\"             \"phylum\"              \"class\"               #> [52] \"order\"               \"family\"              \"genus\"               #> [55] \"species\"             \"subspecies\"          \"rank\"                #> [58] \"ref\"                 \"oxygen_tolerance\"    \"source\"              #> [61] \"lpsn\"                \"lpsn_parent\"         \"lpsn_renamed_to\"     #> [64] \"mycobank\"            \"mycobank_parent\"     \"mycobank_renamed_to\" #> [67] \"gbif\"                \"gbif_parent\"         \"gbif_renamed_to\"     #> [70] \"prevalence\"          \"snomed\"              # }"},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":null,"dir":"Reference","previous_headings":"","what":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"functions can used determine first weighted isolates considering phenotype isolate selection (see first_isolate()). Using phenotype-based method determine first isolates reliable methods disregard phenotypes.","code":""},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"","code":"key_antimicrobials(x = NULL, col_mo = NULL, universal = c(\"ampicillin\",   \"amoxicillin/clavulanic acid\", \"cefuroxime\", \"piperacillin/tazobactam\",   \"ciprofloxacin\", \"trimethoprim/sulfamethoxazole\"),   gram_negative = c(\"gentamicin\", \"tobramycin\", \"colistin\", \"cefotaxime\",   \"ceftazidime\", \"meropenem\"), gram_positive = c(\"vancomycin\", \"teicoplanin\",   \"tetracycline\", \"erythromycin\", \"oxacillin\", \"rifampin\"),   antifungal = c(\"anidulafungin\", \"caspofungin\", \"fluconazole\", \"miconazole\",   \"nystatin\", \"voriconazole\"), only_sir_columns = FALSE, ...)  all_antimicrobials(x = NULL, only_sir_columns = FALSE, ...)  antimicrobials_equal(y, z, type = c(\"points\", \"keyantimicrobials\"),   ignore_I = TRUE, points_threshold = 2, ...)"},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"x data.frame antibiotics columns, like AMX amox. Can left blank determine automatically col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). universal names broad-spectrum antimicrobial drugs, case-insensitive. Set NULL ignore. See Details default antimicrobial drugs gram_negative names antibiotic drugs Gram-positives, case-insensitive. Set NULL ignore. See Details default antibiotic drugs gram_positive names antibiotic drugs Gram-negatives, case-insensitive. Set NULL ignore. See Details default antibiotic drugs antifungal names antifungal drugs fungi, case-insensitive. Set NULL ignore. See Details default antifungal drugs only_sir_columns logical indicate whether columns must included transformed class sir (see .sir()) beforehand (default FALSE) ... ignored, place allow future extensions y, z character vectors compare type type determine weighed isolates; can \"keyantimicrobials\" \"points\", see Details ignore_I logical indicate whether antibiotic interpretations \"\" ignored type = \"keyantimicrobials\", see Details points_threshold minimum number points require differences antibiogram lead inclusion isolate type = \"points\", see Details","code":""},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"key_antimicrobials() all_antimicrobials() functions context-aware. means x argument can left blank used inside data.frame call, see Examples. function key_antimicrobials() returns character vector 12 antimicrobial results every isolate. function all_antimicrobials() returns character vector antimicrobial drug results every isolate. vectors can compared using antimicrobials_equal(), check two isolates generally antibiogram. Missing invalid values replaced dot (\".\") key_antimicrobials() ignored antimicrobials_equal(). Please see first_isolate() function important functions enable 'phenotype-based' method determination first isolates. default antimicrobial drugs used rows (set universal) : Ampicillin Amoxicillin/clavulanic acid Cefuroxime Ciprofloxacin Piperacillin/tazobactam Trimethoprim/sulfamethoxazole default antimicrobial drugs used Gram-negative bacteria (set gram_negative) : Cefotaxime Ceftazidime Colistin Gentamicin Meropenem Tobramycin default antimicrobial drugs used Gram-positive bacteria (set gram_positive) : Erythromycin Oxacillin Rifampin Teicoplanin Tetracycline Vancomycin default antimicrobial drugs used fungi (set antifungal) : Anidulafungin Caspofungin Fluconazole Miconazole Nystatin Voriconazole","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # output of the `key_antimicrobials()` function could be like this: strainA <- \"SSSRR.S.R..S\" strainB <- \"SSSIRSSSRSSS\"  # those strings can be compared with: antimicrobials_equal(strainA, strainB, type = \"keyantimicrobials\") #> [1] TRUE # TRUE, because I is ignored (as well as missing values)  antimicrobials_equal(strainA, strainB, type = \"keyantimicrobials\", ignore_I = FALSE) #> [1] FALSE # FALSE, because I is not ignored and so the 4th [character] differs  # \\donttest{ if (require(\"dplyr\")) {   # set key antibiotics to a new variable   my_patients <- example_isolates %>%     mutate(keyab = key_antimicrobials(antifungal = NULL)) %>% # no need to define `x`     mutate(       # now calculate first isolates       first_regular = first_isolate(col_keyantimicrobials = FALSE),       # and first WEIGHTED isolates       first_weighted = first_isolate(col_keyantimicrobials = \"keyab\")     )    # Check the difference in this data set, 'weighted' results in more isolates:   sum(my_patients$first_regular, na.rm = TRUE)   sum(my_patients$first_weighted, na.rm = TRUE) } #> [1] 1383 # }"},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":null,"dir":"Reference","previous_headings":"","what":"Kurtosis of the Sample — kurtosis","title":"Kurtosis of the Sample — kurtosis","text":"Kurtosis measure \"tailedness\" probability distribution real-valued random variable. normal distribution kurtosis 3 excess kurtosis 0.","code":""},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kurtosis of the Sample — kurtosis","text":"","code":"kurtosis(x, na.rm = FALSE, excess = FALSE)  # Default S3 method kurtosis(x, na.rm = FALSE, excess = FALSE)  # S3 method for class 'matrix' kurtosis(x, na.rm = FALSE, excess = FALSE)  # S3 method for class 'data.frame' kurtosis(x, na.rm = FALSE, excess = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kurtosis of the Sample — kurtosis","text":"x vector values, matrix data.frame na.rm logical indicate whether NA values stripped computation proceeds excess logical indicate whether excess kurtosis returned, defined kurtosis minus 3.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kurtosis of the Sample — kurtosis","text":"","code":"kurtosis(rnorm(10000)) #> [1] 2.971316 kurtosis(rnorm(10000), excess = TRUE) #> [1] -0.05342143"},{"path":"https://msberends.github.io/AMR/reference/like.html","id":null,"dir":"Reference","previous_headings":"","what":"Vectorised Pattern Matching with Keyboard Shortcut — like","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"Convenient wrapper around grepl() match pattern: x %like% pattern. always returns logical vector always case-insensitive (use x %like_case% pattern case-sensitive matching). Also, pattern can long x compare items index vectors, can length iterate cases.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"","code":"like(x, pattern, ignore.case = TRUE)  x %like% pattern  x %unlike% pattern  x %like_case% pattern  x %unlike_case% pattern"},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"Idea like function data.table package, although altered explained Details.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"x character vector matches sought, object can coerced .character() character vector. pattern character vector containing regular expressions (character string fixed = TRUE) matched given character vector. Coerced .character() character string possible. ignore.case FALSE, pattern matching case sensitive TRUE, case ignored matching.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"like() %like%/%unlike% functions: case-insensitive (use %like_case%/%unlike_case% case-sensitive matching) Support multiple patterns Check pattern valid regular expression sets fixed = TRUE , greatly improve speed (vectorised pattern) Always use compatibility Perl unless fixed = TRUE, greatly improve speed Using RStudio? %like%/%unlike% functions can also directly inserted code Addins menu can keyboard shortcut like Shift+Ctrl+L Shift+Cmd+L (see menu Tools > Modify Keyboard Shortcuts...). keep pressing shortcut, inserted text iterated %like% -> %unlike% -> %like_case% -> %unlike_case%.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"","code":"# data.table has a more limited version of %like%, so unload it: try(detach(\"package:data.table\", unload = TRUE), silent = TRUE)  a <- \"This is a test\" b <- \"TEST\" a %like% b #> [1] TRUE b %like% a #> [1] FALSE  # also supports multiple patterns a <- c(\"Test case\", \"Something different\", \"Yet another thing\") b <- c(\"case\", \"diff\", \"yet\") a %like% b #> [1] TRUE TRUE TRUE a %unlike% b #> [1] FALSE FALSE FALSE  a[1] %like% b #> [1]  TRUE FALSE FALSE a %like% b[1] #> [1]  TRUE FALSE FALSE  # \\donttest{ # get isolates whose name start with 'Entero' (case-insensitive) example_isolates[which(mo_name() %like% \"^entero\"), ] #> ℹ Using column 'mo' as input for mo_name() #> # A tibble: 106 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinic… B_ENTRC_FACM    NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-08-13 F35553     52 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU     B_ENTRC         NA    NA    NA    NA  #> 10 2003-09-05 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #> # ℹ 96 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_name() %like% \"^ent\") } #> ℹ Using column 'mo' as input for mo_name() #> # A tibble: 106 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinic… B_ENTRC_FACM    NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-08-13 F35553     52 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU     B_ENTRC         NA    NA    NA    NA  #> 10 2003-09-05 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #> # ℹ 96 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Determine isolates multidrug-resistant organisms (MDRO) according international, national, custom guidelines.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"","code":"mdro(x = NULL, guideline = \"CMI2012\", col_mo = NULL, esbl = NA,   carbapenemase = NA, mecA = NA, mecC = NA, vanA = NA, vanB = NA,   info = interactive(), pct_required_classes = 0.5, combine_SI = TRUE,   verbose = FALSE, only_sir_columns = FALSE, ...)  custom_mdro_guideline(..., as_factor = TRUE)  brmo(x = NULL, only_sir_columns = FALSE, ...)  mrgn(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  mdr_tb(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  mdr_cmi2012(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  eucast_exceptional_phenotypes(x = NULL, only_sir_columns = FALSE,   verbose = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"See supported guidelines list publications used function.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"x data.frame antibiotics columns, like AMX amox. Can left blank automatic determination. guideline specific guideline follow, see sections Supported international / national guidelines Using Custom Guidelines . left empty, publication Magiorakos et al. (see ) followed. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). esbl logical values, column name containing logical values, indicating presence ESBL gene (production proteins) carbapenemase logical values, column name containing logical values, indicating presence carbapenemase gene (production proteins) mecA logical values, column name containing logical values, indicating presence mecA gene (production proteins) mecC logical values, column name containing logical values, indicating presence mecC gene (production proteins) vanA logical values, column name containing logical values, indicating presence vanA gene (production proteins) vanB logical values, column name containing logical values, indicating presence vanB gene (production proteins) info logical indicate whether progress printed console - default print interactive sessions pct_required_classes minimal required percentage antimicrobial classes must available per isolate, rounded . example, default guideline, 17 antimicrobial classes must available S. aureus. Setting pct_required_classes argument 0.5 (default) means every S. aureus isolate least 8 different classes must available. lower number available classes return NA isolate. combine_SI logical indicate whether values S must merged one, resistance considered isolates R, . default behaviour mdro() function, follows redefinition EUCAST interpretation (increased exposure) 2019, see section 'Interpretation S, R' . using combine_SI = FALSE, resistance considered isolates R . verbose logical turn Verbose mode (default ). Verbose mode, function return MDRO results, instead returns data set logbook form extensive info isolates MDRO-positive, . only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE) ... case custom_mdro_guideline(): set rules, see section Using Custom Guidelines . Otherwise: column name antibiotic, see section Antibiotics . as_factor logical indicate whether returned value ordered factor (TRUE, default), otherwise character vector","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"CMI 2012 paper - function mdr_cmi2012() mdro(): Ordered factor levels Negative < Multi-drug-resistant (MDR) < Extensively drug-resistant (XDR) < Pandrug-resistant (PDR) TB guideline - function mdr_tb() mdro(..., guideline = \"TB\"): Ordered factor levels Negative < Mono-resistant < Poly-resistant < Multi-drug-resistant < Extensively drug-resistant German guideline - function mrgn() mdro(..., guideline = \"MRGN\"): Ordered factor levels Negative < 3MRGN < 4MRGN Everything else, except custom guidelines: Ordered factor levels Negative < Positive, unconfirmed < Positive. value \"Positive, unconfirmed\" means , according guideline, entirely sure isolate multi-drug resistant confirmed additional (e.g. molecular) tests","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"functions context-aware. means x argument can left blank used inside data.frame call, see Examples. pct_required_classes argument, values 1 divided 100. support fractions (0.75 3/4) percentages (75). Note: Every test involves Enterobacteriaceae family, internally performed using newly named order Enterobacterales, since Enterobacteriaceae family taxonomically reclassified Adeolu et al. 2016. , Enterobacteriaceae family Enterobacteriales () order. species old Enterobacteriaceae family still new Enterobacterales (without ) order, divided multiple families. way tests performed now mdro() function makes sure results 2016 2016 identical.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"supported-international-national-guidelines","dir":"Reference","previous_headings":"","what":"Supported International / National Guidelines","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Currently supported guidelines (case-insensitive): guideline = \"CMI2012\" (default) Magiorakos AP, Srinivasan et al. \"Multidrug-resistant, extensively drug-resistant pandrug-resistant bacteria: international expert proposal interim standard definitions acquired resistance.\" Clinical Microbiology Infection (2012) (doi:10.1111/j.1469-0691.2011.03570.x ) guideline = \"EUCAST3.3\" (simply guideline = \"EUCAST\") European international guideline - EUCAST Expert Rules Version 3.3 \"Intrinsic Resistance Unusual Phenotypes\" (link) guideline = \"EUCAST3.2\" European international guideline - EUCAST Expert Rules Version 3.2 \"Intrinsic Resistance Unusual Phenotypes\" (link) guideline = \"EUCAST3.1\" European international guideline - EUCAST Expert Rules Version 3.1 \"Intrinsic Resistance Exceptional Phenotypes Tables\" (link) guideline = \"TB\" international guideline multi-drug resistant tuberculosis - World Health Organization \"Companion handbook guidelines programmatic management drug-resistant tuberculosis\" (link) guideline = \"MRGN\" German national guideline - Mueller et al. (2015) Antimicrobial Resistance Infection Control 4:7; doi:10.1186/s13756-015-0047-6 guideline = \"BRMO\" Dutch national guideline - Samenwerkingverband Richtlijnen Infectiepreventie (SRI) (2024) \"Bijzonder Resistente Micro-Organismen (BRMO)\" (link) Also: guideline = \"BRMO 2017\" former Dutch national guideline - Werkgroep Infectiepreventie (WIP), RIVM, last revision 2017: \"Bijzonder Resistente Micro-Organismen (BRMO)\" Please suggest implement guidelines letting us know: https://github.com/msberends/AMR/issues/new.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"using-custom-guidelines","dir":"Reference","previous_headings":"","what":"Using Custom Guidelines","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Custom guidelines can set custom_mdro_guideline() function. great importance custom rules determine MDROs hospital, e.g., rules dependent ward, state contact isolation variables data. familiar case_when() function dplyr package, recognise input method set rules. Rules must set using R considers 'formula notation'. rule written tilde (~) consequence rule written tilde:   row/isolate matches first rule, value first ~ (case 'Elderly Type ') set MDRO value. Otherwise, second rule tried . number rules unlimited. can print rules set console overview. Colours help reading console supports colours.   outcome function can used guideline argument mdro() function:   Rules can also combined custom rules using c():   rules set (custom object case) exported shared file location using saveRDS() collaborate multiple users. custom rules set imported using readRDS().","code":"custom <- custom_mdro_guideline(CIP == \"R\" & age > 60 ~ \"Elderly Type A\",                                 ERY == \"R\" & age > 60 ~ \"Elderly Type B\") custom #> A set of custom MDRO rules: #>   1. CIP is \"R\" and age is higher than 60 -> Elderly Type A #>   2. ERY is \"R\" and age is higher than 60 -> Elderly Type B #>   3. Otherwise -> Negative #> #> Unmatched rows will return NA. x <- mdro(example_isolates,           guideline = custom) table(x) #>       Negative Elderly Type A Elderly Type B #>           1070            198            732 x <- mdro(example_isolates,           guideline = c(custom,                         custom_mdro_guideline(ERY == \"R\" & age > 50 ~ \"Elderly Type C\"))) table(x) #>       Negative Elderly Type A Elderly Type B Elderly Type C #>            961            198            732            109"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"antibiotics","dir":"Reference","previous_headings":"","what":"Antibiotics","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"define antibiotics column names, leave determine automatically guess_ab_col() input text (case-insensitive), use NULL skip column (e.g. TIC = NULL skip ticarcillin). Manually defined non-existing columns skipped warning. following antibiotics eligible functions eucast_rules() mdro(). shown format 'name (antimicrobial ID, ATC code)', sorted alphabetically: Amikacin (AMK, J01GB06), amoxicillin (AMX, J01CA04), amoxicillin/clavulanic acid (AMC, J01CR02), ampicillin (AMP, J01CA01), ampicillin/sulbactam (SAM, J01CR01), apramycin (APR, QA07AA92), arbekacin (ARB, J01GB12), aspoxicillin (APX, J01CA19), azidocillin (AZD, J01CE04), azithromycin (AZM, J01FA10), azlocillin (AZL, J01CA09), aztreonam (ATM, J01DF01), bacampicillin (BAM, J01CA06), bekanamycin (BEK, J01GB13), benzathine benzylpenicillin (BNB, J01CE08), benzathine phenoxymethylpenicillin (BNP, J01CE10), benzylpenicillin (PEN, J01CE01), besifloxacin (BES, S01AE08), biapenem (BIA, J01DH05), carbenicillin (CRB, J01CA03), carindacillin (CRN, J01CA05), carumonam (CAR, J01DF02), cefacetrile (CAC, J01DB10), cefaclor (CEC, J01DC04), cefadroxil (CFR, J01DB05), cefalexin (LEX, J01DB01), cefaloridine (RID, J01DB02), cefalotin (CEP, J01DB03), cefamandole (MAN, J01DC03), cefapirin (HAP, J01DB08), cefatrizine (CTZ, J01DB07), cefazedone (CZD, J01DB06), cefazolin (CZO, J01DB04), cefcapene (CCP, J01DD17), cefdinir (CDR, J01DD15), cefditoren (DIT, J01DD16), cefepime (FEP, J01DE01), cefepime/amikacin (CFA, J01RA06), cefetamet (CAT, J01DD10), cefiderocol (FDC, J01DI04), cefixime (CFM, J01DD08), cefmenoxime (CMX, J01DD05), cefmetazole (CMZ, J01DC09), cefodizime (DIZ, J01DD09), cefonicid (CID, J01DC06), cefoperazone (CFP, J01DD12), cefoperazone/sulbactam (CSL, J01DD62), ceforanide (CND, J01DC11), cefotaxime (CTX, J01DD01), cefotaxime/clavulanic acid (CTC, J01DD51), cefotetan (CTT, J01DC05), cefotiam (CTF, J01DC07), cefovecin (FOV, QJ01DD91), cefoxitin (FOX, J01DC01), cefozopran (ZOP, J01DE03), cefpiramide (CPM, J01DD11), cefpirome (CPO, J01DE02), cefpodoxime (CPD, J01DD13), cefprozil (CPR, J01DC10), cefquinome (CEQ, QG51AA07), cefroxadine (CRD, J01DB11), cefsulodin (CFS, J01DD03), ceftaroline (CPT, J01DI02), ceftazidime (CAZ, J01DD02), ceftazidime/clavulanic acid (CCV, J01DD52), cefteram (CEM, J01DD18), ceftezole (CTL, J01DB12), ceftibuten (CTB, J01DD14), ceftiofur (TIO, QJ01DD90), ceftizoxime (CZX, J01DD07), ceftobiprole medocaril (CFM1, J01DI01), ceftolozane/tazobactam (CZT, J01DI54), ceftriaxone (CRO, J01DD04), ceftriaxone/beta-lactamase inhibitor (CEB, J01DD63), cefuroxime (CXM, J01DC02), cephradine (CED, J01DB09), chloramphenicol (CHL, J01BA01), ciprofloxacin (CIP, J01MA02), clarithromycin (CLR, J01FA09), clindamycin (CLI, J01FF01), clometocillin (CLM, J01CE07), cloxacillin (CLO, J01CF02), colistin (COL, J01XB01), cycloserine (CYC, J04AB01), dalbavancin (DAL, J01XA04), danofloxacin (DAN, QJ01MA92), daptomycin (DAP, J01XX09), delafloxacin (DFX, J01MA23), dibekacin (DKB, J01GB09), dicloxacillin (DIC, J01CF01), difloxacin (DIF, QJ01MA94), dirithromycin (DIR, J01FA13), doripenem (DOR, J01DH04), doxycycline (DOX, J01AA02), enoxacin (ENX, J01MA04), enrofloxacin (ENR, QJ01MA90), epicillin (EPC, J01CA07), ertapenem (ETP, J01DH03), erythromycin (ERY, J01FA01), fleroxacin (FLE, J01MA08), flucloxacillin (FLC, J01CF05), flurithromycin (FLR1, J01FA14), fosfomycin (FOS, J01XX01), framycetin (FRM, D09AA01), fusidic acid (FUS, J01XC01), gamithromycin (GAM, QJ01FA95), garenoxacin (GRN, J01MA19), gatifloxacin (GAT, J01MA16), gemifloxacin (GEM, J01MA15), gentamicin (GEN, J01GB03), grepafloxacin (GRX, J01MA11), hetacillin (HET, J01CA18), imipenem (IPM, J01DH51), imipenem/relebactam (IMR, J01DH56), isepamicin (ISE, J01GB11), josamycin (JOS, J01FA07), kanamycin (KAN, J01GB04), kitasamycin (KIT, QJ01FA93), lascufloxacin (LSC, J01MA25), latamoxef (LTM, J01DD06), levofloxacin (LVX, J01MA12), levonadifloxacin (LND, J01MA24), lincomycin (LIN, J01FF02), linezolid (LNZ, J01XX08), lomefloxacin (LOM, J01MA07), loracarbef (LOR, J01DC08), marbofloxacin (MAR, QJ01MA93), mecillinam (MEC, J01CA11), meropenem (MEM, J01DH02), meropenem/vaborbactam (MEV, J01DH52), metampicillin (MTM, J01CA14), meticillin (MET, J01CF03), mezlocillin (MEZ, J01CA10), micronomicin (MCR, S01AA22), midecamycin (MID, J01FA03), minocycline (MNO, J01AA08), miocamycin (MCM, J01FA11), moxifloxacin (MFX, J01MA14), nadifloxacin (NAD, D10AF05), nafcillin (NAF, J01CF06), nalidixic acid (NAL, J01MB02), neomycin (NEO, J01GB05), netilmicin (NET, J01GB07), nitrofurantoin (NIT, J01XE01), norfloxacin (, J01MA06), novobiocin (NOV, QJ01XX95), ofloxacin (OFX, J01MA01), oleandomycin (OLE, J01FA05), orbifloxacin (ORB, QJ01MA95), oritavancin (ORI, J01XA05), oxacillin (OXA, J01CF04), panipenem (PAN, J01DH55), pazufloxacin (PAZ, J01MA18), pefloxacin (PEF, J01MA03), penamecillin (PNM, J01CE06), pheneticillin (PHE, J01CE05), phenoxymethylpenicillin (PHN, J01CE02), piperacillin (PIP, J01CA12), piperacillin/tazobactam (TZP, J01CR05), pirlimycin (PRL, QJ51FF90), pivampicillin (PVM, J01CA02), pivmecillinam (PME, J01CA08), plazomicin (PLZ, J01GB14), polymyxin B (PLB, J01XB02), pradofloxacin (PRA, QJ01MA97), pristinamycin (PRI, J01FG01), procaine benzylpenicillin (PRB, J01CE09), propicillin (PRP, J01CE03), prulifloxacin (PRU, J01MA17), quinupristin/dalfopristin (QDA, QJ01FG02), ribostamycin (RST, J01GB10), rifampicin (RIF, J04AB02), rokitamycin (ROK, J01FA12), roxithromycin (RXT, J01FA06), rufloxacin (RFL, J01MA10), sarafloxacin (SAR, QJ01MA98), sisomicin (SIS, J01GB08), sitafloxacin (SIT, J01MA21), solithromycin (SOL, J01FA16), sparfloxacin (SPX, J01MA09), spiramycin (SPI, J01FA02), streptoduocin (STR, J01GA02), streptomycin (STR1, J01GA01), sulbenicillin (SBC, J01CA16), sulfadiazine (SDI, J01EC02), sulfadiazine/trimethoprim (SLT1, J01EE02), sulfadimethoxine (SUD, J01ED01), sulfadimidine (SDM, J01EB03), sulfadimidine/trimethoprim (SLT2, J01EE05), sulfafurazole (SLF, J01EB05), sulfaisodimidine (SLF1, J01EB01), sulfalene (SLF2, J01ED02), sulfamazone (SZO, J01ED09), sulfamerazine (SLF3, J01ED07), sulfamerazine/trimethoprim (SLT3, J01EE07), sulfamethizole (SLF4, J01EB02), sulfamethoxazole (SMX, J01EC01), sulfamethoxypyridazine (SLF5, J01ED05), sulfametomidine (SLF6, J01ED03), sulfametoxydiazine (SLF7, J01ED04), sulfametrole/trimethoprim (SLT4, J01EE03), sulfamoxole (SLF8, J01EC03), sulfamoxole/trimethoprim (SLT5, J01EE04), sulfanilamide (SLF9, J01EB06), sulfaperin (SLF10, J01ED06), sulfaphenazole (SLF11, J01ED08), sulfapyridine (SLF12, J01EB04), sulfathiazole (SUT, J01EB07), sulfathiourea (SLF13, J01EB08), sultamicillin (SLT6, J01CR04), talampicillin (TAL, J01CA15), tebipenem (TBP, J01DH06), tedizolid (TZD, J01XX11), teicoplanin (TEC, J01XA02), telavancin (TLV, J01XA03), telithromycin (TLT, J01FA15), temafloxacin (TMX, J01MA05), temocillin (TEM, J01CA17), tetracycline (TCY, J01AA07), ticarcillin (TIC, J01CA13), ticarcillin/clavulanic acid (TCC, J01CR03), tigecycline (TGC, J01AA12), tilbroquinol (TBQ, P01AA05), tildipirosin (TIP, QJ01FA96), tilmicosin (TIL, QJ01FA91), tobramycin (TOB, J01GB01), tosufloxacin (TFX, J01MA22), trimethoprim (TMP, J01EA01), trimethoprim/sulfamethoxazole (SXT, J01EE01), troleandomycin (TRL, J01FA08), trovafloxacin (TVA, J01MA13), tulathromycin (TUL, QJ01FA94), tylosin (TYL, QJ01FA90), tylvalosin (TYL1, QJ01FA92), vancomycin (VAN, J01XA01)","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"","code":"out <- mdro(example_isolates, guideline = \"EUCAST\") str(out) #>  Ord.factor w/ 3 levels \"Negative\"<\"Positive, unconfirmed\"<..: 1 1 1 1 1 1 1 1 1 1 ... table(out) #> out #>              Negative Positive, unconfirmed              Positive  #>                  1994                     0                     6   out <- mdro(example_isolates,   guideline = custom_mdro_guideline(     AMX == \"R\" ~ \"Custom MDRO 1\",     VAN == \"R\" ~ \"Custom MDRO 2\"   ) ) table(out) #> out #>      Negative Custom MDRO 1 Custom MDRO 2  #>           870           804           326   # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     mdro() %>%     table()    # no need to define `x` when used inside dplyr verbs:   example_isolates %>%     mutate(MDRO = mdro()) %>%     pull(MDRO) %>%     table() } #> Warning: in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `MDRO = mdro()`. #> Caused by warning: #> ! in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) #> . #>                         Negative       Multi-drug-resistant (MDR)  #>                             1617                              128  #> Extensively drug-resistant (XDR)          Pandrug-resistant (PDR)  #>                                0                                0  # }"},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate the Mean AMR Distance — mean_amr_distance","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"Calculates normalised mean antimicrobial resistance multiple observations, help identify similar isolates without comparing antibiograms hand.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"","code":"mean_amr_distance(x, ...)  # S3 method for class 'sir' mean_amr_distance(x, ..., combine_SI = TRUE)  # S3 method for class 'data.frame' mean_amr_distance(x, ..., combine_SI = TRUE)  amr_distance_from_row(amr_distance, row)"},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"x vector class sir, mic disk, data.frame containing columns classes ... variables select (supports tidyselect language column1:column4 (.mic), can thus also antimicrobial selectors combine_SI logical indicate whether values S, SDD, must merged one, input consists S+vs. R (susceptible vs. resistant) - default TRUE amr_distance outcome mean_amr_distance() row index, row number","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"mean AMR distance effectively Z-score; normalised numeric value compare AMR test results can help identify similar isolates, without comparing antibiograms hand. MIC values (see .mic()) transformed log2() first; distance thus calculated (log2(x) - mean(log2(x))) / sd(log2(x)). SIR values (see .sir()) transformed using \"S\" = 1, \"\" = 2, \"R\" = 3. combine_SI TRUE (default), \"\" considered 1. data sets, mean AMR distance calculated per column, mean per row returned, see Examples. Use amr_distance_from_row() subtract distances distance one row, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"interpretation","dir":"Reference","previous_headings":"","what":"Interpretation","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"Isolates distances less 0.01 difference considered similar. Differences lower 0.025 considered suspicious.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"","code":"sir <- random_sir(10) sir #> Class 'sir' #>  [1] S R S S S R R R R R mean_amr_distance(sir) #>  [1] -1.1618950  0.7745967 -1.1618950 -1.1618950 -1.1618950  0.7745967 #>  [7]  0.7745967  0.7745967  0.7745967  0.7745967  mic <- random_mic(10) mic #> Class 'mic' #>  [1] 0.005 0.25  1     2     >=16  8     1     1     0.025 4     mean_amr_distance(mic) #>  [1] -1.9714967 -0.4147020  0.1369756  0.4128144  1.2403309  0.9644921 #>  [7]  0.1369756  0.1369756 -1.3310188  0.6886533 # equal to the Z-score of their log2: (log2(mic) - mean(log2(mic))) / sd(log2(mic)) #>  [1] -1.9714967 -0.4147020  0.1369756  0.4128144  1.2403309  0.9644921 #>  [7]  0.1369756  0.1369756 -1.3310188  0.6886533  disk <- random_disk(10) disk #> Class 'disk' #>  [1] 15 15 50 41 10 21 44 42 10 41 mean_amr_distance(disk) #>  [1] -0.8693350 -0.8693350  1.3196381  0.7567593 -1.1820455 -0.4940825 #>  [7]  0.9443855  0.8193014 -1.1820455  0.7567593  y <- data.frame(   id = LETTERS[1:10],   amox = random_sir(10, ab = \"amox\", mo = \"Escherichia coli\"),   cipr = random_disk(10, ab = \"cipr\", mo = \"Escherichia coli\"),   gent = random_mic(10, ab = \"gent\", mo = \"Escherichia coli\"),   tobr = random_mic(10, ab = \"tobr\", mo = \"Escherichia coli\") ) y #>    id amox cipr   gent tobr #> 1   A    S   26    >=4 0.25 #> 2   B    S   20      1  >=4 #> 3   C    S   28      2 0.25 #> 4   D    S   26 <=0.25  0.5 #> 5   E    S   23      2 0.25 #> 6   F    I   31    0.5    1 #> 7   G    R   31      2  >=4 #> 8   H    R   29      1    2 #> 9   I    R   27 <=0.25  >=4 #> 10  J    I   22    0.5 0.25 mean_amr_distance(y) #> ℹ Calculating mean AMR distance based on columns \"amox\", \"cipr\", \"gent\", #>   and \"tobr\" #>  [1] -0.04047504 -0.25789134 -0.08826444 -0.63237868 -0.42488867  0.02453739 #>  [7]  1.18267023  0.72772870  0.36605356 -0.85709171 y$amr_distance <- mean_amr_distance(y, where(is.mic)) #> Error in .subset(x, j): invalid subscript type 'list' y[order(y$amr_distance), ] #> Error in order(y$amr_distance): argument 1 is not a vector  if (require(\"dplyr\")) {   y %>%     mutate(       amr_distance = mean_amr_distance(y),       check_id_C = amr_distance_from_row(amr_distance, id == \"C\")     ) %>%     arrange(check_id_C) } #> ℹ Calculating mean AMR distance based on columns \"amox\", \"cipr\", \"gent\", #>   and \"tobr\" #>    id amox cipr   gent tobr amr_distance check_id_C #> 1   C    S   28      2 0.25  -0.08826444  0.0000000 #> 2   A    S   26    >=4 0.25  -0.04047504  0.0477894 #> 3   F    I   31    0.5    1   0.02453739  0.1128018 #> 4   B    S   20      1  >=4  -0.25789134  0.1696269 #> 5   E    S   23      2 0.25  -0.42488867  0.3366242 #> 6   I    R   27 <=0.25  >=4   0.36605356  0.4543180 #> 7   D    S   26 <=0.25  0.5  -0.63237868  0.5441142 #> 8   J    I   22    0.5 0.25  -0.85709171  0.7688273 #> 9   H    R   29      1    2   0.72772870  0.8159931 #> 10  G    R   31      2  >=4   1.18267023  1.2709347 if (require(\"dplyr\")) {   # support for groups   example_isolates %>%     filter(mo_genus() == \"Enterococcus\" & mo_species() != \"\") %>%     select(mo, TCY, carbapenems()) %>%     group_by(mo) %>%     mutate(dist = mean_amr_distance(.)) %>%     arrange(mo, dist) } #> ℹ Using column 'mo' as input for mo_genus() #> ℹ Using column 'mo' as input for mo_species() #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Calculating mean AMR distance based on columns \"TCY\", \"IPM\", and \"MEM\" #> # A tibble: 63 × 5 #> # Groups:   mo [4] #>    mo           TCY   IPM   MEM     dist #>    <mo>         <sir> <sir> <sir>  <dbl> #>  1 B_ENTRC_AVIM   S     S     NA   0     #>  2 B_ENTRC_AVIM   S     S     NA   0     #>  3 B_ENTRC_CSSL   NA    S     NA  NA     #>  4 B_ENTRC_FACM   S     S     NA  -2.66  #>  5 B_ENTRC_FACM   S     R     R   -0.423 #>  6 B_ENTRC_FACM   S     R     R   -0.423 #>  7 B_ENTRC_FACM   NA    R     R    0.224 #>  8 B_ENTRC_FACM   NA    R     R    0.224 #>  9 B_ENTRC_FACM   NA    R     R    0.224 #> 10 B_ENTRC_FACM   NA    R     R    0.224 #> # ℹ 53 more rows"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"data set containing commonly used codes microorganisms, laboratory systems WHONET. Define set_mo_source(). searched using .mo() consequently mo_* functions.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"","code":"microorganisms.codes"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"tibble 4 971 observations 2 variables: code Commonly used code microorganism. unique identifier. mo ID microorganism microorganisms data set","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"","code":"microorganisms.codes #> # A tibble: 4,971 × 2 #>    code  mo                #>    <chr> <mo>              #>  1 1011  B_GRAMP           #>  2 1012  B_GRAMP           #>  3 1013  B_GRAMN           #>  4 1014  B_GRAMN           #>  5 1015  F_YEAST           #>  6 103   B_ESCHR_COLI      #>  7 104   B_SLMNL_ENTR_ENTR #>  8 1100  B_STRPT           #>  9 1101  B_STRPT_VIRI      #> 10 1102  B_STRPT_HAEM      #> # ℹ 4,961 more rows  # 'ECO' or 'eco' is the WHONET code for E. coli: microorganisms.codes[microorganisms.codes$code == \"ECO\", ] #> # A tibble: 1 × 2 #>   code  mo           #>   <chr> <mo>         #> 1 ECO   B_ESCHR_COLI  # and therefore, 'eco' will be understood as E. coli in this package: mo_info(\"eco\") #> $mo #> [1] \"B_ESCHR_COLI\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Escherichia\" #>  #> $species #> [1] \"coli\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] \"facultative anaerobe\" #>  #> $url #> [1] \"https://lpsn.dsmz.de/species/escherichia-coli\" #>  #> $ref #> [1] \"Castellani et al., 1919\" #>  #> $snomed #>  [1] \"1095001000112106\" \"715307006\"        \"737528008\"        \"416989002\"        #>  [5] \"116397003\"        \"414097009\"        \"414098004\"        \"414099007\"        #>  [9] \"414100004\"        \"116395006\"        \"735270003\"        \"116396007\"        #> [13] \"83285000\"         \"116394005\"        \"112283007\"        \"710886005\"        #> [17] \"710887001\"        \"710888006\"        \"710889003\"        \"414132004\"        #> [21] \"721892009\"        \"416812001\"        \"416740004\"        \"417216001\"        #> [25] \"457541006\"        \"710253004\"        \"416530004\"        \"417189006\"        #> [29] \"409800005\"        \"713925008\"        \"444771000124108\"  \"838549008\"        #>  #> $lpsn #> [1] \"776057\" #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] \"11286021\" #>  #> $group_members #> character(0) #>   # works for all AMR functions: mo_is_intrinsic_resistant(\"eco\", ab = \"vancomycin\") #> [1] TRUE"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"data set containing species groups microbiological complexes, used clinical breakpoints table.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"","code":"microorganisms.groups"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"tibble 521 observations 4 variables: mo_group ID species group / microbiological complex mo ID microorganism belonging species group / microbiological complex mo_group_name Name species group / microbiological complex, retrieved mo_name() mo_name Name microorganism belonging species group / microbiological complex, retrieved mo_name()","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"","code":"microorganisms.groups #> # A tibble: 521 × 4 #>    mo_group       mo           mo_group_name                   mo_name           #>    <mo>           <mo>         <chr>                           <chr>             #>  1 B_ACNTB_BMNN-C B_ACNTB_BMNN Acinetobacter baumannii complex Acinetobacter ba… #>  2 B_ACNTB_BMNN-C B_ACNTB_CLCC Acinetobacter baumannii complex Acinetobacter ca… #>  3 B_ACNTB_BMNN-C B_ACNTB_LCTC Acinetobacter baumannii complex Acinetobacter di… #>  4 B_ACNTB_BMNN-C B_ACNTB_NSCM Acinetobacter baumannii complex Acinetobacter no… #>  5 B_ACNTB_BMNN-C B_ACNTB_PITT Acinetobacter baumannii complex Acinetobacter pi… #>  6 B_ACNTB_BMNN-C B_ACNTB_SFRT Acinetobacter baumannii complex Acinetobacter se… #>  7 B_BCTRD_FRGL-C B_BCTRD_FRGL Bacteroides fragilis complex    Bacteroides frag… #>  8 B_BCTRD_FRGL-C B_BCTRD_OVTS Bacteroides fragilis complex    Bacteroides ovat… #>  9 B_BCTRD_FRGL-C B_BCTRD_THTT Bacteroides fragilis complex    Bacteroides thet… #> 10 B_BCTRD_FRGL-C B_PHCCL_VLGT Bacteroides fragilis complex    Bacteroides vulg… #> # ℹ 511 more rows  # these are all species in the Bacteroides fragilis group, as per WHONET: microorganisms.groups[microorganisms.groups$mo_group == \"B_BCTRD_FRGL-C\", ] #> # A tibble: 5 × 4 #>   mo_group       mo           mo_group_name                mo_name               #>   <mo>           <mo>         <chr>                        <chr>                 #> 1 B_BCTRD_FRGL-C B_BCTRD_FRGL Bacteroides fragilis complex Bacteroides fragilis  #> 2 B_BCTRD_FRGL-C B_BCTRD_OVTS Bacteroides fragilis complex Bacteroides ovatus    #> 3 B_BCTRD_FRGL-C B_BCTRD_THTT Bacteroides fragilis complex Bacteroides thetaiot… #> 4 B_BCTRD_FRGL-C B_PHCCL_VLGT Bacteroides fragilis complex Bacteroides vulgatus  #> 5 B_BCTRD_FRGL-C B_PRBCT_DSTS Bacteroides fragilis complex Parabacteroides dist…"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"data set containing full microbial taxonomy (last updated: June 24th, 2024) six kingdoms. data set backbone AMR package. MO codes can looked using .mo() microorganism properties can looked using mo_* functions. data set carefully crafted, yet made 100% reproducible public authoritative taxonomic sources (using script), namely: List Prokaryotic names Standing Nomenclature (LPSN) bacteria, MycoBank fungi, Global Biodiversity Information Facility (GBIF) others taxons.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"","code":"microorganisms"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"tibble 78 678 observations 26 variables: mo ID microorganism used package. unique identifier. fullname Full name, like \"Escherichia coli\". taxonomic ranks genus, species subspecies, 'pasted' text genus, species, subspecies. taxonomic ranks higher genus, name taxon. unique identifier. status  Status taxon, either \"accepted\", \"validly published\", \"synonym\", \"unknown\" kingdom, phylum, class, order, family, genus, species, subspecies Taxonomic rank microorganism. Note fungi, phylum equal taxonomic division. Also, fungi, subkingdom subdivision left since occur bacterial taxonomy. rank Text taxonomic rank microorganism, \"species\" \"genus\" ref Author(s) year related scientific publication. contains first surname year latest authors, e.g. \"Wallis et al. 2006 emend. Smith Jones 2018\" becomes \"Smith et al., 2018\". field directly retrieved source specified column source. Moreover, accents removed comply CRAN allows ASCII characters. oxygen_tolerance  Oxygen tolerance, either \"aerobe\", \"anaerobe\", \"anaerobe/microaerophile\", \"facultative anaerobe\", \"likely facultative anaerobe\", \"microaerophile\". data retrieved BacDive (see Source). Items contain \"likely\" missing BacDive extrapolated species within genus guess oxygen tolerance. Currently 68.3% ~39 000 bacteria data set contain oxygen tolerance. source Either \"GBIF\", \"LPSN\", \"MycoBank\", \"manually added\" (see Source) lpsn Identifier ('Record number') List Prokaryotic names Standing Nomenclature (LPSN). first/highest LPSN identifier keep one identifier per row. example, Acetobacter ascendens LPSN Record number 7864 11011. first available microorganisms data set. unique identifier, though available ~33 000 records. lpsn_parent LPSN identifier parent taxon lpsn_renamed_to LPSN identifier currently valid taxon mycobank Identifier ('MycoBank #') MycoBank. unique identifier, though available ~18 000 records. mycobank_parent MycoBank identifier parent taxon mycobank_renamed_to MycoBank identifier currently valid taxon gbif Identifier ('taxonID') Global Biodiversity Information Facility (GBIF). unique identifier, though available ~49 000 records. gbif_parent GBIF identifier parent taxon gbif_renamed_to GBIF identifier currently valid taxon prevalence Prevalence microorganism based Bartlett et al. (2022, doi:10.1099/mic.0.001269 ), see mo_matching_score() full explanation snomed Systematized Nomenclature Medicine (SNOMED) code microorganism, version July 16th, 2024 (see Source). Use mo_snomed() retrieve quickly, see mo_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Taxonomic entries imported order importance: List Prokaryotic names Standing Nomenclature (LPSN): Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. MycoBank: Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. Global Biodiversity Information Facility (GBIF): GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Furthermore, sources used additional details: BacDive: Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Systematized Nomenclature Medicine - Clinical Terms (SNOMED-CT): Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). Accessed https://www.cdc.gov/phin/php/phinvads/ July 16th, 2024. Grimont et al. (2007). Antigenic Formulae Salmonella Serovars, 9th Edition. Collaborating Centre Reference Research Salmonella (WHOCC-SALM). Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Please note entries based LPSN, MycoBank, GBIF (see ). Since sources incorporate entries based (recent) publications International Journal Systematic Evolutionary Microbiology (IJSEM), can happen year publication sometimes later one might expect. example, Staphylococcus pettenkoferi described first time Diagnostic Microbiology Infectious Disease 2002 (doi:10.1016/s0732-8893(02)00399-1 ), 2007 publication IJSEM followed (doi:10.1099/ijs.0.64381-0 ). Consequently, AMR package returns 2007 mo_year(\"S. pettenkoferi\").","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"included-taxa","dir":"Reference","previous_headings":"","what":"Included Taxa","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Included taxonomic data LPSN, MycoBank, GBIF : ~39 000 (sub)species kingdoms Archaea Bacteria ~28 000 species kingdom Fungi. kingdom Fungi large taxon almost 300,000 different (sub)species, microbial (rather macroscopic, like mushrooms). , fungi fit scope package. relevant fungi covered (species Aspergillus, Candida, Cryptococcus, Histoplasma, Pneumocystis, Saccharomyces Trichophyton). ~8 100 (sub)species kingdom Protozoa ~1 600 (sub)species 39 relevant genera kingdom Animalia (Strongyloides Taenia) ~22 000 previously accepted names included (sub)species (taxonomically renamed) complete taxonomic tree included (sub)species: kingdom subspecies identifier parent taxons year first author related scientific publication","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"manual-additions","dir":"Reference","previous_headings":"","what":"Manual additions","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"convenience, entries added manually: ~1 500 entries Salmonella, city-like serovars groups H 36 species groups (beta-haemolytic Streptococcus groups K, coagulase-negative Staphylococcus (CoNS), Mycobacterium tuberculosis complex, etc.), group compositions stored microorganisms.groups data set 1 entry Blastocystis (B.  hominis), although officially exist (Noel et al. 2005, PMID 15634993) 1 entry Moraxella (M. catarrhalis), formally named Branhamella catarrhalis (Catlin, 1970) though change never accepted within field clinical microbiology 8 'undefined' entries (unknown, unknown Gram-negatives, unknown Gram-positives, unknown yeast, unknown fungus, unknown anaerobic Gram-pos/Gram-neg bacteria) syntax used transform original data cleansed R format, can found .","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"","code":"microorganisms #> # A tibble: 78,678 × 26 #>    mo          fullname   status kingdom phylum class order family genus species #>    <mo>        <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #>  1 B_GRAMN     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  2 B_GRAMP     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  3 B_ANAER-NEG (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  4 B_ANAER-POS (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  5 B_ANAER     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  6 F_FUNGUS    (unknown … unkno… Fungi   (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  7   UNKNOWN   (unknown … unkno… (unkno… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  8 P_PROTOZOAN (unknown … unkno… Protoz… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  9 F_YEAST     (unknown … unkno… Fungi   (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #> 10 F_AABRN     Aabaarnia  unkno… Fungi   Ascom… Leca… Ostr… \"\"     Aaba… \"\"      #> # ℹ 78,668 more rows #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list>"},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate the Matching Score for Microorganisms — mo_matching_score","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"algorithm used .mo() mo_* functions determine probable match taxonomic records based user input.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"","code":"mo_matching_score(x, n)"},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"x user input value(s) n full taxonomic name, exists microorganisms$fullname","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"algorithm originally developed 2018 subsequently described : Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 . Later, work Bartlett et al. bacterial pathogens infecting humans (2022, doi:10.1099/mic.0.001269 ) incorporated, optimalisations algorithm made.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"ambiguous user input .mo() mo_* functions, returned results chosen based matching score using mo_matching_score(). matching score \\(m\\), calculated : $$m_{(x, n)} = \\frac{l_{n} - 0.5 \\cdot \\min \\begin{cases}l_{n} \\\\ \\textrm{lev}(x, n)\\end{cases}}{l_{n} \\cdot p_{n} \\cdot k_{n}}$$ : \\(x\\) user input; \\(n\\) taxonomic name (genus, species, subspecies); \\(l_n\\) length \\(n\\); \\(lev\\) Levenshtein distance function (counting insertion 1, deletion substitution 2) needed change \\(x\\) \\(n\\); \\(p_n\\) human pathogenic prevalence group \\(n\\), described ; \\(k_n\\) taxonomic kingdom \\(n\\), set Bacteria = 1, Fungi = 1.25, Protozoa = 1.5, Chromista = 1.75, Archaea = 2, others = 3. grouping human pathogenic prevalence \\(p\\) based recent work Bartlett et al. (2022, doi:10.1099/mic.0.001269 ) extensively studied medical-scientific literature categorise bacterial species groups: Established, taxonomic species infected least three persons three references. records prevalence = 1.15 microorganisms data set; Putative, taxonomic species fewer three known cases. records prevalence = 1.25 microorganisms data set. Furthermore, Genera World Health Organization's () Priority Pathogen List prevalence = 1.0 microorganisms data set; genus present established list also prevalence = 1.15 microorganisms data set; genus present putative list prevalence = 1.25 microorganisms data set; species subspecies genus present two aforementioned groups, prevalence = 1.5 microorganisms data set; non-bacterial genus, species subspecies genus present following list, prevalence = 1.25 microorganisms data set: Absidia, Acanthamoeba, Acremonium, Actinomucor, Aedes, Alternaria, Amoeba, Ancylostoma, Angiostrongylus, Anisakis, Anopheles, Apophysomyces, Arthroderma, Aspergillus, Aureobasidium, Basidiobolus, Beauveria, Bipolaris, Blastobotrys, Blastocystis, Blastomyces, Candida, Capillaria, Chaetomium, Chilomastix, Chrysonilia, Chrysosporium, Cladophialophora, Cladosporium, Clavispora, Coccidioides, Cokeromyces, Conidiobolus, Coniochaeta, Contracaecum, Cordylobia, Cryptococcus, Cryptosporidium, Cunninghamella, Curvularia, Cyberlindnera, Debaryozyma, Demodex, Dermatobia, Dientamoeba, Diphyllobothrium, Dirofilaria, Echinostoma, Entamoeba, Enterobius, Epidermophyton, Exidia, Exophiala, Exserohilum, Fasciola, Fonsecaea, Fusarium, Geotrichum, Giardia, Graphium, Haloarcula, Halobacterium, Halococcus, Hansenula, Hendersonula, Heterophyes, Histomonas, Histoplasma, Hortaea, Hymenolepis, Hypomyces, Hysterothylacium, Kloeckera, Kluyveromyces, Kodamaea, Lacazia, Leishmania, Lichtheimia, Lodderomyces, Lomentospora, Madurella, Malassezia, Malbranchea, Metagonimus, Meyerozyma, Microsporidium, Microsporum, Millerozyma, Mortierella, Mucor, Mycocentrospora, Nannizzia, Necator, Nectria, Ochroconis, Oesophagostomum, Oidiodendron, Opisthorchis, Paecilomyces, Paracoccidioides, Pediculus, Penicillium, Phaeoacremonium, Phaeomoniella, Phialophora, Phlebotomus, Phoma, Pichia, Piedraia, Pithomyces, Pityrosporum, Pneumocystis, Pseudallescheria, Pseudoscopulariopsis, Pseudoterranova, Pulex, Purpureocillium, Quambalaria, Rhinocladiella, Rhizomucor, Rhizopus, Rhodotorula, Saccharomyces, Saksenaea, Saprochaete, Sarcoptes, Scedosporium, Schistosoma, Schizosaccharomyces, Scolecobasidium, Scopulariopsis, Scytalidium, Spirometra, Sporobolomyces, Sporopachydermia, Sporothrix, Sporotrichum, Stachybotrys, Strongyloides, Syncephalastrum, Syngamus, Taenia, Talaromyces, Teleomorph, Toxocara, Trichinella, Trichobilharzia, Trichoderma, Trichomonas, Trichophyton, Trichosporon, Trichostrongylus, Trichuris, Tritirachium, Trombicula, Trypanosoma, Tunga, Ulocladium, Ustilago, Verticillium, Wallemia, Wangiella, Wickerhamomyces, Wuchereria, Yarrowia, Zygosaccharomyces; records prevalence = 2.0 microorganisms data set. calculating matching score, characters \\(x\\) \\(n\\) ignored -Z, -z, 0-9, spaces parentheses. matches sorted descending matching score user input values, top match returned. lead effect e.g., \"E. coli\" return microbial ID Escherichia coli (\\(m = 0.688\\), highly prevalent microorganism found humans) Entamoeba coli (\\(m = 0.381\\), less prevalent microorganism humans), although latter alphabetically come first.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"","code":"mo_reset_session() #> ℹ Reset 17 previously matched input values.  as.mo(\"E. coli\") #> Class 'mo' #> [1] B_ESCHR_COLI mo_uncertainties() #> Matching scores are based on the resemblance between the input and the full #> taxonomic name, and the pathogenicity in humans. See ?mo_matching_score. #> Colour keys:  0.000-0.549  0.550-0.649  0.650-0.749  0.750-1.000  #>  #> -------------------------------------------------------------------------------- #> \"E. coli\" -> Escherichia coli (B_ESCHR_COLI, 0.688) #> Also matched: Enterococcus crotali (0.650), Escherichia coli coli #>               (0.643), Escherichia coli expressing (0.611), Enterobacter cowanii #>               (0.600), Enterococcus columbae (0.595), Enterococcus camelliae (0.591), #>               Enterococcus casseliflavus (0.577), Enterobacter cloacae cloacae #>               (0.571), Enterobacter cloacae complex (0.571), and Enterobacter cloacae #>               dissolvens (0.565) #>  #> Only the first 10 other matches of each record are shown. Run #> print(mo_uncertainties(), n = ...) to view more entries, or save #> mo_uncertainties() to an object.  mo_matching_score(   x = \"E. coli\",   n = c(\"Escherichia coli\", \"Entamoeba coli\") ) #> [1] 0.6875000 0.3809524"},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of a Microorganism — mo_property","title":"Get Properties of a Microorganism — mo_property","text":"Use functions return specific property microorganism based latest accepted taxonomy. input values evaluated internally .mo(), makes possible use microbial abbreviations, codes names input. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of a Microorganism — mo_property","text":"","code":"mo_name(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_fullname(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_shortname(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_subspecies(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_species(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_genus(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_family(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_order(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_class(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_phylum(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_kingdom(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_domain(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_type(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_status(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_pathogenicity(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_gramstain(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_gram_negative(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_gram_positive(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_yeast(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_intrinsic_resistant(x, ab, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_oxygen_tolerance(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_anaerobic(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_snomed(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_ref(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_authors(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_year(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_lpsn(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_mycobank(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_gbif(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_rank(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_taxonomy(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_synonyms(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_current(x, language = get_AMR_locale(), ...)  mo_group_members(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_info(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_url(x, open = FALSE, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_property(x, property = \"fullname\", language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)"},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of a Microorganism — mo_property","text":"x character (vector) can coerced valid microorganism code .mo(). Can left blank auto-guessing column containing microorganism codes used data set, see Examples. language language translate text like \"growth\", defaults system language (see get_AMR_locale()) keep_synonyms logical indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE, return note old taxonomic names processed. default can set package option AMR_keep_synonyms, .e. options(AMR_keep_synonyms = TRUE) options(AMR_keep_synonyms = FALSE). ... arguments passed .mo(), 'minimum_matching_score', 'ignore_pattern', 'remove_from_input' ab (vector ) text can coerced valid antibiotic drug code .ab() open browse URL using browseURL() property one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\", must \"shortname\"","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of a Microorganism — mo_property","text":"integer case mo_year() ordered factor case mo_pathogenicity() list case mo_taxonomy(), mo_synonyms(), mo_snomed(), mo_info() logical case mo_is_anaerobic(), mo_is_gram_negative(), mo_is_gram_positive(), mo_is_intrinsic_resistant(), mo_is_yeast() named character case mo_synonyms() mo_url() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of a Microorganism — mo_property","text":"functions , default, keep old taxonomic properties, synonyms automatically replaced current taxonomy. Take example Enterobacter aerogenes, initially named 1960 renamed Klebsiella aerogenes 2017: mo_genus(\"Enterobacter aerogenes\") return \"Klebsiella\" (note renaming) mo_genus(\"Enterobacter aerogenes\", keep_synonyms = TRUE) return \"Enterobacter\" (-per-session warning name outdated) mo_ref(\"Enterobacter aerogenes\") return \"Tindall et al., 2017\" (note renaming) mo_ref(\"Enterobacter aerogenes\", keep_synonyms = TRUE) return \"Hormaeche et al., 1960\" (-per-session warning name outdated) short name (mo_shortname()) returns first character genus full species, \"E. coli\", species subspecies. Exceptions abbreviations staphylococci (\"CoNS\", Coagulase-Negative Staphylococci) beta-haemolytic streptococci (\"GBS\", Group B Streptococci). Please bear mind e.g. E. coli mean Escherichia coli (kingdom Bacteria) well Entamoeba coli (kingdom Protozoa). Returning full name done using .mo() internally, giving priority bacteria human pathogens, .e. \"E. coli\" considered Escherichia coli. result, mo_fullname(mo_shortname(\"Entamoeba coli\")) returns \"Escherichia coli\". Since top-level taxonomy sometimes referred 'kingdom' sometimes 'domain', functions mo_kingdom() mo_domain() return exact results. Determination human pathogenicity (mo_pathogenicity()) strongly based Bartlett et al. (2022, doi:10.1099/mic.0.001269 ). function returns factor levels Pathogenic, Potentially pathogenic, Non-pathogenic, Unknown. Determination Gram stain (mo_gramstain()) based taxonomic kingdom phylum. Originally, Cavalier-Smith defined -called subkingdoms Negibacteria Posibacteria (2002, PMID 11837318), considered phyla Posibacteria: Actinobacteria, Chloroflexi, Firmicutes, Tenericutes. phyla later renamed Actinomycetota, Chloroflexota, Bacillota, Mycoplasmatota (2021, PMID 34694987). Bacteria phyla considered Gram-positive AMR package, except members class Negativicutes (within phylum Bacillota) Gram-negative. bacteria considered Gram-negative. Species outside kingdom Bacteria return value NA. Functions mo_is_gram_negative() mo_is_gram_positive() always return TRUE FALSE (NA input NA MO code UNKNOWN), thus always return FALSE species outside taxonomic kingdom Bacteria. Determination yeasts (mo_is_yeast()) based taxonomic kingdom class. Budding yeasts yeasts reproduce asexually process called budding, new cell develops small protrusion parent cell. Taxonomically, members phylum Ascomycota, class Saccharomycetes (also called Hemiascomycetes) Pichiomycetes. True yeasts quite specifically refers yeasts underlying order Saccharomycetales (Saccharomyces cerevisiae). Thus, microorganisms member taxonomic class Saccharomycetes Pichiomycetes, function return TRUE. returns FALSE otherwise (NA input NA MO code UNKNOWN). Determination intrinsic resistance (mo_is_intrinsic_resistant()) based intrinsic_resistant data set, based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021). mo_is_intrinsic_resistant() function can vectorised argument x (input microorganisms) ab (input antibiotics). Determination bacterial oxygen tolerance (mo_oxygen_tolerance()) based BacDive, see Source. function mo_is_anaerobic() returns TRUE oxygen tolerance \"anaerobe\", indicting obligate anaerobic species genus. always returns FALSE species outside taxonomic kingdom Bacteria. function mo_url() return direct URL online database entry, also shows scientific reference concerned species. MycoBank URL used fungi wherever available , LPSN URL bacteria wherever available, GBIF link otherwise. SNOMED codes (mo_snomed()) last updated July 16th, 2024. See Source microorganisms data set info. Old taxonomic names (-called 'synonyms') can retrieved mo_synonyms() (scientific reference name), current taxonomic name can retrieved mo_current(). functions return full names. output translated possible.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Get Properties of a Microorganism — mo_property","text":"function uses .mo() internally, uses advanced algorithm translate arbitrary user input valid taxonomy using -called matching score. can read public algorithm MO matching score page.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of a Microorganism — mo_property","text":"Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/ Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of a Microorganism — mo_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of a Microorganism — mo_property","text":"","code":"# taxonomic tree -----------------------------------------------------------  mo_kingdom(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_phylum(\"Klebsiella pneumoniae\") #> [1] \"Pseudomonadota\" mo_class(\"Klebsiella pneumoniae\") #> [1] \"Gammaproteobacteria\" mo_order(\"Klebsiella pneumoniae\") #> [1] \"Enterobacterales\" mo_family(\"Klebsiella pneumoniae\") #> [1] \"Enterobacteriaceae\" mo_genus(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella\" mo_species(\"Klebsiella pneumoniae\") #> [1] \"pneumoniae\" mo_subspecies(\"Klebsiella pneumoniae\") #> [1] \"\"   # full names and short names -----------------------------------------------  mo_name(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella pneumoniae\" mo_fullname(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella pneumoniae\" mo_shortname(\"Klebsiella pneumoniae\") #> [1] \"K. pneumoniae\"   # other properties ---------------------------------------------------------  mo_pathogenicity(\"Klebsiella pneumoniae\") #> [1] Pathogenic #> Levels: Pathogenic < Potentially pathogenic < Non-pathogenic < Unknown mo_gramstain(\"Klebsiella pneumoniae\") #> [1] \"Gram-negative\" mo_snomed(\"Klebsiella pneumoniae\") #> [[1]] #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\"        #>  mo_type(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_rank(\"Klebsiella pneumoniae\") #> [1] \"species\" mo_url(\"Klebsiella pneumoniae\") #>                                Klebsiella pneumoniae  #> \"https://lpsn.dsmz.de/species/klebsiella-pneumoniae\"  mo_is_yeast(c(\"Candida\", \"Trichophyton\", \"Klebsiella\")) #> [1]  TRUE FALSE FALSE  mo_group_members(c(\"Streptococcus group A\",                    \"Streptococcus group C\",                    \"Streptococcus group G\",                    \"Streptococcus group L\")) #> $`Streptococcus Group A` #> [1] \"Streptococcus pyogenes\" #>  #> $`Streptococcus Group C` #> [1] \"Streptococcus dysgalactiae\"              #> [2] \"Streptococcus dysgalactiae dysgalactiae\" #> [3] \"Streptococcus dysgalactiae equisimilis\"  #> [4] \"Streptococcus equi\"                      #> [5] \"Streptococcus equi equi\"                 #> [6] \"Streptococcus equi ruminatorum\"          #> [7] \"Streptococcus equi zooepidemicus\"        #>  #> $`Streptococcus Group G` #> [1] \"Streptococcus canis\"                     #> [2] \"Streptococcus dysgalactiae\"              #> [3] \"Streptococcus dysgalactiae dysgalactiae\" #> [4] \"Streptococcus dysgalactiae equisimilis\"  #>  #> $`Streptococcus Group L` #> [1] \"Streptococcus dysgalactiae\"              #> [2] \"Streptococcus dysgalactiae dysgalactiae\" #> [3] \"Streptococcus dysgalactiae equisimilis\"  #>    # scientific reference -----------------------------------------------------  mo_ref(\"Klebsiella aerogenes\") #> [1] \"Tindall et al., 2017\" mo_authors(\"Klebsiella aerogenes\") #> [1] \"Tindall et al.\" mo_year(\"Klebsiella aerogenes\") #> [1] 2017 mo_synonyms(\"Klebsiella aerogenes\") #>   Hormaeche et al., 1960     Bascomb et al., 1971  #> \"Enterobacter aerogenes\"     \"Klebsiella mobilis\"  mo_lpsn(\"Klebsiella aerogenes\") #> [1] \"777146\" mo_gbif(\"Klebsiella aerogenes\") #> [1] \"9281703\" mo_mycobank(\"Candida albicans\") #> [1] \"256187\" mo_mycobank(\"Candida krusei\") #> [1] \"337013\" mo_mycobank(\"Candida krusei\", keep_synonyms = TRUE) #> Warning: Function as.mo() returned one old taxonomic name. Use as.mo(..., #> keep_synonyms = FALSE) to clean the input to currently accepted taxonomic #> names, or set the R option AMR_keep_synonyms to FALSE. This warning #> will be shown once per session. #> [1] \"268707\"   # abbreviations known in the field -----------------------------------------  mo_genus(\"MRSA\") #> [1] \"Staphylococcus\" mo_species(\"MRSA\") #> [1] \"aureus\" mo_shortname(\"VISA\") #> [1] \"S. aureus\" mo_gramstain(\"VISA\") #> [1] \"Gram-positive\"  mo_genus(\"EHEC\") #> [1] \"Escherichia\" mo_species(\"EIEC\") #> [1] \"coli\" mo_name(\"UPEC\") #> [1] \"Escherichia coli\"   # known subspecies ---------------------------------------------------------  mo_fullname(\"K. pneu rh\") #> [1] \"Klebsiella pneumoniae rhinoscleromatis\" mo_shortname(\"K. pneu rh\") #> [1] \"K. pneumoniae\"  # \\donttest{ # Becker classification, see ?as.mo ----------------------------------------  mo_fullname(\"Staph epidermidis\") #> [1] \"Staphylococcus epidermidis\" mo_fullname(\"Staph epidermidis\", Becker = TRUE) #> [1] \"Coagulase-negative Staphylococcus (CoNS)\" mo_shortname(\"Staph epidermidis\") #> [1] \"S. epidermidis\" mo_shortname(\"Staph epidermidis\", Becker = TRUE) #> [1] \"CoNS\"   # Lancefield classification, see ?as.mo ------------------------------------  mo_fullname(\"Strep agalactiae\") #> [1] \"Streptococcus agalactiae\" mo_fullname(\"Strep agalactiae\", Lancefield = TRUE) #> [1] \"Streptococcus Group B\" mo_shortname(\"Strep agalactiae\") #> [1] \"S. agalactiae\" mo_shortname(\"Strep agalactiae\", Lancefield = TRUE) #> [1] \"GBS\"   # language support  --------------------------------------------------------  mo_gramstain(\"Klebsiella pneumoniae\", language = \"de\") # German #> [1] \"Gramnegativ\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"nl\") # Dutch #> [1] \"Gram-negatief\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"es\") # Spanish #> [1] \"Gram negativo\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"el\") # Greek #> [1] \"Αρνητικό κατά Gram\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"uk\") # Ukrainian #> [1] \"Грамнегативні\"  # mo_type is equal to mo_kingdom, but mo_kingdom will remain untranslated mo_kingdom(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_type(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_kingdom(\"Klebsiella pneumoniae\", language = \"zh\") # Chinese, no effect #> [1] \"Bacteria\" mo_type(\"Klebsiella pneumoniae\", language = \"zh\") # Chinese, translated #> [1] \"细菌\"  mo_fullname(\"S. pyogenes\", Lancefield = TRUE, language = \"de\") #> [1] \"Streptococcus Gruppe A\" mo_fullname(\"S. pyogenes\", Lancefield = TRUE, language = \"uk\") #> [1] \"Streptococcus Група A\"   # other --------------------------------------------------------------------  # gram stains and intrinsic resistance can be used as a filter in dplyr verbs if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_is_gram_positive()) %>%     count(mo_genus(), sort = TRUE) } #> ℹ Using column 'mo' as input for mo_is_gram_positive() #> ℹ Using column 'mo' as input for mo_genus() #> # A tibble: 18 × 2 #>    `mo_genus()`        n #>    <chr>           <int> #>  1 Staphylococcus    840 #>  2 Streptococcus     275 #>  3 Enterococcus       83 #>  4 Corynebacterium    17 #>  5 Micrococcus         6 #>  6 Gemella             3 #>  7 Aerococcus          2 #>  8 Cutibacterium       1 #>  9 Dermabacter         1 #> 10 Fusibacter          1 #> 11 Globicatella        1 #> 12 Granulicatella      1 #> 13 Lactobacillus       1 #> 14 Leuconostoc         1 #> 15 Listeria            1 #> 16 Paenibacillus       1 #> 17 Rothia              1 #> 18 Schaalia            1 if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_is_intrinsic_resistant(ab = \"vanco\")) %>%     count(mo_genus(), sort = TRUE) } #> ℹ Using column 'mo' as input for mo_is_intrinsic_resistant() #> ℹ Using column 'mo' as input for mo_genus() #> # A tibble: 20 × 2 #>    `mo_genus()`         n #>    <chr>            <int> #>  1 Escherichia        467 #>  2 Klebsiella          77 #>  3 Proteus             39 #>  4 Pseudomonas         30 #>  5 Serratia            25 #>  6 Enterobacter        23 #>  7 Citrobacter         11 #>  8 Haemophilus          8 #>  9 Acinetobacter        6 #> 10 Morganella           6 #> 11 Pantoea              4 #> 12 Salmonella           3 #> 13 Neisseria            2 #> 14 Stenotrophomonas     2 #> 15 Campylobacter        1 #> 16 Enterococcus         1 #> 17 Hafnia               1 #> 18 Lactobacillus        1 #> 19 Leuconostoc          1 #> 20 Pseudescherichia     1  # get a list with the complete taxonomy (from kingdom to subspecies) mo_taxonomy(\"Klebsiella pneumoniae\") #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Klebsiella\" #>  #> $species #> [1] \"pneumoniae\" #>  #> $subspecies #> [1] \"\" #>   # get a list with the taxonomy, the authors, Gram-stain, # SNOMED codes, and URL to the online database mo_info(\"Klebsiella pneumoniae\") #> $mo #> [1] \"B_KLBSL_PNMN\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Klebsiella\" #>  #> $species #> [1] \"pneumoniae\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] \"facultative anaerobe\" #>  #> $url #> [1] \"https://lpsn.dsmz.de/species/klebsiella-pneumoniae\" #>  #> $ref #> [1] \"Trevisan, 1887\" #>  #> $snomed #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\"        #>  #> $lpsn #> [1] \"777151\" #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] \"3221874\" #>  #> $group_members #> character(0) #>  # }"},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":null,"dir":"Reference","previous_headings":"","what":"User-Defined Reference Data Set for Microorganisms — mo_source","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"functions can used predefine reference used .mo() consequently mo_* functions (mo_genus() mo_gramstain()). fastest way organisation (analysis) specific codes picked translated package, since bother setting .","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"","code":"set_mo_source(path, destination = getOption(\"AMR_mo_source\",   \"~/mo_source.rds\"))  get_mo_source(destination = getOption(\"AMR_mo_source\", \"~/mo_source.rds\"))"},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"path location reference file, can text file (comma-, tab- pipe-separated) Excel file (see Details). Can also \"\", NULL FALSE delete reference file. destination destination compressed data file - default user's home directory.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"reference file can text file separated commas (CSV) tabs pipes, Excel file (either 'xls' 'xlsx' format) R object file (extension '.rds'). use Excel file, need readxl package installed. set_mo_source() check file validity: must data.frame, must column named \"mo\" contains values microorganisms$mo microorganisms$fullname must reference column defined values. tests pass, set_mo_source() read file R ask export \"~/mo_source.rds\". CRAN policy disallows packages write file system, although 'exceptions may allowed interactive sessions package obtains confirmation user'. reason, function works interactive sessions user can specifically confirm allow file created. destination file can set destination argument defaults user's home directory. can also set package option AMR_mo_source, e.g. options(AMR_mo_source = \"/location/file.rds\"). created compressed data file \"mo_source.rds\" used default MO determination (function .mo() consequently mo_* functions like mo_genus() mo_gramstain()). location timestamp original file saved attribute compressed data file. function get_mo_source() return data set reading \"mo_source.rds\" readRDS(). original file changed (checking location timestamp original file), call set_mo_source() update data file automatically used interactive session. Reading Excel file (.xlsx) one row size 8-9 kB. compressed file created set_mo_source() size 0.1 kB can read get_mo_source() couple microseconds (millionths second).","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"how-to-setup","dir":"Reference","previous_headings":"","what":"How to Setup","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"Imagine data sheet Excel file. first column contains organisation specific codes, second column contains valid taxonomic names:   save \"/Users//Documents/ourcodes.xlsx\". Now set source:   now created file \"~/mo_source.rds\" contents Excel file. first column foreign values 'mo' column kept creating RDS file. now can use functions:   edit Excel file , say, adding row 4 like :   ...new usage MO function package update data file:   delete reference data file, just use \"\", NULL FALSE input set_mo_source():   original file (previous case Excel file) moved deleted, mo_source.rds file removed upon next use .mo() mo_* function.","code":"|         A          |            B          | --|--------------------|-----------------------| 1 | Organisation XYZ   | mo                    | 2 | lab_mo_ecoli       | Escherichia coli      | 3 | lab_mo_kpneumoniae | Klebsiella pneumoniae | 4 |                    |                       | set_mo_source(\"/Users/me/Documents/ourcodes.xlsx\") #> NOTE: Created mo_source file '/Users/me/mo_source.rds' (0.3 kB) from #>       '/Users/me/Documents/ourcodes.xlsx' (9 kB), columns #>       \"Organisation XYZ\" and \"mo\" as.mo(\"lab_mo_ecoli\") #> Class 'mo' #> [1] B_ESCHR_COLI  mo_genus(\"lab_mo_kpneumoniae\") #> [1] \"Klebsiella\"  # other input values still work too as.mo(c(\"Escherichia coli\", \"E. coli\", \"lab_mo_ecoli\")) #> NOTE: Translation to one microorganism was guessed with uncertainty. #>       Use mo_uncertainties() to review it. #> Class 'mo' #> [1] B_ESCHR_COLI B_ESCHR_COLI B_ESCHR_COLI |         A          |            B          | --|--------------------|-----------------------| 1 | Organisation XYZ   | mo                    | 2 | lab_mo_ecoli       | Escherichia coli      | 3 | lab_mo_kpneumoniae | Klebsiella pneumoniae | 4 | lab_Staph_aureus   | Staphylococcus aureus | 5 |                    |                       | as.mo(\"lab_mo_ecoli\") #> NOTE: Updated mo_source file '/Users/me/mo_source.rds' (0.3 kB) from #>       '/Users/me/Documents/ourcodes.xlsx' (9 kB), columns #>        \"Organisation XYZ\" and \"mo\" #> Class 'mo' #> [1] B_ESCHR_COLI  mo_genus(\"lab_Staph_aureus\") #> [1] \"Staphylococcus\" set_mo_source(NULL) #> Removed mo_source file '/Users/me/mo_source.rds'"},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Component Analysis (for AMR) — pca","title":"Principal Component Analysis (for AMR) — pca","text":"Performs principal component analysis (PCA) based data set automatic determination afterwards plotting groups labels, automatic filtering suitable (.e. non-empty numeric) variables.","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Component Analysis (for AMR) — pca","text":"","code":"pca(x, ..., retx = TRUE, center = TRUE, scale. = TRUE, tol = NULL,   rank. = NULL)"},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Component Analysis (for AMR) — pca","text":"x data.frame containing numeric columns ... columns x selected PCA, can unquoted since supports quasiquotation. retx logical value indicating whether rotated variables     returned. center logical value indicating whether variables     shifted zero centered. Alternately, vector     length equal number columns x can supplied.     value passed scale. scale. logical value indicating whether variables     scaled unit variance analysis takes     place.  default FALSE consistency S,     general scaling advisable.  Alternatively, vector length     equal number columns x can supplied.      value passed scale. tol value indicating magnitude components     omitted. (Components omitted     standard deviations less equal tol times     standard deviation first component.)  default null     setting, components omitted (unless rank. specified     less min(dim(x)).).  settings tol     tol = 0 tol = sqrt(.Machine$double.eps),     omit essentially constant components. rank. optionally, number specifying maximal rank, .e.,     maximal number principal components used.  Can set     alternative addition tol, useful notably     desired rank considerably smaller dimensions matrix.","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Component Analysis (for AMR) — pca","text":"object classes pca prcomp","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Component Analysis (for AMR) — pca","text":"pca() function takes data.frame input performs actual PCA R function prcomp(). result pca() function prcomp object, additional attribute non_numeric_cols vector column names columns contain numeric values. probably groups labels, used ggplot_pca().","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Component Analysis (for AMR) — pca","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # \\donttest{ if (require(\"dplyr\")) {   # calculate the resistance per group first   resistance_data <- example_isolates %>%     group_by(       order = mo_order(mo), # group on anything, like order       genus = mo_genus(mo)     ) %>% #   and genus as we do here;     filter(n() >= 30) %>% # filter on only 30 results per group     summarise_if(is.sir, resistance) # then get resistance of all drugs    # now conduct PCA for certain antimicrobial drugs   pca_result <- resistance_data %>%     pca(AMC, CXM, CTX, CAZ, GEN, TOB, TMP, SXT)    pca_result   summary(pca_result)    # old base R plotting method:   biplot(pca_result)   # new ggplot2 plotting method using this package:   if (require(\"ggplot2\")) {     ggplot_pca(pca_result)      ggplot_pca(pca_result) +       scale_colour_viridis_d() +       labs(title = \"Title here\")   } } #> Warning: There were 73 warnings in `summarise()`. #> The first warning was: #> ℹ In argument: `PEN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 5: `order = \"Lactobacillales\"` `genus = \"Enterococcus\"`. #> Caused by warning: #> ! Introducing NA: only 14 results available for PEN in group: order = #> \"Lactobacillales\", genus = \"Enterococcus\" (minimum = 30). #> ℹ Run `dplyr::last_dplyr_warnings()` to see the 72 remaining warnings. #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\"  #>    # }"},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":null,"dir":"Reference","previous_headings":"","what":"Plotting Helpers for AMR Data Analysis — plot","title":"Plotting Helpers for AMR Data Analysis — plot","text":"Functions plot classes sir, mic disk, support base R ggplot2. Especially scale_*_mic() functions relevant wrappers plot MIC values ggplot2. allows custom MIC ranges plot intermediate log2 levels missing MIC values.","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plotting Helpers for AMR Data Analysis — plot","text":"","code":"scale_x_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_y_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_colour_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_fill_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_x_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  scale_colour_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  scale_fill_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  # S3 method for class 'mic' plot(x, mo = NULL, ab = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   main = deparse(substitute(x)), ylab = translate_AMR(\"Frequency\", language   = language),   xlab = translate_AMR(\"Minimum Inhibitory Concentration (mg/L)\", language =   language), colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'mic' autoplot(object, mo = NULL, ab = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   title = deparse(substitute(object)), ylab = translate_AMR(\"Frequency\",   language = language),   xlab = translate_AMR(\"Minimum Inhibitory Concentration (mg/L)\", language =   language), colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'disk' plot(x, main = deparse(substitute(x)),   ylab = translate_AMR(\"Frequency\", language = language),   xlab = translate_AMR(\"Disk diffusion diameter (mm)\", language = language),   mo = NULL, ab = NULL, guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'disk' autoplot(object, mo = NULL, ab = NULL,   title = deparse(substitute(object)), ylab = translate_AMR(\"Frequency\",   language = language), xlab = translate_AMR(\"Disk diffusion diameter (mm)\",   language = language), guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'sir' plot(x, ylab = translate_AMR(\"Percentage\", language =   language), xlab = translate_AMR(\"Antimicrobial Interpretation\", language =   language), main = deparse(substitute(x)), language = get_AMR_locale(),   ...)  # S3 method for class 'sir' autoplot(object, title = deparse(substitute(object)),   xlab = translate_AMR(\"Antimicrobial Interpretation\", language = language),   ylab = translate_AMR(\"Frequency\", language = language),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), ...)  facet_sir(facet = c(\"interpretation\", \"antibiotic\"), nrow = NULL)  scale_y_percent(breaks = function(x) seq(0, max(x, na.rm = TRUE), 0.1),   limits = c(0, NA))  scale_sir_colours(..., aesthetics, colours_SIR = c(\"#3CAEA3\", \"#F6D55C\",   \"#ED553B\"))  theme_sir()  labels_sir_count(position = NULL, x = \"antibiotic\",   translate_ab = \"name\", minimum = 30, language = get_AMR_locale(),   combine_SI = TRUE, datalabels.size = 3, datalabels.colour = \"grey15\")"},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plotting Helpers for AMR Data Analysis — plot","text":"keep_operators character specifying handle operators (> <=) input. Accepts one three values: \"\" (TRUE) keep operators, \"none\" (FALSE) remove operators, \"edges\" keep operators ends range. mic_range manual range rescale MIC values (using rescale_mic()), e.g., mic_range = c(0.001, 32). Use NA prevent rescaling one side, e.g., mic_range = c(NA, 32). Note: rescales values filter - use ggplot2 limits argument separately exclude values plot. ... arguments passed methods colours_SIR colours use filling bars, must vector three values (order S, R). default colours colour-blind friendly. language language used translate 'Susceptible', 'Increased exposure'/'Intermediate' 'Resistant' - default system language (see get_AMR_locale()) can overwritten setting package option AMR_locale, e.g. options(AMR_locale = \"de\"), see translate. Use language = NULL prevent translation. eucast_I logical indicate whether '' must interpreted \"Susceptible, increased exposure\". TRUE default AMR interpretation guideline set EUCAST (default). FALSE, interpreted \"Intermediate\". x, object values created .mic(), .disk() .sir() (random_* variants, random_mic()) mo (vector ) text can coerced valid microorganism code .mo() ab (vector ) text can coerced valid antimicrobial drug code .ab() guideline interpretation guideline use - default latest included EUCAST guideline, see Details main, title title plot xlab, ylab axis title expand logical indicate whether range x axis expanded lowest highest value. MIC values, intermediate values factors 2 starting highest MIC value. disk diameters, whole diameter range filled. include_PKPD logical indicate PK/PD clinical breakpoints must applied last resort - default TRUE. Can also set package option AMR_include_PKPD. breakpoint_type type breakpoints use, either \"ECOFF\", \"animal\", \"human\". ECOFF stands Epidemiological Cut-values. default \"human\", can also set package option AMR_breakpoint_type. host set values veterinary species, automatically set \"animal\". facet variable split plots , either \"interpretation\" (default) \"antibiotic\" grouping variable nrow (using facet) number rows breaks numeric vector positions limits numeric vector length two providing limits scale, use NA refer existing minimum maximum aesthetics aesthetics apply colours - default \"fill\" can also (combination ) \"alpha\", \"colour\", \"fill\", \"linetype\", \"shape\" \"size\" position position adjustment bars, either \"fill\", \"stack\" \"dodge\" translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE datalabels.size size datalabels datalabels.colour colour datalabels","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plotting Helpers for AMR Data Analysis — plot","text":"autoplot() functions return ggplot model extendible ggplot2 function.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"the-scale-mic-functions","dir":"Reference","previous_headings":"","what":"The scale_*_mic() Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"functions scale_x_mic(), scale_y_mic(), scale_colour_mic(), scale_fill_mic() functions allow plot mic class (MIC values) continuous, logarithmic scale. also allow rescale MIC range 'inside' 'outside' range required, retain signs MIC values desired. Missing intermediate log2 levels plotted .","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"the-scale-sir-functions","dir":"Reference","previous_headings":"","what":"The scale_*_sir() Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"functions scale_x_sir(), scale_colour_sir(), scale_fill_sir() functions allow plot sir class right order (S < SDD < < R < NI). default, translate S//R values interpretative text (\"Susceptible\", \"Resistant\", etc.) 20 supported languages (use language = NULL keep S//R). Also, except scale_x_sir(), set colour-blind friendly colours colour fill aesthetics.","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"additional-ggplot-functions","dir":"Reference","previous_headings":"","what":"Additional ggplot2 Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"package contains functions extend ggplot2 package, help visualising AMR data results. functions internally used ggplot_sir() . facet_sir() creates 2d plots (default based S//R) using ggplot2::facet_wrap(). scale_y_percent() transforms y axis 0 100% range using ggplot2::scale_y_continuous(). scale_sir_colours() allows set colours aesthetic, even shape linetype. theme_sir() [ggplot2 theme][ggplot2::theme() minimal distraction. labels_sir_count() print datalabels bars percentage number isolates, using ggplot2::geom_text(). interpretation \"\" named \"Increased exposure\" EUCAST guidelines since 2019, named \"Intermediate\" cases. interpreting MIC values well disk diffusion diameters, default guideline EUCAST 2024, unless package option AMR_guideline set. See .sir() information.","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Plotting Helpers for AMR Data Analysis — plot","text":"","code":"some_mic_values <- random_mic(size = 100) some_disk_values <- random_disk(size = 100, mo = \"Escherichia coli\", ab = \"cipro\") some_sir_values <- random_sir(50, prob_SIR = c(0.55, 0.05, 0.30))   # Plotting using base R's plot() ---------------------------------------  plot(some_mic_values)  plot(some_disk_values)  plot(some_sir_values)   # when providing the microorganism and antibiotic, colours will show interpretations: plot(some_mic_values, mo = \"S. aureus\", ab = \"ampicillin\")  plot(some_disk_values, mo = \"Escherichia coli\", ab = \"cipro\")  plot(some_disk_values, mo = \"Escherichia coli\", ab = \"cipro\", language = \"nl\")    # Plotting using scale_x_mic() ----------------------------------------- # \\donttest{ if (require(\"ggplot2\")) {   mic_plot <- ggplot(data.frame(mics = as.mic(c(0.25, \"<=4\", 4, 8, 32, \">=32\")),                                 counts = c(1, 1, 2, 2, 3, 3)),                      aes(mics, counts)) +     geom_col()   mic_plot +     labs(title = \"without scale_x_mic()\") }  if (require(\"ggplot2\")) {   mic_plot +     scale_x_mic() +     labs(title = \"with scale_x_mic()\") }  if (require(\"ggplot2\")) {   mic_plot +     scale_x_mic(keep_operators = \"all\") +     labs(title = \"with scale_x_mic() keeping all operators\") }  if (require(\"ggplot2\")) {   mic_plot +     scale_x_mic(mic_range = c(1, 16)) +     labs(title = \"with scale_x_mic() using a manual 'within' range\") }  if (require(\"ggplot2\")) {   mic_plot +     scale_x_mic(mic_range = c(0.032, 256)) +     labs(title = \"with scale_x_mic() using a manual 'outside' range\") }    # Plotting using scale_y_mic() ----------------------------------------- some_groups <- sample(LETTERS[1:5], 20, replace = TRUE)  if (require(\"ggplot2\")) {   ggplot(data.frame(mic = some_mic_values,                     group = some_groups),          aes(group, mic)) +     geom_boxplot() +     geom_violin(linetype = 2, colour = \"grey\", fill = NA) +     scale_y_mic() }  if (require(\"ggplot2\")) {   ggplot(data.frame(mic = some_mic_values,                     group = some_groups),          aes(group, mic)) +     geom_boxplot() +     geom_violin(linetype = 2, colour = \"grey\", fill = NA) +     scale_y_mic(mic_range = c(NA, 2)) }    # Plotting using scale_fill_mic() ----------------------------------------- some_counts <- as.integer(runif(20, 5, 50))  if (require(\"ggplot2\")) {   ggplot(data.frame(mic = some_mic_values,                    group = some_groups,                    counts = some_counts),          aes(group, counts, fill = mic)) +     geom_col() +     scale_fill_mic(mic_range = c(0.5, 16)) }   # Plotting using scale_x_sir() ----------------------------------------- if (require(\"ggplot2\")) {   ggplot(data.frame(x = c(\"I\", \"R\", \"S\"),                     y = c(45,323, 573)),          aes(x, y)) +     geom_col() +     scale_x_sir() }    # Plotting using scale_y_mic() and scale_colour_sir() ------------------ if (require(\"ggplot2\")) {   plain <- ggplot(data.frame(mic = some_mic_values,                              group = some_groups,                              sir = as.sir(some_mic_values,                                           mo = \"E. coli\",                                           ab = \"cipro\")),                   aes(x = group, y = mic, colour = sir)) +     theme_minimal() +     geom_boxplot(fill = NA, colour = \"grey\") +     geom_jitter(width = 0.25)      plain } #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'.  if (require(\"ggplot2\")) {   # and now with our MIC and SIR scale functions:   plain +     scale_y_mic() +     scale_colour_sir() }  if (require(\"ggplot2\")) {   plain +     scale_y_mic(mic_range = c(0.005, 32), name = \"Our MICs!\") +     scale_colour_sir(language = \"nl\", eucast_I = FALSE,                      name = \"In Dutch!\") }    # Plotting using ggplot2's autoplot() ---------------------------------- if (require(\"ggplot2\")) {   autoplot(some_mic_values) }  if (require(\"ggplot2\")) {   autoplot(some_disk_values, mo = \"Escherichia coli\", ab = \"cipro\") }  if (require(\"ggplot2\")) {   autoplot(some_sir_values) }    # Plotting using scale_y_percent() ------------------------------------- if (require(\"ggplot2\")) {   p <- ggplot(data.frame(mics = as.mic(c(0.25, \"<=4\", 4, 8, 32, \">=32\")),                                counts = c(1, 1, 2, 2, 3, 3)),                          aes(mics, counts / sum(counts))) +     geom_col()   print(p)      p2 <- p +     scale_y_percent() +     theme_sir()   print(p2)      p +     scale_y_percent(breaks = seq(from = 0, to = 1, by = 0.1),                     limits = c(0, 1)) +     theme_sir() }    # }"},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate Antimicrobial Resistance — proportion","title":"Calculate Antimicrobial Resistance — proportion","text":"functions can used calculate (co-)resistance susceptibility microbial isolates (.e. percentage S, SI, , IR R). functions support quasiquotation pipes, can used summarise() dplyr package also support grouped variables, see Examples. resistance() used calculate resistance, susceptibility() used calculate susceptibility.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate Antimicrobial Resistance — proportion","text":"","code":"resistance(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  susceptibility(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  sir_confidence_interval(..., ab_result = \"R\", minimum = 30,   as_percent = FALSE, only_all_tested = FALSE, confidence_level = 0.95,   side = \"both\", collapse = FALSE)  proportion_R(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_IR(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_I(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_SI(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_S(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_df(data, translate_ab = \"name\", language = get_AMR_locale(),   minimum = 30, as_percent = FALSE, combine_SI = TRUE,   confidence_level = 0.95)  sir_df(data, translate_ab = \"name\", language = get_AMR_locale(),   minimum = 30, as_percent = FALSE, combine_SI = TRUE,   confidence_level = 0.95)"},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Calculate Antimicrobial Resistance — proportion","text":"M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate Antimicrobial Resistance — proportion","text":"... one vectors (columns) antibiotic interpretations. transformed internally .sir() needed. Use multiple columns calculate (lack ) co-resistance: probability one two drugs resistant susceptible result. See Examples. minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. as_percent logical indicate whether output must returned hundred fold % sign (character). value 0.123456 returned \"12.3%\". only_all_tested (combination therapies, .e. using one variable ...): logical indicate isolates must tested antibiotics, see section Combination Therapy ab_result antibiotic results test , must one values \"S\", \"SDD\", \"\", \"R\" confidence_level confidence level returned confidence interval. calculation, number S SI isolates, R isolates compared total number available isolates R, S, using binom.test(), .e., Clopper-Pearson method. side side confidence interval return. default \"\" length 2 vector, can also (abbreviated ) \"min\"/\"left\"/\"lower\"/\"less\" \"max\"/\"right\"/\"higher\"/\"greater\". collapse logical indicate whether output values 'collapsed', .e. merged together one value, character value use collapsing data data.frame containing columns class sir (see .sir()) translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate Antimicrobial Resistance — proportion","text":"double , as_percent = TRUE, character.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate Antimicrobial Resistance — proportion","text":"automated comprehensive analysis, consider using antibiogram() wisca(), streamline many aspects susceptibility reporting , importantly, also support WISCA. functions described offer hands-, manual approach greater customisation. Remember filter data let contain first isolates! needed exclude duplicates reduce selection bias. Use first_isolate() determine data set one four available algorithms. function resistance() equal function proportion_R(). function susceptibility() equal function proportion_SI(). Since AMR v3.0, proportion_SI() proportion_I() include dose-dependent susceptibility ('SDD'). Use sir_confidence_interval() calculate confidence interval, relies binom.test(), .e., Clopper-Pearson method. function returns vector length 2 default antimicrobial resistance. Change side argument \"left\"/\"min\" \"right\"/\"max\" return single value, change ab_result argument e.g. c(\"S\", \"\") test antimicrobial susceptibility, see Examples. functions meant count isolates, calculate proportion resistance/susceptibility. Use count_*() functions count isolates. function susceptibility() essentially equal count_susceptible()/count_all(). Low counts can influence outcome - proportion_*() functions may camouflage , since return proportion (albeit dependent minimum argument). function proportion_df() takes variable data sir class (created .sir()) calculates proportions S, , R. also supports grouped variables. function sir_df() works exactly like proportion_df(), adds number isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"combination-therapy","dir":"Reference","previous_headings":"","what":"Combination Therapy","title":"Calculate Antimicrobial Resistance — proportion","text":"using one variable ... (= combination therapy), use only_all_tested count isolates tested antibiotics/variables test . See example two antibiotics, Drug Drug B, susceptibility() works calculate %SI:   Please note , combination therapies, only_all_tested = TRUE applies :   , combination therapies, only_all_tested = FALSE applies :   Using only_all_tested impact using one antibiotic input.","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - -------------------------------------------------------------------- count_S()    +   count_I()    +   count_R()    = count_all()   proportion_S() + proportion_I() + proportion_R() = 1 count_S()    +   count_I()    +   count_R()    >= count_all()   proportion_S() + proportion_I() + proportion_R() >= 1"},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Calculate Antimicrobial Resistance — proportion","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate Antimicrobial Resistance — proportion","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …   # base R ------------------------------------------------------------ # determines %R resistance(example_isolates$AMX) #> [1] 0.5955556 sir_confidence_interval(example_isolates$AMX) #> [1] 0.5688204 0.6218738 sir_confidence_interval(example_isolates$AMX,   confidence_level = 0.975 ) #> [1] 0.5650148 0.6255670 sir_confidence_interval(example_isolates$AMX,   confidence_level = 0.975,   collapse = \", \" ) #> [1] \"0.565, 0.626\"  # determines %S+I: susceptibility(example_isolates$AMX) #> [1] 0.4044444 sir_confidence_interval(example_isolates$AMX,   ab_result = c(\"S\", \"I\") ) #> [1] 0.3781262 0.4311796  # be more specific proportion_S(example_isolates$AMX) #> [1] 0.4022222 proportion_SI(example_isolates$AMX) #> Note that proportion_SI() will also include dose-dependent #> susceptibility, 'SDD'. This note will be shown once for this session. #> [1] 0.4044444 proportion_I(example_isolates$AMX) #> Note that proportion_I() will also include dose-dependent susceptibility, #> 'SDD'. This note will be shown once for this session. #> [1] 0.002222222 proportion_IR(example_isolates$AMX) #> Note that proportion_IR() will also include dose-dependent #> susceptibility, 'SDD'. This note will be shown once for this session. #> [1] 0.5977778 proportion_R(example_isolates$AMX) #> [1] 0.5955556  # dplyr ------------------------------------------------------------- # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       r = resistance(CIP),       n = n_sir(CIP)     ) # n_sir works like n_distinct in dplyr, see ?n_sir } #> # A tibble: 3 × 3 #>   ward           r     n #>   <chr>      <dbl> <int> #> 1 Clinical   0.147   869 #> 2 ICU        0.190   447 #> 3 Outpatient 0.161    93 if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       cipro_R = resistance(CIP),       ci_min = sir_confidence_interval(CIP, side = \"min\"),       ci_max = sir_confidence_interval(CIP, side = \"max\"),     ) } #> # A tibble: 3 × 4 #>   ward       cipro_R ci_min ci_max #>   <chr>        <dbl>  <dbl>  <dbl> #> 1 Clinical     0.147 0.124   0.173 #> 2 ICU          0.190 0.155   0.230 #> 3 Outpatient   0.161 0.0932  0.252 if (require(\"dplyr\")) {   # scoped dplyr verbs with antimicrobial selectors   # (you could also use across() of course)   example_isolates %>%     group_by(ward) %>%     summarise_at(       c(aminoglycosides(), carbapenems()),       resistance     ) } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> Warning: There was 1 warning in `summarise()`. #> ℹ In argument: `KAN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 3: `ward = \"Outpatient\"`. #> Caused by warning: #> ! Introducing NA: only 23 results available for KAN in group: ward = #> \"Outpatient\" (minimum = 30). #> # A tibble: 3 × 7 #>   ward         GEN   TOB   AMK   KAN    IPM    MEM #>   <chr>      <dbl> <dbl> <dbl> <dbl>  <dbl>  <dbl> #> 1 Clinical   0.229 0.315 0.626     1 0.0498 0.0458 #> 2 ICU        0.290 0.400 0.662     1 0.0862 0.0894 #> 3 Outpatient 0.2   0.368 0.605    NA 0.0541 0.0541 if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       R = resistance(CIP, as_percent = TRUE),       SI = susceptibility(CIP, as_percent = TRUE),       n1 = count_all(CIP), # the actual total; sum of all three       n2 = n_sir(CIP), # same - analogous to n_distinct       total = n()     ) # NOT the number of tested isolates!    # Calculate co-resistance between amoxicillin/clav acid and gentamicin,   # so we can see that combination therapy does a lot more than mono therapy:   example_isolates %>% susceptibility(AMC) # %SI = 76.3%   example_isolates %>% count_all(AMC) #   n = 1879    example_isolates %>% susceptibility(GEN) # %SI = 75.4%   example_isolates %>% count_all(GEN) #   n = 1855    example_isolates %>% susceptibility(AMC, GEN) # %SI = 94.1%   example_isolates %>% count_all(AMC, GEN) #   n = 1939     # See Details on how `only_all_tested` works. Example:   example_isolates %>%     summarise(       numerator = count_susceptible(AMC, GEN),       denominator = count_all(AMC, GEN),       proportion = susceptibility(AMC, GEN)     )    example_isolates %>%     summarise(       numerator = count_susceptible(AMC, GEN, only_all_tested = TRUE),       denominator = count_all(AMC, GEN, only_all_tested = TRUE),       proportion = susceptibility(AMC, GEN, only_all_tested = TRUE)     )     example_isolates %>%     group_by(ward) %>%     summarise(       cipro_p = susceptibility(CIP, as_percent = TRUE),       cipro_n = count_all(CIP),       genta_p = susceptibility(GEN, as_percent = TRUE),       genta_n = count_all(GEN),       combination_p = susceptibility(CIP, GEN, as_percent = TRUE),       combination_n = count_all(CIP, GEN)     )    # Get proportions S/I/R immediately of all sir columns   example_isolates %>%     select(AMX, CIP) %>%     proportion_df(translate = FALSE)    # It also supports grouping variables   # (use sir_df to also include the count)   example_isolates %>%     select(ward, AMX, CIP) %>%     group_by(ward) %>%     sir_df(translate = FALSE) } #> # A tibble: 12 × 7 #>    ward       antibiotic interpretation value ci_min ci_max isolates #>  * <chr>      <chr>      <ord>          <dbl>  <dbl>  <dbl>    <int> #>  1 Clinical   AMX        SI             0.423 0.389   0.457      357 #>  2 Clinical   AMX        R              0.577 0.543   0.611      487 #>  3 Clinical   CIP        SI             0.853 0.827   0.876      741 #>  4 Clinical   CIP        R              0.147 0.124   0.173      128 #>  5 ICU        AMX        SI             0.369 0.323   0.417      158 #>  6 ICU        AMX        R              0.631 0.583   0.677      270 #>  7 ICU        CIP        SI             0.810 0.770   0.845      362 #>  8 ICU        CIP        R              0.190 0.155   0.230       85 #>  9 Outpatient AMX        SI             0.397 0.288   0.515       31 #> 10 Outpatient AMX        R              0.603 0.485   0.712       47 #> 11 Outpatient CIP        SI             0.839 0.748   0.907       78 #> 12 Outpatient CIP        R              0.161 0.0932  0.252       15 # }"},{"path":"https://msberends.github.io/AMR/reference/random.html","id":null,"dir":"Reference","previous_headings":"","what":"Random MIC Values/Disk Zones/SIR Generation — random","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"functions can used generating random MIC values disk diffusion diameters, AMR data analysis practice. providing microorganism antimicrobial drug, generated results reflect reality much possible.","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"","code":"random_mic(size = NULL, mo = NULL, ab = NULL, ...)  random_disk(size = NULL, mo = NULL, ab = NULL, ...)  random_sir(size = NULL, prob_SIR = c(0.33, 0.33, 0.33), ...)"},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"size desired size returned vector. used data.frame call dplyr verb, get current (group) size left blank. mo character can coerced valid microorganism code .mo() ab character can coerced valid antimicrobial drug code .ab() ... ignored, place allow future extensions prob_SIR vector length 3: probabilities \"S\" (1st value), \"\" (2nd value) \"R\" (3rd value)","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"class mic random_mic() (see .mic()) class disk random_disk() (see .disk())","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"base R function sample() used generating values. Generated values based EUCAST 2024 guideline implemented clinical_breakpoints data set. create specific generated values per bug drug, set mo /ab argument.","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"","code":"random_mic(25) #> Class 'mic' #>  [1] 0.001  0.005  0.0625 0.001  0.001  2      >=256  128    0.005  0.002  #> [11] 1      0.002  32     16     64     0.005  0.005  0.025  32     0.5    #> [21] 0.25   0.005  8      >=256  0.01   random_disk(25) #> Class 'disk' #>  [1] 16  6 43 36 43 50 45 43 25 34 37  9 25 46 39 40 45 47 17  9 22 45 14 47 41 random_sir(25) #> Class 'sir' #>  [1] S S S R R I S R S S I S R S R I I R I I R I S S S  # \\donttest{ # make the random generation more realistic by setting a bug and/or drug: random_mic(25, \"Klebsiella pneumoniae\") # range 0.0625-64 #> Class 'mic' #>  [1] 0.125 4     0.01  >=256 0.25  0.25  1     0.002 0.25  2     0.005 0.001 #> [13] 64    0.001 0.025 0.025 64    128   0.5   16    2     0.001 0.125 8     #> [25] 0.025 random_mic(25, \"Klebsiella pneumoniae\", \"meropenem\") # range 0.0625-16 #> Class 'mic' #>  [1] >=32  16    8     2     8     8     1     1     8     >=32  16    <=0.5 #> [13] 2     >=32  4     >=32  >=32  >=32  4     8     2     16    4     >=32  #> [25] 2     random_mic(25, \"Streptococcus pneumoniae\", \"meropenem\") # range 0.0625-4 #> Class 'mic' #>  [1] 2      0.125  0.125  0.25   1      1      0.0625 0.5    1      0.5    #> [11] 4      0.125  0.25   0.25   0.0625 4      0.25   0.125  4      1      #> [21] 4      0.5    0.0625 4      0.125   random_disk(25, \"Klebsiella pneumoniae\") # range 8-50 #> Class 'disk' #>  [1] 47 49 10 40 23 16 20 34 45  8 42 10 20 11 30 36 40 35 14 24  8 38 10 10 15 random_disk(25, \"Klebsiella pneumoniae\", \"ampicillin\") # range 11-17 #> Class 'disk' #>  [1] 14 12 14 11 13 15 13 11 17 15 14 13 11 15 17 14 16 17 17 17 11 12 12 12 14 random_disk(25, \"Streptococcus pneumoniae\", \"ampicillin\") # range 12-27 #> Class 'disk' #>  [1] 27 15 17 23 23 18 18 17 15 18 27 15 25 16 21 15 24 23 22 22 15 23 25 23 24 # }"},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":null,"dir":"Reference","previous_headings":"","what":"Predict Antimicrobial Resistance — resistance_predict","title":"Predict Antimicrobial Resistance — resistance_predict","text":"Create prediction model predict antimicrobial resistance next years statistical solid ground. Standard errors (SE) returned columns se_min se_max. See Examples real live example.","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Predict Antimicrobial Resistance — resistance_predict","text":"","code":"resistance_predict(x, col_ab, col_date = NULL, year_min = NULL,   year_max = NULL, year_every = 1, minimum = 30, model = NULL,   I_as_S = TRUE, preserve_measurements = TRUE, info = interactive(), ...)  sir_predict(x, col_ab, col_date = NULL, year_min = NULL, year_max = NULL,   year_every = 1, minimum = 30, model = NULL, I_as_S = TRUE,   preserve_measurements = TRUE, info = interactive(), ...)  # S3 method for class 'resistance_predict' plot(x, main = paste(\"Resistance Prediction of\",   x_name), ...)  ggplot_sir_predict(x, main = paste(\"Resistance Prediction of\", x_name),   ribbon = TRUE, ...)  # S3 method for class 'resistance_predict' autoplot(object,   main = paste(\"Resistance Prediction of\", x_name), ribbon = TRUE, ...)"},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Predict Antimicrobial Resistance — resistance_predict","text":"x data.frame containing isolates. Can left blank automatic determination, see Examples. col_ab column name x containing antimicrobial interpretations (\"R\", \"\" \"S\") col_date column name date, used calculate years column consist years already - default first column date class year_min lowest year use prediction model, dafaults lowest year col_date year_max highest year use prediction model - default 10 years today year_every unit sequence lowest year found data year_max minimum minimal amount available isolates per year include. Years containing less observations estimated model. model statistical model choice. generalised linear regression model binomial distribution (.e. using glm(..., family = binomial), assuming period zero resistance followed period increasing resistance leading slowly resistance. See Details valid options. I_as_S logical indicate whether values \"\" treated \"S\" (otherwise treated \"R\"). default, TRUE, follows redefinition EUCAST interpretation (increased exposure) 2019, see section Interpretation S, R . preserve_measurements logical indicate whether predictions years actually available data overwritten original data. standard errors years NA. info logical indicate whether textual analysis printed name summary() statistical model. ... arguments passed functions main title plot ribbon logical indicate whether ribbon shown (default) error bars object model data plotted","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Predict Antimicrobial Resistance — resistance_predict","text":"data.frame extra class resistance_predict columns: year value, estimated preserve_measurements = FALSE, combination observed estimated otherwise se_min, lower bound standard error minimum 0 (standard error never go 0%) se_max upper bound standard error maximum 1 (standard error never go 100%) observations, total number available observations year, .e. \\(S + + R\\) observed, original observed resistant percentages estimated, estimated resistant percentages, calculated model Furthermore, model available attribute: attributes(x)$model, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Predict Antimicrobial Resistance — resistance_predict","text":"Valid options statistical model (argument model) : \"binomial\" \"binom\" \"logit\": generalised linear regression model binomial distribution \"loglin\" \"poisson\": generalised log-linear regression model poisson distribution \"lin\" \"linear\": linear regression model","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Predict Antimicrobial Resistance — resistance_predict","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Predict Antimicrobial Resistance — resistance_predict","text":"","code":"x <- resistance_predict(example_isolates,   col_ab = \"AMX\",   year_min = 2010,   model = \"binomial\" ) plot(x)  # \\donttest{ if (require(\"ggplot2\")) {   ggplot_sir_predict(x) }   # using dplyr: if (require(\"dplyr\")) {   x <- example_isolates %>%     filter_first_isolate() %>%     filter(mo_genus(mo) == \"Staphylococcus\") %>%     resistance_predict(\"PEN\", model = \"binomial\")   print(plot(x))    # get the model from the object   mymodel <- attributes(x)$model   summary(mymodel) }  #> NULL #>  #> Call: #> glm(formula = df_matrix ~ year, family = binomial) #>  #> Coefficients: #>             Estimate Std. Error z value Pr(>|z|) #> (Intercept) 35.76101   72.29172   0.495    0.621 #> year        -0.01720    0.03603  -0.477    0.633 #>  #> (Dispersion parameter for binomial family taken to be 1) #>  #>     Null deviance: 5.3681  on 11  degrees of freedom #> Residual deviance: 5.1408  on 10  degrees of freedom #> AIC: 50.271 #>  #> Number of Fisher Scoring iterations: 4 #>   # create nice plots with ggplot2 yourself if (require(\"dplyr\") && require(\"ggplot2\")) {   data <- example_isolates %>%     filter(mo == as.mo(\"E. coli\")) %>%     resistance_predict(       col_ab = \"AMX\",       col_date = \"date\",       model = \"binomial\",       info = FALSE,       minimum = 15     )   head(data)   autoplot(data) }  # }"},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":null,"dir":"Reference","previous_headings":"","what":"Skewness of the Sample — skewness","title":"Skewness of the Sample — skewness","text":"Skewness measure asymmetry probability distribution real-valued random variable mean. negative ('left-skewed'): left tail longer; mass distribution concentrated right histogram. positive ('right-skewed'): right tail longer; mass distribution concentrated left histogram. normal distribution skewness 0.","code":""},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Skewness of the Sample — skewness","text":"","code":"skewness(x, na.rm = FALSE)  # Default S3 method skewness(x, na.rm = FALSE)  # S3 method for class 'matrix' skewness(x, na.rm = FALSE)  # S3 method for class 'data.frame' skewness(x, na.rm = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Skewness of the Sample — skewness","text":"x vector values, matrix data.frame na.rm logical value indicating whether NA values stripped computation proceeds","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Skewness of the Sample — skewness","text":"","code":"skewness(runif(1000)) #> [1] -0.0199978"},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Filter Top n Microorganisms — top_n_microorganisms","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"function filters data set include top n microorganisms based specified property, taxonomic family genus. example, can filter data set top 3 species, species top 5 genera, top 3 species top 5 genera.","code":""},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"","code":"top_n_microorganisms(x, n, property = \"fullname\", n_for_each = NULL,   col_mo = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"x data frame containing microbial data n integer specifying maximum number unique values property include output property character string indicating microorganism property use filtering. Must one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". NULL, raw values col_mo used without transformation. n_for_each optional integer specifying maximum number rows retain value selected property. NULL, rows within top n groups included. col_mo character string indicating column x contains microorganism names codes. Defaults first column class mo. Values coerced using .mo(). ... Additional arguments passed mo_property() property NULL.","code":""},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"function useful preprocessing data creating antibiograms analyses require focused subsets microbial data. example, can filter data set include isolates top 10 species.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"","code":"# filter to the top 3 species: top_n_microorganisms(example_isolates,                      n = 3) #> # A tibble: 1,015 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  5 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  6 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  7 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #>  8 2002-02-14 067927     45 F      ICU      B_STPHY_CONS   R     NA    R     NA  #>  9 2002-02-14 067927     45 F      ICU      B_STPHY_CONS   S     NA    S     NA  #> 10 2002-02-21 A56499     64 M      Clinical B_STPHY_CONS   S     NA    S     NA  #> # ℹ 1,005 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter to any species in the top 5 genera: top_n_microorganisms(example_isolates,                      n = 5, property = \"genus\") #> # A tibble: 1,742 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,732 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter to the top 3 species in each of the top 5 genera: top_n_microorganisms(example_isolates,                      n = 5, property = \"genus\", n_for_each = 3) #> # A tibble: 1,497 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinical B_ENTRC_FCLS   NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinical B_ENTRC_FCLS   NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #>  8 2003-09-05 F35553     52 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #> 10 2003-09-28 1B0933     80 M      Clinical B_ENTRC        NA    NA    NA    NA  #> # ℹ 1,487 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":null,"dir":"Reference","previous_headings":"","what":"Translate Strings from the AMR Package — translate","title":"Translate Strings from the AMR Package — translate","text":"language-dependent output AMR functions, mo_name(), mo_gramstain(), mo_type() ab_name().","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Translate Strings from the AMR Package — translate","text":"","code":"get_AMR_locale()  set_AMR_locale(language)  reset_AMR_locale()  translate_AMR(x, language = get_AMR_locale())"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Translate Strings from the AMR Package — translate","text":"language language choose. Use one supported language names ISO-639-1 codes: English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). x text translate","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Translate Strings from the AMR Package — translate","text":"currently 20 supported languages English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). languages translations available antimicrobial drugs colloquial microorganism names. permanently silence -per-session language note non-English operating system, can set package option AMR_locale .Rprofile file like :   save file. Please read adding updating language Wiki.","code":"# Open .Rprofile file utils::file.edit(\"~/.Rprofile\")  # Then add e.g. Italian support to that file using: options(AMR_locale = \"Italian\")"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"changing-the-default-language","dir":"Reference","previous_headings":"","what":"Changing the Default Language","title":"Translate Strings from the AMR Package — translate","text":"system language used default (returned Sys.getenv(\"LANG\") , LANG set, Sys.getlocale(\"LC_COLLATE\")), language supported. language used can overwritten two ways checked order: Setting package option AMR_locale, either using e.g. set_AMR_locale(\"German\") running e.g. options(AMR_locale = \"German\"). Note setting R option works session. Save command options(AMR_locale = \"(language)\") .Rprofile file apply every session. Run utils::file.edit(\"~/.Rprofile\") edit .Rprofile file. Setting system variable LANGUAGE LANG, e.g. adding LANGUAGE=\"de_DE.utf8\" .Renviron file home directory. Thus, package option AMR_locale set, system variables LANGUAGE LANG ignored.","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Translate Strings from the AMR Package — translate","text":"","code":"# Current settings (based on system language) ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacin\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Coagulase-negative Staphylococcus (CoNS)\"  # setting another language set_AMR_locale(\"Dutch\") #> ℹ Using Dutch (Nederlands) for the AMR package for this session. ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacine\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Coagulase-negatieve Staphylococcus (CNS)\"  # setting yet another language set_AMR_locale(\"German\") #> ℹ Using German (Deutsch) for the AMR package for this session. ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacin\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Koagulase-negative Staphylococcus (KNS)\"  # set_AMR_locale() understands endonyms, English exonyms, and ISO-639-1: set_AMR_locale(\"Deutsch\") #> ℹ Using German (Deutsch) for the AMR package for this session. set_AMR_locale(\"German\") #> ℹ Using German (Deutsch) for the AMR package for this session. set_AMR_locale(\"de\") #> ℹ Using German (Deutsch) for the AMR package for this session. ab_name(\"amox/clav\") #> ℹ More than one result was found for item 1: amoxicillin and clavulansäure #> [1] \"Amoxicillin/Clavulansäure\"  # reset to system default reset_AMR_locale() #> ℹ Using the English language (English) for the AMR package for this #>   session. ab_name(\"amox/clav\") #> [1] \"Amoxicillin/clavulanic acid\""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"amr-2119153","dir":"Changelog","previous_headings":"","what":"AMR 2.1.1.9153","title":"AMR 2.1.1.9153","text":"(beta version eventually become v3.0. ’re happy reach new major milestone soon, new One Health support! Install beta using instructions .)","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"A New Milestone: AMR v3.0 with One Health Support (= Human + Veterinary + Environmental)","title":"AMR 2.1.1.9153","text":"package now supports tools AMR data analysis clinical settings, also veterinary environmental microbiology. made possible collaboration University Prince Edward Island’s Atlantic Veterinary College, Canada. celebrate great improvement package, also updated package logo reflect change.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"breaking-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"Breaking","title":"AMR 2.1.1.9153","text":"Removed functions references used deprecated rsi class, replaced sir equivalents two years ago","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"new-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"New","title":"AMR 2.1.1.9153","text":"Function .sir() now extensive support veterinary breakpoints CLSI. Use breakpoint_type = \"animal\" set host argument variable contains animal species names. CLSI VET09 guideline implemented address cases veterinary breakpoints missing (applies guideline set CLSI) clinical_breakpoints data set contains breakpoints, can downloaded download page. antibiotics data set contains veterinary antibiotics, pradofloxacin enrofloxacin. WHOCC codes veterinary use added well. ab_atc() now supports ATC codes veterinary antibiotics (start “Q”) ab_url() now supports retrieving WHOCC url ATCvet pages antibiogram() function now supports creating true Weighted-Incidence Syndromic Combination Antibiograms (WISCA), powerful Bayesian method estimating regimen coverage probabilities using pathogen incidence antimicrobial susceptibility data. WISCA offers improved precision syndrome-specific treatment, even datasets sparse data. dedicated wisca() function also available easy usage. MycoBank now integrated primary taxonomic source fungi. microorganisms data set enriched new columns (mycobank, mycobank_parent, mycobank_renamed_to) provide detailed information fungal species. remarkable addition 20,000 new fungal records New function mo_mycobank() retrieve MycoBank record number, analogous existing functions mo_lpsn() mo_gbif(). .mo() function mo_*() functions now include only_fungi argument, allowing users restrict results solely fungal species. ensures fungi prioritised bacteria microorganism identification. can also set globally new AMR_only_fungi option. Also updated kingdoms, welcoming total 2,149 new records 2023 927 2024. EUCAST 2024 CLSI 2024 now supported, adding 4,000 new clinical breakpoints clinical_breakpoints data set usage .sir(). EUCAST 2024 now new default guideline MIC disk diffusion interpretations. .sir() now brings additional factor levels: “NI” non-interpretable “SDD” susceptible dose-dependent. Currently, clinical_breakpoints data set contains 24 breakpoints can return value “SDD” instead “”. New function group scale_*_mic(), namely: scale_x_mic(), scale_y_mic(), scale_colour_mic() scale_fill_mic(). allow easy plotting MIC values. allow manual range definition plotting missing intermediate log2 levels. New function group scale_*_sir(), namely: scale_x_sir(), scale_colour_sir() scale_fill_sir(). allow plot sir class, translates system language default. also set colourblind-safe colours plots. New function rescale_mic(), allows users rescale MIC values manually set range. powerhouse behind scale_*_mic() functions, can used independently , instance, compare equality MIC distributions rescaling range first. using R heavy lifting, ‘AMR’ Python Package developed run AMR R package natively Python. Python package always version number R package, built automatically every code change. antimicrobial selectors (aminoglycosides() betalactams()) now supported tidymodels packages recipe parsnip. See info tutorial using AMR functions predictive modelling. New function top_n_microorganisms() filter data set top n taxonomic property, e.g., filter top 3 species, filter species top 5 genera, filter top 3 species top 5 genera New function mo_group_members() retrieve member microorganisms microorganism group. example, mo_group_members(\"Strep group C\") returns vector microorganisms belong group.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"changed-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"Changed","title":"AMR 2.1.1.9153","text":"now possible use column names argument ab, mo, uti: .sir(..., ab = \"column1\", mo = \"column2\", uti = \"column3\"). greatly improves flexibility users. Users can now set criteria (using regular expressions) considered S, , R, SDD, NI. get quantitative values, .double() sir object return 1 S, 2 SDD/, 3 R (NI become NA). functions using sir classes (e.g., summary()) updated reflect change contain NI SDD. Fix conserve_capped_values, now works expected: MIC values, <x always S, >x always R New argument formatting_type set 22 options formatting ‘cells’. defaults 10 non-WISCA 14 WISCA, changing output antibiograms cells info. reason, add_total_n now FALSE default since denominators added cells ab_transform argument now defaults \"name\", displaying antibiotic column names instead codes ‘Antibiotic selectors’ now called ‘antimicrobial selectors’ since scope broader just antibiotics. documentation updated, ab_class() ab_selector() replaced amr_class() amr_selector(). old functions now deprecated removed future version. Added selectors isoxazolylpenicillins(), monobactams(), nitrofurans(), phenicols(), rifamycins() using antimicrobial selectors exclude non-treatable drugs (gentamicin-high using aminoglycosides()), function now always returns warning can included using only_treatable = FALSE Added new argument return_all selectors, defaults TRUE include match. FALSE, old behaviour, first hit unique antimicrobial returned. selectors can now run separate command retrieve vector possible antimicrobials selector can select selectors lincosamides() macrolides() overlap anymore - antibiotic now classified either Added “clindamycin inducible screening” CLI1. Since clindamycin lincosamide, antimicrobial selector lincosamides() now contains argument only_treatable = TRUE (similar antibiotic selectors contain non-treatable drugs) Added Amorolfine (AMO, D01AE16), now also part antifungals() selector Added Efflux (EFF), allow mapping AMRFinderPlus Added Tigemonam (TNM), monobactam Added valid levels: 4096, 6 powers 0.0625, 5 powers 192 (192, 384, 576, 768, 960) Fixed bug .mic() failed translation scientifically formatted numbers Added new argument keep_operators .mic(). can \"\" (default), \"none\", \"edges\". argument also available new rescale_mic() scale_*_mic() functions. Comparisons MIC values now strict. example, >32 higher (never equal ) 32. Thus, .mic(\">32\") == .mic(32) now returns FALSE, .mic(\">32\") > .mic(32) now returns TRUE. Sorting MIC values (using sort()) fixed manner; <0.001 now gets sorted 0.001, >0.001 gets sorted 0.001. Intermediate log2 levels used MIC plotting now common values instead following strict dilution range eucast_rules() now argument overwrite (default: TRUE) indicate whether non-NA values overwritten Disks 0 5 mm now allowed, newly allowed range disk diffusion (.disk()) now 0 50 mm Updated italicise_taxonomy() support HTML output custom_eucast_rules() now supports multiple antibiotics antibiotic groups affected single rule mo_info() now contains extra element rank group_members (contents new mo_group_members() function) Updated ATC codes WHOCC Updated antibiotic DDDs WHOCC Added 1,500 trade names antibiotics Fix using manual value mo_transform antibiogram() Fixed bug antibiogram() returns empty data set Fix mapping ‘high level’ antibiotics .ab() (amphotericin B-high, gentamicin-high, kanamycin-high, streptomycin-high, tobramycin-high) Improved overall algorithm .ab() better performance accuracy weight given genus species combinations cases subspecies miswritten, result correct genus species Genera World Health Organization’s () Priority Pathogen List now highest prevalence Fixed bug sir_confidence_interval() isolates available Updated prevalence calculation include genera World Health Organization’s () Priority Pathogen List Improved algorithm first_isolate() using phenotype-based method, prioritise records highest availability SIR values scale_y_percent() can now cope ranges outside 0-100% range Implemented new Dutch national MDRO guideline (SRI-richtlijn BRMO, Nov 2024) Added arguments esbl, carbapenemase, mecA, mecC, vanA, vanB denote column names logical values indicating presence genes (production proteins) Added console colours support sir class Positron","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"other-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"Other","title":"AMR 2.1.1.9153","text":"Added Dr. Larisse Bolton contributor fantastic implementation WISCA mathematically solid way Added Matthew Saab, Dr. Jordan Stull, Prof. Javier Sanchez contributors tremendous input veterinary breakpoints interpretations Greatly improved vctrs integration, Tidyverse package working background many Tidyverse functions. users, means functions dplyr’s bind_rows(), rowwise() c_across() now supported e.g. columns class mic. Despite , AMR package still zero-dependent package, including dplyr vctrs. Greatly updated expanded documentation Stopped support SAS (.xpt) files, since file structure extremely inefficient requires disk space GitHub allows single commit.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"older-versions-2-1-1-9153","dir":"Changelog","previous_headings":"","what":"Older Versions","title":"AMR 2.1.1.9153","text":"changelog contains changes AMR v3.0 (February 2025) later. prior v2 versions, please see v2 archive. prior v1 versions, please see v1 archive.","code":""}]
+[{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"How to conduct AMR data analysis","text":"Conducting AMR data analysis unfortunately requires -depth knowledge different scientific fields, makes hard right. least, requires: Good questions (always start !) reliable data thorough understanding (clinical) epidemiology, understand clinical epidemiological relevance possible bias results thorough understanding (clinical) microbiology/infectious diseases, understand microorganisms causal infections implications pharmaceutical treatment, well understanding intrinsic acquired microbial resistance Experience data analysis microbiological tests results, understand determination limitations MIC values interpretations SIR values Availability biological taxonomy microorganisms probably normalisation factors pharmaceuticals, defined daily doses (DDD) Available (inter-)national guidelines, profound methods apply course, instantly provide knowledge experience. AMR package, aimed providing (1) tools simplify antimicrobial resistance data cleaning, transformation analysis, (2) methods easily incorporate international guidelines (3) scientifically reliable reference data, including requirements mentioned . AMR package enables standardised reproducible AMR data analysis, application evidence-based rules, determination first isolates, translation various codes microorganisms antimicrobial agents, determination (multi-drug) resistant microorganisms, calculation antimicrobial resistance, prevalence future trends.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"preparation","dir":"Articles","previous_headings":"","what":"Preparation","title":"How to conduct AMR data analysis","text":"tutorial, create fake demonstration data work . can skip Cleaning data already data ready. start analysis, try make structure data generally look like :","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"needed-r-packages","dir":"Articles","previous_headings":"Preparation","what":"Needed R packages","title":"How to conduct AMR data analysis","text":"many uses R, need additional packages AMR data analysis. package works closely together tidyverse packages dplyr ggplot2 RStudio. tidyverse tremendously improves way conduct data science - allows natural way writing syntaxes creating beautiful plots R. also use cleaner package, can used cleaning data creating frequency tables. AMR package contains data set example_isolates_unclean, might look data users extracted laboratory systems: AMR data analysis, like microorganism column contain valid, --date taxonomy, antibiotic columns cleaned SIR values well.","code":"library(dplyr) library(ggplot2) library(AMR)  # (if not yet installed, install with:) # install.packages(c(\"dplyr\", \"ggplot2\", \"AMR\")) example_isolates_unclean #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria      AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <chr>         <chr> <chr> <chr> <chr> #>  1 J3         A        2012-11-21 E. coli       R     I     S     S     #>  2 R7         A        2018-04-03 K. pneumoniae R     I     S     S     #>  3 P3         A        2014-09-19 E. coli       R     S     S     S     #>  4 P10        A        2015-12-10 E. coli       S     I     S     S     #>  5 B7         A        2015-03-02 E. coli       S     S     S     S     #>  6 W3         A        2018-03-31 S. aureus     R     S     R     S     #>  7 J8         A        2016-06-14 E. coli       R     S     S     S     #>  8 M3         A        2015-10-25 E. coli       R     S     S     S     #>  9 J3         A        2019-06-19 E. coli       S     S     S     S     #> 10 G6         A        2015-04-27 S. aureus     S     S     S     S     #> # ℹ 2,990 more rows  # we will use 'our_data' as the data set name for this tutorial our_data <- example_isolates_unclean"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"taxonomy-of-microorganisms","dir":"Articles","previous_headings":"Preparation","what":"Taxonomy of microorganisms","title":"How to conduct AMR data analysis","text":".mo(), users can transform arbitrary microorganism names codes current taxonomy. AMR package contains --date taxonomic data. specific, currently included data retrieved 24 Jun 2024. codes AMR packages come .mo() short, still human readable. importantly, .mo() supports kinds input: first character codes denote taxonomic kingdom, Bacteria (B), Fungi (F), Protozoa (P). AMR package also contain functions directly retrieve taxonomic properties, name, genus, species, family, order, even Gram-stain. start mo_ use .mo() internally, still arbitrary user input can used: Now can thus clean data: Apparently, uncertainty translation taxonomic codes. Let’s check : ’s good.","code":"as.mo(\"Klebsiella pneumoniae\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"K. pneumoniae\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"KLEPNE\") #> Class 'mo' #> [1] B_KLBSL_PNMN as.mo(\"KLPN\") #> Class 'mo' #> [1] B_KLBSL_PNMN mo_family(\"K. pneumoniae\") #> [1] \"Enterobacteriaceae\" mo_genus(\"K. pneumoniae\") #> [1] \"Klebsiella\" mo_species(\"K. pneumoniae\") #> [1] \"pneumoniae\"  mo_gramstain(\"Klebsiella pneumoniae\") #> [1] \"Gram-negative\"  mo_ref(\"K. pneumoniae\") #> [1] \"Trevisan, 1887\"  mo_snomed(\"K. pneumoniae\") #> [[1]] #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\" our_data$bacteria <- as.mo(our_data$bacteria, info = TRUE) #> ℹ Microorganism translation was uncertain for four microorganisms. Run #>   mo_uncertainties() to review these uncertainties, or use #>   add_custom_microorganisms() to add custom entries. mo_uncertainties() #> Matching scores are based on the resemblance between the input and the full #> taxonomic name, and the pathogenicity in humans. See ?mo_matching_score. #> Colour keys:  0.000-0.549  0.550-0.649  0.650-0.749  0.750-1.000  #>  #> -------------------------------------------------------------------------------- #> \"E. coli\" -> Escherichia coli (B_ESCHR_COLI, 0.688) #> Also matched: Enterococcus crotali (0.650), Escherichia coli coli #>               (0.643), Escherichia coli expressing (0.611), Enterobacter cowanii #>               (0.600), Enterococcus columbae (0.595), Enterococcus camelliae (0.591), #>               Enterococcus casseliflavus (0.577), Enterobacter cloacae cloacae #>               (0.571), Enterobacter cloacae complex (0.571), and Enterobacter cloacae #>               dissolvens (0.565) #> -------------------------------------------------------------------------------- #> \"K. pneumoniae\" -> Klebsiella pneumoniae (B_KLBSL_PNMN, 0.786) #> Also matched: Klebsiella pneumoniae ozaenae (0.707), Klebsiella #>               pneumoniae pneumoniae (0.688), Klebsiella pneumoniae rhinoscleromatis #>               (0.658), Klebsiella pasteurii (0.500), Klebsiella planticola (0.500), #>               Kingella potus (0.400), Kluyveromyces pseudotropicale (0.386), #>               Kluyveromyces pseudotropicalis (0.363), Kosakonia pseudosacchari #>               (0.361), and Kluyveromyces pseudotropicalis pseudotropicalis (0.361) #> -------------------------------------------------------------------------------- #> \"S. aureus\" -> Staphylococcus aureus (B_STPHY_AURS, 0.690) #> Also matched: Staphylococcus aureus aureus (0.643), Staphylococcus #>               argenteus (0.625), Staphylococcus aureus anaerobius (0.625), #>               Staphylococcus auricularis (0.615), Salmonella Aurelianis (0.595), #>               Salmonella Aarhus (0.588), Salmonella Amounderness (0.587), #>               Staphylococcus argensis (0.587), Streptococcus australis (0.587), and #>               Salmonella choleraesuis arizonae (0.562) #> -------------------------------------------------------------------------------- #> \"S. pneumoniae\" -> Streptococcus pneumoniae (B_STRPT_PNMN, 0.750) #> Also matched: Streptococcus pseudopneumoniae (0.700), Streptococcus #>               phocae salmonis (0.552), Serratia proteamaculans quinovora (0.545), #>               Streptococcus pseudoporcinus (0.536), Staphylococcus piscifermentans #>               (0.533), Staphylococcus pseudintermedius (0.532), Serratia #>               proteamaculans proteamaculans (0.526), Streptococcus gallolyticus #>               pasteurianus (0.526), Salmonella Portanigra (0.524), and Streptococcus #>               periodonticum (0.519) #>  #> Only the first 10 other matches of each record are shown. Run #> print(mo_uncertainties(), n = ...) to view more entries, or save #> mo_uncertainties() to an object."},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"antibiotic-results","dir":"Articles","previous_headings":"Preparation","what":"Antibiotic results","title":"How to conduct AMR data analysis","text":"column antibiotic test results must also cleaned. AMR package comes three new data types work test results: mic minimal inhibitory concentrations (MIC), disk disk diffusion diameters, sir SIR data interpreted already. package can also determine SIR values based MIC disk diffusion values, read .sir() page. now, just clean SIR columns data using dplyr: basically cleaning, time start data inclusion.","code":"# method 1, be explicit about the columns: our_data <- our_data %>%   mutate_at(vars(AMX:GEN), as.sir)  # method 2, let the AMR package determine the eligible columns our_data <- our_data %>%   mutate_if(is_sir_eligible, as.sir)  # result: our_data #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> #>  1 J3         A        2012-11-21 B_ESCHR_COLI   R     I     S     S   #>  2 R7         A        2018-04-03 B_KLBSL_PNMN   R     I     S     S   #>  3 P3         A        2014-09-19 B_ESCHR_COLI   R     S     S     S   #>  4 P10        A        2015-12-10 B_ESCHR_COLI   S     I     S     S   #>  5 B7         A        2015-03-02 B_ESCHR_COLI   S     S     S     S   #>  6 W3         A        2018-03-31 B_STPHY_AURS   R     S     R     S   #>  7 J8         A        2016-06-14 B_ESCHR_COLI   R     S     S     S   #>  8 M3         A        2015-10-25 B_ESCHR_COLI   R     S     S     S   #>  9 J3         A        2019-06-19 B_ESCHR_COLI   S     S     S     S   #> 10 G6         A        2015-04-27 B_STPHY_AURS   S     S     S     S   #> # ℹ 2,990 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"first-isolates","dir":"Articles","previous_headings":"Preparation","what":"First isolates","title":"How to conduct AMR data analysis","text":"need know isolates can actually use analysis without repetition bias. conduct analysis antimicrobial resistance, must include first isolate every patient per episode (Hindler et al., Clin Infect Dis. 2007). , easily get overestimate underestimate resistance antibiotic. Imagine patient admitted MRSA found 5 different blood cultures following weeks (yes, countries like Netherlands blood drawing policies). resistance percentage oxacillin isolates overestimated, included MRSA . clearly selection bias. Clinical Laboratory Standards Institute (CLSI) appoints follows: (…) preparing cumulative antibiogram guide clinical decisions empirical antimicrobial therapy initial infections, first isolate given species per patient, per analysis period (eg, one year) included, irrespective body site, antimicrobial susceptibility profile, phenotypical characteristics (eg, biotype). first isolate easily identified, cumulative antimicrobial susceptibility test data prepared using first isolate generally comparable cumulative antimicrobial susceptibility test data calculated methods, providing duplicate isolates excluded. M39-A4 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 4th Edition. CLSI, 2014. Chapter 6.4 AMR package includes methodology first_isolate() function able apply four different methods defined Hindler et al. 2007: phenotype-based, episode-based, patient-based, isolate-based. right method depends goals analysis, default phenotype-based method case method properly correct duplicate isolates. Read methods first_isolate() page. outcome function can easily added data: 91% suitable resistance analysis! can now filter filter() function, also dplyr package: future use, two syntaxes can shortened: end 2 724 isolates analysis. Now data looks like: Time analysis.","code":"our_data <- our_data %>%   mutate(first = first_isolate(info = TRUE)) #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Using column 'bacteria' as input for col_mo. #> ℹ Using column 'date' as input for col_date. #> ℹ Using column 'patient_id' as input for col_patient_id. #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> => Found 2,724 'phenotype-based' first isolates (90.8% of total where a #>    microbial ID was available) our_data_1st <- our_data %>%   filter(first == TRUE) our_data_1st <- our_data %>%   filter_first_isolate() our_data_1st #> # A tibble: 2,724 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 J3         A        2012-11-21 B_ESCHR_COLI   R     I     S     S   TRUE  #>  2 R7         A        2018-04-03 B_KLBSL_PNMN   R     I     S     S   TRUE  #>  3 P3         A        2014-09-19 B_ESCHR_COLI   R     S     S     S   TRUE  #>  4 P10        A        2015-12-10 B_ESCHR_COLI   S     I     S     S   TRUE  #>  5 B7         A        2015-03-02 B_ESCHR_COLI   S     S     S     S   TRUE  #>  6 W3         A        2018-03-31 B_STPHY_AURS   R     S     R     S   TRUE  #>  7 M3         A        2015-10-25 B_ESCHR_COLI   R     S     S     S   TRUE  #>  8 J3         A        2019-06-19 B_ESCHR_COLI   S     S     S     S   TRUE  #>  9 G6         A        2015-04-27 B_STPHY_AURS   S     S     S     S   TRUE  #> 10 P4         A        2011-06-21 B_ESCHR_COLI   S     S     S     S   TRUE  #> # ℹ 2,714 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"analysing-the-data","dir":"Articles","previous_headings":"","what":"Analysing the data","title":"How to conduct AMR data analysis","text":"base R summary() function gives good first impression, comes support new mo sir classes now data set:","code":"summary(our_data_1st) #>   patient_id          hospital              date            #>  Length:2724        Length:2724        Min.   :2011-01-01   #>  Class :character   Class :character   1st Qu.:2013-04-07   #>  Mode  :character   Mode  :character   Median :2015-06-03   #>                                        Mean   :2015-06-09   #>                                        3rd Qu.:2017-08-11   #>                                        Max.   :2019-12-27   #>    bacteria               AMX                    AMC                 #>  Class :mo             Class:sir              Class:sir              #>  <NA>  :0              %S   :41.6% (n=1133)   %S   :52.6% (n=1432)   #>  Unique:4              %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)      #>  #1    :B_ESCHR_COLI   %I   :16.4% (n=446)    %I   :12.2% (n=333)    #>  #2    :B_STPHY_AURS   %R   :42.0% (n=1145)   %R   :35.2% (n=959)    #>  #3    :B_STRPT_PNMN   %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)      #>     CIP                    GEN                  first         #>  Class:sir              Class:sir              Mode:logical   #>  %S   :52.5% (n=1431)   %S   :61.0% (n=1661)   TRUE:2724      #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)                     #>  %I   : 6.5% (n=176)    %I   : 3.0% (n=82)                    #>  %R   :41.0% (n=1117)   %R   :36.0% (n=981)                   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)  glimpse(our_data_1st) #> Rows: 2,724 #> Columns: 9 #> $ patient_id <chr> \"J3\", \"R7\", \"P3\", \"P10\", \"B7\", \"W3\", \"M3\", \"J3\", \"G6\", \"P4\"… #> $ hospital   <chr> \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\", \"A\",… #> $ date       <date> 2012-11-21, 2018-04-03, 2014-09-19, 2015-12-10, 2015-03-02… #> $ bacteria   <mo> \"B_ESCHR_COLI\", \"B_KLBSL_PNMN\", \"B_ESCHR_COLI\", \"B_ESCHR_COL… #> $ AMX        <sir> R, R, R, S, S, R, R, S, S, S, S, R, S, S, R, R, R, R, S, R,… #> $ AMC        <sir> I, I, S, I, S, S, S, S, S, S, S, S, S, S, S, S, S, R, S, S,… #> $ CIP        <sir> S, S, S, S, S, R, S, S, S, S, S, S, S, S, S, S, S, S, S, S,… #> $ GEN        <sir> S, S, S, S, S, S, S, S, S, S, S, R, S, S, S, S, S, S, S, S,… #> $ first      <lgl> TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE,…  # number of unique values per column: sapply(our_data_1st, n_distinct) #> patient_id   hospital       date   bacteria        AMX        AMC        CIP  #>        260          3       1854          4          3          3          3  #>        GEN      first  #>          3          1"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"availability-of-species","dir":"Articles","previous_headings":"Analysing the data","what":"Availability of species","title":"How to conduct AMR data analysis","text":"just get idea species distributed, create frequency table count() based name microorganisms:","code":"our_data %>%   count(mo_name(bacteria), sort = TRUE) #> # A tibble: 4 × 2 #>   `mo_name(bacteria)`          n #>   <chr>                    <int> #> 1 Escherichia coli          1518 #> 2 Staphylococcus aureus      730 #> 3 Streptococcus pneumoniae   426 #> 4 Klebsiella pneumoniae      326  our_data_1st %>%   count(mo_name(bacteria), sort = TRUE) #> # A tibble: 4 × 2 #>   `mo_name(bacteria)`          n #>   <chr>                    <int> #> 1 Escherichia coli          1321 #> 2 Staphylococcus aureus      682 #> 3 Streptococcus pneumoniae   402 #> 4 Klebsiella pneumoniae      319"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"select-and-filter-with-antibiotic-selectors","dir":"Articles","previous_headings":"Analysing the data","what":"Select and filter with antibiotic selectors","title":"How to conduct AMR data analysis","text":"Using -called antibiotic class selectors, can select filter columns based antibiotic class antibiotic results :","code":"our_data_1st %>%   select(date, aminoglycosides()) #> ℹ For aminoglycosides() using column 'GEN' (gentamicin) #> # A tibble: 2,724 × 2 #>    date       GEN   #>    <date>     <sir> #>  1 2012-11-21   S   #>  2 2018-04-03   S   #>  3 2014-09-19   S   #>  4 2015-12-10   S   #>  5 2015-03-02   S   #>  6 2018-03-31   S   #>  7 2015-10-25   S   #>  8 2019-06-19   S   #>  9 2015-04-27   S   #> 10 2011-06-21   S   #> # ℹ 2,714 more rows  our_data_1st %>%   select(bacteria, betalactams()) #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 2,724 × 3 #>    bacteria     AMX   AMC   #>    <mo>         <sir> <sir> #>  1 B_ESCHR_COLI   R     I   #>  2 B_KLBSL_PNMN   R     I   #>  3 B_ESCHR_COLI   R     S   #>  4 B_ESCHR_COLI   S     I   #>  5 B_ESCHR_COLI   S     S   #>  6 B_STPHY_AURS   R     S   #>  7 B_ESCHR_COLI   R     S   #>  8 B_ESCHR_COLI   S     S   #>  9 B_STPHY_AURS   S     S   #> 10 B_ESCHR_COLI   S     S   #> # ℹ 2,714 more rows  our_data_1st %>%   select(bacteria, where(is.sir)) #> # A tibble: 2,724 × 5 #>    bacteria     AMX   AMC   CIP   GEN   #>    <mo>         <sir> <sir> <sir> <sir> #>  1 B_ESCHR_COLI   R     I     S     S   #>  2 B_KLBSL_PNMN   R     I     S     S   #>  3 B_ESCHR_COLI   R     S     S     S   #>  4 B_ESCHR_COLI   S     I     S     S   #>  5 B_ESCHR_COLI   S     S     S     S   #>  6 B_STPHY_AURS   R     S     R     S   #>  7 B_ESCHR_COLI   R     S     S     S   #>  8 B_ESCHR_COLI   S     S     S     S   #>  9 B_STPHY_AURS   S     S     S     S   #> 10 B_ESCHR_COLI   S     S     S     S   #> # ℹ 2,714 more rows  # filtering using AB selectors is also possible: our_data_1st %>%   filter(any(aminoglycosides() == \"R\")) #> ℹ For aminoglycosides() using column 'GEN' (gentamicin) #> # A tibble: 981 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 J5         A        2017-12-25 B_STRPT_PNMN   R     S     S     R   TRUE  #>  2 X1         A        2017-07-04 B_STPHY_AURS   R     S     S     R   TRUE  #>  3 B3         A        2016-07-24 B_ESCHR_COLI   S     S     S     R   TRUE  #>  4 V7         A        2012-04-03 B_ESCHR_COLI   S     S     S     R   TRUE  #>  5 C9         A        2017-03-23 B_ESCHR_COLI   S     S     S     R   TRUE  #>  6 R1         A        2018-06-10 B_STPHY_AURS   S     S     S     R   TRUE  #>  7 S2         A        2013-07-19 B_STRPT_PNMN   S     S     S     R   TRUE  #>  8 P5         A        2019-03-09 B_STPHY_AURS   S     S     S     R   TRUE  #>  9 Q8         A        2019-08-10 B_STPHY_AURS   S     S     S     R   TRUE  #> 10 K5         A        2013-03-15 B_STRPT_PNMN   S     S     S     R   TRUE  #> # ℹ 971 more rows  our_data_1st %>%   filter(all(betalactams() == \"R\")) #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 462 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 M7         A        2013-07-22 B_STRPT_PNMN   R     R     S     S   TRUE  #>  2 R10        A        2013-12-20 B_STPHY_AURS   R     R     S     S   TRUE  #>  3 R7         A        2015-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  4 R8         A        2019-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  5 B6         A        2016-11-20 B_ESCHR_COLI   R     R     R     R   TRUE  #>  6 I7         A        2015-08-19 B_ESCHR_COLI   R     R     S     S   TRUE  #>  7 N3         A        2014-12-29 B_STRPT_PNMN   R     R     R     S   TRUE  #>  8 Q2         A        2019-09-22 B_ESCHR_COLI   R     R     S     S   TRUE  #>  9 X7         A        2011-03-20 B_ESCHR_COLI   R     R     S     R   TRUE  #> 10 V1         A        2018-08-07 B_STPHY_AURS   R     R     S     S   TRUE  #> # ℹ 452 more rows  # even works in base R (since R 3.0): our_data_1st[all(betalactams() == \"R\"), ] #> ℹ For betalactams() using columns 'AMX' (amoxicillin) and 'AMC' #>   (amoxicillin/clavulanic acid) #> # A tibble: 462 × 9 #>    patient_id hospital date       bacteria     AMX   AMC   CIP   GEN   first #>    <chr>      <chr>    <date>     <mo>         <sir> <sir> <sir> <sir> <lgl> #>  1 M7         A        2013-07-22 B_STRPT_PNMN   R     R     S     S   TRUE  #>  2 R10        A        2013-12-20 B_STPHY_AURS   R     R     S     S   TRUE  #>  3 R7         A        2015-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  4 R8         A        2019-10-25 B_STPHY_AURS   R     R     S     S   TRUE  #>  5 B6         A        2016-11-20 B_ESCHR_COLI   R     R     R     R   TRUE  #>  6 I7         A        2015-08-19 B_ESCHR_COLI   R     R     S     S   TRUE  #>  7 N3         A        2014-12-29 B_STRPT_PNMN   R     R     R     S   TRUE  #>  8 Q2         A        2019-09-22 B_ESCHR_COLI   R     R     S     S   TRUE  #>  9 X7         A        2011-03-20 B_ESCHR_COLI   R     R     S     R   TRUE  #> 10 V1         A        2018-08-07 B_STPHY_AURS   R     R     S     S   TRUE  #> # ℹ 452 more rows"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"generate-antibiograms","dir":"Articles","previous_headings":"Analysing the data","what":"Generate antibiograms","title":"How to conduct AMR data analysis","text":"Since AMR v2.0 (March 2023), easy create different types antibiograms, support 20 different languages. four antibiogram types, proposed Klinker et al. (2021, DOI 10.1177/20499361211011373), supported new antibiogram() function: Traditional Antibiogram (TA) e.g, susceptibility Pseudomonas aeruginosa piperacillin/tazobactam (TZP) Combination Antibiogram (CA) e.g, sdditional susceptibility Pseudomonas aeruginosa TZP + tobramycin versus TZP alone Syndromic Antibiogram (SA) e.g, susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) Weighted-Incidence Syndromic Combination Antibiogram (WISCA) e.g, susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) male patients age >=65 years heart failure section, show use antibiogram() function create antibiogram types. starters, included example_isolates data set looks like:","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"traditional-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Traditional Antibiogram","title":"How to conduct AMR data analysis","text":"create traditional antibiogram, simply state antibiotics used. antibiotics argument antibiogram() function supports (combination) previously mentioned antibiotic class selectors: Notice antibiogram() function automatically prints right format using Quarto R Markdown (page), even applies italics taxonomic names (using italicise_taxonomy() internally). also uses language OS either English, Chinese, Czech, Danish, Dutch, Finnish, French, German, Greek, Italian, Japanese, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Turkish, Ukrainian. next example, force language Spanish using language argument:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems())) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) antibiogram(example_isolates,             mo_transform = \"gramstain\",             antibiotics = aminoglycosides(),             ab_transform = \"name\",             language = \"es\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"combined-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Combined Antibiogram","title":"How to conduct AMR data analysis","text":"create combined antibiogram, use antibiotic codes names plus + character like :","code":"antibiogram(example_isolates,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"syndromic-antibiogram","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Syndromic Antibiogram","title":"How to conduct AMR data analysis","text":"create syndromic antibiogram, syndromic_group argument must used. can column data, e.g. ifelse() calculations based certain columns:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems()),             syndromic_group = \"ward\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"weighted-incidence-syndromic-combination-antibiogram-wisca","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Weighted-Incidence Syndromic Combination Antibiogram (WISCA)","title":"How to conduct AMR data analysis","text":"create WISCA, must state combination therapy antibiotics argument (similar Combination Antibiogram), define syndromic group syndromic_group argument (similar Syndromic Antibiogram) cases predefined based clinical demographic characteristics (e.g., endocarditis 75+ females). next example simplification without clinical characteristics, just gives idea WISCA can created:","code":"wisca <- antibiogram(example_isolates,                      antibiotics = c(\"AMC\", \"AMC+CIP\", \"TZP\", \"TZP+TOB\"),                      mo_transform = \"gramstain\",                      minimum = 10, # this should be >= 30, but now just as example                      syndromic_group = ifelse(example_isolates$age >= 65 &                                                 example_isolates$gender == \"M\",                                               \"WISCA Group 1\", \"WISCA Group 2\")) wisca"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"plotting-antibiograms","dir":"Articles","previous_headings":"Analysing the data > Generate antibiograms","what":"Plotting antibiograms","title":"How to conduct AMR data analysis","text":"Antibiograms can plotted using autoplot() ggplot2 packages, since AMR package provides extension function:  calculate antimicrobial resistance sensible way, also correcting results, use resistance() susceptibility() functions.","code":"autoplot(wisca)"},{"path":"https://msberends.github.io/AMR/articles/AMR.html","id":"resistance-percentages","dir":"Articles","previous_headings":"Analysing the data","what":"Resistance percentages","title":"How to conduct AMR data analysis","text":"functions resistance() susceptibility() can used calculate antimicrobial resistance susceptibility. specific analyses, functions proportion_S(), proportion_SI(), proportion_I(), proportion_IR() proportion_R() can used determine proportion specific antimicrobial outcome. functions contain minimum argument, denoting minimum required number test results returning value. functions otherwise return NA. default minimum = 30, following CLSI M39-A4 guideline applying microbial epidemiology. per EUCAST guideline 2019, calculate resistance proportion R (proportion_R(), equal resistance()) susceptibility proportion S (proportion_SI(), equal susceptibility()). functions can used : can used conjunction group_by() summarise(), dplyr package: Author: Dr. Matthijs Berends, 26th Feb 2023","code":"our_data_1st %>% resistance(AMX) #> [1] 0.4203377 our_data_1st %>%   group_by(hospital) %>%   summarise(amoxicillin = resistance(AMX)) #> # A tibble: 3 × 2 #>   hospital amoxicillin #>   <chr>          <dbl> #> 1 A              0.340 #> 2 B              0.551 #> 3 C              0.370"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"AMR for Python","text":"AMR package R powerful tool antimicrobial resistance (AMR) analysis. provides extensive features handling microbial antimicrobial data. However, work primarily Python, now intuitive option available: AMR Python Package Index. Python package wrapper round AMR R package. uses rpy2 package internally. Despite need R installed, Python users can now easily work AMR data directly Python code.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"install","dir":"Articles","previous_headings":"","what":"Install","title":"AMR for Python","text":"Since Python package available official Python Package Index, can just run: Make sure R installed. need install AMR R package, installed automatically. Linux: macOS (using Homebrew): Windows, visit CRAN download page download install R.","code":"pip install AMR # Ubuntu / Debian sudo apt install r-base # Fedora: sudo dnf install R # CentOS/RHEL sudo yum install R brew install r"},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"cleaning-taxonomy","dir":"Articles","previous_headings":"Examples of Usage","what":"Cleaning Taxonomy","title":"AMR for Python","text":"’s example demonstrates clean microorganism drug names using AMR Python package:","code":"import pandas as pd import AMR  # Sample data data = {     \"MOs\": ['E. coli', 'ESCCOL', 'esco', 'Esche coli'],     \"Drug\": ['Cipro', 'CIP', 'J01MA02', 'Ciproxin'] } df = pd.DataFrame(data)  # Use AMR functions to clean microorganism and drug names df['MO_clean'] = AMR.mo_name(df['MOs']) df['Drug_clean'] = AMR.ab_name(df['Drug'])  # Display the results print(df)"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"explanation","dir":"Articles","previous_headings":"Examples of Usage > Cleaning Taxonomy","what":"Explanation","title":"AMR for Python","text":"mo_name: function standardises microorganism names. , different variations Escherichia coli (“E. coli”, “ESCCOL”, “esco”, “Esche coli”) converted correct, standardised form, “Escherichia coli”. ab_name: Similarly, function standardises antimicrobial names. different representations ciprofloxacin (e.g., “Cipro”, “CIP”, “J01MA02”, “Ciproxin”) converted standard name, “Ciprofloxacin”.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"calculating-amr","dir":"Articles","previous_headings":"Examples of Usage","what":"Calculating AMR","title":"AMR for Python","text":"","code":"import AMR import pandas as pd  df = AMR.example_isolates result = AMR.resistance(df[\"AMX\"]) print(result) [0.59555556]"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"generating-antibiograms","dir":"Articles","previous_headings":"Examples of Usage","what":"Generating Antibiograms","title":"AMR for Python","text":"One core functions AMR package generating antibiogram, table summarises antimicrobial susceptibility bacterial isolates. ’s can generate antibiogram Python: example, generate antibiogram selecting various antibiotics.","code":"result2a = AMR.antibiogram(df[[\"mo\", \"AMX\", \"CIP\", \"TZP\"]]) print(result2a) result2b = AMR.antibiogram(df[[\"mo\", \"AMX\", \"CIP\", \"TZP\"]], mo_transform = \"gramstain\") print(result2b)"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"taxonomic-data-sets-now-in-python","dir":"Articles","previous_headings":"Examples of Usage","what":"Taxonomic Data Sets Now in Python!","title":"AMR for Python","text":"Python user, might like important data sets AMR R package, microorganisms, antibiotics, clinical_breakpoints, example_isolates, now available regular Python data frames:","code":"AMR.microorganisms AMR.antibiotics"},{"path":"https://msberends.github.io/AMR/articles/AMR_for_Python.html","id":"conclusion","dir":"Articles","previous_headings":"","what":"Conclusion","title":"AMR for Python","text":"AMR Python package, Python users can now effortlessly call R functions AMR R package. eliminates need complex rpy2 configurations provides clean, easy--use interface antimicrobial resistance analysis. examples provided demonstrate can applied typical workflows, standardising microorganism antimicrobial names calculating resistance. just running import AMR, users can seamlessly integrate robust features R AMR package Python workflows. Whether ’re cleaning data analysing resistance patterns, AMR Python package makes easy work AMR data Python.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"objective","dir":"Articles","previous_headings":"","what":"Objective","title":"AMR with tidymodels","text":"goal build predictive model using tidymodels framework determine Gramstain microorganism based microbial data. : Preprocess data using selector functions aminoglycosides() betalactams(). Define logistic regression model prediction. Use structured tidymodels workflow preprocess, train, evaluate model.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"data-preparation","dir":"Articles","previous_headings":"","what":"Data Preparation","title":"AMR with tidymodels","text":"begin loading required libraries preparing example_isolates dataset AMR package. Explanation: aminoglycosides() betalactams() dynamically select columns antibiotics classes. drop_na() ensures model receives complete cases training.","code":"# Load required libraries library(tidymodels)   # For machine learning workflows, and data manipulation (dplyr, tidyr, ...) #> ── Attaching packages ────────────────────────────────────── tidymodels 1.3.0 ── #> ✔ broom        1.0.7     ✔ recipes      1.1.1 #> ✔ dials        1.4.0     ✔ rsample      1.2.1 #> ✔ dplyr        1.1.4     ✔ tibble       3.2.1 #> ✔ ggplot2      3.5.1     ✔ tidyr        1.3.1 #> ✔ infer        1.0.7     ✔ tune         1.3.0 #> ✔ modeldata    1.4.0     ✔ workflows    1.2.0 #> ✔ parsnip      1.3.0     ✔ workflowsets 1.1.0 #> ✔ purrr        1.0.4     ✔ yardstick    1.3.2 #> ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ── #> ✖ purrr::discard() masks scales::discard() #> ✖ dplyr::filter()  masks stats::filter() #> ✖ dplyr::lag()     masks stats::lag() #> ✖ recipes::step()  masks stats::step() library(AMR)          # For AMR data analysis  # Load the example_isolates dataset data(\"example_isolates\")  # Preloaded dataset with AMR results  # Select relevant columns for prediction data <- example_isolates %>%   # select AB results dynamically   select(mo, aminoglycosides(), betalactams()) %>%   # replace NAs with NI (not-interpretable)    mutate(across(where(is.sir),                  ~replace_na(.x, \"NI\")),           # make factors of SIR columns           across(where(is.sir),                  as.integer),           # get Gramstain of microorganisms           mo = as.factor(mo_gramstain(mo))) %>%   # drop NAs - the ones without a Gramstain (fungi, etc.)   drop_na() #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For betalactams() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'TZP' #>   (piperacillin/tazobactam), 'CZO' (cefazolin), 'FEP' (cefepime), 'CXM' #>   (cefuroxime), 'FOX' (cefoxitin), 'CTX' (cefotaxime), 'CAZ' (ceftazidime), #>   'CRO' (ceftriaxone), 'IPM' (imipenem), and 'MEM' (meropenem)"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"defining-the-workflow","dir":"Articles","previous_headings":"","what":"Defining the Workflow","title":"AMR with tidymodels","text":"now define tidymodels workflow, consists three steps: preprocessing, model specification, fitting.","code":""},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"preprocessing-with-a-recipe","dir":"Articles","previous_headings":"Defining the Workflow","what":"1. Preprocessing with a Recipe","title":"AMR with tidymodels","text":"create recipe preprocess data modelling. Explanation: recipe(mo ~ ., data = data) take mo column outcome columns predictors. step_corr() removes predictors (.e., antibiotic columns) higher correlation 90%. Notice recipe contains just antibiotic selector functions - need define columns specifically.","code":"# Define the recipe for data preprocessing resistance_recipe <- recipe(mo ~ ., data = data) %>%   step_corr(c(aminoglycosides(), betalactams()), threshold = 0.9) resistance_recipe #>  #> ── Recipe ────────────────────────────────────────────────────────────────────── #>  #> ── Inputs #> Number of variables by role #> outcome:    1 #> predictor: 20 #>  #> ── Operations #> • Correlation filter on: c(aminoglycosides(), betalactams())"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"specifying-the-model","dir":"Articles","previous_headings":"Defining the Workflow","what":"2. Specifying the Model","title":"AMR with tidymodels","text":"define logistic regression model since resistance prediction binary classification task. Explanation: logistic_reg() sets logistic regression model. set_engine(\"glm\") specifies use R’s built-GLM engine.","code":"# Specify a logistic regression model logistic_model <- logistic_reg() %>%   set_engine(\"glm\") # Use the Generalized Linear Model engine logistic_model #> Logistic Regression Model Specification (classification) #>  #> Computational engine: glm"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"building-the-workflow","dir":"Articles","previous_headings":"Defining the Workflow","what":"3. Building the Workflow","title":"AMR with tidymodels","text":"bundle recipe model together workflow, organizes entire modeling process.","code":"# Combine the recipe and model into a workflow resistance_workflow <- workflow() %>%   add_recipe(resistance_recipe) %>% # Add the preprocessing recipe   add_model(logistic_model) # Add the logistic regression model"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"training-and-evaluating-the-model","dir":"Articles","previous_headings":"","what":"Training and Evaluating the Model","title":"AMR with tidymodels","text":"train model, split data training testing sets. , fit workflow training set evaluate performance. Explanation: initial_split() splits data training testing sets. fit() trains workflow training set. Notice fit(), antibiotic selector functions internally called . training, functions called since stored recipe. Next, evaluate model testing data. Explanation: predict() generates predictions testing set. metrics() computes evaluation metrics like accuracy kappa. appears can predict Gram based AMR results 0.995 accuracy based AMR results aminoglycosides beta-lactam antibiotics. ROC curve looks like :","code":"# Split data into training and testing sets set.seed(123) # For reproducibility data_split <- initial_split(data, prop = 0.8) # 80% training, 20% testing training_data <- training(data_split) # Training set testing_data <- testing(data_split)   # Testing set  # Fit the workflow to the training data fitted_workflow <- resistance_workflow %>%   fit(training_data) # Train the model #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For betalactams() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'TZP' #>   (piperacillin/tazobactam), 'CZO' (cefazolin), 'FEP' (cefepime), 'CXM' #>   (cefuroxime), 'FOX' (cefoxitin), 'CTX' (cefotaxime), 'CAZ' (ceftazidime), #>   'CRO' (ceftriaxone), 'IPM' (imipenem), and 'MEM' (meropenem) # Make predictions on the testing set predictions <- fitted_workflow %>%   predict(testing_data)                # Generate predictions probabilities <- fitted_workflow %>%   predict(testing_data, type = \"prob\") # Generate probabilities  predictions <- predictions %>%   bind_cols(probabilities) %>%   bind_cols(testing_data) # Combine with true labels  predictions #> # A tibble: 394 × 24 #>    .pred_class   `.pred_Gram-negative` `.pred_Gram-positive` mo        GEN   TOB #>    <fct>                         <dbl>                 <dbl> <fct>   <int> <int> #>  1 Gram-positive              1.07e- 1              8.93e- 1 Gram-p…     5     5 #>  2 Gram-positive              3.17e- 8              1.00e+ 0 Gram-p…     5     1 #>  3 Gram-negative              9.99e- 1              1.42e- 3 Gram-n…     5     5 #>  4 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     5     5 #>  5 Gram-negative              9.46e- 1              5.42e- 2 Gram-n…     5     5 #>  6 Gram-positive              1.07e- 1              8.93e- 1 Gram-p…     5     5 #>  7 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     1     5 #>  8 Gram-positive              2.22e-16              1   e+ 0 Gram-p…     4     4 #>  9 Gram-negative              1   e+ 0              2.22e-16 Gram-n…     1     1 #> 10 Gram-positive              6.05e-11              1.00e+ 0 Gram-p…     4     4 #> # ℹ 384 more rows #> # ℹ 18 more variables: AMK <int>, KAN <int>, PEN <int>, OXA <int>, FLC <int>, #> #   AMX <int>, AMC <int>, AMP <int>, TZP <int>, CZO <int>, FEP <int>, #> #   CXM <int>, FOX <int>, CTX <int>, CAZ <int>, CRO <int>, IPM <int>, MEM <int>  # Evaluate model performance metrics <- predictions %>%   metrics(truth = mo, estimate = .pred_class) # Calculate performance metrics  metrics #> # A tibble: 2 × 3 #>   .metric  .estimator .estimate #>   <chr>    <chr>          <dbl> #> 1 accuracy binary         0.995 #> 2 kap      binary         0.989 predictions %>%   roc_curve(mo, `.pred_Gram-negative`) %>%   autoplot()"},{"path":"https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html","id":"conclusion","dir":"Articles","previous_headings":"","what":"Conclusion","title":"AMR with tidymodels","text":"post, demonstrated build machine learning pipeline tidymodels framework AMR package. combining selector functions like aminoglycosides() betalactams() tidymodels, efficiently prepared data, trained model, evaluated performance. workflow extensible antibiotic classes resistance patterns, empowering users analyse AMR data systematically reproducibly.","code":""},{"path":"https://msberends.github.io/AMR/articles/EUCAST.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"How to apply EUCAST rules","text":"EUCAST rules? European Committee Antimicrobial Susceptibility Testing (EUCAST) states website: EUCAST expert rules tabulated collection expert knowledge intrinsic resistances, exceptional resistance phenotypes interpretive rules may applied antimicrobial susceptibility testing order reduce errors make appropriate recommendations reporting particular resistances. Europe, lot medical microbiological laboratories already apply rules (Brown et al., 2015). package features latest insights intrinsic resistance unusual phenotypes (v3.1, 2016). Moreover, eucast_rules() function use purpose can also apply additional rules, like forcing ampicillin = R isolates amoxicillin/clavulanic acid = R.","code":""},{"path":"https://msberends.github.io/AMR/articles/EUCAST.html","id":"examples","dir":"Articles","previous_headings":"","what":"Examples","title":"How to apply EUCAST rules","text":"rules can used discard impossible bug-drug combinations data. example, Klebsiella produces beta-lactamase prevents ampicillin (amoxicillin) working . words, practically every strain Klebsiella resistant ampicillin. Sometimes, laboratory data can still contain strains ampicillin susceptible ampicillin. antibiogram available identification available, antibiogram re-interpreted based identification (namely, Klebsiella). EUCAST expert rules solve , can applied using eucast_rules(): convenient function mo_is_intrinsic_resistant() uses guideline, allows check one specific microorganisms antibiotics: EUCAST rules can used correction, can also used filling known resistance susceptibility based results antimicrobials drugs. process called interpretive reading, basically form imputation, part eucast_rules() function well:","code":"oops <- data.frame(   mo = c(     \"Klebsiella\",     \"Escherichia\"   ),   ampicillin = \"S\" ) oops #>            mo ampicillin #> 1  Klebsiella          S #> 2 Escherichia          S  eucast_rules(oops, info = FALSE) #>            mo ampicillin #> 1  Klebsiella          R #> 2 Escherichia          S mo_is_intrinsic_resistant(   c(\"Klebsiella\", \"Escherichia\"),   \"ampicillin\" ) #> [1]  TRUE FALSE  mo_is_intrinsic_resistant(   \"Klebsiella\",   c(\"ampicillin\", \"kanamycin\") ) #> [1]  TRUE FALSE data <- data.frame(   mo = c(     \"Staphylococcus aureus\",     \"Enterococcus faecalis\",     \"Escherichia coli\",     \"Klebsiella pneumoniae\",     \"Pseudomonas aeruginosa\"   ),   VAN = \"-\", # Vancomycin   AMX = \"-\", # Amoxicillin   COL = \"-\", # Colistin   CAZ = \"-\", # Ceftazidime   CXM = \"-\", # Cefuroxime   PEN = \"S\", # Benzylenicillin   FOX = \"S\", # Cefoxitin   stringsAsFactors = FALSE ) data eucast_rules(data)"},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"type-of-input","dir":"Articles","previous_headings":"","what":"Type of input","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function takes data set input, regular data.frame. tries automatically determine right columns info isolates, name species columns results antimicrobial agents. See help page info set right settings data command ?mdro. WHONET data (data), settings automatically set correctly.","code":""},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"guidelines","dir":"Articles","previous_headings":"","what":"Guidelines","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function support multiple guidelines. can select guideline guideline parameter. Currently supported guidelines (case-insensitive): guideline = \"CMI2012\" (default) Magiorakos AP, Srinivasan et al. “Multidrug-resistant, extensively drug-resistant pandrug-resistant bacteria: international expert proposal interim standard definitions acquired resistance.” Clinical Microbiology Infection (2012) (link) guideline = \"EUCAST3.2\" (simply guideline = \"EUCAST\") European international guideline - EUCAST Expert Rules Version 3.2 “Intrinsic Resistance Unusual Phenotypes” (link) guideline = \"EUCAST3.1\" European international guideline - EUCAST Expert Rules Version 3.1 “Intrinsic Resistance Exceptional Phenotypes Tables” (link) guideline = \"TB\" international guideline multi-drug resistant tuberculosis - World Health Organization “Companion handbook guidelines programmatic management drug-resistant tuberculosis” (link) guideline = \"MRGN\" German national guideline - Mueller et al. (2015) Antimicrobial Resistance Infection Control 4:7. DOI: 10.1186/s13756-015-0047-6 guideline = \"BRMO\" Dutch national guideline - Rijksinstituut voor Volksgezondheid en Milieu “WIP-richtlijn BRMO (Bijzonder Resistente Micro-Organismen) (ZKH)” (link) Please suggest (country-specific) guidelines letting us know: https://github.com/msberends/AMR/issues/new.","code":""},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"custom-guidelines","dir":"Articles","previous_headings":"Guidelines","what":"Custom Guidelines","title":"How to determine multi-drug resistance (MDR)","text":"can also use custom guideline. Custom guidelines can set custom_mdro_guideline() function. great importance custom rules determine MDROs hospital, e.g., rules dependent ward, state contact isolation variables data. familiar case_when() dplyr package, recognise input method set rules. Rules must set using R considers ‘formula notation’: row/isolate matches first rule, value first ~ (case ‘Elderly Type ’) set MDRO value. Otherwise, second rule tried . maximum number rules unlimited. can print rules set console overview. Colours help reading console supports colours. outcome function can used guideline argument mdro() function: rules set (custom object case) exported shared file location using saveRDS() collaborate multiple users. custom rules set imported using readRDS().","code":"custom <- custom_mdro_guideline(   CIP == \"R\" & age > 60 ~ \"Elderly Type A\",   ERY == \"R\" & age > 60 ~ \"Elderly Type B\" ) custom #> A set of custom MDRO rules: #>   1. If CIP is  R  and age is higher than 60 then: Elderly Type A #>   2. If ERY is  R  and age is higher than 60 then: Elderly Type B #>   3. Otherwise: Negative #>  #> Unmatched rows will return NA. #> Results will be of class 'factor', with ordered levels: Negative < Elderly Type A < Elderly Type B x <- mdro(example_isolates, guideline = custom) table(x) #> x #>       Negative Elderly Type A Elderly Type B  #>           1070            198            732"},{"path":"https://msberends.github.io/AMR/articles/MDR.html","id":"examples","dir":"Articles","previous_headings":"","what":"Examples","title":"How to determine multi-drug resistance (MDR)","text":"mdro() function always returns ordered factor predefined guidelines. example, output default guideline Magiorakos et al. returns factor levels ‘Negative’, ‘MDR’, ‘XDR’ ‘PDR’ order. next example uses example_isolates data set. data set included package contains full antibiograms 2,000 microbial isolates. reflects reality can used practise AMR data analysis. test MDR/XDR/PDR guideline data set, get: Frequency table Class: factor > ordered (numeric) Length: 2,000 Levels: 4: Negative < Multi-drug-resistant (MDR) < Extensively drug-resistant … Available: 1,745 (87.25%, NA: 255 = 12.75%) Unique: 2 another example, create data set determine multi-drug resistant TB: column names automatically verified valid drug names codes, worked exactly way: data set now looks like : can now add interpretation MDR-TB data set. can use: shortcut mdr_tb(): Create frequency table results: Frequency table Class: factor > ordered (numeric) Length: 5,000 Levels: 5: Negative < Mono-resistant < Poly-resistant < Multi-drug-resistant <… Available: 5,000 (100%, NA: 0 = 0%) Unique: 5","code":"library(dplyr) # to support pipes: %>% library(cleaner) # to create frequency tables example_isolates %>%   mdro() %>%   freq() # show frequency table of the result #> Warning: in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) # random_sir() is a helper function to generate # a random vector with values S, I and R my_TB_data <- data.frame(   rifampicin = random_sir(5000),   isoniazid = random_sir(5000),   gatifloxacin = random_sir(5000),   ethambutol = random_sir(5000),   pyrazinamide = random_sir(5000),   moxifloxacin = random_sir(5000),   kanamycin = random_sir(5000) ) my_TB_data <- data.frame(   RIF = random_sir(5000),   INH = random_sir(5000),   GAT = random_sir(5000),   ETH = random_sir(5000),   PZA = random_sir(5000),   MFX = random_sir(5000),   KAN = random_sir(5000) ) head(my_TB_data) #>   rifampicin isoniazid gatifloxacin ethambutol pyrazinamide moxifloxacin #> 1          I         R            S          S            S            S #> 2          S         S            I          R            R            S #> 3          R         I            I          I            R            I #> 4          I         S            S          S            S            S #> 5          I         I            I          S            I            S #> 6          R         S            R          S            I            I #>   kanamycin #> 1         R #> 2         I #> 3         S #> 4         I #> 5         I #> 6         I mdro(my_TB_data, guideline = \"TB\") my_TB_data$mdr <- mdr_tb(my_TB_data) #> ℹ No column found as input for col_mo, assuming all rows contain #>   Mycobacterium tuberculosis. freq(my_TB_data$mdr)"},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"transforming","dir":"Articles","previous_headings":"","what":"Transforming","title":"How to conduct principal component analysis (PCA) for AMR","text":"PCA, need transform AMR data first. example_isolates data set package looks like: Now transform data set resistance percentages per taxonomic order genus:","code":"library(AMR) library(dplyr) glimpse(example_isolates) #> Rows: 2,000 #> Columns: 46 #> $ date    <date> 2002-01-02, 2002-01-03, 2002-01-07, 2002-01-07, 2002-01-13, 2… #> $ patient <chr> \"A77334\", \"A77334\", \"067927\", \"067927\", \"067927\", \"067927\", \"4… #> $ age     <dbl> 65, 65, 45, 45, 45, 45, 78, 78, 45, 79, 67, 67, 71, 71, 75, 50… #> $ gender  <chr> \"F\", \"F\", \"F\", \"F\", \"F\", \"F\", \"M\", \"M\", \"F\", \"F\", \"M\", \"M\", \"M… #> $ ward    <chr> \"Clinical\", \"Clinical\", \"ICU\", \"ICU\", \"ICU\", \"ICU\", \"Clinical\"… #> $ mo      <mo> \"B_ESCHR_COLI\", \"B_ESCHR_COLI\", \"B_STPHY_EPDR\", \"B_STPHY_EPDR\",… #> $ PEN     <sir> R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, R, S,… #> $ OXA     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ FLC     <sir> NA, NA, R, R, R, R, S, S, R, S, S, S, NA, NA, NA, NA, NA, R, R… #> $ AMX     <sir> NA, NA, NA, NA, NA, NA, R, R, NA, NA, NA, NA, NA, NA, R, NA, N… #> $ AMC     <sir> I, I, NA, NA, NA, NA, S, S, NA, NA, S, S, I, I, R, I, I, NA, N… #> $ AMP     <sir> NA, NA, NA, NA, NA, NA, R, R, NA, NA, NA, NA, NA, NA, R, NA, N… #> $ TZP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CZO     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, NA,… #> $ FEP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CXM     <sir> I, I, R, R, R, R, S, S, R, S, S, S, S, S, NA, S, S, R, R, S, S… #> $ FOX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, NA,… #> $ CTX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ CAZ     <sir> NA, NA, R, R, R, R, R, R, R, R, R, R, NA, NA, NA, S, S, R, R, … #> $ CRO     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ GEN     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ TOB     <sir> NA, NA, NA, NA, NA, NA, S, S, NA, NA, NA, NA, S, S, NA, NA, NA… #> $ AMK     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ KAN     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ TMP     <sir> R, R, S, S, R, R, R, R, S, S, NA, NA, S, S, S, S, S, R, R, R, … #> $ SXT     <sir> R, R, S, S, NA, NA, NA, NA, S, S, NA, NA, S, S, S, S, S, NA, N… #> $ NIT     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R,… #> $ FOS     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ LNZ     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… #> $ CIP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, NA, NA, NA, NA, S, S… #> $ MFX     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ VAN     <sir> R, R, S, S, S, S, S, S, S, S, NA, NA, R, R, R, R, R, S, S, S, … #> $ TEC     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… #> $ TCY     <sir> R, R, S, S, S, S, S, S, S, I, S, S, NA, NA, I, R, R, S, I, R, … #> $ TGC     <sir> NA, NA, S, S, S, S, S, S, S, NA, S, S, NA, NA, NA, R, R, S, NA… #> $ DOX     <sir> NA, NA, S, S, S, S, S, S, S, NA, S, S, NA, NA, NA, R, R, S, NA… #> $ ERY     <sir> R, R, R, R, R, R, S, S, R, S, S, S, R, R, R, R, R, R, R, R, S,… #> $ CLI     <sir> R, R, NA, NA, NA, R, NA, NA, NA, NA, NA, NA, R, R, R, R, R, NA… #> $ AZM     <sir> R, R, R, R, R, R, S, S, R, S, S, S, R, R, R, R, R, R, R, R, S,… #> $ IPM     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, S, S, NA, S, S… #> $ MEM     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ MTR     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ CHL     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ COL     <sir> NA, NA, R, R, R, R, R, R, R, R, R, R, NA, NA, NA, R, R, R, R, … #> $ MUP     <sir> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA… #> $ RIF     <sir> R, R, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, R, R, R, R, R, N… resistance_data <- example_isolates %>%   group_by(     order = mo_order(mo), # group on anything, like order     genus = mo_genus(mo)   ) %>% #  and genus as we do here   summarise_if(is.sir, resistance) %>% # then get resistance of all drugs   select(     order, genus, AMC, CXM, CTX,     CAZ, GEN, TOB, TMP, SXT   ) # and select only relevant columns  head(resistance_data) #> # A tibble: 6 × 10 #> # Groups:   order [5] #>   order             genus          AMC   CXM   CTX   CAZ   GEN   TOB   TMP   SXT #>   <chr>             <chr>        <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 (unknown order)   (unknown ge…    NA    NA    NA    NA    NA    NA    NA    NA #> 2 Actinomycetales   Schaalia        NA    NA    NA    NA    NA    NA    NA    NA #> 3 Bacteroidales     Bacteroides     NA    NA    NA    NA    NA    NA    NA    NA #> 4 Campylobacterales Campylobact…    NA    NA    NA    NA    NA    NA    NA    NA #> 5 Caryophanales     Gemella         NA    NA    NA    NA    NA    NA    NA    NA #> 6 Caryophanales     Listeria        NA    NA    NA    NA    NA    NA    NA    NA"},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"perform-principal-component-analysis","dir":"Articles","previous_headings":"","what":"Perform principal component analysis","title":"How to conduct principal component analysis (PCA) for AMR","text":"new pca() function automatically filter rows contain numeric values selected variables, now need : result can reviewed good old summary() function: Good news. first two components explain total 93.3% variance (see PC1 PC2 values Proportion Variance. can create -called biplot base R biplot() function, see antimicrobial resistance per drug explain difference per microorganism.","code":"pca_result <- pca(resistance_data) #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. summary(pca_result) #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\" #> Importance of components: #>                           PC1    PC2    PC3     PC4     PC5     PC6       PC7 #> Standard deviation     2.1539 1.6807 0.6138 0.33879 0.20808 0.03140 1.232e-16 #> Proportion of Variance 0.5799 0.3531 0.0471 0.01435 0.00541 0.00012 0.000e+00 #> Cumulative Proportion  0.5799 0.9330 0.9801 0.99446 0.99988 1.00000 1.000e+00 #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\""},{"path":"https://msberends.github.io/AMR/articles/PCA.html","id":"plotting-the-results","dir":"Articles","previous_headings":"","what":"Plotting the results","title":"How to conduct principal component analysis (PCA) for AMR","text":"can’t see explanation points. Perhaps works better new ggplot_pca() function, automatically adds right labels even groups:  can also print ellipse per group, edit appearance:","code":"biplot(pca_result) ggplot_pca(pca_result) ggplot_pca(pca_result, ellipse = TRUE) +   ggplot2::labs(title = \"An AMR/PCA biplot!\")"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"import-of-data","dir":"Articles","previous_headings":"","what":"Import of data","title":"How to work with WHONET data","text":"tutorial assumes already imported WHONET data e.g. readxl package. RStudio, can done using menu button ‘Import Dataset’ tab ‘Environment’. Choose option ‘Excel’ select exported file. Make sure date fields imported correctly. example syntax look like : package comes example data set WHONET. use analysis.","code":"library(readxl) data <- read_excel(path = \"path/to/your/file.xlsx\")"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"preparation","dir":"Articles","previous_headings":"","what":"Preparation","title":"How to work with WHONET data","text":"First, load relevant packages yet . use tidyverse analyses. . don’t know yet, suggest read website: https://www.tidyverse.org/. transform variables simplify automate analysis: Microorganisms transformed microorganism codes (called mo) using Catalogue Life reference data set, contains ~70,000 microorganisms taxonomic kingdoms Bacteria, Fungi Protozoa. tranformation .mo(). function also recognises almost WHONET abbreviations microorganisms. Antimicrobial results interpretations clean valid. words, contain values \"S\", \"\" \"R\". exactly .sir() function . errors warnings, values transformed succesfully. also created package dedicated data cleaning checking, called cleaner package. freq() function can used create frequency tables. let’s check data, couple frequency tables: Frequency table Class: character Length: 500 Available: 500 (100%, NA: 0 = 0%) Unique: 38 Shortest: 11 Longest: 40 (omitted 28 entries, n = 57 [11.4%]) Frequency table Class: factor > ordered > sir (numeric) Length: 500 Levels: 5: S < SDD < < R < NI Available: 481 (96.2%, NA: 19 = 3.8%) Unique: 3 Drug: Amoxicillin/clavulanic acid (AMC, J01CR02) Drug group: Beta-lactams/penicillins %SI: 78.59%","code":"library(dplyr) # part of tidyverse library(ggplot2) # part of tidyverse library(AMR) # this package library(cleaner) # to create frequency tables # transform variables data <- WHONET %>%   # get microbial ID based on given organism   mutate(mo = as.mo(Organism)) %>%   # transform everything from \"AMP_ND10\" to \"CIP_EE\" to the new `sir` class   mutate_at(vars(AMP_ND10:CIP_EE), as.sir) # our newly created `mo` variable, put in the mo_name() function data %>% freq(mo_name(mo), nmax = 10) # our transformed antibiotic columns # amoxicillin/clavulanic acid (J01CR02) as an example data %>% freq(AMC_ND2)"},{"path":"https://msberends.github.io/AMR/articles/WHONET.html","id":"a-first-glimpse-at-results","dir":"Articles","previous_headings":"","what":"A first glimpse at results","title":"How to work with WHONET data","text":"easy ggplot already give lot information, using included ggplot_sir() function:","code":"data %>%   group_by(Country) %>%   select(Country, AMP_ND2, AMC_ED20, CAZ_ED10, CIP_ED5) %>%   ggplot_sir(translate_ab = \"ab\", facet = \"Country\", datalabels = FALSE)"},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-full-microbial-taxonomy","dir":"Articles","previous_headings":"","what":"microorganisms: Full Microbial Taxonomy","title":"Data sets for download / own use","text":"data set 78 678 rows 26 columns, containing following column names:mo, fullname, status, kingdom, phylum, class, order, family, genus, species, subspecies, rank, ref, oxygen_tolerance, source, lpsn, lpsn_parent, lpsn_renamed_to, mycobank, mycobank_parent, mycobank_renamed_to, gbif, gbif_parent, gbif_renamed_to, prevalence, snomed. data set R available microorganisms, load AMR package. last updated 4 October 2024 13:28:44 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (1.8 MB) Download tab-separated text file (17.7 MB) Download Microsoft Excel workbook (8.7 MB) Download Apache Feather file (8.3 MB) Download Apache Parquet file (3.8 MB) Download IBM SPSS Statistics data file (29 MB) Download Stata DTA file (92.5 MB) NOTE: exported files SPSS Stata contain first 50 SNOMED codes per record, file size otherwise exceed 100 MB; file size limit GitHub. file structures compression techniques inefficient. Advice? Use R instead. ’s free much better many ways. tab-separated text file Microsoft Excel workbook contain SNOMED codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source","dir":"Articles","previous_headings":"microorganisms: Full Microbial Taxonomy","what":"Source","title":"Data sets for download / own use","text":"data set contains full microbial taxonomy six kingdoms List Prokaryotic names Standing Nomenclature (LPSN), MycoBank, Global Biodiversity Information Facility (GBIF): Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name ‘Microorganism’, OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example-content","dir":"Articles","previous_headings":"microorganisms: Full Microbial Taxonomy","what":"Example content","title":"Data sets for download / own use","text":"Included (sub)species per taxonomic kingdom: Example rows filtering genus Escherichia:","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"antibiotics-antibiotic-antifungal-drugs","dir":"Articles","previous_headings":"","what":"antibiotics: Antibiotic (+Antifungal) Drugs","title":"Data sets for download / own use","text":"data set 487 rows 14 columns, containing following column names:ab, cid, name, group, atc, atc_group1, atc_group2, abbreviations, synonyms, oral_ddd, oral_units, iv_ddd, iv_units, loinc. data set R available antibiotics, load AMR package. last updated 7 February 2025 17:01:22 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (42 kB) Download tab-separated text file (0.1 MB) Download Microsoft Excel workbook (74 kB) Download Apache Feather file (0.1 MB) Download Apache Parquet file (0.1 MB) Download IBM SPSS Statistics data file (0.4 MB) Download Stata DTA file (0.4 MB) tab-separated text, Microsoft Excel, SPSS, Stata files contain ATC codes, common abbreviations, trade names LOINC codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-1","dir":"Articles","previous_headings":"antibiotics: Antibiotic (+Antifungal) Drugs","what":"Source","title":"Data sets for download / own use","text":"data set contains EARS-Net ATC codes gathered WHONET, compound IDs PubChem. also contains brand names (synonyms) found PubChem Defined Daily Doses (DDDs) oral parenteral administration. ATC/DDD index Collaborating Centre Drug Statistics Methodology (note: may used commercial purposes, freely available CC website personal use) PubChem US National Library Medicine WHONET software 2019 LOINC (Logical Observation Identifiers Names Codes)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"antivirals-antiviral-drugs","dir":"Articles","previous_headings":"","what":"antivirals: Antiviral Drugs","title":"Data sets for download / own use","text":"data set 120 rows 11 columns, containing following column names:av, name, atc, cid, atc_group, synonyms, oral_ddd, oral_units, iv_ddd, iv_units, loinc. data set R available antivirals, load AMR package. last updated 20 October 2023 12:51:48 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (6 kB) Download tab-separated text file (17 kB) Download Microsoft Excel workbook (16 kB) Download Apache Feather file (16 kB) Download Apache Parquet file (13 kB) Download IBM SPSS Statistics data file (32 kB) Download Stata DTA file (78 kB) tab-separated text, Microsoft Excel, SPSS, Stata files contain trade names LOINC codes comma separated values.","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-2","dir":"Articles","previous_headings":"antivirals: Antiviral Drugs","what":"Source","title":"Data sets for download / own use","text":"data set contains ATC codes gathered compound IDs PubChem. also contains brand names (synonyms) found PubChem Defined Daily Doses (DDDs) oral parenteral administration. ATC/DDD index Collaborating Centre Drug Statistics Methodology (note: may used commercial purposes, freely available CC website personal use) PubChem US National Library Medicine LOINC (Logical Observation Identifiers Names Codes)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"clinical_breakpoints-interpretation-from-mic-values-disk-diameters-to-sir","dir":"Articles","previous_headings":"","what":"clinical_breakpoints: Interpretation from MIC values & disk diameters to SIR","title":"Data sets for download / own use","text":"data set 34 063 rows 14 columns, containing following column names:guideline, type, host, method, site, mo, rank_index, ab, ref_tbl, disk_dose, breakpoint_S, breakpoint_R, uti, is_SDD. data set R available clinical_breakpoints, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (70 kB) Download tab-separated text file (3.1 MB) Download Microsoft Excel workbook (2 MB) Download Apache Feather file (1.5 MB) Download Apache Parquet file (0.1 MB) Download IBM SPSS Statistics data file (5.6 MB) Download Stata DTA file (9.3 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-3","dir":"Articles","previous_headings":"clinical_breakpoints: Interpretation from MIC values & disk diameters to SIR","what":"Source","title":"Data sets for download / own use","text":"data set contains interpretation rules MIC values disk diffusion diameters. Included guidelines CLSI (2011-2024) EUCAST (2011-2024). Clinical breakpoints package validated imported WHONET, free desktop Windows application developed supported Collaborating Centre Surveillance Antimicrobial Resistance. can read website. developers WHONET AMR package contact sharing work. highly appreciate development WHONET software. CEO CLSI chairman EUCAST endorsed work public use AMR package (consequently use breakpoints) June 2023, future development distributing clinical breakpoints discussed meeting CLSI, EUCAST, , developers WHONET AMR package. NOTE: AMR package (WHONET software well) contains internal methods apply guidelines, rather complex. example, breakpoints must applied certain species groups (case package available microorganisms.groups data set). important considered using breakpoints use.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"intrinsic_resistant-intrinsic-bacterial-resistance","dir":"Articles","previous_headings":"","what":"intrinsic_resistant: Intrinsic Bacterial Resistance","title":"Data sets for download / own use","text":"data set 301 583 rows 2 columns, containing following column names:mo ab. data set R available intrinsic_resistant, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (0.1 MB) Download tab-separated text file (10.9 MB) Download Microsoft Excel workbook (3 MB) Download Apache Feather file (2.5 MB) Download Apache Parquet file (0.3 MB) Download IBM SPSS Statistics data file (16.2 MB) Download Stata DTA file (25 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-4","dir":"Articles","previous_headings":"intrinsic_resistant: Intrinsic Bacterial Resistance","what":"Source","title":"Data sets for download / own use","text":"data set contains defined intrinsic resistance EUCAST bug-drug combinations, based ‘EUCAST Expert Rules’ ‘EUCAST Intrinsic Resistance Unusual Phenotypes’ v3.3 (2021).","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example-content-4","dir":"Articles","previous_headings":"intrinsic_resistant: Intrinsic Bacterial Resistance","what":"Example content","title":"Data sets for download / own use","text":"Example rows filtering Enterobacter cloacae:","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"dosage-dosage-guidelines-from-eucast","dir":"Articles","previous_headings":"","what":"dosage: Dosage Guidelines from EUCAST","title":"Data sets for download / own use","text":"data set 503 rows 9 columns, containing following column names:ab, name, type, dose, dose_times, administration, notes, original_txt, eucast_version. data set R available dosage, load AMR package. last updated 22 June 2023 13:10:59 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (3 kB) Download tab-separated text file (43 kB) Download Microsoft Excel workbook (25 kB) Download Apache Feather file (21 kB) Download Apache Parquet file (9 kB) Download IBM SPSS Statistics data file (64 kB) Download Stata DTA file (0.1 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-5","dir":"Articles","previous_headings":"dosage: Dosage Guidelines from EUCAST","what":"Source","title":"Data sets for download / own use","text":"EUCAST breakpoints used package based dosages data set. Currently included dosages data set meant :  (), ‘EUCAST Clinical Breakpoint Tables’ v11.0 (2021), ‘EUCAST Clinical Breakpoint Tables’ v12.0 (2022).","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example_isolates-example-data-for-practice","dir":"Articles","previous_headings":"","what":"example_isolates: Example Data for Practice","title":"Data sets for download / own use","text":"data set 2 000 rows 46 columns, containing following column names:date, patient, age, gender, ward, mo, PEN, OXA, FLC, AMX, AMC, AMP, TZP, CZO, FEP, CXM, FOX, CTX, CAZ, CRO, GEN, TOB, AMK, KAN, TMP, SXT, NIT, FOS, LNZ, CIP, MFX, VAN, TEC, TCY, TGC, DOX, ERY, CLI, AZM, IPM, MEM, MTR, CHL, COL, MUP, RIF. data set R available example_isolates, load AMR package. last updated 15 June 2024 13:33:49 UTC. Find info structure data set .","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-6","dir":"Articles","previous_headings":"example_isolates: Example Data for Practice","what":"Source","title":"Data sets for download / own use","text":"data set contains randomised fictitious data, reflects reality can used practise AMR data analysis.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"example_isolates_unclean-example-data-for-practice","dir":"Articles","previous_headings":"","what":"example_isolates_unclean: Example Data for Practice","title":"Data sets for download / own use","text":"data set 3 000 rows 8 columns, containing following column names:patient_id, hospital, date, bacteria, AMX, AMC, CIP, GEN. data set R available example_isolates_unclean, load AMR package. last updated 27 August 2022 18:49:37 UTC. Find info structure data set .","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-7","dir":"Articles","previous_headings":"example_isolates_unclean: Example Data for Practice","what":"Source","title":"Data sets for download / own use","text":"data set contains randomised fictitious data, reflects reality can used practise AMR data analysis.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-groups-species-groups-and-microbiological-complexes","dir":"Articles","previous_headings":"","what":"microorganisms.groups: Species Groups and Microbiological Complexes","title":"Data sets for download / own use","text":"data set 521 rows 4 columns, containing following column names:mo_group, mo, mo_group_name, mo_name. data set R available microorganisms.groups, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (5 kB) Download tab-separated text file (49 kB) Download Microsoft Excel workbook (19 kB) Download Apache Feather file (19 kB) Download Apache Parquet file (13 kB) Download IBM SPSS Statistics data file (63 kB) Download Stata DTA file (81 kB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-8","dir":"Articles","previous_headings":"microorganisms.groups: Species Groups and Microbiological Complexes","what":"Source","title":"Data sets for download / own use","text":"data set contains species groups microbiological complexes, used clinical_breakpoints data set.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"microorganisms-codes-common-laboratory-codes","dir":"Articles","previous_headings":"","what":"microorganisms.codes: Common Laboratory Codes","title":"Data sets for download / own use","text":"data set 4 971 rows 2 columns, containing following column names:code mo. data set R available microorganisms.codes, load AMR package. last updated 29 September 2024 20:17:56 UTC. Find info structure data set . Direct download links: Download original R Data Structure (RDS) file (22 kB) Download tab-separated text file (0.1 MB) Download Microsoft Excel workbook (82 kB) Download Apache Feather file (85 kB) Download Apache Parquet file (56 kB) Download IBM SPSS Statistics data file (0.1 MB) Download Stata DTA file (0.1 MB)","code":""},{"path":"https://msberends.github.io/AMR/articles/datasets.html","id":"source-9","dir":"Articles","previous_headings":"microorganisms.codes: Common Laboratory Codes","what":"Source","title":"Data sets for download / own use","text":"data set contains commonly used codes microorganisms, laboratory systems WHONET.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"needed-r-packages","dir":"Articles","previous_headings":"","what":"Needed R packages","title":"How to predict antimicrobial resistance","text":"many uses R, need additional packages AMR data analysis. package works closely together tidyverse packages dplyr ggplot2. tidyverse tremendously improves way conduct data science - allows natural way writing syntaxes creating beautiful plots R. AMR package depends packages even extends use functions.","code":"library(dplyr) library(ggplot2) library(AMR)  # (if not yet installed, install with:) # install.packages(c(\"tidyverse\", \"AMR\"))"},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"prediction-analysis","dir":"Articles","previous_headings":"","what":"Prediction analysis","title":"How to predict antimicrobial resistance","text":"package contains function resistance_predict(), takes input functions AMR data analysis. Based date column, calculates cases per year uses regression model predict antimicrobial resistance. basically easy : function look date column col_date set. running commands, summary regression model printed unless using resistance_predict(..., info = FALSE). text printed summary - actual result (output) function data.frame containing year: number observations, actual observed resistance, estimated resistance standard error estimation: function plot available base R, can extended packages depend output based type input. extended function cope resistance predictions:  fastest way plot result. automatically adds right axes, error bars, titles, number available observations type model. also support ggplot2 package custom function ggplot_sir_predict() create appealing plots:","code":"# resistance prediction of piperacillin/tazobactam (TZP): resistance_predict(tbl = example_isolates, col_date = \"date\", col_ab = \"TZP\", model = \"binomial\")  # or: example_isolates %>%   resistance_predict(     col_ab = \"TZP\",     model = \"binomial\"   )  # to bind it to object 'predict_TZP' for example: predict_TZP <- example_isolates %>%   resistance_predict(     col_ab = \"TZP\",     model = \"binomial\"   ) predict_TZP #> # A tibble: 34 × 7 #>     year  value se_min se_max observations observed estimated #>  * <dbl>  <dbl>  <dbl>  <dbl>        <int>    <dbl>     <dbl> #>  1  2002 0.2        NA     NA           15   0.2       0.0562 #>  2  2003 0.0625     NA     NA           32   0.0625    0.0616 #>  3  2004 0.0854     NA     NA           82   0.0854    0.0676 #>  4  2005 0.05       NA     NA           60   0.05      0.0741 #>  5  2006 0.0508     NA     NA           59   0.0508    0.0812 #>  6  2007 0.121      NA     NA           66   0.121     0.0889 #>  7  2008 0.0417     NA     NA           72   0.0417    0.0972 #>  8  2009 0.0164     NA     NA           61   0.0164    0.106  #>  9  2010 0.0566     NA     NA           53   0.0566    0.116  #> 10  2011 0.183      NA     NA           93   0.183     0.127  #> # ℹ 24 more rows plot(predict_TZP) ggplot_sir_predict(predict_TZP) # choose for error bars instead of a ribbon ggplot_sir_predict(predict_TZP, ribbon = FALSE)"},{"path":"https://msberends.github.io/AMR/articles/resistance_predict.html","id":"choosing-the-right-model","dir":"Articles","previous_headings":"Prediction analysis","what":"Choosing the right model","title":"How to predict antimicrobial resistance","text":"Resistance easily predicted; look vancomycin resistance Gram-positive bacteria, spread (.e. standard error) enormous:  Vancomycin resistance 100% ten years, might remain low. can define model model parameter. model chosen generalised linear regression model using binomial distribution, assuming period zero resistance followed period increasing resistance leading slowly resistance. Valid values : vancomycin resistance Gram-positive bacteria, linear model might appropriate:  model also available object, attribute:","code":"example_isolates %>%   filter(mo_gramstain(mo, language = NULL) == \"Gram-positive\") %>%   resistance_predict(col_ab = \"VAN\", year_min = 2010, info = FALSE, model = \"binomial\") %>%   ggplot_sir_predict() example_isolates %>%   filter(mo_gramstain(mo, language = NULL) == \"Gram-positive\") %>%   resistance_predict(col_ab = \"VAN\", year_min = 2010, info = FALSE, model = \"linear\") %>%   ggplot_sir_predict() model <- attributes(predict_TZP)$model  summary(model)$family #>  #> Family: binomial  #> Link function: logit  summary(model)$coefficients #>                  Estimate  Std. Error   z value     Pr(>|z|) #> (Intercept) -200.67944891 46.17315349 -4.346237 1.384932e-05 #> year           0.09883005  0.02295317  4.305725 1.664395e-05"},{"path":"https://msberends.github.io/AMR/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Matthijs S. Berends. Author, maintainer. Dennis Souverein. Author, contributor. Erwin E. . Hassing. Author, contributor. Andrew P. Norgan. Contributor. Anita Williams. Contributor. Annick Lenglet. Contributor. Anthony Underwood. Contributor. Anton Mymrikov. Contributor. Bart C. Meijer. Contributor. Christian F. Luz. Contributor. Dmytro Mykhailenko. Contributor. Eric H. L. C. M. Hazenberg. Contributor. Gwen Knight. Contributor. Jason Stull. Contributor. Javier Sanchez. Contributor. Jonas Salm. Contributor. Judith M. Fonville. Contributor. Larisse Bolton. Contributor. Matthew Saab. Contributor. Peter Dutey-Magni. Contributor. Rogier P. Schade. Contributor. Sofia Ny. Contributor. Alex W. Friedrich. Thesis advisor. Bhanu N. M. Sinha. Thesis advisor. Casper J. Albers. Thesis advisor. Corinna Glasner. Thesis advisor.","code":""},{"path":"https://msberends.github.io/AMR/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Berends MS, Luz CF, Friedrich AW, Sinha BNM, Albers CJ, Glasner C (2022). “AMR: R Package Working Antimicrobial Resistance Data.” Journal Statistical Software, 104(3), 1–31. doi:10.18637/jss.v104.i03.","code":"@Article{,   title = {{AMR}: An {R} Package for Working with Antimicrobial Resistance Data},   author = {Matthijs S. Berends and Christian F. Luz and Alexander W. Friedrich and Bhanu N. M. Sinha and Casper J. Albers and Corinna Glasner},   journal = {Journal of Statistical Software},   year = {2022},   volume = {104},   number = {3},   pages = {1--31},   doi = {10.18637/jss.v104.i03}, }"},{"path":"https://msberends.github.io/AMR/index.html","id":"the-amr-package-for-r-","dir":"","previous_headings":"","what":"Antimicrobial Resistance Data Analysis","title":"Antimicrobial Resistance Data Analysis","text":"Provides --one solution antimicrobial resistance (AMR) data analysis One Health approach Used 175 countries, available 20 languages Generates antibiograms - traditional, combined, syndromic, even WISCA Provides full microbiological taxonomy extensive info antimicrobial drugs Applies recent CLSI EUCAST clinical veterinary breakpoints MICs, disk zones ECOFFs Corrects duplicate isolates, calculates predicts AMR per antimicrobial class Integrates WHONET, ATC, EARS-Net, PubChem, LOINC, SNOMED CT, NCBI 100% free costs dependencies, highly suitable places limited resources Now available Python ! Click read . https://msberends.github.io/AMR https://doi.org/10.18637/jss.v104.i03","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"introduction","dir":"","previous_headings":"","what":"Introduction","title":"Antimicrobial Resistance Data Analysis","text":"AMR package free open-source R package zero dependencies simplify analysis prediction Antimicrobial Resistance (AMR) work microbial antimicrobial data properties, using evidence-based methods. aim provide standard clean reproducible AMR data analysis, can therefore empower epidemiological analyses continuously enable surveillance treatment evaluation setting. Many different researchers around globe continually helping us make successful durable project! work published Journal Statistical Software (Volume 104(3); DOI 10.18637/jss.v104.i03) formed basis two PhD theses (DOI 10.33612/diss.177417131 DOI 10.33612/diss.192486375). installing package, R knows ~52,000 distinct microbial species (updated December 2022) ~600 antibiotic, antimycotic antiviral drugs name code (including ATC, EARS-Net, ASIARS-Net, PubChem, LOINC SNOMED CT), knows valid SIR MIC values. integral clinical breakpoint guidelines CLSI EUCAST included, even epidemiological cut-(ECOFF) values. supports can read data format, including WHONET data. package works Windows, macOS Linux versions R since R-3.0 (April 2013). designed work setting, including limited resources. created routine data analysis academic research Faculty Medical Sciences University Groningen, collaboration non-profit organisations Certe Medical Diagnostics Advice Foundation University Medical Center Groningen.","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"used-in-over-175-countries-available-in-20-languages","dir":"","previous_headings":"Introduction","what":"Used in over 175 countries, available in 20 languages","title":"Antimicrobial Resistance Data Analysis","text":"Since first public release early 2018, R package used almost countries world. Click map enlarge see country names. help contributors corners world, AMR package available  English,  Czech,  Chinese,  Danish,  Dutch,  Finnish,  French,  German,  Greek,  Italian,  Japanese,  Norwegian,  Polish,  Portuguese,  Romanian,  Russian,  Spanish,  Swedish,  Turkish,  Ukrainian. Antimicrobial drug (group) names colloquial microorganism names provided languages.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/index.html","id":"filtering-and-selecting-data","dir":"","previous_headings":"Practical examples","what":"Filtering and selecting data","title":"Antimicrobial Resistance Data Analysis","text":"One powerful functions package, aside calculating plotting AMR, selecting filtering based antimicrobial columns. can done using -called antimicrobial selectors, work base R, dplyr data.table. defined row filter Gram-negative bacteria intrinsic resistance cefotaxime (mo_is_gram_negative() mo_is_intrinsic_resistant()) column selection two antibiotic groups (aminoglycosides() carbapenems()), reference data microorganisms antibiotics AMR package make sure get meant:","code":"# AMR works great with dplyr, but it's not required or neccesary library(AMR) library(dplyr)  example_isolates %>%   mutate(bacteria = mo_fullname()) %>%   # filtering functions for microorganisms:   filter(mo_is_gram_negative(),          mo_is_intrinsic_resistant(ab = \"cefotax\")) %>%   # antimicrobial selectors:   select(bacteria,          aminoglycosides(),          carbapenems())"},{"path":"https://msberends.github.io/AMR/index.html","id":"generating-antibiograms","dir":"","previous_headings":"Practical examples","what":"Generating antibiograms","title":"Antimicrobial Resistance Data Analysis","text":"AMR package supports generating traditional, combined, syndromic, even weighted-incidence syndromic combination antibiograms (WISCA). used inside R Markdown Quarto, table printed right output format automatically (markdown, LaTeX, HTML, etc.). combination antibiograms, clear combined antibiotics yield higher empiric coverage: Like many functions package, antibiogram() comes support 20 languages often detected automatically based system language:","code":"antibiogram(example_isolates,             antibiotics = c(aminoglycosides(), carbapenems())) antibiogram(example_isolates,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),             mo_transform = \"gramstain\") antibiogram(example_isolates,             antibiotics = c(\"cipro\", \"tobra\", \"genta\"), # any arbitrary name or code will work             mo_transform = \"gramstain\",             ab_transform = \"name\",             language = \"uk\") # Ukrainian"},{"path":"https://msberends.github.io/AMR/index.html","id":"interpreting-and-plotting-mic-and-sir-values","dir":"","previous_headings":"Practical examples","what":"Interpreting and plotting MIC and SIR values","title":"Antimicrobial Resistance Data Analysis","text":"AMR package allows interpretation MIC disk diffusion values based CLSI EUCAST. Moreover, ggplot2 package extended new scale functions, allow plotting log2-distributed MIC values SIR values.","code":"library(ggplot2) library(AMR)  # generate some random values some_mic_values <- random_mic(size = 100) some_groups <- sample(LETTERS[1:5], 20, replace = TRUE) interpretation <- as.sir(some_mic_values,                          guideline = \"EUCAST 2024\",                          mo = \"E. coli\", # or any code or name resembling a known species                          ab = \"Cipro\")   # or any code or name resembling an antibiotic  # create the plot ggplot(data.frame(mic = some_mic_values,                   group = some_groups,                   sir = interpretation),        aes(x = group, y = mic, colour = sir)) +   theme_minimal() +   geom_boxplot(fill = NA, colour = \"grey\") +   geom_jitter(width = 0.25) +      # NEW scale function: plot MIC values to x, y, colour or fill   scale_y_mic() +      # NEW scale function: write out S/I/R in any of the 20 supported languages   #                     and set colourblind-friendly colours   scale_colour_sir()"},{"path":"https://msberends.github.io/AMR/index.html","id":"calculating-resistance-per-group","dir":"","previous_headings":"Practical examples","what":"Calculating resistance per group","title":"Antimicrobial Resistance Data Analysis","text":"manual approach, can use resistance susceptibility() function: use antimicrobial selectors select series antibiotic columns:","code":"example_isolates %>%   # group by ward:   group_by(ward) %>%   # calculate AMR using resistance() for gentamicin and tobramycin   # and get their 95% confidence intervals using sir_confidence_interval():   summarise(across(c(GEN, TOB),                    list(total_R = resistance,                         conf_int = function(x) sir_confidence_interval(x, collapse = \"-\")))) library(AMR) library(dplyr)  out <- example_isolates %>%   # group by ward:   group_by(ward) %>%   # calculate AMR using resistance(), over all aminoglycosides and polymyxins:   summarise(across(c(aminoglycosides(), polymyxins()),             resistance)) out # transform the antibiotic columns to names: out %>% set_ab_names() # transform the antibiotic column to ATC codes: out %>% set_ab_names(property = \"atc\")"},{"path":"https://msberends.github.io/AMR/index.html","id":"what-else-can-you-do-with-this-package","dir":"","previous_headings":"","what":"What else can you do with this package?","title":"Antimicrobial Resistance Data Analysis","text":"package intended comprehensive toolbox integrated AMR data analysis. package can used : Reference taxonomy microorganisms, since package contains microbial (sub)species List Prokaryotic names Standing Nomenclature (LPSN) Global Biodiversity Information Facility (GBIF) (manual) Interpreting raw MIC disk diffusion values, based CLSI EUCAST guideline (manual) Retrieving antimicrobial drug names, doses forms administration clinical health care records (manual) Determining first isolates used AMR data analysis (manual) Calculating antimicrobial resistance (tutorial) Determining multi-drug resistance (MDR) / multi-drug resistant organisms (MDRO) (tutorial) Calculating (empirical) susceptibility mono therapy combination therapies (tutorial) Predicting future antimicrobial resistance using regression models (tutorial) Getting properties microorganism (like Gram stain, species, genus family) (manual) Getting properties antimicrobial (like name, code EARS-Net/ATC/LOINC/PubChem, defined daily dose trade name) (manual) Plotting antimicrobial resistance (tutorial) Applying EUCAST expert rules (manual) Getting SNOMED codes microorganism, getting properties microorganism based SNOMED code (manual) Getting LOINC codes antibiotic, getting properties antibiotic based LOINC code (manual) Machine reading EUCAST CLSI guidelines 2011-2021 translate MIC values disk diffusion diameters SIR (link) Principal component analysis AMR (tutorial)","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/index.html","id":"latest-official-version","dir":"","previous_headings":"Get this package","what":"Latest official version","title":"Antimicrobial Resistance Data Analysis","text":"package available official R network (CRAN). Install package R CRAN using command: downloaded installed automatically. RStudio, click menu Tools > Install Packages… type “AMR” press Install. Note: functions website may available latest release. use functions data sets mentioned website, install latest development version.","code":"install.packages(\"AMR\")"},{"path":"https://msberends.github.io/AMR/index.html","id":"latest-development-version","dir":"","previous_headings":"Get this package","what":"Latest development version","title":"Antimicrobial Resistance Data Analysis","text":"Please read Developer Guideline . latest unpublished development version can installed GitHub two ways: Manually, using: Automatically, using rOpenSci R-universe platform, adding R-universe address list repositories (‘repos’): , can install update AMR package like official release (e.g., using install.packages(\"AMR\") RStudio via Tools > Check Package Updates…).","code":"install.packages(\"remotes\") # if you haven't already remotes::install_github(\"msberends/AMR\") options(repos = c(getOption(\"repos\"),                   msberends = \"https://msberends.r-universe.dev\"))"},{"path":"https://msberends.github.io/AMR/index.html","id":"get-started","dir":"","previous_headings":"","what":"Get started","title":"Antimicrobial Resistance Data Analysis","text":"find conduct AMR data analysis, please continue reading get started click link ‘’ menu.","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"partners","dir":"","previous_headings":"","what":"Partners","title":"Antimicrobial Resistance Data Analysis","text":"development package part , related , made possible following non-profit organisations initiatives:","code":""},{"path":"https://msberends.github.io/AMR/index.html","id":"copyright","dir":"","previous_headings":"","what":"Copyright","title":"Antimicrobial Resistance Data Analysis","text":"R package free, open-source software licensed GNU General Public License v2.0 (GPL-2). nutshell, means package: May used commercial purposes May used private purposes May used patent purposes May modified, although: Modifications must released license distributing package Changes made code must documented May distributed, although: Source code must made available package distributed copy license copyright notice must included package. Comes LIMITATION liability Comes warranty","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-deprecated.html","id":null,"dir":"Reference","previous_headings":"","what":"Deprecated Functions — AMR-deprecated","title":"Deprecated Functions — AMR-deprecated","text":"functions -called 'Deprecated'. removed future version package. Using functions give warning name function replaced (one).","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-deprecated.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Deprecated Functions — AMR-deprecated","text":"","code":"ab_class(...)  ab_selector(...)"},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":null,"dir":"Reference","previous_headings":"","what":"Options for the AMR package — AMR-options","title":"Options for the AMR package — AMR-options","text":"overview package-specific options() can set AMR package.","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"options","dir":"Reference","previous_headings":"","what":"Options","title":"Options for the AMR package — AMR-options","text":"AMR_antibiogram_formatting_type  numeric (1-12) use antibiogram(), indicate formatting type use. AMR_breakpoint_type  character use .sir(), indicate breakpoint type use. must either \"ECOFF\", \"animal\", \"human\". AMR_cleaning_regex  regular expression (case-insensitive) use .mo() mo_* functions, clean user input. default outcome mo_cleaning_regex(), removes texts brackets texts \"species\" \"serovar\". AMR_custom_ab  file location RDS file, use custom antimicrobial drugs package. explained add_custom_antimicrobials(). AMR_custom_mo  file location RDS file, use custom microorganisms package. explained add_custom_microorganisms(). AMR_eucastrules  character set default types rules eucast_rules() function, must one : \"breakpoints\", \"expert\", \"\", \"custom\", \"\", defaults c(\"breakpoints\", \"expert\"). AMR_guideline  character set default guideline interpreting MIC values disk diffusion diameters .sir(). Can guideline name (e.g., \"CLSI\") name year (e.g. \"CLSI 2019\"). default latest implemented EUCAST guideline, currently \"EUCAST 2024\". Supported guideline currently EUCAST (2011-2024) CLSI (2011-2024). AMR_ignore_pattern  regular expression ignore (.e., make NA) match given .mo() mo_* functions. AMR_include_PKPD  logical use .sir(), indicate PK/PD clinical breakpoints must applied last resort - default TRUE. AMR_include_screening  logical use .sir(), indicate clinical breakpoints screening allowed - default FALSE. AMR_keep_synonyms  logical use .mo() mo_* functions, indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE. AMR_locale  character set language AMR package, can one supported language names ISO-639-1 codes: English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). default current system language (supported, English otherwise). AMR_mo_source  file location manual code list used .mo() mo_* functions. explained set_mo_source().","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"saving-settings-between-sessions","dir":"Reference","previous_headings":"","what":"Saving Settings Between Sessions","title":"Options for the AMR package — AMR-options","text":"Settings R saved globally thus lost R exited. can save options .Rprofile file, user-specific file. can edit using:   file, can set options ...   ...add Portuguese language support antibiotics, allow PK/PD rules interpreting MIC values .sir().","code":"utils::file.edit(\"~/.Rprofile\") options(AMR_locale = \"pt\")  options(AMR_include_PKPD = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/AMR-options.html","id":"share-options-within-team","dir":"Reference","previous_headings":"","what":"Share Options Within Team","title":"Options for the AMR package — AMR-options","text":"global approach, e.g. within (data) team, save options file remote file location, shared network drive, user read file automatically start-. work way: Save plain text file e.g. \"X:/team_folder/R_options.R\" fill preferred settings. user, open .Rprofile file using utils::file.edit(\"~/.Rprofile\") put :   Reload R/RStudio check settings getOption(), e.g. getOption(\"AMR_locale\") set value. Now team settings configured one place, can maintained .","code":"source(\"X:/team_folder/R_options.R\")"},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":null,"dir":"Reference","previous_headings":"","what":"The AMR Package — AMR","title":"The AMR Package — AMR","text":"Welcome AMR package. AMR package free open-source R package zero dependencies simplify analysis prediction Antimicrobial Resistance (AMR) work microbial antimicrobial data properties, using evidence-based methods. aim provide standard clean reproducible AMR data analysis, can therefore empower epidemiological analyses continuously enable surveillance treatment evaluation setting. Many different researchers around globe continually helping us make successful durable project! work published Journal Statistical Software (Volume 104(3); doi:10.18637/jss.v104.i03 ) formed basis two PhD theses (doi:10.33612/diss.177417131  doi:10.33612/diss.192486375 ). installing package, R knows ~79 000 microorganisms (updated June 2024) ~610 antibiotic, antimycotic antiviral drugs name code (including ATC, EARS-Net, ASIARS-Net, PubChem, LOINC SNOMED CT), knows valid SIR MIC values. integral clinical breakpoint guidelines CLSI EUCAST included, even epidemiological cut-(ECOFF) values. supports can read data format, including WHONET data. package works Windows, macOS Linux versions R since R-3.0 (April 2013). designed work setting, including limited resources. created routine data analysis academic research Faculty Medical Sciences public University Groningen, collaboration non-profit organisations Certe Medical Diagnostics Advice Foundation University Medical Center Groningen. AMR package available English, Chinese, Czech, Danish, Dutch, Finnish, French, German, Greek, Italian, Japanese, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Turkish, Ukrainian. Antimicrobial drug (group) names colloquial microorganism names provided languages.","code":""},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"The AMR Package — AMR","text":"cite AMR publications use: Berends MS, Luz CF, Friedrich AW, Sinha BNM, Albers CJ, Glasner C (2022). \"AMR: R Package Working Antimicrobial Resistance Data.\" Journal Statistical Software, 104(3), 1-31. doi:10.18637/jss.v104.i03 BibTeX entry LaTeX users :","code":"@Article{,   title = {{AMR}: An {R} Package for Working with Antimicrobial Resistance Data},   author = {Matthijs S. Berends and Christian F. Luz and Alexander W. Friedrich and Bhanu N. M. Sinha and Casper J. Albers and Corinna Glasner},   journal = {Journal of Statistical Software},   year = {2022},   volume = {104},   number = {3},   pages = {1--31},   doi = {10.18637/jss.v104.i03}, }"},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"The AMR Package — AMR","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/AMR.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"The AMR Package — AMR","text":"Maintainer: Matthijs S. Berends m.s.berends@umcg.nl (ORCID) Authors: Dennis Souverein (ORCID) [contributor] Erwin E. . Hassing [contributor] contributors: Andrew P. Norgan (ORCID) [contributor] Anita Williams (ORCID) [contributor] Annick Lenglet (ORCID) [contributor] Anthony Underwood (ORCID) [contributor] Anton Mymrikov [contributor] Bart C. Meijer [contributor] Christian F. Luz (ORCID) [contributor] Dmytro Mykhailenko [contributor] Eric H. L. C. M. Hazenberg [contributor] Gwen Knight (ORCID) [contributor] Jason Stull (ORCID) [contributor] Javier Sanchez (ORCID) [contributor] Jonas Salm [contributor] Judith M. Fonville [contributor] Larisse Bolton (ORCID) [contributor] Matthew Saab [contributor] Peter Dutey-Magni (ORCID) [contributor] Rogier P. Schade [contributor] Sofia Ny (ORCID) [contributor] Alex W. Friedrich (ORCID) [thesis advisor] Bhanu N. M. Sinha (ORCID) [thesis advisor] Casper J. Albers (ORCID) [thesis advisor] Corinna Glasner (ORCID) [thesis advisor]","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":null,"dir":"Reference","previous_headings":"","what":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"antimicrobial drugs official names, ATC codes, ATC groups defined daily dose (DDD) included package, using Collaborating Centre Drug Statistics Methodology.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHOCC.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"WHOCC: WHO Collaborating Centre for Drug Statistics Methodology — WHOCC","text":"","code":"as.ab(\"meropenem\") #> Class 'ab' #> [1] MEM ab_name(\"J01DH02\") #> [1] \"Meropenem\"  ab_tradenames(\"flucloxacillin\") #>  [1] \"culpen\"               \"floxacillin\"          \"floxacillinsodium\"    #>  [4] \"floxapen\"             \"floxapensodiumsalt\"   \"fluclox\"              #>  [7] \"flucloxacilina\"       \"flucloxacilline\"      \"flucloxacillinum\"     #> [10] \"fluorochloroxacillin\" \"staphylex\""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 500 Isolates - WHONET Example — WHONET","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"example data set exact structure export file WHONET. files can used package, example data set shows. antibiotic results example_isolates data set. patient names created using online surname generators place practice purposes.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"","code":"WHONET"},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"tibble 500 observations 53 variables: Identification number ID sample Specimen number ID specimen Organism Name microorganism. analysis, transform valid microbial class, using .mo(). Country Country origin Laboratory Name laboratory Last name Fictitious last name patient First name Fictitious initial patient Sex Fictitious gender patient Age Fictitious age patient Age category Age group, can also looked using age_groups() Date admissionDate hospital admission Specimen dateDate specimen received laboratory Specimen type Specimen type group Specimen type (Numeric) Translation \"Specimen type\" Reason Reason request Differential Diagnosis Isolate number ID isolate Organism type Type microorganism, can also looked using mo_type() Serotype Serotype microorganism Beta-lactamase Microorganism produces beta-lactamase? ESBL Microorganism produces extended spectrum beta-lactamase? Carbapenemase Microorganism produces carbapenemase? MRSA screening test Microorganism possible MRSA? Inducible clindamycin resistance Clindamycin can induced? Comment comments Date data entryDate data entered WHONET AMP_ND10:CIP_EE 28 different antibiotics. can lookup abbreviations antibiotics data set, use e.g. ab_name(\"AMP\") get official name immediately. analysis, transform valid antibiotic class, using .sir().","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/WHONET.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 500 Isolates - WHONET Example — WHONET","text":"","code":"WHONET #> # A tibble: 500 × 53 #>    `Identification number` `Specimen number` Organism Country         Laboratory #>    <chr>                               <int> <chr>    <chr>           <chr>      #>  1 fe41d7bafa                           1748 SPN      Belgium         National … #>  2 91f175ec37                           1767 eco      The Netherlands National … #>  3 cc4015056e                           1343 eco      The Netherlands National … #>  4 e864b692f5                           1894 MAP      Denmark         National … #>  5 3d051fe345                           1739 PVU      Belgium         National … #>  6 c80762a08d                           1846 103      The Netherlands National … #>  7 8022d3727c                           1628 103      Denmark         National … #>  8 f3dc5f553d                           1493 eco      The Netherlands National … #>  9 15add38f6c                           1847 eco      France          National … #> 10 fd41248def                           1458 eco      Germany         National … #> # ℹ 490 more rows #> # ℹ 48 more variables: `Last name` <chr>, `First name` <chr>, Sex <chr>, #> #   Age <dbl>, `Age category` <chr>, `Date of admission` <date>, #> #   `Specimen date` <date>, `Specimen type` <chr>, #> #   `Specimen type (Numeric)` <dbl>, Reason <chr>, `Isolate number` <int>, #> #   `Organism type` <chr>, Serotype <chr>, `Beta-lactamase` <lgl>, ESBL <lgl>, #> #   Carbapenemase <lgl>, `MRSA screening test` <lgl>, …"},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":null,"dir":"Reference","previous_headings":"","what":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"Use function e.g. clinical texts health care records. returns list antimicrobial drugs, doses forms administration found texts.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"","code":"ab_from_text(text, type = c(\"drug\", \"dose\", \"administration\"),   collapse = NULL, translate_ab = FALSE, thorough_search = NULL,   info = interactive(), ...)"},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"text text analyse type type property search , either \"drug\", \"dose\" \"administration\", see Examples collapse character pass paste(, collapse = ...) return one character per element text, see Examples translate_ab type = \"drug\": column name antibiotics data set translate antibiotic abbreviations , using ab_property(). default FALSE. Using TRUE equal using \"name\". thorough_search logical indicate whether input must extensively searched misspelling faulty input values. Setting TRUE take considerably time using FALSE. default, turn TRUE input elements contain maximum three words. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed .ab()","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"list,  character collapse NULL","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"function also internally used .ab(), although searches first drug name throw note drug names returned. Note: .ab() function may use long regular expression match brand names antimicrobial drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"argument-type","dir":"Reference","previous_headings":"","what":"Argument type","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"default, function search antimicrobial drug names. text elements searched official names, ATC codes brand names. uses .ab() internally, correct misspelling. type = \"dose\" (similar, like \"dosing\", \"doses\"), text elements searched numeric values higher 100 resemble years. output numeric. supports unit (g, mg, IE, etc.) multiple values one clinical text, see Examples. type = \"administration\" (abbreviations, like \"admin\", \"adm\"), text elements searched form drug administration. supports following forms (including common abbreviations): buccal, implant, inhalation, instillation, intravenous, nasal, oral, parenteral, rectal, sublingual, transdermal vaginal. Abbreviations oral ('po', 'per os') become \"oral\", values intravenous ('iv', 'intraven') become \"iv\". supports multiple values one clinical text, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"argument-collapse","dir":"Reference","previous_headings":"","what":"Argument collapse","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"Without using collapse, function return list. can convenient use e.g. inside mutate()):df %>% mutate(abx = ab_from_text(clinical_text)) returned AB codes can transformed official names, groups, etc. ab_* functions ab_name() ab_group(), using translate_ab argument. using collapse, function return character:df %>% mutate(abx = ab_from_text(clinical_text, collapse = \"|\"))","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_from_text.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Retrieve Antimicrobial Drug Names and Doses from Clinical Text — ab_from_text","text":"","code":"# mind the bad spelling of amoxicillin in this line, # straight from a true health care record: ab_from_text(\"28/03/2020 regular amoxicilliin 500mg po tid\") #> [[1]] #> Class 'ab' #> [1] AMX #>   ab_from_text(\"500 mg amoxi po and 400mg cipro iv\") #> [[1]] #> Class 'ab' #> [1] AMX CIP #>  ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", type = \"dose\") #> [[1]] #> [1] 500 400 #>  ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", type = \"admin\") #> [[1]] #> [1] \"oral\" \"iv\"   #>   ab_from_text(\"500 mg amoxi po and 400mg cipro iv\", collapse = \", \") #> [1] \"AMX, CIP\" # \\donttest{ # if you want to know which antibiotic groups were administered, do e.g.: abx <- ab_from_text(\"500 mg amoxi po and 400mg cipro iv\") ab_group(abx[[1]]) #> [1] \"Beta-lactams/penicillins\" \"Quinolones\"                if (require(\"dplyr\")) {   tibble(clinical_text = c(     \"given 400mg cipro and 500 mg amox\",     \"started on doxy iv today\"   )) %>%     mutate(       abx_codes = ab_from_text(clinical_text),       abx_doses = ab_from_text(clinical_text, type = \"doses\"),       abx_admin = ab_from_text(clinical_text, type = \"admin\"),       abx_coll = ab_from_text(clinical_text, collapse = \"|\"),       abx_coll_names = ab_from_text(clinical_text,         collapse = \"|\",         translate_ab = \"name\"       ),       abx_coll_doses = ab_from_text(clinical_text,         type = \"doses\",         collapse = \"|\"       ),       abx_coll_admin = ab_from_text(clinical_text,         type = \"admin\",         collapse = \"|\"       )     ) } #> Loading required package: dplyr #>  #> Attaching package: ‘dplyr’ #> The following objects are masked from ‘package:stats’: #>  #>     filter, lag #> The following objects are masked from ‘package:base’: #>  #>     intersect, setdiff, setequal, union #> # A tibble: 2 × 8 #>   clinical_text            abx_codes abx_doses abx_admin abx_coll abx_coll_names #>   <chr>                    <list>    <list>    <list>    <chr>    <chr>          #> 1 given 400mg cipro and 5… <ab [2]>  <dbl [2]> <chr [1]> CIP|AMX  Ciprofloxacin… #> 2 started on doxy iv today <ab [1]>  <dbl [1]> <chr [1]> DOX      Doxycycline    #> # ℹ 2 more variables: abx_coll_doses <chr>, abx_coll_admin <chr> # }"},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of an Antibiotic — ab_property","title":"Get Properties of an Antibiotic — ab_property","text":"Use functions return specific property antibiotic antibiotics data set. input values evaluated internally .ab().","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of an Antibiotic — ab_property","text":"","code":"ab_name(x, language = get_AMR_locale(), tolower = FALSE, ...)  ab_cid(x, ...)  ab_synonyms(x, ...)  ab_tradenames(x, ...)  ab_group(x, language = get_AMR_locale(), ...)  ab_atc(x, only_first = FALSE, ...)  ab_atc_group1(x, language = get_AMR_locale(), ...)  ab_atc_group2(x, language = get_AMR_locale(), ...)  ab_loinc(x, ...)  ab_ddd(x, administration = \"oral\", ...)  ab_ddd_units(x, administration = \"oral\", ...)  ab_info(x, language = get_AMR_locale(), ...)  ab_url(x, open = FALSE, ...)  ab_property(x, property = \"name\", language = get_AMR_locale(), ...)  set_ab_names(data, ..., property = \"name\", language = get_AMR_locale(),   snake_case = NULL)"},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of an Antibiotic — ab_property","text":"x (vector ) text can coerced valid antibiotic drug code .ab() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. tolower logical indicate whether first character every output transformed lower case character. lead e.g. \"polymyxin B\" \"polymyxin b\". ... case set_ab_names() data data.frame: columns select (supports tidy selection column1:column4), otherwise arguments passed .ab() only_first logical indicate whether first ATC code must returned, giving preference J0-codes (.e., antimicrobial drug group) administration way administration, either \"oral\" \"iv\" open browse URL using utils::browseURL() property one column names one antibiotics data set: vector_or(colnames(antibiotics), sort = FALSE). data data.frame columns need renamed, character vector column names snake_case logical indicate whether names -called snake case: lower case spaces/slashes replaced underscore (_)","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of an Antibiotic — ab_property","text":"integer case ab_cid() named list case ab_info() multiple ab_atc()/ab_synonyms()/ab_tradenames() double case ab_ddd() data.frame case set_ab_names() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of an Antibiotic — ab_property","text":"output translated possible. function ab_url() return direct URL official website. warning returned required ATC code available. function set_ab_names() special column renaming function data.frames. renames columns names resemble antimicrobial drugs. always makes sure new column names unique. property = \"atc\" set, preference given ATC codes J-group.","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of an Antibiotic — ab_property","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of an Antibiotic — ab_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/ab_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of an Antibiotic — ab_property","text":"","code":"# all properties: ab_name(\"AMX\") #> [1] \"Amoxicillin\" ab_atc(\"AMX\") #> [1] \"J01CA04\" ab_cid(\"AMX\") #> [1] 33613 ab_synonyms(\"AMX\") #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           ab_tradenames(\"AMX\") #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           ab_group(\"AMX\") #> [1] \"Beta-lactams/penicillins\" ab_atc_group1(\"AMX\") #> [1] \"Beta-lactam antibacterials, penicillins\" ab_atc_group2(\"AMX\") #> [1] \"Penicillins with extended spectrum\" ab_url(\"AMX\") #>                                                             Amoxicillin  #> \"https://atcddd.fhi.no/atc_ddd_index//?code=J01CA04&showdescription=no\"   # smart lowercase transformation ab_name(x = c(\"AMC\", \"PLB\")) #> [1] \"Amoxicillin/clavulanic acid\" \"Polymyxin B\"                 ab_name(x = c(\"AMC\", \"PLB\"), tolower = TRUE) #> [1] \"amoxicillin/clavulanic acid\" \"polymyxin B\"                  # defined daily doses (DDD) ab_ddd(\"AMX\", \"oral\") #> [1] 1.5 ab_ddd_units(\"AMX\", \"oral\") #> [1] \"g\" ab_ddd(\"AMX\", \"iv\") #> [1] 3 ab_ddd_units(\"AMX\", \"iv\") #> [1] \"g\"  ab_info(\"AMX\") # all properties as a list #> $ab #> [1] \"AMX\" #>  #> $cid #> [1] 33613 #>  #> $name #> [1] \"Amoxicillin\" #>  #> $group #> [1] \"Beta-lactams/penicillins\" #>  #> $atc #> [1] \"J01CA04\" #>  #> $atc_group1 #> [1] \"Beta-lactam antibacterials, penicillins\" #>  #> $atc_group2 #> [1] \"Penicillins with extended spectrum\" #>  #> $tradenames #>  [1] \"actimoxi\"           \"amoclen\"            \"amolin\"             #>  [4] \"amopen\"             \"amopenixin\"         \"amoxibiotic\"        #>  [7] \"amoxicaps\"          \"amoxicilina\"        \"amoxicilline\"       #> [10] \"amoxicillinhydrate\" \"amoxicillinum\"      \"amoxiden\"           #> [13] \"amoxil\"             \"amoxivet\"           \"amoxy\"              #> [16] \"amoxycillin\"        \"amoxyke\"            \"anemolin\"           #> [19] \"aspenil\"            \"atoksilin\"          \"biomox\"             #> [22] \"bristamox\"          \"cemoxin\"            \"clamoxyl\"           #> [25] \"damoxy\"             \"delacillin\"         \"demoksil\"           #> [28] \"dispermox\"          \"efpenix\"            \"flemoxin\"           #> [31] \"hiconcil\"           \"histocillin\"        \"hydroxyampicillin\"  #> [34] \"ibiamox\"            \"imacillin\"          \"lamoxy\"             #> [37] \"largopen\"           \"metafarmacapsules\"  \"metifarmacapsules\"  #> [40] \"moksilin\"           \"moxacin\"            \"moxatag\"            #> [43] \"ospamox\"            \"pamoxicillin\"       \"piramox\"            #> [46] \"promoxil\"           \"remoxil\"            \"robamox\"            #> [49] \"sawamoxpm\"          \"tolodina\"           \"topramoxin\"         #> [52] \"unicillin\"          \"utimox\"             \"vetramox\"           #>  #> $loinc #>  [1] \"101498-4\" \"15-8\"     \"16-6\"     \"16365-9\"  \"17-4\"     \"18-2\"     #>  [7] \"18861-5\"  \"18862-3\"  \"19-0\"     \"20-8\"     \"21-6\"     \"22-4\"     #> [13] \"25274-2\"  \"25310-4\"  \"3344-9\"   \"55614-2\"  \"55615-9\"  \"55616-7\"  #> [19] \"6976-5\"   \"6977-3\"   \"80133-2\"  #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] 1.5 #>  #> $ddd$oral$units #> [1] \"g\" #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] 3 #>  #> $ddd$iv$units #> [1] \"g\" #>  #>  #>   # all ab_* functions use as.ab() internally, so you can go from 'any' to 'any': ab_atc(\"AMP\") #> [1] \"J01CA01\" \"S01AA19\" ab_group(\"J01CA01\") #> [1] \"Beta-lactams/penicillins\" ab_loinc(\"ampicillin\") #>  [1] \"101477-8\" \"101478-6\" \"18864-9\"  \"18865-6\"  \"20374-5\"  \"21066-6\"  #>  [7] \"23618-2\"  \"27-3\"     \"28-1\"     \"29-9\"     \"30-7\"     \"31-5\"     #> [13] \"32-3\"     \"33-1\"     \"3355-5\"   \"33562-0\"  \"33919-2\"  \"34-9\"     #> [19] \"43883-8\"  \"43884-6\"  \"6979-9\"   \"6980-7\"   \"87604-5\"  ab_name(\"21066-6\") #> [1] \"Ampicillin\" ab_name(6249) #> [1] \"Ampicillin\" ab_name(\"J01CA01\") #> [1] \"Ampicillin\"  # spelling from different languages and dyslexia are no problem ab_atc(\"ceftriaxon\") #> [1] \"J01DD04\" ab_atc(\"cephtriaxone\") #> [1] \"J01DD04\" ab_atc(\"cephthriaxone\") #> [1] \"J01DD04\" ab_atc(\"seephthriaaksone\") #> [1] \"J01DD04\"  # use set_ab_names() for renaming columns colnames(example_isolates) #>  [1] \"date\"    \"patient\" \"age\"     \"gender\"  \"ward\"    \"mo\"      \"PEN\"     #>  [8] \"OXA\"     \"FLC\"     \"AMX\"     \"AMC\"     \"AMP\"     \"TZP\"     \"CZO\"     #> [15] \"FEP\"     \"CXM\"     \"FOX\"     \"CTX\"     \"CAZ\"     \"CRO\"     \"GEN\"     #> [22] \"TOB\"     \"AMK\"     \"KAN\"     \"TMP\"     \"SXT\"     \"NIT\"     \"FOS\"     #> [29] \"LNZ\"     \"CIP\"     \"MFX\"     \"VAN\"     \"TEC\"     \"TCY\"     \"TGC\"     #> [36] \"DOX\"     \"ERY\"     \"CLI\"     \"AZM\"     \"IPM\"     \"MEM\"     \"MTR\"     #> [43] \"CHL\"     \"COL\"     \"MUP\"     \"RIF\"     colnames(set_ab_names(example_isolates)) #>  [1] \"date\"                          \"patient\"                       #>  [3] \"age\"                           \"gender\"                        #>  [5] \"ward\"                          \"mo\"                            #>  [7] \"benzylpenicillin\"              \"oxacillin\"                     #>  [9] \"flucloxacillin\"                \"amoxicillin\"                   #> [11] \"amoxicillin_clavulanic_acid\"   \"ampicillin\"                    #> [13] \"piperacillin_tazobactam\"       \"cefazolin\"                     #> [15] \"cefepime\"                      \"cefuroxime\"                    #> [17] \"cefoxitin\"                     \"cefotaxime\"                    #> [19] \"ceftazidime\"                   \"ceftriaxone\"                   #> [21] \"gentamicin\"                    \"tobramycin\"                    #> [23] \"amikacin\"                      \"kanamycin\"                     #> [25] \"trimethoprim\"                  \"trimethoprim_sulfamethoxazole\" #> [27] \"nitrofurantoin\"                \"fosfomycin\"                    #> [29] \"linezolid\"                     \"ciprofloxacin\"                 #> [31] \"moxifloxacin\"                  \"vancomycin\"                    #> [33] \"teicoplanin\"                   \"tetracycline\"                  #> [35] \"tigecycline\"                   \"doxycycline\"                   #> [37] \"erythromycin\"                  \"clindamycin\"                   #> [39] \"azithromycin\"                  \"imipenem\"                      #> [41] \"meropenem\"                     \"metronidazole\"                 #> [43] \"chloramphenicol\"               \"colistin\"                      #> [45] \"mupirocin\"                     \"rifampicin\"                    colnames(set_ab_names(example_isolates, NIT:VAN)) #>  [1] \"date\"           \"patient\"        \"age\"            \"gender\"         #>  [5] \"ward\"           \"mo\"             \"PEN\"            \"OXA\"            #>  [9] \"FLC\"            \"AMX\"            \"AMC\"            \"AMP\"            #> [13] \"TZP\"            \"CZO\"            \"FEP\"            \"CXM\"            #> [17] \"FOX\"            \"CTX\"            \"CAZ\"            \"CRO\"            #> [21] \"GEN\"            \"TOB\"            \"AMK\"            \"KAN\"            #> [25] \"TMP\"            \"SXT\"            \"nitrofurantoin\" \"fosfomycin\"     #> [29] \"linezolid\"      \"ciprofloxacin\"  \"moxifloxacin\"   \"vancomycin\"     #> [33] \"TEC\"            \"TCY\"            \"TGC\"            \"DOX\"            #> [37] \"ERY\"            \"CLI\"            \"AZM\"            \"IPM\"            #> [41] \"MEM\"            \"MTR\"            \"CHL\"            \"COL\"            #> [45] \"MUP\"            \"RIF\"            # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     set_ab_names()    # this does the same:   example_isolates %>%     rename_with(set_ab_names)    # set_ab_names() works with any AB property:   example_isolates %>%     set_ab_names(property = \"atc\")    example_isolates %>%     set_ab_names(where(is.sir)) %>%     colnames()    example_isolates %>%     set_ab_names(NIT:VAN) %>%     colnames() } #>  [1] \"date\"           \"patient\"        \"age\"            \"gender\"         #>  [5] \"ward\"           \"mo\"             \"PEN\"            \"OXA\"            #>  [9] \"FLC\"            \"AMX\"            \"AMC\"            \"AMP\"            #> [13] \"TZP\"            \"CZO\"            \"FEP\"            \"CXM\"            #> [17] \"FOX\"            \"CTX\"            \"CAZ\"            \"CRO\"            #> [21] \"GEN\"            \"TOB\"            \"AMK\"            \"KAN\"            #> [25] \"TMP\"            \"SXT\"            \"nitrofurantoin\" \"fosfomycin\"     #> [29] \"linezolid\"      \"ciprofloxacin\"  \"moxifloxacin\"   \"vancomycin\"     #> [33] \"TEC\"            \"TCY\"            \"TGC\"            \"DOX\"            #> [37] \"ERY\"            \"CLI\"            \"AZM\"            \"IPM\"            #> [41] \"MEM\"            \"MTR\"            \"CHL\"            \"COL\"            #> [45] \"MUP\"            \"RIF\"            # }"},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Custom Antimicrobials — add_custom_antimicrobials","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"add_custom_antimicrobials() can add custom antimicrobial drug names codes.","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"","code":"add_custom_antimicrobials(x)  clear_custom_antimicrobials()"},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"x data.frame resembling antibiotics data set, least containing columns \"ab\" \"name\"","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"Important: Due R works, add_custom_antimicrobials() function run every R session - added antimicrobials stored sessions thus lost R exited. two ways circumvent automate process adding antimicrobials: Method 1: Using package option AMR_custom_ab, preferred method. use method: Create data set structure antibiotics data set (containing least columns \"ab\" \"name\") save saveRDS() location choice, e.g. \"~/my_custom_ab.rds\", remote location. Set file location package option AMR_custom_ab: options(AMR_custom_ab = \"~/my_custom_ab.rds\"). can even remote file location, https URL. Since options saved R sessions, best save option .Rprofile file loaded start-R. , open .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"), add text save file:   Upon package load, file loaded run add_custom_antimicrobials() function. Method 2: Loading antimicrobial additions directly .Rprofile file. Note definitions stored user-specific R file, suboptimal workflow. use method: Edit .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"). Add text like save file:   Use clear_custom_antimicrobials() clear previously added antimicrobials.","code":"# Add custom antimicrobial codes: options(AMR_custom_ab = \"~/my_custom_ab.rds\") # Add custom antibiotic drug codes:  AMR::add_custom_antimicrobials(    data.frame(ab = \"TESTAB\",               name = \"Test Antibiotic\",               group = \"Test Group\")  )"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/add_custom_antimicrobials.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Custom Antimicrobials — add_custom_antimicrobials","text":"","code":"# \\donttest{  # returns NA and throws a warning (which is suppressed here): suppressWarnings(   as.ab(\"testab\") ) #> Class 'ab' #> [1] <NA>  # now add a custom entry - it will be considered by as.ab() and # all ab_*() functions add_custom_antimicrobials(   data.frame(     ab = \"TESTAB\",     name = \"Test Antibiotic\",     # you can add any property present in the     # 'antibiotics' data set, such as 'group':     group = \"Test Group\"   ) ) #> ℹ Added one record to the internal antibiotics data set.  # \"testab\" is now a new antibiotic: as.ab(\"testab\") #> Class 'ab' #> [1] TESTAB ab_name(\"testab\") #> [1] \"Test Antibiotic\" ab_group(\"testab\") #> [1] \"Test Group\"  ab_info(\"testab\") #> $ab #> [1] \"TESTAB\" #>  #> $cid #> [1] NA #>  #> $name #> [1] \"Test Antibiotic\" #>  #> $group #> [1] \"Test Group\" #>  #> $atc #> [1] NA #>  #> $atc_group1 #> [1] NA #>  #> $atc_group2 #> [1] NA #>  #> $tradenames #> [1] NA #>  #> $loinc #> [1] NA #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] NA #>  #> $ddd$oral$units #> [1] NA #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] NA #>  #> $ddd$iv$units #> [1] NA #>  #>  #>    # Add Co-fluampicil, which is one of the many J01CR50 codes, see # https://atcddd.fhi.no/ddd/list_of_ddds_combined_products/ add_custom_antimicrobials(   data.frame(     ab = \"COFLU\",     name = \"Co-fluampicil\",     atc = \"J01CR50\",     group = \"Beta-lactams/penicillins\"   ) ) #> ℹ Added one record to the internal antibiotics data set. ab_atc(\"Co-fluampicil\") #> [1] \"J01CR50\" ab_name(\"J01CR50\") #> [1] \"Co-fluampicil\"  # even antimicrobial selectors work x <- data.frame(   random_column = \"some value\",   coflu = as.sir(\"S\"),   ampicillin = as.sir(\"R\") ) x #>   random_column coflu ampicillin #> 1    some value     S          R x[, betalactams()] #> ℹ For betalactams() using columns 'coflu' (co-fluampicil) and #>   'ampicillin' #>   coflu ampicillin #> 1     S          R # }"},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Add Custom Microorganisms — add_custom_microorganisms","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"add_custom_microorganisms() can add custom microorganisms, non-taxonomic outcome laboratory analysis.","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"","code":"add_custom_microorganisms(x)  clear_custom_microorganisms()"},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"x data.frame resembling microorganisms data set, least containing column \"genus\" (case-insensitive)","code":""},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"function fill missing taxonomy , specific taxonomic columns missing, see Examples. Important: Due R works, add_custom_microorganisms() function run every R session - added microorganisms stored sessions thus lost R exited. two ways circumvent automate process adding microorganisms: Method 1: Using package option AMR_custom_mo, preferred method. use method: Create data set structure microorganisms data set (containing least column \"genus\") save saveRDS() location choice, e.g. \"~/my_custom_mo.rds\", remote location. Set file location package option AMR_custom_mo: options(AMR_custom_mo = \"~/my_custom_mo.rds\"). can even remote file location, https URL. Since options saved R sessions, best save option .Rprofile file loaded start-R. , open .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"), add text save file:   Upon package load, file loaded run add_custom_microorganisms() function. Method 2: Loading microorganism directly .Rprofile file. Note definitions stored user-specific R file, suboptimal workflow. use method: Edit .Rprofile file using e.g. utils::file.edit(\"~/.Rprofile\"). Add text like save file:   Use clear_custom_microorganisms() clear previously added microorganisms.","code":"# Add custom microorganism codes: options(AMR_custom_mo = \"~/my_custom_mo.rds\") # Add custom antibiotic drug codes:  AMR::add_custom_microorganisms(    data.frame(genus = \"Enterobacter\",               species = \"asburiae/cloacae\")  )"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/add_custom_microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add Custom Microorganisms — add_custom_microorganisms","text":"","code":"# \\donttest{ # a combination of species is not formal taxonomy, so # this will result in \"Enterobacter cloacae cloacae\", # since it resembles the input best: mo_name(\"Enterobacter asburiae/cloacae\") #> [1] \"Enterobacter asburiae\"  # now add a custom entry - it will be considered by as.mo() and # all mo_*() functions add_custom_microorganisms(   data.frame(     genus = \"Enterobacter\",     species = \"asburiae/cloacae\"   ) ) #> ℹ Added Enterobacter asburiae/cloacae to the internal microorganisms data #>   set.  # E. asburiae/cloacae is now a new microorganism: mo_name(\"Enterobacter asburiae/cloacae\") #> [1] \"Enterobacter asburiae/cloacae\"  # its code: as.mo(\"Enterobacter asburiae/cloacae\") #> Class 'mo' #> [1] CUSTOM1_ENTRB_ASB/  # all internal algorithms will work as well: mo_name(\"Ent asburia cloacae\") #> [1] \"Enterobacter asburiae/cloacae\"  # and even the taxonomy was added based on the genus! mo_family(\"E. asburiae/cloacae\") #> [1] \"Enterobacteriaceae\" mo_gramstain(\"Enterobacter asburiae/cloacae\") #> [1] \"Gram-negative\"  mo_info(\"Enterobacter asburiae/cloacae\") #> $mo #> [1] \"CUSTOM1_ENTRB_ASB/\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Enterobacter\" #>  #> $species #> [1] \"asburiae/cloacae\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] NA #>  #> $url #> [1] \"\" #>  #> $ref #> [1] \"Self-added, 2025\" #>  #> $snomed #> [1] NA #>  #> $lpsn #> [1] NA #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] NA #>  #> $group_members #> character(0) #>    # the function tries to be forgiving: add_custom_microorganisms(   data.frame(     GENUS = \"BACTEROIDES / PARABACTEROIDES SLASHLINE\",     SPECIES = \"SPECIES\"   ) ) #> ℹ Added Bacteroides/Parabacteroides to the internal microorganisms data #>   set. mo_name(\"BACTEROIDES / PARABACTEROIDES\") #> [1] \"Bacteroides/Parabacteroides\" mo_rank(\"BACTEROIDES / PARABACTEROIDES\") #> [1] \"genus\"  # taxonomy still works, even though a slashline genus was given as input: mo_family(\"Bacteroides/Parabacteroides\") #> [1] \"Bacteroidaceae\"   # for groups and complexes, set them as species or subspecies: add_custom_microorganisms(   data.frame(     genus = \"Citrobacter\",     species = c(\"freundii\", \"braakii complex\"),     subspecies = c(\"complex\", \"\")   ) ) #> ℹ Added Citrobacter braakii complex and Citrobacter freundii complex to the #>   internal microorganisms data set. mo_name(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"Citrobacter freundii complex\" \"Citrobacter braakii complex\"  mo_species(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"freundii complex\" \"braakii complex\"  mo_gramstain(c(\"C. freundii complex\", \"C. braakii complex\")) #> [1] \"Gram-negative\" \"Gram-negative\" # }"},{"path":"https://msberends.github.io/AMR/reference/age.html","id":null,"dir":"Reference","previous_headings":"","what":"Age in Years of Individuals — age","title":"Age in Years of Individuals — age","text":"Calculates age years based reference date, system date default.","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Age in Years of Individuals — age","text":"","code":"age(x, reference = Sys.Date(), exact = FALSE, na.rm = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Age in Years of Individuals — age","text":"x date(s), character (vectors) coerced .POSIXlt() reference reference date(s) (default today), character (vectors) coerced .POSIXlt() exact logical indicate whether age calculation exact, .e. decimals. divides number days year--date (YTD) x number days year reference (either 365 366). na.rm logical indicate whether missing values removed ... arguments passed .POSIXlt(), origin","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Age in Years of Individuals — age","text":"integer (decimals) exact = FALSE, double (decimals) otherwise","code":""},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Age in Years of Individuals — age","text":"Ages 0 returned NA warning. Ages 120 give warning. function vectorises x reference, meaning either can length 1 argument larger length.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/age.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Age in Years of Individuals — age","text":"","code":"# 10 random pre-Y2K birth dates df <- data.frame(birth_date = as.Date(\"2000-01-01\") - runif(10) * 25000)  # add ages df$age <- age(df$birth_date)  # add exact ages df$age_exact <- age(df$birth_date, exact = TRUE)  # add age at millenium switch df$age_at_y2k <- age(df$birth_date, \"2000-01-01\")  df #>    birth_date age age_exact age_at_y2k #> 1  1965-12-05  59  59.21644         34 #> 2  1980-03-01  44  44.98082         19 #> 3  1949-11-01  75  75.30959         50 #> 4  1947-02-14  78  78.02192         52 #> 5  1940-02-19  85  85.00822         59 #> 6  1988-01-10  37  37.11781         11 #> 7  1997-08-27  27  27.49041          2 #> 8  1978-01-26  47  47.07397         21 #> 9  1972-06-17  52  52.68493         27 #> 10 1986-08-10  38  38.53699         13"},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":null,"dir":"Reference","previous_headings":"","what":"Split Ages into Age Groups — age_groups","title":"Split Ages into Age Groups — age_groups","text":"Split ages age groups defined split argument. allows easier demographic (antimicrobial resistance) analysis.","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Split Ages into Age Groups — age_groups","text":"","code":"age_groups(x, split_at = c(12, 25, 55, 75), na.rm = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Split Ages into Age Groups — age_groups","text":"x age, e.g. calculated age() split_at values split x - default age groups 0-11, 12-24, 25-54, 55-74 75+. See Details. na.rm logical indicate whether missing values removed","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Split Ages into Age Groups — age_groups","text":"Ordered factor","code":""},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Split Ages into Age Groups — age_groups","text":"split ages, input split_at argument can : numeric vector. value e.g. c(10, 20) split x 0-9, 10-19 20+. value 50 split x 0-49 50+. default split young children (0-11), youth (12-24), young adults (25-54), middle-aged adults (55-74) elderly (75+). character: \"children\" \"kids\", equivalent : c(0, 1, 2, 4, 6, 13, 18). split 0, 1, 2-3, 4-5, 6-12, 13-17 18+. \"elderly\" \"seniors\", equivalent : c(65, 75, 85). split 0-64, 65-74, 75-84, 85+. \"fives\", equivalent : 1:20 * 5. split 0-4, 5-9, ..., 95-99, 100+. \"tens\", equivalent : 1:10 * 10. split 0-9, 10-19, ..., 90-99, 100+.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/age_groups.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Split Ages into Age Groups — age_groups","text":"","code":"ages <- c(3, 8, 16, 54, 31, 76, 101, 43, 21)  # split into 0-49 and 50+ age_groups(ages, 50) #> [1] 0-49 0-49 0-49 50+  0-49 50+  50+  0-49 0-49 #> Levels: 0-49 < 50+  # split into 0-19, 20-49 and 50+ age_groups(ages, c(20, 50)) #> [1] 0-19  0-19  0-19  50+   20-49 50+   50+   20-49 20-49 #> Levels: 0-19 < 20-49 < 50+  # split into groups of ten years age_groups(ages, 1:10 * 10) #> [1] 0-9   0-9   10-19 50-59 30-39 70-79 100+  40-49 20-29 #> 11 Levels: 0-9 < 10-19 < 20-29 < 30-39 < 40-49 < 50-59 < 60-69 < ... < 100+ age_groups(ages, split_at = \"tens\") #> [1] 0-9   0-9   10-19 50-59 30-39 70-79 100+  40-49 20-29 #> 11 Levels: 0-9 < 10-19 < 20-29 < 30-39 < 40-49 < 50-59 < 60-69 < ... < 100+  # split into groups of five years age_groups(ages, 1:20 * 5) #> [1] 0-4   5-9   15-19 50-54 30-34 75-79 100+  40-44 20-24 #> 21 Levels: 0-4 < 5-9 < 10-14 < 15-19 < 20-24 < 25-29 < 30-34 < ... < 100+ age_groups(ages, split_at = \"fives\") #> [1] 0-4   5-9   15-19 50-54 30-34 75-79 100+  40-44 20-24 #> 21 Levels: 0-4 < 5-9 < 10-14 < 15-19 < 20-24 < 25-29 < 30-34 < ... < 100+  # split specifically for children age_groups(ages, c(1, 2, 4, 6, 13, 18)) #> [1] 2-3   6-12  13-17 18+   18+   18+   18+   18+   18+   #> Levels: 0 < 1 < 2-3 < 4-5 < 6-12 < 13-17 < 18+ age_groups(ages, \"children\") #> [1] 2-3   6-12  13-17 18+   18+   18+   18+   18+   18+   #> Levels: 0 < 1 < 2-3 < 4-5 < 6-12 < 13-17 < 18+  # \\donttest{ # resistance of ciprofloxacin per age group if (require(\"dplyr\") && require(\"ggplot2\")) {   example_isolates %>%     filter_first_isolate() %>%     filter(mo == as.mo(\"Escherichia coli\")) %>%     group_by(age_group = age_groups(age)) %>%     select(age_group, CIP) %>%     ggplot_sir(       x = \"age_group\",       minimum = 0,       x.title = \"Age Group\",       title = \"Ciprofloxacin resistance per age group\"     ) } #> Loading required package: ggplot2  # }"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Create detailed antibiograms options traditional, combination, syndromic, Bayesian WISCA methods. Adhering previously described approaches (see Source) especially Bayesian WISCA model (Weighted-Incidence Syndromic Combination Antibiogram) Bielicki et al., functions provide flexible output formats including plots tables, ideal integration R Markdown Quarto reports.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"","code":"antibiogram(x, antibiotics = where(is.sir), mo_transform = \"shortname\",   ab_transform = \"name\", syndromic_group = NULL, add_total_n = FALSE,   only_all_tested = FALSE, digits = ifelse(wisca, 1, 0),   formatting_type = getOption(\"AMR_antibiogram_formatting_type\", 14),   col_mo = NULL, language = get_AMR_locale(), minimum = 30,   combine_SI = TRUE, sep = \" + \", wisca = FALSE, simulations = 1000,   conf_interval = 0.95, interval_side = \"two-tailed\",   info = interactive())  wisca(x, antibiotics = where(is.sir), ab_transform = \"name\",   syndromic_group = NULL, add_total_n = FALSE, only_all_tested = FALSE,   digits = 1,   formatting_type = getOption(\"AMR_antibiogram_formatting_type\", 14),   col_mo = NULL, language = get_AMR_locale(), minimum = 30,   combine_SI = TRUE, sep = \" + \", simulations = 1000,   conf_interval = 0.95, interval_side = \"two-tailed\",   info = interactive())  retrieve_wisca_parameters(wisca_model, ...)  # S3 method for class 'antibiogram' plot(x, ...)  # S3 method for class 'antibiogram' autoplot(object, ...)  # S3 method for class 'antibiogram' knit_print(x, italicise = TRUE,   na = getOption(\"knitr.kable.NA\", default = \"\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Bielicki JA et al. (2016). Selecting appropriate empirical antibiotic regimens paediatric bloodstream infections: application Bayesian decision model local pooled antimicrobial resistance surveillance data Journal Antimicrobial Chemotherapy 71(3); doi:10.1093/jac/dkv397 Bielicki JA et al. (2020). Evaluation coverage 3 antibiotic regimens neonatal sepsis hospital setting across Asian countries JAMA Netw Open. 3(2):e1921124; doi:10.1001.jamanetworkopen.2019.21124 Klinker KP et al. (2021). Antimicrobial stewardship antibiograms: importance moving beyond traditional antibiograms. Therapeutic Advances Infectious Disease, May 5;8:20499361211011373; doi:10.1177/20499361211011373 Barbieri E et al. (2021). Development Weighted-Incidence Syndromic Combination Antibiogram (WISCA) guide choice empiric antibiotic treatment urinary tract infection paediatric patients: Bayesian approach Antimicrobial Resistance & Infection Control May 1;10(1):74; doi:10.1186/s13756-021-00939-2 M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"x data.frame containing least column microorganisms columns antimicrobial results (class 'sir', see .sir()) antibiotics vector antimicrobial name code (evaluated .ab(), column name x, (combinations ) antimicrobial selectors aminoglycosides() carbapenems(). combination antibiograms, can also set values separated \"+\", \"TZP+TOB\" \"cipro + genta\", given columns resembling antimicrobials exist x. See Examples. mo_transform character transform microorganism input - must \"name\", \"shortname\" (default), \"gramstain\", one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". Can also NULL transform input NA consider microorganisms 'unknown'. ab_transform character transform antimicrobial input - must one column names antibiotics data set (defaults \"name\"): \"ab\", \"cid\", \"name\", \"group\", \"atc\", \"atc_group1\", \"atc_group2\", \"abbreviations\", \"synonyms\", \"oral_ddd\", \"oral_units\", \"iv_ddd\", \"iv_units\", \"loinc\". Can also NULL transform input. syndromic_group column name x, values calculated split rows x, e.g. using ifelse() case_when(). See Examples. add_total_n logical indicate whether n_tested available numbers per pathogen added table (default TRUE). add lowest highest number available isolates per antimicrobial (e.g, E. coli 200 isolates available ciprofloxacin 150 amoxicillin, returned number \"150-200\"). option unavailable wisca = TRUE; case, use retrieve_wisca_parameters() get parameters used WISCA. only_all_tested (combination antibiograms): logical indicate isolates must tested antimicrobials, see Details digits number digits use rounding antimicrobial coverage, defaults 1 WISCA 0 otherwise formatting_type numeric value (1–22 WISCA, 1-12 non-WISCA) indicating 'cells' antibiogram table formatted. See Details > Formatting Type list options. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). language language translate text, defaults system language (see get_AMR_locale()) minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether susceptibility determined results either S, SDD, , instead S (default TRUE) sep separating character antimicrobial columns combination antibiograms wisca logical indicate whether Weighted-Incidence Syndromic Combination Antibiogram (WISCA) must generated (default FALSE). use Bayesian decision model estimate regimen coverage probabilities using Monte Carlo simulations. Set simulations, conf_interval, interval_side adjust. simulations (WISCA) numerical value set number Monte Carlo simulations conf_interval (WISCA) numerical value set confidence interval (default 0.95) interval_side (WISCA) side confidence interval, either \"two-tailed\" (default), \"left\" \"right\" info logical indicate info printed - default TRUE interactive mode wisca_model outcome wisca() antibiogram(..., wisca = TRUE) ... used R Markdown Quarto: arguments passed knitr::kable() (otherwise, use) object antibiogram() object italicise logical indicate whether microorganism names knitr table made italic, using italicise_taxonomy(). na character use showing NA values","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"functions return table values 0 100 susceptibility, resistance. Remember filter data let contain first isolates! needed exclude duplicates reduce selection bias. Use first_isolate() determine one four available algorithms: isolate-based, patient-based, episode-based, phenotype-based. estimating antimicrobial coverage, especially creating WISCA, outcome might become reliable including top n species encountered data. can filter top n using top_n_microorganisms(). example, use top_n_microorganisms(your_data, n = 10) pre-processing step include top 10 species data. numeric values antibiogram stored long format attribute long_numeric. can retrieve using attributes(x)$long_numeric, x outcome antibiogram() wisca(). ideal e.g. advanced plotting.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"formatting-type","dir":"Reference","previous_headings":"","what":"Formatting Type","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"formatting 'cells' table can set argument formatting_type. examples, 5 antimicrobial coverage (4-6 indicates confidence level), 15 number susceptible isolates, 300 number tested (.e., available) isolates: 5 15 300 15/300 5 (300) 5% (300) 5 (N=300) 5% (N=300) 5 (15/300) 5% (15/300) 5 (N=15/300) 5% (N=15/300) 5 (4-6) 5% (4-6%) - default 5 (4-6,300) 5% (4-6%,300) 5 (4-6,N=300) 5% (4-6%,N=300) 5 (4-6,15/300) 5% (4-6%,15/300) 5 (4-6,N=15/300) 5% (4-6%,N=15/300) default 14, can set globally package option AMR_antibiogram_formatting_type, e.g. options(AMR_antibiogram_formatting_type = 5). note WISCA, total numbers tested susceptible isolates less useful report, since included Bayesian model apparent susceptibility confidence level. Set digits (defaults 0) alter rounding susceptibility percentages.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"antibiogram-types","dir":"Reference","previous_headings":"","what":"Antibiogram Types","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"various antibiogram types, summarised Klinker et al. (2021, doi:10.1177/20499361211011373 ), supported antibiogram(). clinical coverage estimations, use WISCA whenever possible, since provides precise coverage estimates accounting pathogen incidence antimicrobial susceptibility, shown Bielicki et al. (2020, doi:10.1001.jamanetworkopen.2019.21124 ). See section Explaining WISCA page. note WISCA pathogen-agnostic, meaning outcome stratied pathogen, rather syndrome. Traditional Antibiogram Case example: Susceptibility Pseudomonas aeruginosa piperacillin/tazobactam (TZP) Code example:   Combination Antibiogram Case example: Additional susceptibility Pseudomonas aeruginosa TZP + tobramycin versus TZP alone Code example:   Syndromic Antibiogram Case example: Susceptibility Pseudomonas aeruginosa TZP among respiratory specimens (obtained among ICU patients ) Code example:   Weighted-Incidence Syndromic Combination Antibiogram (WISCA) WISCA can applied antibiogram, see section Explaining WISCA page information. Code example:   WISCA uses sophisticated Bayesian decision model combine local pooled antimicrobial resistance data. approach evaluates local patterns can also draw multi-centre datasets improve regimen accuracy, even low-incidence infections like paediatric bloodstream infections (BSIs).","code":"antibiogram(your_data,             antibiotics = \"TZP\") antibiogram(your_data,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\")) antibiogram(your_data,             antibiotics = penicillins(),             syndromic_group = \"ward\") antibiogram(your_data,             antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),             wisca = TRUE)  # this is equal to: wisca(your_data,       antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"grouped-tibbles","dir":"Reference","previous_headings":"","what":"Grouped tibbles","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"type antibiogram, grouped tibbles can also used calculate susceptibilities various groups. Code example:","code":"library(dplyr) your_data %>%   group_by(has_sepsis, is_neonate, sex) %>%   wisca(antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"))"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"stepped-approach-for-clinical-insight","dir":"Reference","previous_headings":"","what":"Stepped Approach for Clinical Insight","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"clinical practice, antimicrobial coverage decisions evolve microbiological data becomes available. theoretical stepped approach ensures empirical coverage can continuously assessed improve patient outcomes: Initial Empirical Therapy (Admission / Pre-Culture Data) admission, pathogen information available. Action: broad-spectrum coverage based local resistance patterns syndromic antibiograms. Using pathogen-agnostic yet incidence-weighted WISCA preferred. Code example:   Refinement Gram Stain Results blood culture becomes positive, Gram stain provides initial crucial first stratification (Gram-positive vs. Gram-negative). Action: narrow coverage based Gram stain-specific resistance patterns. Code example:   Definitive Therapy Based Species Identification cultivation pathogen, full pathogen identification allows precise targeting therapy. Action: adjust treatment pathogen-specific antibiograms, minimizing resistance risks. Code example:   structuring antibiograms around stepped approach, clinicians can make data-driven adjustments stage, ensuring optimal empirical targeted therapy reducing unnecessary broad-spectrum antimicrobial use.","code":"antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = NA) # all pathogens set to `NA`  # preferred: use WISCA wisca(your_data,       antibiotics = selected_regimens) antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = \"gramstain\") # all pathogens set to Gram-pos/Gram-neg antibiogram(your_data,             antibiotics = selected_regimens,             mo_transform = \"shortname\") # all pathogens set to 'G. species', e.g., E. coli"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"inclusion-in-combination-antibiograms","dir":"Reference","previous_headings":"","what":"Inclusion in Combination Antibiograms","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Note combination antibiograms, important realise susceptibility can calculated two ways, can set only_all_tested argument (default FALSE). See example two antimicrobials, Drug Drug B, antibiogram() works calculate %SI:","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - --------------------------------------------------------------------"},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"plotting","dir":"Reference","previous_headings":"","what":"Plotting","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"types antibiograms listed can plotted (using ggplot2::autoplot() base R's plot() barplot()). mentioned , numeric values antibiogram stored long format attribute long_numeric. can retrieve using attributes(x)$long_numeric, x outcome antibiogram() wisca(). outcome antibiogram() can also used directly R Markdown / Quarto (.e., knitr) reports. case, knitr::kable() applied automatically microorganism names even printed italics default (see argument italicise). can also use functions specific 'table reporting' packages transform output antibiogram() needs, e.g. flextable::as_flextable() gt::gt().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"explaining-wisca","dir":"Reference","previous_headings":"","what":"Explaining WISCA","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"WISCA, outlined Bielicki et al. (doi:10.1093/jac/dkv397 ), stands Weighted-Incidence Syndromic Combination Antibiogram, estimates probability adequate empirical antimicrobial regimen coverage specific infection syndromes. method leverages Bayesian decision model random effects pathogen incidence susceptibility, enabling robust estimates presence sparse data. Bayesian model assumes conjugate priors parameter estimation. example, coverage probability \\(\\theta\\) given antimicrobial regimen modelled using Beta distribution prior: $$\\theta \\sim \\text{Beta}(\\alpha_0, \\beta_0)$$ \\(\\alpha_0\\) \\(\\beta_0\\) represent prior successes failures, respectively, informed expert knowledge weakly informative priors (e.g., \\(\\alpha_0 = 1, \\beta_0 = 1\\)). likelihood function constructed based observed data, number covered cases regimen follows binomial distribution: $$y \\sim \\text{Binomial}(n, \\theta)$$ Posterior parameter estimates obtained combining prior likelihood using Bayes' theorem. posterior distribution \\(\\theta\\) also Beta distribution: $$\\theta | y \\sim \\text{Beta}(\\alpha_0 + y, \\beta_0 + n - y)$$ Pathogen incidence, representing proportion infections caused different pathogens, modelled using Dirichlet distribution, natural conjugate prior multinomial outcomes. Dirichlet distribution parameterised vector concentration parameters \\(\\alpha\\), \\(\\alpha_i\\) corresponds specific pathogen. prior typically chosen uniform (\\(\\alpha_i = 1\\)), reflecting assumption equal prior probability across pathogens. posterior distribution pathogen incidence given : $$\\text{Dirichlet}(\\alpha_1 + n_1, \\alpha_2 + n_2, \\dots, \\alpha_K + n_K)$$ \\(n_i\\) number infections caused pathogen \\(\\) observed data. practical implementation, pathogen incidences sampled posterior using normalised Gamma-distributed random variables: $$x_i \\sim \\text{Gamma}(\\alpha_i + n_i, 1)$$ $$p_i = \\frac{x_i}{\\sum_{j=1}^K x_j}$$ \\(x_i\\) represents unnormalised pathogen counts, \\(p_i\\) normalised proportion pathogen \\(\\). hierarchical modelling, pathogen-level effects (e.g., differences resistance patterns) regimen-level effects modelled using Gaussian priors log-odds. hierarchical structure ensures partial pooling estimates across groups, improving stability strata small sample sizes. model implemented using Hamiltonian Monte Carlo (HMC) sampling. Stratified results can provided based covariates age, sex, clinical complexity (e.g., prior antimicrobial treatments renal/urological comorbidities) using dplyr's group_by() pre-processing step running wisca(). Posterior odds ratios (ORs) derived quantify effect covariates coverage probabilities: $$\\text{}_{\\text{covariate}} = \\frac{\\exp(\\beta_{\\text{covariate}})}{\\exp(\\beta_0)}$$ combining empirical data prior knowledge, WISCA overcomes limitations traditional combination antibiograms, offering disease-specific, patient-stratified estimates robust uncertainty quantification. tool invaluable antimicrobial stewardship programs empirical treatment guideline refinement. Note: WISCA never gives output pathogen/species level, incidences susceptibilities already weighted species.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"Implementation: Dr. Larisse Bolton Dr. Matthijs Berends","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiogram.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Generate Traditional, Combination, Syndromic, or WISCA Antibiograms — antibiogram","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # \\donttest{ # Traditional antibiogram ----------------------------------------------  antibiogram(example_isolates,   antibiotics = c(aminoglycosides(), carbapenems()) ) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 10 × 7 #>    Pathogen       Amikacin    Gentamicin Imipenem Kanamycin Meropenem Tobramycin #>  * <chr>          <chr>       <chr>      <chr>    <chr>     <chr>     <chr>      #>  1 CoNS           0% (0-8%)   86% (82-9… 52% (37… 0% (0-8%) 52% (37-… 22% (12-3… #>  2 E. coli        100% (98-1… 98% (96-9… 100% (9… NA        100% (99… 97% (96-9… #>  3 E. faecalis    0% (0-9%)   0% (0-9%)  100% (9… 0% (0-9%) NA        0% (0-9%)  #>  4 K. pneumoniae  NA          90% (79-9… 100% (9… NA        100% (93… 90% (79-9… #>  5 P. aeruginosa  NA          100% (88-… NA       0% (0-12… NA        100% (88-… #>  6 P. mirabilis   NA          94% (80-9… 94% (79… NA        NA        94% (80-9… #>  7 S. aureus      NA          99% (97-1… NA       NA        NA        98% (92-1… #>  8 S. epidermidis 0% (0-8%)   79% (71-8… NA       0% (0-8%) NA        51% (40-6… #>  9 S. hominis     NA          92% (84-9… NA       NA        NA        85% (74-9… #> 10 S. pneumoniae  0% (0-3%)   0% (0-3%)  NA       0% (0-3%) NA        0% (0-3%)  #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  antibiogram(example_isolates,   antibiotics = aminoglycosides(),   ab_transform = \"atc\",   mo_transform = \"gramstain\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # An Antibiogram (non-WISCA / 95% CI): 2 × 5 #>   Pathogen      J01GB01      J01GB03      J01GB04    J01GB06      #> * <chr>         <chr>        <chr>        <chr>      <chr>        #> 1 Gram-negative 96% (94-97%) 96% (95-98%) 0% (0-10%) 98% (96-99%) #> 2 Gram-positive 34% (31-38%) 63% (60-66%) 0% (0-1%)  0% (0-1%)    #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  antibiogram(example_isolates,   antibiotics = carbapenems(),   ab_transform = \"name\",   mo_transform = \"name\") #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 5 × 3 #>   Pathogen                                 Imipenem       Meropenem      #> * <chr>                                    <chr>          <chr>          #> 1 Coagulase-negative Staphylococcus (CoNS) 52% (37-67%)   52% (37-67%)   #> 2 Enterococcus faecalis                    100% (91-100%) NA             #> 3 Escherichia coli                         100% (99-100%) 100% (99-100%) #> 4 Klebsiella pneumoniae                    100% (93-100%) 100% (93-100%) #> 5 Proteus mirabilis                        94% (79-99%)   NA             #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Combined antibiogram -------------------------------------------------  # combined antibiotics yield higher empiric coverage antibiogram(example_isolates,   antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),   mo_transform = \"gramstain\") #> # An Antibiogram (non-WISCA / 95% CI): 2 × 4 #>   Pathogen  Piperacillin/tazobac…¹ Piperacillin/tazobac…² Piperacillin/tazobac…³ #> * <chr>     <chr>                  <chr>                  <chr>                  #> 1 Gram-neg… 88% (85-91%)           99% (97-99%)           98% (97-99%)           #> 2 Gram-pos… 86% (82-89%)           98% (96-98%)           95% (93-97%)           #> # ℹ abbreviated names: ¹​`Piperacillin/tazobactam`, #> #   ²​`Piperacillin/tazobactam + Gentamicin`, #> #   ³​`Piperacillin/tazobactam + Tobramycin` #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  # names of antibiotics do not need to resemble columns exactly: antibiogram(example_isolates,   antibiotics = c(\"Cipro\", \"cipro + genta\"),   mo_transform = \"gramstain\",   ab_transform = \"name\",   sep = \" & \") #> # An Antibiogram (non-WISCA / 95% CI): 2 × 3 #>   Pathogen      Ciprofloxacin `Ciprofloxacin & Gentamicin` #> * <chr>         <chr>         <chr>                        #> 1 Gram-negative 91% (88-93%)  99% (97-99%)                 #> 2 Gram-positive 77% (74-80%)  93% (91-94%)                 #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Syndromic antibiogram ------------------------------------------------  # the data set could contain a filter for e.g. respiratory specimens antibiogram(example_isolates,   antibiotics = c(aminoglycosides(), carbapenems()),   syndromic_group = \"ward\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # An Antibiogram (non-WISCA / 95% CI): 14 × 8 #>    `Syndromic Group` Pathogen   Amikacin Gentamicin Imipenem Kanamycin Meropenem #>  * <chr>             <chr>      <chr>    <chr>      <chr>    <chr>     <chr>     #>  1 Clinical          CoNS       NA       89% (84-9… 57% (39… NA        57% (39-… #>  2 ICU               CoNS       NA       79% (68-8… NA       NA        NA        #>  3 Outpatient        CoNS       NA       84% (66-9… NA       NA        NA        #>  4 Clinical          E. coli    100% (9… 98% (96-9… 100% (9… NA        100% (99… #>  5 ICU               E. coli    100% (9… 99% (95-1… 100% (9… NA        100% (97… #>  6 Clinical          K. pneumo… NA       92% (81-9… 100% (9… NA        100% (92… #>  7 Clinical          P. mirabi… NA       100% (88-… NA       NA        NA        #>  8 Clinical          S. aureus  NA       99% (95-1… NA       NA        NA        #>  9 ICU               S. aureus  NA       100% (95-… NA       NA        NA        #> 10 Clinical          S. epider… NA       82% (72-9… NA       NA        NA        #> 11 ICU               S. epider… NA       72% (60-8… NA       NA        NA        #> 12 Clinical          S. hominis NA       96% (85-9… NA       NA        NA        #> 13 Clinical          S. pneumo… 0% (0-5… 0% (0-5%)  NA       0% (0-5%) NA        #> 14 ICU               S. pneumo… 0% (0-1… 0% (0-12%) NA       0% (0-12… NA        #> # ℹ 1 more variable: Tobramycin <chr> #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram  # now define a data set with only E. coli ex1 <- example_isolates[which(mo_genus() == \"Escherichia\"), ] #> ℹ Using column 'mo' as input for mo_genus()  # with a custom language, though this will be determined automatically # (i.e., this table will be in Spanish on Spanish systems) antibiogram(ex1,   antibiotics = aminoglycosides(),   ab_transform = \"name\",   syndromic_group = ifelse(ex1$ward == \"ICU\",     \"UCI\", \"No UCI\"   ),   language = \"es\") #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # An Antibiogram (non-WISCA / 95% CI): 2 × 5 #>   `Grupo sindrómico` Patógeno Amikacina      Gentamicina   Tobramicina  #> * <chr>              <chr>    <chr>          <chr>         <chr>        #> 1 No UCI             E. coli  100% (97-100%) 98% (96-99%)  98% (96-99%) #> 2 UCI                E. coli  100% (93-100%) 99% (95-100%) 96% (92-99%) #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # WISCA antibiogram ----------------------------------------------------  # WISCA are not stratified by species, but rather on syndromes antibiogram(example_isolates,   antibiotics = c(\"TZP\", \"TZP+TOB\", \"TZP+GEN\"),   syndromic_group = \"ward\",   wisca = TRUE) #> Warning: WISCA must be based on the species level as WISCA parameters are based on #> this. For that reason, mo_transform will be ignored. #> # An Antibiogram (WISCA / 95% CI): 3 × 4 #>   `Syndromic Group` `Piperacillin/tazobactam` Piperacillin/tazobactam + Gentam…¹ #> * <chr>             <chr>                     <chr>                              #> 1 Clinical          78% (70-85.4%)            98.3% (94.9-99.7%)                 #> 2 ICU               90.6% (84.4-94.9%)        97.6% (92.7-99.7%)                 #> 3 Outpatient        92.2% (79.7-99%)          94.7% (82.7-99.5%)                 #> # ℹ abbreviated name: ¹​`Piperacillin/tazobactam + Gentamicin` #> # ℹ 1 more variable: `Piperacillin/tazobactam + Tobramycin` <chr> #> # Use `plot()` or `ggplot2::autoplot()` to create a plot of this antibiogram, #> # or use it directly in R Markdown or https://quarto.org, see ?antibiogram   # Print the output for R Markdown / Quarto -----------------------------  ureido <- antibiogram(example_isolates,   antibiotics = ureidopenicillins(),   syndromic_group = \"ward\",   wisca = TRUE) #> Warning: WISCA must be based on the species level as WISCA parameters are based on #> this. For that reason, mo_transform will be ignored. #> ℹ For ureidopenicillins() using column 'TZP' (piperacillin/tazobactam)  # in an Rmd file, you would just need to return `ureido` in a chunk, # but to be explicit here: if (requireNamespace(\"knitr\")) {   cat(knitr::knit_print(ureido)) } #>  #>  #> |Syndromic Group |Piperacillin/tazobactam | #> |:---------------|:-----------------------| #> |Clinical        |78% (70.1-85.7%)        | #> |ICU             |90.4% (83.6-95%)        | #> |Outpatient      |91.8% (78.5-98.9%)      |   # Generate plots with ggplot2 or base R --------------------------------  ab1 <- antibiogram(example_isolates,   antibiotics = c(\"AMC\", \"CIP\", \"TZP\", \"TZP+TOB\"),   mo_transform = \"gramstain\") ab2 <- antibiogram(example_isolates,   antibiotics = c(\"AMC\", \"CIP\", \"TZP\", \"TZP+TOB\"),   mo_transform = \"gramstain\",   syndromic_group = \"ward\")  if (requireNamespace(\"ggplot2\")) {   ggplot2::autoplot(ab1) }  if (requireNamespace(\"ggplot2\")) {   ggplot2::autoplot(ab2) }   plot(ab1)  plot(ab2)  # }"},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Sets with 607 Antimicrobial Drugs — antibiotics","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Two data sets containing antibiotics/antimycotics antivirals. Use .ab() one ab_* functions retrieve values antibiotics data set. Three identifiers included data set: antibiotic ID (ab, primarily used package) defined WHONET/EARS-Net, ATC code (atc) defined , Compound ID (cid) found PubChem. properties data set derived one codes. Note drugs multiple ATC codes.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"","code":"antibiotics  antivirals"},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"object class tbl_df (inherits tbl, data.frame) 120 rows 11 columns.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"for-the-antibiotics-data-set-a-tibble-with-observations-and-variables-","dir":"Reference","previous_headings":"","what":"For the antibiotics data set: a tibble with 487 observations and 14 variables:","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"ab Antibiotic ID used package (AMC), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available. unique identifier. cid Compound ID found PubChem. unique identifier. name Official name used WHONET/EARS-Net . unique identifier. group short concise group name, based WHONET WHOCC definitions atc ATC codes (Anatomical Therapeutic Chemical) defined WHOCC, like J01CR02 atc_group1 Official pharmacological subgroup (3rd level ATC code) defined WHOCC, like \"Macrolides, lincosamides streptogramins\" atc_group2 Official chemical subgroup (4th level ATC code) defined WHOCC, like \"Macrolides\" abbr List abbreviations used many countries, also antibiotic susceptibility testing (AST) synonyms Synonyms (often trade names) drug, found PubChem based compound ID oral_ddd Defined Daily Dose (DDD), oral treatment, currently available 179 drugs oral_units Units oral_ddd iv_ddd Defined Daily Dose (DDD), parenteral (intravenous) treatment, currently available 153 drugs iv_units Units iv_ddd loinc codes associated name antimicrobial drug Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Use ab_loinc() retrieve quickly, see ab_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"for-the-antivirals-data-set-a-tibble-with-observations-and-variables-","dir":"Reference","previous_headings":"","what":"For the antivirals data set: a tibble with 120 observations and 11 variables:","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"av Antiviral ID used package (ACI), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available. unique identifier. Combinations codes contain + indicate , ATA+COBI atazanavir/cobicistat. name Official name used WHONET/EARS-Net . unique identifier. atc ATC codes (Anatomical Therapeutic Chemical) defined WHOCC cid Compound ID found PubChem. unique identifier. atc_group Official pharmacological subgroup (3rd level ATC code) defined WHOCC synonyms Synonyms (often trade names) drug, found PubChem based compound ID oral_ddd Defined Daily Dose (DDD), oral treatment oral_units Units oral_ddd iv_ddd Defined Daily Dose (DDD), parenteral treatment iv_units Units iv_ddd loinc codes associated name antiviral drug Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Use av_loinc() retrieve quickly, see av_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology (WHOCC): https://atcddd.fhi./atc_ddd_index/ Logical Observation Identifiers Names Codes (LOINC), Version 2.76 (18 September, 2023). Accessed https://loinc.org October 19th, 2023. European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Properties based ATC code available ATC available. properties : atc_group1, atc_group2, oral_ddd, oral_units, iv_ddd iv_units. Synonyms (.e. trade names) derived PubChem Compound ID (column cid) consequently available CID available.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"Like data sets package, data sets publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/antibiotics.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Sets with 607 Antimicrobial Drugs — antibiotics","text":"","code":"antibiotics #> # A tibble: 487 × 14 #>    ab        cid name   group atc   atc_group1 atc_group2 abbreviations synonyms #>    <ab>    <dbl> <chr>  <chr> <lis> <chr>      <chr>      <list>        <named > #>  1 AMA      4649 4-ami… Anti… <chr> Drugs for… Aminosali… <chr [1]>     <chr>    #>  2 ACM   6450012 Acety… Macr… <chr> NA         NA         <chr [1]>     <chr>    #>  3 ASP  49787020 Acety… Macr… <chr> NA         NA         <chr [1]>     <chr>    #>  4 ALS      8954 Aldes… Othe… <chr> Drugs for… Drugs for… <chr [1]>     <chr>    #>  5 AMK     37768 Amika… Amin… <chr> Aminoglyc… Other ami… <chr [5]>     <chr>    #>  6 AKF        NA Amika… Amin… <chr> NA         NA         <chr [1]>     <chr>    #>  7 AMO     54260 Amoro… Anti… <chr> Antifunga… Other ant… <chr [1]>     <chr>    #>  8 AMX     33613 Amoxi… Beta… <chr> Beta-lact… Penicilli… <chr [3]>     <chr>    #>  9 AMC  23665637 Amoxi… Beta… <chr> Beta-lact… Combinati… <chr [5]>     <chr>    #> 10 AXS    465441 Amoxi… Beta… <chr> NA         NA         <chr [1]>     <chr>    #> # ℹ 477 more rows #> # ℹ 5 more variables: oral_ddd <dbl>, oral_units <chr>, iv_ddd <dbl>, #> #   iv_units <chr>, loinc <list> antivirals #> # A tibble: 120 × 11 #>    av       name      atc      cid atc_group synonyms oral_ddd oral_units iv_ddd #>    <av>     <chr>     <chr>  <dbl> <chr>     <list>      <dbl> <chr>       <dbl> #>  1 ABA      Abacavir  J05A… 4.41e5 Nucleosi… <chr>         0.6 g              NA #>  2 ACI      Aciclovir J05A… 1.35e8 Nucleosi… <chr>         4   g               4 #>  3 ADD      Adefovir… J05A… 6.09e4 Nucleosi… <chr>        10   mg             NA #>  4 AME      Amenamev… J05A… 1.14e7 Other an… <chr>         0.4 g              NA #>  5 AMP      Amprenav… J05A… 6.50e4 Protease… <chr>         1.2 g              NA #>  6 ASU      Asunapre… J05A… 1.61e7 Antivira… <chr>         0.2 g              NA #>  7 ATA      Atazanav… J05A… 1.48e5 Protease… <chr>         0.3 g              NA #>  8 ATA+COBI Atazanav… J05A… 8.66e7 Antivira… <chr>        NA   NA             NA #>  9 ATA+RIT  Atazanav… J05A… 2.51e7 Antivira… <chr>         0.3 g              NA #> 10 BAM      Baloxavi… J05A… 1.24e8 Other an… <chr>        40   mg             NA #> # ℹ 110 more rows #> # ℹ 2 more variables: iv_units <chr>, loinc <list>"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":null,"dir":"Reference","previous_headings":"","what":"Antimicrobial Selectors — antimicrobial_selectors","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"functions allow filtering rows selecting columns based antimicrobial test results specific antimicrobial class group, without need define columns antimicrobial abbreviations. short, column name resembles antimicrobial drug, picked functions matches pharmaceutical class: \"cefazolin\", \"kefzol\", \"CZO\" \"J01DB04\" picked using:","code":"library(dplyr) my_data_with_all_these_columns %>%   select(cephalosporins())"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"","code":"amr_class(amr_class, only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  amr_selector(filter, only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  aminoglycosides(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  aminopenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  antifungals(only_sir_columns = FALSE, return_all = TRUE, ...)  antimycobacterials(only_sir_columns = FALSE, return_all = TRUE, ...)  betalactams(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  betalactams_with_inhibitor(only_sir_columns = FALSE, return_all = TRUE,   ...)  carbapenems(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  cephalosporins(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_1st(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_2nd(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_3rd(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_4th(only_sir_columns = FALSE, return_all = TRUE, ...)  cephalosporins_5th(only_sir_columns = FALSE, return_all = TRUE, ...)  fluoroquinolones(only_sir_columns = FALSE, return_all = TRUE, ...)  glycopeptides(only_sir_columns = FALSE, return_all = TRUE, ...)  isoxazolylpenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  lincosamides(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  lipoglycopeptides(only_sir_columns = FALSE, return_all = TRUE, ...)  macrolides(only_sir_columns = FALSE, return_all = TRUE, ...)  monobactams(only_sir_columns = FALSE, return_all = TRUE, ...)  nitrofurans(only_sir_columns = FALSE, return_all = TRUE, ...)  oxazolidinones(only_sir_columns = FALSE, return_all = TRUE, ...)  penicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  phenicols(only_sir_columns = FALSE, return_all = TRUE, ...)  polymyxins(only_sir_columns = FALSE, only_treatable = TRUE,   return_all = TRUE, ...)  quinolones(only_sir_columns = FALSE, return_all = TRUE, ...)  rifamycins(only_sir_columns = FALSE, return_all = TRUE, ...)  streptogramins(only_sir_columns = FALSE, return_all = TRUE, ...)  tetracyclines(only_sir_columns = FALSE, return_all = TRUE, ...)  trimethoprims(only_sir_columns = FALSE, return_all = TRUE, ...)  ureidopenicillins(only_sir_columns = FALSE, return_all = TRUE, ...)  administrable_per_os(only_sir_columns = FALSE, return_all = TRUE, ...)  administrable_iv(only_sir_columns = FALSE, return_all = TRUE, ...)  not_intrinsic_resistant(only_sir_columns = FALSE, col_mo = NULL,   version_expertrules = 3.3, ...)"},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"amr_class antimicrobial class part , \"carba\" \"carbapenems\". columns group, atc_group1 atc_group2 antibiotics data set searched (case-insensitive) value. only_sir_columns logical indicate whether columns class sir must selected (default FALSE), see .sir() only_treatable logical indicate whether antimicrobial drugs excluded laboratory tests (default TRUE), gentamicin-high (GEH) imipenem/EDTA (IPE) return_all logical indicate whether matched columns must returned (default TRUE). FALSE, first unique antimicrobial returned, e.g. columns \"genta\" \"gentamicin\" exist data, first hit gentamicin returned. ... ignored, place allow future extensions filter expression evaluated antibiotics data set, name %like% \"trim\" col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). version_expertrules version number use EUCAST Expert Rules Intrinsic Resistance guideline. Can \"3.3\", \"3.2\", \"3.1\".","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"used inside selecting filtering, returns character vector column names, additional class \"amr_selector\". used individually, returns 'ab' vector possible antimicrobials function able select filter.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"functions can used data set calls selecting columns filtering rows. work base R, Tidyverse, data.table. heavily inspired Tidyverse selection helpers everything(), limited dplyr verbs. Nonetheless, convenient use dplyr functions select(), filter() summarise(), see Examples. selectors can also used tidymodels packages recipe parsnip. See info tutorial using AMR functions predictive modelling. columns data functions called searched known antimicrobial names, abbreviations, brand names, codes (ATC, EARS-Net, , etc.) according antibiotics data set. means selector aminoglycosides() pick column names like 'gen', 'genta', 'J01GB03', 'tobra', 'Tobracin', etc. amr_class() function can used filter/select manually defined antimicrobial class. searches results antibiotics data set within columns group, atc_group1 atc_group2. amr_selector() function can used internally filter antibiotics data set results, see Examples. allows filtering (part ) certain name, /group name even minimum DDDs oral treatment. function yields highest flexibility, also least user-friendly, since requires hard-coded filter set. administrable_per_os() administrable_iv() functions also rely antibiotics data set - antimicrobials matched DDD (defined daily dose) resp. oral IV treatment available antibiotics data set. not_intrinsic_resistant() function can used select antimicrobials pose intrinsic resistance microorganisms data set. example, data set contains microorganism codes names E. coli K. pneumoniae contains column \"vancomycin\", column removed (rather, unselected) using function. currently applies 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021) determine intrinsic resistance, using eucast_rules() function internally. determination, function quite slow terms performance.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"full-list-of-supported-antimicrobial-classes","dir":"Reference","previous_headings":"","what":"Full list of supported (antimicrobial) classes","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"aminoglycosides() can select:  amikacin (AMK), amikacin/fosfomycin (AKF), apramycin (APR), arbekacin (ARB), astromicin (AST), bekanamycin (BEK), dibekacin (DKB), framycetin (FRM), gentamicin (GEN), gentamicin-high (GEH), habekacin (HAB), hygromycin (HYG), isepamicin (ISE), kanamycin (KAN), kanamycin-high (KAH), kanamycin/cephalexin (KAC), micronomicin (MCR), neomycin (NEO), netilmicin (NET), pentisomicin (PIM), plazomicin (PLZ), propikacin (PKA), ribostamycin (RST), sisomicin (SIS), streptoduocin (STR), streptomycin (STR1), streptomycin-high (STH), tobramycin (TOB), tobramycin-high (TOH) aminopenicillins() can select:  amoxicillin (AMX) ampicillin (AMP) antifungals() can select:  amorolfine (AMO), amphotericin B (AMB), amphotericin B-high (AMH), anidulafungin (ANI), butoconazole (), caspofungin (CAS), ciclopirox (CIX), clotrimazole (CTR), econazole (ECO), fluconazole (FLU), flucytosine (FCT), fosfluconazole (FFL), griseofulvin (GRI), hachimycin (HCH), ibrexafungerp (IBX), isavuconazole (ISV), isoconazole (ISO), itraconazole (ITR), ketoconazole (KET), manogepix (MGX), micafungin (MIF), miconazole (MCZ), nystatin (NYS), oteseconazole (OTE), pimaricin (PMR), posaconazole (POS), rezafungin (RZF), ribociclib (RBC), sulconazole (SUC), terbinafine (TRB), terconazole (TRC), voriconazole (VOR) antimycobacterials() can select:  4-aminosalicylic acid (AMA), calcium aminosalicylate (CLA), capreomycin (CAP), clofazimine (CLF), delamanid (DLM), enviomycin (ENV), ethambutol (ETH), ethambutol/isoniazid (ETI), ethionamide (ETI1), isoniazid (INH), isoniazid/sulfamethoxazole/trimethoprim/pyridoxine (IST), morinamide (MRN), p-aminosalicylic acid (PAS), pretomanid (PMD), protionamide (PTH), pyrazinamide (PZA), rifabutin (RIB), rifampicin (RIF), rifampicin/ethambutol/isoniazid (REI), rifampicin/isoniazid (RFI), rifampicin/pyrazinamide/ethambutol/isoniazid (RPEI), rifampicin/pyrazinamide/isoniazid (RPI), rifamycin (RFM), rifapentine (RFP), simvastatin/fenofibrate (SMF), sodium aminosalicylate (SDA), streptomycin/isoniazid (STI), terizidone (TRZ), thioacetazone (TAT), thioacetazone/isoniazid (THI1), tiocarlide (TCR), viomycin (VIO) betalactams() can select:  amoxicillin (AMX), amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin (AMP), ampicillin/sulbactam (SAM), apalcillin (APL), aspoxicillin (APX), azidocillin (AZD), azlocillin (AZL), aztreonam (ATM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), bacampicillin (BAM), benzathine benzylpenicillin (BNB), benzathine phenoxymethylpenicillin (BNP), benzylpenicillin (PEN), biapenem (BIA), carbenicillin (CRB), carindacillin (CRN), carumonam (CAR), cefacetrile (CAC), cefaclor (CEC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefamandole (MAN), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetamet (CAT), cefetamet pivoxil (CPI), cefetecol (CCL), cefetrizole (CZL), cefiderocol (FDC), cefixime (CFM), cefmenoxime (CMX), cefmetazole (CMZ), cefodizime (DIZ), cefonicid (CID), cefoperazone (CFP), cefoperazone/sulbactam (CSL), ceforanide (CND), cefoselis (CSE), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotetan (CTT), cefotiam (CTF), cefotiam hexetil (CHE), cefovecin (FOV), cefoxitin (FOX), cefoxitin screening (FOX1), cefozopran (ZOP), cefpimizole (CFZ), cefpiramide (CPM), cefpirome (CPO), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefprozil (CPR), cefquinome (CEQ), cefroxadine (CRD), cefsulodin (CFS), cefsumide (CSU), ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftezole (CTL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), cefuroxime (CXM), cefuroxime axetil (CXA), cephradine (CED), ciclacillin (CIC), clometocillin (CLM), cloxacillin (CLO), dicloxacillin (DIC), doripenem (DOR), epicillin (EPC), ertapenem (ETP), flucloxacillin (FLC), hetacillin (HET), imipenem (IPM), imipenem/EDTA (IPE), imipenem/relebactam (IMR), latamoxef (LTM), lenampicillin (LEN), loracarbef (LOR), mecillinam (MEC), meropenem (MEM), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), metampicillin (MTM), meticillin (MET), mezlocillin (MEZ), mezlocillin/sulbactam (MSU), nafcillin (NAF), oxacillin (OXA), panipenem (PAN), penamecillin (PNM), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), pheneticillin (PHE), phenoxymethylpenicillin (PHN), piperacillin (PIP), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), piridicillin (PRC), pivampicillin (PVM), pivmecillinam (PME), procaine benzylpenicillin (PRB), propicillin (PRP), razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), sarmoxicillin (SRX), sulbenicillin (SBC), sultamicillin (SLT6), talampicillin (TAL), tebipenem (TBP), temocillin (TEM), ticarcillin (TIC), ticarcillin/clavulanic acid (TCC), tigemonam (TMN) betalactams_with_inhibitor() can select:  amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin/sulbactam (SAM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefoperazone/sulbactam (CSL), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefpodoxime/clavulanic acid (CDC), ceftaroline/avibactam (CPA), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), ceftolozane/tazobactam (CZT), ceftriaxone/beta-lactamase inhibitor (CEB), imipenem/relebactam (IMR), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), mezlocillin/sulbactam (MSU), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), ticarcillin/clavulanic acid (TCC) carbapenems() can select:  biapenem (BIA), doripenem (DOR), ertapenem (ETP), imipenem (IPM), imipenem/EDTA (IPE), imipenem/relebactam (IMR), meropenem (MEM), meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), panipenem (PAN), razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), tebipenem (TBP) cephalosporins() can select:  cefacetrile (CAC), cefaclor (CEC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefamandole (MAN), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetamet (CAT), cefetamet pivoxil (CPI), cefetecol (CCL), cefetrizole (CZL), cefiderocol (FDC), cefixime (CFM), cefmenoxime (CMX), cefmetazole (CMZ), cefodizime (DIZ), cefonicid (CID), cefoperazone (CFP), cefoperazone/sulbactam (CSL), ceforanide (CND), cefoselis (CSE), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotetan (CTT), cefotiam (CTF), cefotiam hexetil (CHE), cefovecin (FOV), cefoxitin (FOX), cefoxitin screening (FOX1), cefozopran (ZOP), cefpimizole (CFZ), cefpiramide (CPM), cefpirome (CPO), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefprozil (CPR), cefquinome (CEQ), cefroxadine (CRD), cefsulodin (CFS), cefsumide (CSU), ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftezole (CTL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), cefuroxime (CXM), cefuroxime axetil (CXA), cephradine (CED), latamoxef (LTM), loracarbef (LOR) cephalosporins_1st() can select:  cefacetrile (CAC), cefadroxil (CFR), cefalexin (LEX), cefaloridine (RID), cefalotin (CEP), cefapirin (HAP), cefatrizine (CTZ), cefazedone (CZD), cefazolin (CZO), cefroxadine (CRD), ceftezole (CTL), cephradine (CED) cephalosporins_2nd() can select:  cefaclor (CEC), cefamandole (MAN), cefmetazole (CMZ), cefonicid (CID), ceforanide (CND), cefotetan (CTT), cefotiam (CTF), cefoxitin (FOX), cefoxitin screening (FOX1), cefprozil (CPR), cefuroxime (CXM), cefuroxime axetil (CXA), loracarbef (LOR) cephalosporins_3rd() can select:  cefcapene (CCP), cefcapene pivoxil (CCX), cefdinir (CDR), cefditoren (DIT), cefditoren pivoxil (DIX), cefetamet (CAT), cefetamet pivoxil (CPI), cefixime (CFM), cefmenoxime (CMX), cefodizime (DIZ), cefoperazone (CFP), cefoperazone/sulbactam (CSL), cefotaxime (CTX), cefotaxime/clavulanic acid (CTC), cefotaxime/sulbactam (CTS), cefotiam hexetil (CHE), cefovecin (FOV), cefpimizole (CFZ), cefpiramide (CPM), cefpodoxime (CPD), cefpodoxime proxetil (CPX), cefpodoxime/clavulanic acid (CDC), cefsulodin (CFS), ceftazidime (CAZ), ceftazidime/avibactam (CZA), ceftazidime/clavulanic acid (CCV), cefteram (CEM), cefteram pivoxil (CPL), ceftibuten (CTB), ceftiofur (TIO), ceftizoxime (CZX), ceftizoxime alapivoxil (CZP), ceftriaxone (CRO), ceftriaxone/beta-lactamase inhibitor (CEB), latamoxef (LTM) cephalosporins_4th() can select:  cefepime (FEP), cefepime/amikacin (CFA), cefepime/clavulanic acid (CPC), cefepime/nacubactam (FNC), cefepime/tazobactam (FPT), cefepime/zidebactam (FPZ), cefetecol (CCL), cefoselis (CSE), cefozopran (ZOP), cefpirome (CPO), cefquinome (CEQ) cephalosporins_5th() can select:  ceftaroline (CPT), ceftaroline/avibactam (CPA), ceftobiprole (BPR), ceftobiprole medocaril (CFM1), ceftolozane/tazobactam (CZT) fluoroquinolones() can select:  besifloxacin (BES), ciprofloxacin (CIP), clinafloxacin (CLX), danofloxacin (DAN), delafloxacin (DFX), difloxacin (DIF), enoxacin (ENX), enrofloxacin (ENR), finafloxacin (FIN), fleroxacin (FLE), garenoxacin (GRN), gatifloxacin (GAT), gemifloxacin (GEM), grepafloxacin (GRX), lascufloxacin (LSC), levofloxacin (LVX), levonadifloxacin (LND), lomefloxacin (LOM), marbofloxacin (MAR), metioxate (MXT), miloxacin (MIL), moxifloxacin (MFX), nadifloxacin (NAD), nifuroquine (NIF), norfloxacin (), ofloxacin (OFX), orbifloxacin (ORB), pazufloxacin (PAZ), pefloxacin (PEF), pradofloxacin (PRA), premafloxacin (PRX), prulifloxacin (PRU), rufloxacin (RFL), sarafloxacin (SAR), sitafloxacin (SIT), sparfloxacin (SPX), temafloxacin (TMX), tilbroquinol (TBQ), tioxacin (TXC), tosufloxacin (TFX), trovafloxacin (TVA) glycopeptides() can select:  avoparcin (AVO), dalbavancin (DAL), norvancomycin (NVA), oritavancin (ORI), ramoplanin (RAM), teicoplanin (TEC), teicoplanin-macromethod (TCM), telavancin (TLV), vancomycin (VAN), vancomycin-macromethod (VAM) isoxazolylpenicillins() can select:  cloxacillin (CLO), dicloxacillin (DIC), flucloxacillin (FLC), meticillin (MET), oxacillin (OXA) lincosamides() can select:  clindamycin (CLI), lincomycin (LIN), pirlimycin (PRL) lipoglycopeptides() can select:  dalbavancin (DAL), oritavancin (ORI), telavancin (TLV) macrolides() can select:  acetylmidecamycin (ACM), acetylspiramycin (ASP), azithromycin (AZM), clarithromycin (CLR), dirithromycin (DIR), erythromycin (ERY), flurithromycin (FLR1), gamithromycin (GAM), josamycin (JOS), kitasamycin (KIT), meleumycin (MEL), midecamycin (MID), miocamycin (MCM), nafithromycin (ZWK), oleandomycin (OLE), rokitamycin (ROK), roxithromycin (RXT), solithromycin (SOL), spiramycin (SPI), telithromycin (TLT), tildipirosin (TIP), tilmicosin (TIL), troleandomycin (TRL), tulathromycin (TUL), tylosin (TYL), tylvalosin (TYL1) monobactams() can select:  aztreonam (ATM), aztreonam/avibactam (AZA), aztreonam/nacubactam (ANC), carumonam (CAR), tigemonam (TMN) nitrofurans() can select:  furazidin (FUR), furazolidone (FRZ), nifurtoinol (NFR), nitrofurantoin (NIT), nitrofurazone (NIZ) oxazolidinones() can select:  cadazolid (CDZ), cycloserine (CYC), linezolid (LNZ), tedizolid (TZD), thiacetazone (THA) penicillins() can select:  amoxicillin (AMX), amoxicillin/clavulanic acid (AMC), amoxicillin/sulbactam (AXS), ampicillin (AMP), ampicillin/sulbactam (SAM), apalcillin (APL), aspoxicillin (APX), azidocillin (AZD), azlocillin (AZL), bacampicillin (BAM), benzathine benzylpenicillin (BNB), benzathine phenoxymethylpenicillin (BNP), benzylpenicillin (PEN), carbenicillin (CRB), carindacillin (CRN), ciclacillin (CIC), clometocillin (CLM), cloxacillin (CLO), dicloxacillin (DIC), epicillin (EPC), flucloxacillin (FLC), hetacillin (HET), lenampicillin (LEN), mecillinam (MEC), metampicillin (MTM), meticillin (MET), mezlocillin (MEZ), mezlocillin/sulbactam (MSU), nafcillin (NAF), oxacillin (OXA), penamecillin (PNM), penicillin/novobiocin (PNO), penicillin/sulbactam (PSU), pheneticillin (PHE), phenoxymethylpenicillin (PHN), piperacillin (PIP), piperacillin/sulbactam (PIS), piperacillin/tazobactam (TZP), piridicillin (PRC), pivampicillin (PVM), pivmecillinam (PME), procaine benzylpenicillin (PRB), propicillin (PRP), sarmoxicillin (SRX), sulbenicillin (SBC), sultamicillin (SLT6), talampicillin (TAL), temocillin (TEM), ticarcillin (TIC), ticarcillin/clavulanic acid (TCC) phenicols() can select:  chloramphenicol (CHL), florfenicol (FLR), thiamphenicol (THI) polymyxins() can select:  colistin (COL), polymyxin B (PLB), polymyxin B/polysorbate 80 (POP) quinolones() can select:  besifloxacin (BES), cinoxacin (CIN), ciprofloxacin (CIP), ciprofloxacin/metronidazole (CIM), ciprofloxacin/ornidazole (CIO), ciprofloxacin/tinidazole (CIT), clinafloxacin (CLX), danofloxacin (DAN), delafloxacin (DFX), difloxacin (DIF), enoxacin (ENX), enrofloxacin (ENR), finafloxacin (FIN), fleroxacin (FLE), flumequine (FLM), garenoxacin (GRN), gatifloxacin (GAT), gemifloxacin (GEM), grepafloxacin (GRX), lascufloxacin (LSC), levofloxacin (LVX), levonadifloxacin (LND), lomefloxacin (LOM), marbofloxacin (MAR), metioxate (MXT), miloxacin (MIL), moxifloxacin (MFX), nadifloxacin (NAD), nalidixic acid (NAL), nemonoxacin (NEM), nifuroquine (NIF), nitroxoline (NTR), norfloxacin (), ofloxacin (OFX), orbifloxacin (ORB), oxolinic acid (OXO), pazufloxacin (PAZ), pefloxacin (PEF), pipemidic acid (PPA), piromidic acid (PIR), pradofloxacin (PRA), premafloxacin (PRX), prulifloxacin (PRU), rosoxacin (ROS), rufloxacin (RFL), sarafloxacin (SAR), sitafloxacin (SIT), sparfloxacin (SPX), temafloxacin (TMX), tilbroquinol (TBQ), tioxacin (TXC), tosufloxacin (TFX), trovafloxacin (TVA) rifamycins() can select:  rifabutin (RIB), rifampicin (RIF), rifampicin/ethambutol/isoniazid (REI), rifampicin/isoniazid (RFI), rifampicin/pyrazinamide/ethambutol/isoniazid (RPEI), rifampicin/pyrazinamide/isoniazid (RPI), rifamycin (RFM), rifapentine (RFP) streptogramins() can select:  pristinamycin (PRI) quinupristin/dalfopristin (QDA) tetracyclines() can select:  cetocycline (CTO), chlortetracycline (CTE), clomocycline (CLM1), demeclocycline (DEM), doxycycline (DOX), eravacycline (ERV), lymecycline (LYM), metacycline (MTC), minocycline (MNO), omadacycline (OMC), oxytetracycline (OXY), penimepicycline (PNM1), rolitetracycline (RLT), sarecycline (SRC), tetracycline (TCY), tigecycline (TGC) trimethoprims() can select:  brodimoprim (BDP), sulfadiazine (SDI), sulfadiazine/tetroxoprim (SLT), sulfadiazine/trimethoprim (SLT1), sulfadimethoxine (SUD), sulfadimidine (SDM), sulfadimidine/trimethoprim (SLT2), sulfafurazole (SLF), sulfaisodimidine (SLF1), sulfalene (SLF2), sulfamazone (SZO), sulfamerazine (SLF3), sulfamerazine/trimethoprim (SLT3), sulfamethizole (SLF4), sulfamethoxazole (SMX), sulfamethoxypyridazine (SLF5), sulfametomidine (SLF6), sulfametoxydiazine (SLF7), sulfametrole/trimethoprim (SLT4), sulfamoxole (SLF8), sulfamoxole/trimethoprim (SLT5), sulfanilamide (SLF9), sulfaperin (SLF10), sulfaphenazole (SLF11), sulfapyridine (SLF12), sulfathiazole (SUT), sulfathiourea (SLF13), trimethoprim (TMP), trimethoprim/sulfamethoxazole (SXT) ureidopenicillins() can select:  azlocillin (AZL), mezlocillin (MEZ), piperacillin (PIP), piperacillin/tazobactam (TZP)","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/antimicrobial_selectors.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Antimicrobial Selectors — antimicrobial_selectors","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …   # you can use the selectors separately to retrieve all possible antimicrobials: carbapenems() #> ℹ The function carbapenems() should be used inside a dplyr verb or #>   data.frame call, e.g.: #>   • your_data %>% select(carbapenems()) #>   • your_data %>% select(column_a, column_b, carbapenems()) #>   • your_data %>% filter(any(carbapenems() == \"R\")) #>   • your_data[, carbapenems()] #>   • your_data[, c(\"column_a\", \"column_b\", carbapenems())] #>  #> Now returning a vector of all possible antimicrobials that carbapenems() #> can select. #> Class 'ab' #>  [1] BIA DOR ETP IMR IPE IPM MEM MEV MNC PAN RIA RIT RZM TBP   # Though they are primarily intended to use for selections and filters. # Examples sections below are split into 'dplyr', 'base R', and 'data.table':  # \\donttest{ if (FALSE) { # \\dontrun{ # dplyr -------------------------------------------------------------------  library(dplyr, warn.conflicts = FALSE)  example_isolates %>% select(carbapenems())  # select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB' example_isolates %>% select(mo, aminoglycosides())  # you can combine selectors like you are used with tidyverse # e.g., for betalactams, but not the ones with an enzyme inhibitor: example_isolates %>% select(betalactams(), -betalactams_with_inhibitor())  # select only antimicrobials with DDDs for oral treatment example_isolates %>% select(administrable_per_os())  # get AMR for all aminoglycosides e.g., per ward: example_isolates %>%   group_by(ward) %>%   summarise(across(aminoglycosides(),                    resistance))  # You can combine selectors with '&' to be more specific: example_isolates %>%   select(penicillins() & administrable_per_os())  # get AMR for only drugs that matter - no intrinsic resistance: example_isolates %>%   filter(mo_genus() %in% c(\"Escherichia\", \"Klebsiella\")) %>%   group_by(ward) %>%   summarise_at(not_intrinsic_resistant(),                resistance)  # get susceptibility for antimicrobials whose name contains \"trim\": example_isolates %>%   filter(first_isolate()) %>%   group_by(ward) %>%   summarise(across(amr_selector(name %like% \"trim\"), susceptibility))  # this will select columns 'IPM' (imipenem) and 'MEM' (meropenem): example_isolates %>%   select(carbapenems())  # this will select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB': example_isolates %>%   select(mo, aminoglycosides())  # any() and all() work in dplyr's filter() too: example_isolates %>%   filter(     any(aminoglycosides() == \"R\"),     all(cephalosporins_2nd() == \"R\")   )  # also works with c(): example_isolates %>%   filter(any(c(carbapenems(), aminoglycosides()) == \"R\"))  # not setting any/all will automatically apply all(): example_isolates %>%   filter(aminoglycosides() == \"R\")  # this will select columns 'mo' and all antimycobacterial drugs ('RIF'): example_isolates %>%   select(mo, amr_class(\"mycobact\"))  # get bug/drug combinations for only glycopeptides in Gram-positives: example_isolates %>%   filter(mo_is_gram_positive()) %>%   select(mo, glycopeptides()) %>%   bug_drug_combinations() %>%   format()  data.frame(   some_column = \"some_value\",   J01CA01 = \"S\" ) %>% # ATC code of ampicillin   select(penicillins()) # only the 'J01CA01' column will be selected  # with recent versions of dplyr, this is all equal: x <- example_isolates[carbapenems() == \"R\", ] y <- example_isolates %>% filter(carbapenems() == \"R\") z <- example_isolates %>% filter(if_all(carbapenems(), ~ .x == \"R\")) identical(x, y) && identical(y, z)  } # } # base R ------------------------------------------------------------------  # select columns 'IPM' (imipenem) and 'MEM' (meropenem) example_isolates[, carbapenems()] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 2,000 × 2 #>    IPM   MEM   #>    <sir> <sir> #>  1   NA    NA  #>  2   NA    NA  #>  3   NA    NA  #>  4   NA    NA  #>  5   NA    NA  #>  6   NA    NA  #>  7   NA    NA  #>  8   NA    NA  #>  9   NA    NA  #> 10   NA    NA  #> # ℹ 1,990 more rows  # select columns 'mo', 'AMK', 'GEN', 'KAN' and 'TOB' example_isolates[, c(\"mo\", aminoglycosides())] #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 2,000 × 5 #>    mo           GEN   TOB   AMK   KAN   #>    <mo>         <sir> <sir> <sir> <sir> #>  1 B_ESCHR_COLI   NA    NA    NA    NA  #>  2 B_ESCHR_COLI   NA    NA    NA    NA  #>  3 B_STPHY_EPDR   NA    NA    NA    NA  #>  4 B_STPHY_EPDR   NA    NA    NA    NA  #>  5 B_STPHY_EPDR   NA    NA    NA    NA  #>  6 B_STPHY_EPDR   NA    NA    NA    NA  #>  7 B_STPHY_AURS   NA    S     NA    NA  #>  8 B_STPHY_AURS   NA    S     NA    NA  #>  9 B_STPHY_EPDR   NA    NA    NA    NA  #> 10 B_STPHY_EPDR   NA    NA    NA    NA  #> # ℹ 1,990 more rows  # select only antimicrobials with DDDs for oral treatment example_isolates[, administrable_per_os()] #> ℹ For administrable_per_os() using columns 'OXA' (oxacillin), 'FLC' #>   (flucloxacillin), 'AMX' (amoxicillin), 'AMC' (amoxicillin/clavulanic acid), #>   'AMP' (ampicillin), 'CXM' (cefuroxime), 'KAN' (kanamycin), 'TMP' #>   (trimethoprim), 'NIT' (nitrofurantoin), 'FOS' (fosfomycin), 'LNZ' #>   (linezolid), 'CIP' (ciprofloxacin), 'MFX' (moxifloxacin), 'VAN' #>   (vancomycin), 'TCY' (tetracycline), 'DOX' (doxycycline), 'ERY' #>   (erythromycin), 'CLI' (clindamycin), 'AZM' (azithromycin), 'MTR' #>   (metronidazole), 'CHL' (chloramphenicol), 'COL' (colistin), and 'RIF' #>   (rifampicin) #> # A tibble: 2,000 × 23 #>    OXA   FLC   AMX   AMC   AMP   CXM   KAN   TMP   NIT   FOS   LNZ   CIP   MFX   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA    I     NA    R     NA    NA    R     NA    NA  #>  2   NA    NA    NA    I     NA    I     NA    R     NA    NA    R     NA    NA  #>  3   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    NA    NA  #>  4   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    NA    NA  #>  5   NA    R     NA    NA    NA    R     NA    R     NA    NA    NA    NA    NA  #>  6   NA    R     NA    NA    NA    R     NA    R     NA    NA    NA    NA    NA  #>  7   NA    S     R     S     R     S     NA    R     NA    NA    NA    NA    NA  #>  8   NA    S     R     S     R     S     NA    R     NA    NA    NA    NA    NA  #>  9   NA    R     NA    NA    NA    R     NA    S     NA    NA    NA    S     NA  #> 10   NA    S     NA    NA    NA    S     NA    S     NA    NA    NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 10 more variables: VAN <sir>, TCY <sir>, DOX <sir>, ERY <sir>, CLI <sir>, #> #   AZM <sir>, MTR <sir>, CHL <sir>, COL <sir>, RIF <sir>  # filter using any() or all() example_isolates[any(carbapenems() == \"R\"), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 55 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  4 2005-04-21 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  5 2005-04-22 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  8 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  9 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> 10 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> # ℹ 45 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … subset(example_isolates, any(carbapenems() == \"R\")) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> # A tibble: 55 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2004-06-09 529296     69 M      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  4 2005-04-21 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  5 2005-04-22 452212     82 F      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  8 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  9 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> 10 2008-01-22 1710B8     82 M      Clinical B_PROTS_MRBL   R     NA    NA    NA  #> # ℹ 45 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter on any or all results in the carbapenem columns (i.e., IPM, MEM): example_isolates[any(carbapenems()), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Filtering any of columns 'IPM' and 'MEM' to contain value \"S\", \"I\" or \"R\" #> # A tibble: 962 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  4 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  5 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  6 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  7 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  8 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #>  9 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #> 10 2002-03-08 4FC193     69 M      Clinical B_ESCHR_COLI   R     NA    NA    R   #> # ℹ 952 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … example_isolates[all(carbapenems()), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Filtering all of columns 'IPM' and 'MEM' to contain value \"S\", \"I\" or \"R\" #> # A tibble: 756 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-04-14 F30196     73 M      Outpat… B_STRPT_GRPB    S     NA    S     S   #>  2 2003-04-08 114570     74 M      ICU     B_STRPT_PYGN    S     NA    S     S   #>  3 2003-04-08 114570     74 M      ICU     B_STRPT_GRPA    S     NA    S     S   #>  4 2003-04-08 114570     74 M      ICU     B_STRPT_GRPA    S     NA    S     S   #>  5 2003-08-14 F71508      0 F      Clinic… B_STRPT_GRPB    S     NA    S     S   #>  6 2003-10-16 650870     63 F      ICU     B_ESCHR_COLI    R     NA    NA    R   #>  7 2003-10-20 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-10-20 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  9 2003-11-04 2FC253     87 F      ICU     B_ESCHR_COLI    R     NA    NA    NA  #> 10 2003-11-04 2FC253     87 F      ICU     B_ESCHR_COLI    R     NA    NA    NA  #> # ℹ 746 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter with multiple antimicrobial selectors using c() example_isolates[all(c(carbapenems(), aminoglycosides()) == \"R\"), ] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 26 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2004-11-03 D65308     80 F      ICU      B_STNTR_MLTP   R     NA    NA    R   #>  2 2005-04-22 452212     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  3 2007-02-21 8BBC46     61 F      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  4 2007-12-15 401043     72 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  5 2008-12-06 501361     43 F      Clinical B_STNTR_MLTP   R     NA    NA    R   #>  6 2011-05-09 207325     82 F      ICU      B_ENTRC_FACM   NA    NA    NA    NA  #>  7 2012-03-12 582258     80 M      ICU      B_STPHY_CONS   R     R     R     R   #>  8 2012-05-19 C25552     89 F      Outpati… B_STPHY_CONS   R     R     R     R   #>  9 2012-07-17 F05015     83 M      ICU      B_STPHY_CONS   R     R     R     R   #> 10 2012-07-20 404299     66 F      Clinical B_STPHY_CONS   R     R     R     R   #> # ℹ 16 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter + select in one go: get penicillins in carbapenem-resistant strains example_isolates[any(carbapenems() == \"R\"), penicillins()] #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #> # A tibble: 55 × 7 #>    PEN   OXA   FLC   AMX   AMC   AMP   TZP   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    NA    NA    NA    NA  #>  2   NA    NA    NA    NA    NA    NA    NA  #>  3   R     NA    NA    R     R     R     R   #>  4   NA    NA    NA    NA    NA    NA    R   #>  5   NA    NA    NA    NA    NA    NA    R   #>  6   NA    NA    NA    NA    NA    NA    R   #>  7   NA    NA    NA    NA    NA    NA    R   #>  8   NA    NA    NA    NA    NA    NA    R   #>  9   R     NA    NA    NA    S     NA    S   #> 10   R     NA    NA    NA    S     NA    S   #> # ℹ 45 more rows  # You can combine selectors with '&' to be more specific. For example, # penicillins() would select benzylpenicillin ('peni G') and # administrable_per_os() would select erythromycin. Yet, when combined these # drugs are both omitted since benzylpenicillin is not administrable per os # and erythromycin is not a penicillin: example_isolates[, penicillins() & administrable_per_os()] #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #> ℹ For administrable_per_os() using columns 'OXA' (oxacillin), 'FLC' #>   (flucloxacillin), 'AMX' (amoxicillin), 'AMC' (amoxicillin/clavulanic acid), #>   'AMP' (ampicillin), 'CXM' (cefuroxime), 'KAN' (kanamycin), 'TMP' #>   (trimethoprim), 'NIT' (nitrofurantoin), 'FOS' (fosfomycin), 'LNZ' #>   (linezolid), 'CIP' (ciprofloxacin), 'MFX' (moxifloxacin), 'VAN' #>   (vancomycin), 'TCY' (tetracycline), 'DOX' (doxycycline), 'ERY' #>   (erythromycin), 'CLI' (clindamycin), 'AZM' (azithromycin), 'MTR' #>   (metronidazole), 'CHL' (chloramphenicol), 'COL' (colistin), and 'RIF' #>   (rifampicin) #> # A tibble: 2,000 × 5 #>    OXA   FLC   AMX   AMC   AMP   #>    <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA  #>  2   NA    NA    NA    I     NA  #>  3   NA    R     NA    NA    NA  #>  4   NA    R     NA    NA    NA  #>  5   NA    R     NA    NA    NA  #>  6   NA    R     NA    NA    NA  #>  7   NA    S     R     S     R   #>  8   NA    S     R     S     R   #>  9   NA    R     NA    NA    NA  #> 10   NA    S     NA    NA    NA  #> # ℹ 1,990 more rows  # amr_selector() applies a filter in the `antibiotics` data set and is thus # very flexible. For instance, to select antimicrobials with an oral DDD # of at least 1 gram: example_isolates[, amr_selector(oral_ddd > 1 & oral_units == \"g\")] #> ℹ For amr_selector(oral_ddd > 1 & oral_units == \"g\") using columns 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), 'KAN' (kanamycin), 'FOS' #>   (fosfomycin), 'LNZ' (linezolid), 'VAN' (vancomycin), 'ERY' (erythromycin), #>   'CLI' (clindamycin), 'MTR' (metronidazole), and 'CHL' (chloramphenicol) #> # A tibble: 2,000 × 13 #>    OXA   FLC   AMX   AMC   AMP   KAN   FOS   LNZ   VAN   ERY   CLI   MTR   CHL   #>    <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>  1   NA    NA    NA    I     NA    NA    NA    R     R     R     R     NA    NA  #>  2   NA    NA    NA    I     NA    NA    NA    R     R     R     R     NA    NA  #>  3   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  4   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  5   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #>  6   NA    R     NA    NA    NA    NA    NA    NA    S     R     R     NA    NA  #>  7   NA    S     R     S     R     NA    NA    NA    S     S     NA    NA    NA  #>  8   NA    S     R     S     R     NA    NA    NA    S     S     NA    NA    NA  #>  9   NA    R     NA    NA    NA    NA    NA    NA    S     R     NA    NA    NA  #> 10   NA    S     NA    NA    NA    NA    NA    NA    S     S     NA    NA    NA  #> # ℹ 1,990 more rows   # data.table --------------------------------------------------------------  # data.table is supported as well, just use it in the same way as with # base R, but add `with = FALSE` if using a single AB selector.  if (require(\"data.table\")) {   dt <- as.data.table(example_isolates)    # this does not work, it returns column *names*   dt[, carbapenems()] } #> Loading required package: data.table #>  #> Attaching package: ‘data.table’ #> The following objects are masked from ‘package:dplyr’: #>  #>     between, first, last #> The following objects are masked from ‘package:AMR’: #>  #>     %like%, like #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> Warning: It should never be needed to print an antimicrobial selector class. Are you #> using data.table? Then add the argument with = FALSE, see our examples at #> ?amr_selector. #> Class 'amr_selector' #> [1] IPM MEM if (require(\"data.table\")) {   # so `with = FALSE` is required   dt[, carbapenems(), with = FALSE] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #>         IPM   MEM #>       <sir> <sir> #>    1:  <NA>  <NA> #>    2:  <NA>  <NA> #>    3:  <NA>  <NA> #>    4:  <NA>  <NA> #>    5:  <NA>  <NA> #>   ---             #> 1996:  <NA>  <NA> #> 1997:     S     S #> 1998:     S     S #> 1999:     S     S #> 2000:     S     S  # for multiple selections or AB selectors, `with = FALSE` is not needed: if (require(\"data.table\")) {   dt[, c(\"mo\", aminoglycosides())] } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #>                 mo   GEN   TOB   AMK   KAN #>               <mo> <sir> <sir> <sir> <sir> #>    1: B_ESCHR_COLI  <NA>  <NA>  <NA>  <NA> #>    2: B_ESCHR_COLI  <NA>  <NA>  <NA>  <NA> #>    3: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>    4: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>    5: B_STPHY_EPDR  <NA>  <NA>  <NA>  <NA> #>   ---                                      #> 1996: B_STRPT_PNMN     R     R     R     R #> 1997: B_ESCHR_COLI     S     S     S  <NA> #> 1998: B_STPHY_CONS     S  <NA>  <NA>  <NA> #> 1999: B_ESCHR_COLI     S     S  <NA>  <NA> #> 2000: B_KLBSL_PNMN     S     S  <NA>  <NA> if (require(\"data.table\")) {   dt[, c(carbapenems(), aminoglycosides())] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #>         IPM   MEM   GEN   TOB   AMK   KAN #>       <sir> <sir> <sir> <sir> <sir> <sir> #>    1:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    2:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    3:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    4:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>    5:  <NA>  <NA>  <NA>  <NA>  <NA>  <NA> #>   ---                                     #> 1996:  <NA>  <NA>     R     R     R     R #> 1997:     S     S     S     S     S  <NA> #> 1998:     S     S     S  <NA>  <NA>  <NA> #> 1999:     S     S     S     S  <NA>  <NA> #> 2000:     S     S     S     S  <NA>  <NA>  # row filters are also supported: if (require(\"data.table\")) {   dt[any(carbapenems() == \"S\"), ] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #>            date patient   age gender       ward           mo   PEN   OXA   FLC #>          <Date>  <char> <num> <char>     <char>         <mo> <sir> <sir> <sir> #>   1: 2002-01-19  738003    71      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #>   2: 2002-01-19  738003    71      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #>   3: 2002-01-22  F35553    50      M        ICU B_PROTS_MRBL     R  <NA>  <NA> #>   4: 2002-01-22  F35553    50      M        ICU B_PROTS_MRBL     R  <NA>  <NA> #>   5: 2002-02-05  067927    45      F        ICU B_SERRT_MRCS     R  <NA>  <NA> #>  ---                                                                           #> 905: 2005-04-12  D71461    70      M        ICU B_ESCHR_COLI     R  <NA>  <NA> #> 906: 2009-11-12  650870    69      F Outpatient B_ESCHR_COLI     R  <NA>  <NA> #> 907: 2012-06-14  8CBCF2    41      F   Clinical B_STPHY_CONS     R     S     S #> 908: 2012-10-11  175532    78      M   Clinical B_ESCHR_COLI     R  <NA>  <NA> #> 909: 2013-11-23  A97263    77      M   Clinical B_KLBSL_PNMN     R  <NA>  <NA> #>        AMX   AMC   AMP   TZP   CZO   FEP   CXM   FOX   CTX   CAZ   CRO   GEN #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S  <NA>     S  <NA> #>   2:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S  <NA>     S  <NA> #>   3:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S     S     S  <NA> #>   4:  <NA>     I  <NA>  <NA>  <NA>  <NA>     S  <NA>     S     S     S  <NA> #>   5:     R     R     R  <NA>     R  <NA>     R     R  <NA>  <NA>  <NA>  <NA> #>  ---                                                                         #> 905:     S     S     S     S  <NA>     S     S     S     S     S     S     S #> 906:     S     S     S     S     S     S     S     S     S     S     S     S #> 907:  <NA>     S  <NA>  <NA>     S     S     S     S     S     R     S     S #> 908:     R     S     R     S  <NA>     S     R     R     S     S     S     S #> 909:     R     S     R     S  <NA>     S     S     S     S     S     S     S #>        TOB   AMK   KAN   TMP   SXT   NIT   FOS   LNZ   CIP   MFX   VAN   TEC #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:     S  <NA>  <NA>     S     S  <NA>  <NA>     R  <NA>  <NA>     R     R #>   2:     S  <NA>  <NA>     S     S  <NA>  <NA>     R  <NA>  <NA>     R     R #>   3:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>   4:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>   5:  <NA>  <NA>  <NA>     S     S     R  <NA>     R     S  <NA>     R     R #>  ---                                                                         #> 905:     S     S  <NA>  <NA>     S     S  <NA>     R     S  <NA>     R     R #> 906:     S     S  <NA>     S     S     S  <NA>     R     S  <NA>     R     R #> 907:  <NA>  <NA>  <NA>     S     S  <NA>  <NA>  <NA>     S  <NA>     S  <NA> #> 908:     S  <NA>  <NA>     R     R     R  <NA>     R     R     R     R     R #> 909:     S  <NA>  <NA>     S     S     S  <NA>     R     S  <NA>     R     R #>        TCY   TGC   DOX   ERY   CLI   AZM   IPM   MEM   MTR   CHL   COL   MUP #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:  <NA>  <NA>  <NA>     R     R     R     S  <NA>  <NA>  <NA>  <NA>  <NA> #>   2:  <NA>  <NA>  <NA>     R     R     R     S  <NA>  <NA>  <NA>  <NA>  <NA> #>   3:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>   4:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>   5:     R     R     R     R     R     R     S  <NA>  <NA>  <NA>     R  <NA> #>  ---                                                                         #> 905:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>  <NA>  <NA> #> 906:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>  <NA>  <NA> #> 907:  <NA>  <NA>  <NA>     S     S     S     S     S  <NA>  <NA>     R  <NA> #> 908:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>     S  <NA> #> 909:  <NA>  <NA>  <NA>     R     R     R     S     S  <NA>  <NA>     S  <NA> #>        RIF #>      <sir> #>   1:     R #>   2:     R #>   3:     R #>   4:     R #>   5:     R #>  ---       #> 905:     R #> 906:     R #> 907:  <NA> #> 908:     R #> 909:     R if (require(\"data.table\")) {   dt[any(carbapenems() == \"S\"), penicillins(), with = FALSE] } #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ For penicillins() using columns 'PEN' (benzylpenicillin), 'OXA' #>   (oxacillin), 'FLC' (flucloxacillin), 'AMX' (amoxicillin), 'AMC' #>   (amoxicillin/clavulanic acid), 'AMP' (ampicillin), and 'TZP' #>   (piperacillin/tazobactam) #>        PEN   OXA   FLC   AMX   AMC   AMP   TZP #>      <sir> <sir> <sir> <sir> <sir> <sir> <sir> #>   1:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   2:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   3:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   4:     R  <NA>  <NA>  <NA>     I  <NA>  <NA> #>   5:     R  <NA>  <NA>     R     R     R  <NA> #>  ---                                           #> 905:     R  <NA>  <NA>     S     S     S     S #> 906:     R  <NA>  <NA>     S     S     S     S #> 907:     R     S     S  <NA>     S  <NA>  <NA> #> 908:     R  <NA>  <NA>     R     S     R     S #> 909:     R  <NA>  <NA>     R     S     R     S # }"},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to an Antibiotic ID — as.ab","title":"Transform Input to an Antibiotic ID — as.ab","text":"Use function determine antibiotic drug code one antibiotics. data set antibiotics searched abbreviations, official names synonyms (brand names).","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to an Antibiotic ID — as.ab","text":"","code":"as.ab(x, flag_multiple_results = TRUE, info = interactive(), ...)  is.ab(x)"},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to an Antibiotic ID — as.ab","text":"x character vector determine antibiotic ID flag_multiple_results logical indicate whether note printed console probably one antibiotic drug code name can retrieved single input value. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed internal functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to an Antibiotic ID — as.ab","text":"character vector additional class ab","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to an Antibiotic ID — as.ab","text":"entries antibiotics data set three different identifiers: human readable EARS-Net code (column ab, used ECDC WHONET), ATC code (column atc, used ), CID code (column cid, Compound ID, used PubChem). data set contains 5,000 official brand names many different countries, found PubChem. drugs contain multiple ATC codes. properties searched user input. .ab() can correct different forms misspelling: Wrong spelling drug names (\"tobramicin\" \"gentamycin\"), corrects audible similarities f/ph, x/ks, c/z/s, t/th, etc. many vowels consonants Switching two characters (\"mreopenem\", often case clinical data, doctors typed fast) Digitalised paper records, leaving artefacts like 0/o/O (zero O's), B/8, n/r, etc. Use ab_* functions get properties based returned antibiotic ID, see Examples. Note: .ab() ab_* functions may use long regular expression match brand names antimicrobial drugs. may fail systems. can add manual codes considered .ab() ab_* functions, see add_custom_antimicrobials().","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Input to an Antibiotic ID — as.ab","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Transform Input to an Antibiotic ID — as.ab","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Input to an Antibiotic ID — as.ab","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.ab.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to an Antibiotic ID — as.ab","text":"","code":"# these examples all return \"ERY\", the ID of erythromycin: as.ab(\"J01FA01\") #> Class 'ab' #> [1] ERY as.ab(\"J 01 FA 01\") #> Class 'ab' #> [1] ERY as.ab(\"Erythromycin\") #> Class 'ab' #> [1] ERY as.ab(\"eryt\") #> Class 'ab' #> [1] ERY as.ab(\"   eryt 123\") #> Class 'ab' #> [1] ERY as.ab(\"ERYT\") #> Class 'ab' #> [1] ERY as.ab(\"ERY\") #> Class 'ab' #> [1] ERY as.ab(\"eritromicine\") # spelled wrong, yet works #> Class 'ab' #> [1] ERY as.ab(\"Erythrocin\") # trade name #> Class 'ab' #> [1] ERY as.ab(\"Romycin\") # trade name #> Class 'ab' #> [1] ERY  # spelling from different languages and dyslexia are no problem ab_atc(\"ceftriaxon\") #> [1] \"J01DD04\" ab_atc(\"cephtriaxone\") # small spelling error #> [1] \"J01DD04\" ab_atc(\"cephthriaxone\") # or a bit more severe #> [1] \"J01DD04\" ab_atc(\"seephthriaaksone\") # and even this works #> [1] \"J01DD04\"  # use ab_* functions to get a specific properties (see ?ab_property); # they use as.ab() internally: ab_name(\"J01FA01\") #> [1] \"Erythromycin\" ab_name(\"eryt\") #> [1] \"Erythromycin\"  # \\donttest{ if (require(\"dplyr\")) {   # you can quickly rename 'sir' columns using set_ab_names() with dplyr:   example_isolates %>%     set_ab_names(where(is.sir), property = \"atc\") } #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           J01CE01 J01CF04 J01CF05 #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir>   <sir>   <sir>   #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R       NA      NA    #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R       NA      NA    #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R       NA      S     #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R       NA      S     #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R       NA      R     #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R       NA      S     #> # ℹ 1,990 more rows #> # ℹ 37 more variables: J01CA04 <sir>, J01CR02 <sir>, J01CA01 <sir>, #> #   J01CR05 <sir>, J01DB04 <sir>, J01DE01 <sir>, J01DC02 <sir>, J01DC01 <sir>, #> #   J01DD01 <sir>, J01DD02 <sir>, J01DD04 <sir>, J01GB03 <sir>, J01GB01 <sir>, #> #   J01GB06 <sir>, J01GB04 <sir>, J01EA01 <sir>, J01EE01 <sir>, J01XE01 <sir>, #> #   J01XX01 <sir>, J01XX08 <sir>, J01MA02 <sir>, J01MA14 <sir>, J01XA01 <sir>, #> #   J01XA02 <sir>, J01AA07 <sir>, J01AA12 <sir>, J01AA02 <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to an Antiviral Drug ID — as.av","title":"Transform Input to an Antiviral Drug ID — as.av","text":"Use function determine antiviral drug code one antiviral drugs. data set antivirals searched abbreviations, official names synonyms (brand names).","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to an Antiviral Drug ID — as.av","text":"","code":"as.av(x, flag_multiple_results = TRUE, info = interactive(), ...)  is.av(x)"},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to an Antiviral Drug ID — as.av","text":"x character vector determine antiviral drug ID flag_multiple_results logical indicate whether note printed console probably one antiviral drug code name can retrieved single input value. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed internal functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to an Antiviral Drug ID — as.av","text":"character vector additional class ab","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to an Antiviral Drug ID — as.av","text":"entries antivirals data set three different identifiers: human readable EARS-Net code (column ab, used ECDC WHONET), ATC code (column atc, used ), CID code (column cid, Compound ID, used PubChem). data set contains 5,000 official brand names many different countries, found PubChem. drugs contain multiple ATC codes. properties searched user input. .av() can correct different forms misspelling: Wrong spelling drug names (\"acyclovir\"), corrects audible similarities f/ph, x/ks, c/z/s, t/th, etc. many vowels consonants Switching two characters (\"aycclovir\", often case clinical data, doctors typed fast) Digitalised paper records, leaving artefacts like 0/o/O (zero O's), B/8, n/r, etc. Use av_* functions get properties based returned antiviral drug ID, see Examples. Note: .av() av_* functions may use long regular expression match brand names antimicrobial drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Input to an Antiviral Drug ID — as.av","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"whocc","dir":"Reference","previous_headings":"","what":"WHOCC","title":"Transform Input to an Antiviral Drug ID — as.av","text":"package contains ~550 antibiotic, antimycotic antiviral drugs Anatomical Therapeutic Chemical (ATC) codes, ATC groups Defined Daily Dose (DDD) World Health Organization Collaborating Centre Drug Statistics Methodology (WHOCC, https://atcddd.fhi.) Pharmaceuticals Community Register European Commission (https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm). become gold standard international drug utilisation monitoring research. WHOCC located Oslo Norwegian Institute Public Health funded Norwegian government. European Commission executive European Union promotes general interest. NOTE: WHOCC copyright allow use commercial purposes, unlike info package. See https://atcddd.fhi./copyright_disclaimer/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Input to an Antiviral Drug ID — as.av","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.av.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to an Antiviral Drug ID — as.av","text":"","code":"# these examples all return \"ACI\", the ID of aciclovir: as.av(\"J05AB01\") #> Class 'av' #> [1] ACI as.av(\"J 05 AB 01\") #> Class 'av' #> [1] ACI as.av(\"Aciclovir\") #> Class 'av' #> [1] ACI as.av(\"aciclo\") #> Class 'av' #> [1] ACI as.av(\"   aciclo 123\") #> Class 'av' #> [1] ACI as.av(\"ACICL\") #> Class 'av' #> [1] ACI as.av(\"ACI\") #> Class 'av' #> [1] ACI as.av(\"Virorax\") # trade name #> Class 'av' #> [1] ACI as.av(\"Zovirax\") # trade name #> Class 'av' #> [1] ACI  as.av(\"acyklofir\") # severe spelling error, yet works #> Class 'av' #> [1] ACI  # use av_* functions to get a specific properties (see ?av_property); # they use as.av() internally: av_name(\"J05AB01\") #> [1] \"Aciclovir\" av_name(\"acicl\") #> [1] \"Aciclovir\""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to Disk Diffusion Diameters — as.disk","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"transforms vector new class disk, disk diffusion growth zone size (around antibiotic disk) millimetres 0 50.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"","code":"as.disk(x, na.rm = FALSE)  NA_disk_  is.disk(x)"},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"object class disk (inherits integer) length 1.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"x vector na.rm logical indicating whether missing values removed","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"integer additional class disk","code":""},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"Interpret disk values SIR values .sir(). supports guidelines EUCAST CLSI. Disk diffusion growth zone sizes must 0 50 millimetres. Values higher 50 lower 100 maximised 50. others input values outside 0-50 range return NA. NA_disk_ missing value new disk class.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.disk.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to Disk Diffusion Diameters — as.disk","text":"","code":"# transform existing disk zones to the `disk` class (using base R) df <- data.frame(   microorganism = \"Escherichia coli\",   AMP = 20,   CIP = 14,   GEN = 18,   TOB = 16 ) df[, 2:5] <- lapply(df[, 2:5], as.disk) str(df) #> 'data.frame':\t1 obs. of  5 variables: #>  $ microorganism: chr \"Escherichia coli\" #>  $ AMP          : 'disk' int 20 #>  $ CIP          : 'disk' int 14 #>  $ GEN          : 'disk' int 18 #>  $ TOB          : 'disk' int 16  # \\donttest{ # transforming is easier with dplyr: if (require(\"dplyr\")) {   df %>% mutate(across(AMP:TOB, as.disk)) } #>      microorganism AMP CIP GEN TOB #> 1 Escherichia coli  20  14  18  16 # }  # interpret disk values, see ?as.sir as.sir(   x = as.disk(18),   mo = \"Strep pneu\", # `mo` will be coerced with as.mo()   ab = \"ampicillin\", # and `ab` with as.ab()   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: 'ampicillin' (AMP), EUCAST 2024... #>  OK  #> Class 'sir' #> [1] R  # interpret whole data set, pretend to be all from urinary tract infections: as.sir(df, uti = TRUE) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'AMP' (ampicillin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'CIP' (ciprofloxacin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'GEN' (gentamicin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: column 'TOB' (tobramycin), EUCAST #> 2024... #>  OK  #>      microorganism AMP  CIP GEN TOB #> 1 Escherichia coli   S <NA>   S   S"},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"transforms vectors new class mic, treats input decimal numbers, maintaining operators (\">=\") allowing valid MIC values known field (medical) microbiology.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"","code":"as.mic(x, na.rm = FALSE, keep_operators = \"all\")  is.mic(x)  NA_mic_  rescale_mic(x, mic_range, keep_operators = \"edges\", as.mic = TRUE)  # S3 method for class 'mic' droplevels(x, as.mic = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"x character numeric vector na.rm logical indicating whether missing values removed keep_operators character specifying handle operators (> <=) input. Accepts one three values: \"\" (TRUE) keep operators, \"none\" (FALSE) remove operators, \"edges\" keep operators ends range. mic_range manual range rescale MIC values, e.g., mic_range = c(0.001, 32). Use NA prevent rescaling one side, e.g., mic_range = c(NA, 32). .mic logical indicate whether mic class kept - default TRUE rescale_mic() FALSE droplevels(). setting FALSE rescale_mic(), output factor levels acknowledge mic_range. ... arguments passed methods","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"Ordered factor additional class mic, mathematical operations acts numeric vector. Bear mind outcome mathematical operation MICs return numeric value.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"interpret MIC values SIR values, use .sir() MIC values. supports guidelines EUCAST (2011-2024) CLSI (2011-2024). class MIC values quite special data type: formally ordered factor valid MIC values factor levels (make sure valid MIC values retained), mathematical operation acts decimal numbers:   makes possible maintain operators often come MIC values, \">=\" \"<=\", even filtering using numeric values data analysis, e.g.:   -called group generic functions implemented MIC class (!, !=, <, >=, exp(), log2()). functions stats package also implemented (quantile(), median(), fivenum()). Since sd() var() non-generic functions, extended. Use mad() alternative, use e.g. sd(.numeric(x)) x vector MIC values. Using .double() .numeric() MIC values remove operators return numeric vector. use .integer() MIC values R convention factors, return index factor levels (often useless regular users). Use droplevels() drop unused levels. default, return plain factor. Use droplevels(..., .mic = TRUE) maintain mic class. rescale_mic(), existing MIC ranges can limited defined range MIC values. can useful better compare MIC distributions. ggplot2, use one scale_*_mic() functions plot MIC values. allows custom MIC ranges plot intermediate log2 levels missing MIC values. NA_mic_ missing value new mic class, analogous e.g. base R's NA_character_.","code":"x <- random_mic(10) x #> Class 'mic' #>  [1] 16     1      8      8      64     >=128  0.0625 32     32     16  is.factor(x) #> [1] TRUE  x[1] * 2 #> [1] 32  median(x) #> [1] 26 x[x > 4] #> Class 'mic' #> [1] 16    8     8     64    >=128 32    32    16  df <- data.frame(x, hospital = \"A\") subset(df, x > 4) # or with dplyr: df %>% filter(x > 4) #>        x hospital #> 1     16        A #> 5     64        A #> 6  >=128        A #> 8     32        A #> 9     32        A #> 10    16        A"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.mic.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Input to Minimum Inhibitory Concentrations (MIC) — as.mic","text":"","code":"mic_data <- as.mic(c(\">=32\", \"1.0\", \"1\", \"1.00\", 8, \"<=0.128\", \"8\", \"16\", \"16\")) mic_data #> Class 'mic' #> [1] >=32    1       1       1       8       <=0.128 8       16      16      is.mic(mic_data) #> [1] TRUE  # this can also coerce combined MIC/SIR values: as.mic(\"<=0.002; S\") #> Class 'mic' #> [1] <=0.002  # mathematical processing treats MICs as numeric values fivenum(mic_data) #> [1]  0.128  1.000  8.000 16.000 32.000 quantile(mic_data) #>     0%    25%    50%    75%   100%  #>  0.128  1.000  8.000 16.000 32.000  all(mic_data < 512) #> [1] TRUE  # rescale MICs using rescale_mic() rescale_mic(mic_data, mic_range = c(4, 16)) #> Class 'mic' #> [1] >=16 <=4  <=4  <=4  8    <=4  8    >=16 >=16  # interpret MIC values as.sir(   x = as.mic(2),   mo = as.mo(\"Streptococcus pneumoniae\"),   ab = \"AMX\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMX' (amoxicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for amoxicillin (AMX) in Streptococcus pneumoniae - assuming body site 'Meningitis'. #> Class 'sir' #> [1] R as.sir(   x = as.mic(c(0.01, 2, 4, 8)),   mo = as.mo(\"Streptococcus pneumoniae\"),   ab = \"AMX\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMX' (amoxicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for amoxicillin (AMX) in Streptococcus pneumoniae - assuming body site 'Meningitis'. #> Class 'sir' #> [1] S R R R  # plot MIC values, see ?plot plot(mic_data)  plot(mic_data, mo = \"E. coli\", ab = \"cipro\")   if (require(\"ggplot2\")) {   autoplot(mic_data, mo = \"E. coli\", ab = \"cipro\") }  if (require(\"ggplot2\")) {   autoplot(mic_data, mo = \"E. coli\", ab = \"cipro\", language = \"nl\") # Dutch }"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":null,"dir":"Reference","previous_headings":"","what":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Use function get valid microorganism code (mo) based arbitrary user input. Determination done using intelligent rules complete taxonomic tree kingdoms Animalia, Archaea, Bacteria, Chromista, Protozoa, microbial species kingdom Fungi (see Source). input can almost anything: full name (like \"Staphylococcus aureus\"), abbreviated name (\"S. aureus\"), abbreviation known field (\"MRSA\"), just genus. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"","code":"as.mo(x, Becker = FALSE, Lancefield = FALSE,   minimum_matching_score = NULL,   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE),   reference_df = get_mo_source(),   ignore_pattern = getOption(\"AMR_ignore_pattern\", NULL),   cleaning_regex = getOption(\"AMR_cleaning_regex\", mo_cleaning_regex()),   only_fungi = getOption(\"AMR_only_fungi\", FALSE),   language = get_AMR_locale(), info = interactive(), ...)  is.mo(x)  mo_uncertainties()  mo_renamed()  mo_failures()  mo_reset_session()  mo_cleaning_regex()"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"x character vector data.frame one two columns Becker logical indicate whether staphylococci categorised coagulase-negative staphylococci (\"CoNS\") coagulase-positive staphylococci (\"CoPS\") instead species, according Karsten Becker et al. (see Source). Please see Details full list staphylococcal species converted. excludes Staphylococcus aureus default, use Becker = \"\" also categorise S. aureus \"CoPS\". Lancefield logical indicate whether beta-haemolytic Streptococcus categorised Lancefield groups instead species, according Rebecca C. Lancefield (see Source). streptococci categorised first group, e.g. Streptococcus dysgalactiae group C, although officially also categorised groups G L. . Please see Details full list streptococcal species converted. excludes enterococci default (group D), use Lancefield = \"\" also categorise enterococci group D. minimum_matching_score numeric value set lower limit MO matching score. left blank, determined automatically based character length x, taxonomic kingdom human pathogenicity. keep_synonyms logical indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE, return note old taxonomic names processed. default can set package option AMR_keep_synonyms, .e. options(AMR_keep_synonyms = TRUE) options(AMR_keep_synonyms = FALSE). reference_df data.frame used extra reference translating x valid mo. See set_mo_source() get_mo_source() automate usage codes (e.g. used analysis organisation). ignore_pattern Perl-compatible regular expression (case-insensitive) matches x must return NA. can convenient exclude known non-relevant input can also set package option AMR_ignore_pattern, e.g. options(AMR_ignore_pattern = \"(reported|contaminated flora)\"). cleaning_regex Perl-compatible regular expression (case-insensitive) clean input x. Every matched part x removed. default, outcome mo_cleaning_regex(), removes texts brackets texts \"species\" \"serovar\". default can set package option AMR_cleaning_regex. only_fungi logical indicate fungi must found, making sure e.g. misspellings always return records kingdom Fungi. can set globally microorganism functions package option AMR_only_fungi, .e. options(AMR_only_fungi = TRUE). language language translate text like \"growth\", defaults system language (see get_AMR_locale()) info logical indicate info must printed, e.g. progress bar 25 items coerced, list old taxonomic names. default TRUE interactive mode. ... arguments passed functions","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"character vector additional class mo","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"microorganism (MO) code package (class: mo) human-readable typically looks like examples:   Values coerced considered 'unknown' return MO code UNKNOWN warning. Use mo_* functions get properties based returned code, see Examples. .mo() function uses novel scientifically validated (doi:10.18637/jss.v104.i03 ) matching score algorithm (see Matching Score Microorganisms ) match input available microbial taxonomy package. implicates e.g. \"E. coli\" (microorganism highly prevalent humans) return microbial ID Escherichia coli Entamoeba coli (microorganism less prevalent humans), although latter alphabetically come first.","code":"Code               Full name   ---------------    --------------------------------------   B_KLBSL            Klebsiella   B_KLBSL_PNMN       Klebsiella pneumoniae   B_KLBSL_PNMN_RHNS  Klebsiella pneumoniae rhinoscleromatis   |   |    |    |   |   |    |    |   |   |    |    \\---> subspecies, a 3-5 letter acronym   |   |    \\----> species, a 3-6 letter acronym   |   \\----> genus, a 4-8 letter acronym   \\----> kingdom: A (Archaea), AN (Animalia), B (Bacteria),                   C (Chromista), F (Fungi), PL (Plantae),                   P (Protozoa)"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"coping-with-uncertain-results","dir":"Reference","previous_headings":"","what":"Coping with Uncertain Results","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Results non-exact taxonomic input based matching score. lowest allowed score can set minimum_matching_score argument. default determined based character length input, taxonomic kingdom, human pathogenicity taxonomic outcome. values matched uncertainty, message shown suggest user inspect results mo_uncertainties(), returns data.frame specifications. increase quality matching, cleaning_regex argument used clean input. must regular expression matches parts input removed input matched available microbial taxonomy. matched Perl-compatible case-insensitive. default value cleaning_regex outcome helper function mo_cleaning_regex(). three helper functions can run using .mo() function: Use mo_uncertainties() get data.frame prints pretty format taxonomic names guessed. output contains matching score matches (see Matching Score Microorganisms ). Use mo_failures() get character vector values coerced valid value. Use mo_renamed() get data.frame values coerced based old, previously accepted taxonomic names.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"for-mycologists","dir":"Reference","previous_headings":"","what":"For Mycologists","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"matching score algorithm gives precedence bacteria fungi. analysing fungi, sure use only_fungi = TRUE, better yet, add code run every session:   make sure bacteria 'non-fungi' returned .mo(), mo_* functions.","code":"options(AMR_only_fungi = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"coagulase-negative-and-coagulase-positive-staphylococci","dir":"Reference","previous_headings":"","what":"Coagulase-negative and Coagulase-positive Staphylococci","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Becker = TRUE, following staphylococci converted corresponding coagulase group: Coagulase-negative: S. americanisciuri, S. argensis, S. arlettae, S. auricularis, S. borealis, S. brunensis, S. caeli, S. caledonicus, S. canis, S. capitis, S. capitis capitis, S. capitis urealyticus, S. capitis ureolyticus, S. caprae, S. carnosus, S. carnosus carnosus, S. carnosus utilis, S. casei, S. caseolyticus, S. chromogenes, S. cohnii, S. cohnii cohnii, S. cohnii urealyticum, S. cohnii urealyticus, S. condimenti, S. croceilyticus, S. debuckii, S. devriesei, S. durrellii, S. edaphicus, S. epidermidis, S. equorum, S. equorum equorum, S. equorum linens, S. felis, S. fleurettii, S. gallinarum, S. haemolyticus, S. hominis, S. hominis hominis, S. hominis novobiosepticus, S. jettensis, S. kloosii, S. lentus, S. lloydii, S. lugdunensis, S. marylandisciuri, S. massiliensis, S. microti, S. muscae, S. nepalensis, S. pasteuri, S. petrasii, S. petrasii croceilyticus, S. petrasii jettensis, S. petrasii petrasii, S. petrasii pragensis, S. pettenkoferi, S. piscifermentans, S. pragensis, S. pseudoxylosus, S. pulvereri, S. ratti, S. rostri, S. saccharolyticus, S. saprophyticus, S. saprophyticus bovis, S. saprophyticus saprophyticus, S. schleiferi, S. schleiferi schleiferi, S. sciuri, S. sciuri carnaticus, S. sciuri lentus, S. sciuri rodentium, S. sciuri sciuri, S. shinii, S. simulans, S. stepanovicii, S. succinus, S. succinus casei, S. succinus succinus, S. taiwanensis, S. urealyticus, S. ureilyticus, S. veratri, S. vitulinus, S. vitulus, S. warneri, S. xylosus Coagulase-positive: S. agnetis, S. argenteus, S. coagulans, S. cornubiensis, S. delphini, S. hyicus, S. hyicus chromogenes, S. hyicus hyicus, S. intermedius, S. lutrae, S. pseudintermedius, S. roterodami, S. schleiferi coagulans, S. schweitzeri, S. simiae, S. singaporensis based : Becker K et al. (2014). Coagulase-Negative Staphylococci. Clin Microbiol Rev. 27(4): 870-926; doi:10.1128/CMR.00109-13 Becker K et al. (2019). Implications identifying recently defined members S. aureus complex, S. argenteus S. schweitzeri: position paper members ESCMID Study Group staphylococci Staphylococcal Diseases (ESGS). Clin Microbiol Infect; doi:10.1016/j.cmi.2019.02.028 Becker K et al. (2020). Emergence coagulase-negative staphylococci. Expert Rev Anti Infect Ther. 18(4):349-366; doi:10.1080/14787210.2020.1730813 newly named staphylococcal species, S. brunensis (2024) S. shinii (2023), looked scientific reference make sure species considered correct coagulase group.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"lancefield-groups-in-streptococci","dir":"Reference","previous_headings":"","what":"Lancefield Groups in Streptococci","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Lancefield = TRUE, following streptococci converted corresponding Lancefield group: Streptococcus Group : S. pyogenes Streptococcus Group B: S. agalactiae Streptococcus Group C: S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis, S. equi, S. equi equi, S. equi ruminatorum, S. equi zooepidemicus Streptococcus Group F: S. anginosus, S. anginosus anginosus, S. anginosus whileyi, S. constellatus, S. constellatus constellatus, S. constellatus pharyngis, S. constellatus viborgensis, S. intermedius Streptococcus Group G: S. canis, S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis Streptococcus Group H: S. sanguinis Streptococcus Group K: S. salivarius, S. salivarius salivarius, S. salivarius thermophilus Streptococcus Group L: S. dysgalactiae, S. dysgalactiae dysgalactiae, S. dysgalactiae equisimilis based : Lancefield RC (1933). serological differentiation human groups hemolytic streptococci. J Exp Med. 57(4): 571-95; doi:10.1084/jem.57.4.571","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/ Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"ambiguous user input .mo() mo_* functions, returned results chosen based matching score using mo_matching_score(). matching score \\(m\\), calculated : $$m_{(x, n)} = \\frac{l_{n} - 0.5 \\cdot \\min \\begin{cases}l_{n} \\\\ \\textrm{lev}(x, n)\\end{cases}}{l_{n} \\cdot p_{n} \\cdot k_{n}}$$ : \\(x\\) user input; \\(n\\) taxonomic name (genus, species, subspecies); \\(l_n\\) length \\(n\\); \\(lev\\) Levenshtein distance function (counting insertion 1, deletion substitution 2) needed change \\(x\\) \\(n\\); \\(p_n\\) human pathogenic prevalence group \\(n\\), described ; \\(k_n\\) taxonomic kingdom \\(n\\), set Bacteria = 1, Fungi = 1.25, Protozoa = 1.5, Chromista = 1.75, Archaea = 2, others = 3. grouping human pathogenic prevalence \\(p\\) based recent work Bartlett et al. (2022, doi:10.1099/mic.0.001269 ) extensively studied medical-scientific literature categorise bacterial species groups: Established, taxonomic species infected least three persons three references. records prevalence = 1.15 microorganisms data set; Putative, taxonomic species fewer three known cases. records prevalence = 1.25 microorganisms data set. Furthermore, Genera World Health Organization's () Priority Pathogen List prevalence = 1.0 microorganisms data set; genus present established list also prevalence = 1.15 microorganisms data set; genus present putative list prevalence = 1.25 microorganisms data set; species subspecies genus present two aforementioned groups, prevalence = 1.5 microorganisms data set; non-bacterial genus, species subspecies genus present following list, prevalence = 1.25 microorganisms data set: Absidia, Acanthamoeba, Acremonium, Actinomucor, Aedes, Alternaria, Amoeba, Ancylostoma, Angiostrongylus, Anisakis, Anopheles, Apophysomyces, Arthroderma, Aspergillus, Aureobasidium, Basidiobolus, Beauveria, Bipolaris, Blastobotrys, Blastocystis, Blastomyces, Candida, Capillaria, Chaetomium, Chilomastix, Chrysonilia, Chrysosporium, Cladophialophora, Cladosporium, Clavispora, Coccidioides, Cokeromyces, Conidiobolus, Coniochaeta, Contracaecum, Cordylobia, Cryptococcus, Cryptosporidium, Cunninghamella, Curvularia, Cyberlindnera, Debaryozyma, Demodex, Dermatobia, Dientamoeba, Diphyllobothrium, Dirofilaria, Echinostoma, Entamoeba, Enterobius, Epidermophyton, Exidia, Exophiala, Exserohilum, Fasciola, Fonsecaea, Fusarium, Geotrichum, Giardia, Graphium, Haloarcula, Halobacterium, Halococcus, Hansenula, Hendersonula, Heterophyes, Histomonas, Histoplasma, Hortaea, Hymenolepis, Hypomyces, Hysterothylacium, Kloeckera, Kluyveromyces, Kodamaea, Lacazia, Leishmania, Lichtheimia, Lodderomyces, Lomentospora, Madurella, Malassezia, Malbranchea, Metagonimus, Meyerozyma, Microsporidium, Microsporum, Millerozyma, Mortierella, Mucor, Mycocentrospora, Nannizzia, Necator, Nectria, Ochroconis, Oesophagostomum, Oidiodendron, Opisthorchis, Paecilomyces, Paracoccidioides, Pediculus, Penicillium, Phaeoacremonium, Phaeomoniella, Phialophora, Phlebotomus, Phoma, Pichia, Piedraia, Pithomyces, Pityrosporum, Pneumocystis, Pseudallescheria, Pseudoscopulariopsis, Pseudoterranova, Pulex, Purpureocillium, Quambalaria, Rhinocladiella, Rhizomucor, Rhizopus, Rhodotorula, Saccharomyces, Saksenaea, Saprochaete, Sarcoptes, Scedosporium, Schistosoma, Schizosaccharomyces, Scolecobasidium, Scopulariopsis, Scytalidium, Spirometra, Sporobolomyces, Sporopachydermia, Sporothrix, Sporotrichum, Stachybotrys, Strongyloides, Syncephalastrum, Syngamus, Taenia, Talaromyces, Teleomorph, Toxocara, Trichinella, Trichobilharzia, Trichoderma, Trichomonas, Trichophyton, Trichosporon, Trichostrongylus, Trichuris, Tritirachium, Trombicula, Trypanosoma, Tunga, Ulocladium, Ustilago, Verticillium, Wallemia, Wangiella, Wickerhamomyces, Wuchereria, Yarrowia, Zygosaccharomyces; records prevalence = 2.0 microorganisms data set. calculating matching score, characters \\(x\\) \\(n\\) ignored -Z, -z, 0-9, spaces parentheses. matches sorted descending matching score user input values, top match returned. lead effect e.g., \"E. coli\" return microbial ID Escherichia coli (\\(m = 0.688\\), highly prevalent microorganism found humans) Entamoeba coli (\\(m = 0.381\\), less prevalent microorganism humans), although latter alphabetically come first.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.mo.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Transform Arbitrary Input to Valid Microbial Taxonomy — as.mo","text":"","code":"# \\donttest{ # These examples all return \"B_STPHY_AURS\", the ID of S. aureus: as.mo(c(   \"sau\", # WHONET code   \"stau\",   \"STAU\",   \"staaur\",   \"S. aureus\",   \"S aureus\",   \"Sthafilokkockus aureus\", # handles incorrect spelling   \"Staphylococcus aureus (MRSA)\",   \"MRSA\", # Methicillin Resistant S. aureus   \"VISA\", # Vancomycin Intermediate S. aureus   \"VRSA\", # Vancomycin Resistant S. aureus   115329001 # SNOMED CT code )) #> Class 'mo' #>  [1] B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS #>  [6] B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS B_STPHY_AURS #> [11] B_STPHY_AURS B_STPHY_AURS  # Dyslexia is no problem - these all work: as.mo(c(   \"Ureaplasma urealyticum\",   \"Ureaplasma urealyticus\",   \"Ureaplasmium urealytica\",   \"Ureaplazma urealitycium\" )) #> Class 'mo' #> [1] B_URPLS_URLY B_URPLS_URLY B_URPLS_URLY B_URPLS_URLY  # input will get cleaned up with the input given in the `cleaning_regex` argument, # which defaults to `mo_cleaning_regex()`: cat(mo_cleaning_regex(), \"\\n\") #> ([^A-Za-z- \\(\\)\\[\\]{}]+|([({]|\\[).+([})]|\\])|(^| )( ?[a-z-]+[-](resistant|susceptible) ?|e?spp([^a-z]+|$)|e?ssp([^a-z]+|$)|serogr.?up[a-z]*|e?ss([^a-z]+|$)|e?sp([^a-z]+|$)|var([^a-z]+|$)|serovar[a-z]*|sube?species|biovar[a-z]*|e?species|Ig[ADEGM]|e?subsp|biotype|titer|dummy))   as.mo(\"Streptococcus group A\") #> Class 'mo' #> [1] B_STRPT_GRPA  as.mo(\"S. epidermidis\") # will remain species: B_STPHY_EPDR #> Class 'mo' #> [1] B_STPHY_EPDR as.mo(\"S. epidermidis\", Becker = TRUE) # will not remain species: B_STPHY_CONS #> Class 'mo' #> [1] B_STPHY_CONS  as.mo(\"S. pyogenes\") # will remain species: B_STRPT_PYGN #> Class 'mo' #> [1] B_STRPT_PYGN as.mo(\"S. pyogenes\", Lancefield = TRUE) # will not remain species: B_STRPT_GRPA #> Class 'mo' #> [1] B_STRPT_GRPA  # All mo_* functions use as.mo() internally too (see ?mo_property): mo_genus(\"E. coli\") #> [1] \"Escherichia\" mo_gramstain(\"ESCO\") #> [1] \"Gram-negative\" mo_is_intrinsic_resistant(\"ESCCOL\", ab = \"vanco\") #> ℹ Determining intrinsic resistance based on 'EUCAST Expert Rules' and #>   'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3 (2021). This note #>   will be shown once per session. #> [1] TRUE # }"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":null,"dir":"Reference","previous_headings":"","what":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Clean existing SIR values, interpret minimum inhibitory concentration (MIC) values disk diffusion diameters according EUCAST CLSI. .sir() transforms input new class sir, ordered factor containing levels S, SDD, , R, NI. breakpoints currently implemented: clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024; veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024; ECOFFs (Epidemiological Cut-Values): EUCAST 2020-2024 CLSI 2022-2024. breakpoints used interpretation available clinical_breakpoints data set.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"","code":"as.sir(x, ...)  NA_sir_  is.sir(x)  is_sir_eligible(x, threshold = 0.05)  # Default S3 method as.sir(x, S = \"^(S|U)+$\", I = \"^(I)+$\", R = \"^(R)+$\",   NI = \"^(N|NI|V)+$\", SDD = \"^(SDD|D|H)+$\", ...)  # S3 method for class 'mic' as.sir(x, mo = NULL, ab = deparse(substitute(x)),   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   conserve_capped_values = FALSE, add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE, ...)  # S3 method for class 'disk' as.sir(x, mo = NULL, ab = deparse(substitute(x)),   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE, ...)  # S3 method for class 'data.frame' as.sir(x, ..., col_mo = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"), uti = NULL,   conserve_capped_values = FALSE, add_intrinsic_resistance = FALSE,   reference_data = AMR::clinical_breakpoints,   include_screening = getOption(\"AMR_include_screening\", FALSE),   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), host = NULL,   verbose = FALSE)  sir_interpretation_history(clean = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"interpretations minimum inhibitory concentration (MIC) values disk diffusion diameters: CLSI M39: Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 2011-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/. CLSI M100: Performance Standard Antimicrobial Susceptibility Testing, 2011-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m100/. CLSI VET01: Performance Standards Antimicrobial Disk Dilution Susceptibility Tests Bacteria Isolated Animals, 2019-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/veterinary-medicine/documents/vet01/. CLSI VET09: Understanding Susceptibility Test Data Component Antimicrobial Stewardship Veterinary Settings, 2019-2024, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/veterinary-medicine/documents/vet09/. EUCAST Breakpoint tables interpretation MICs zone diameters, 2011-2024, European Committee Antimicrobial Susceptibility Testing (EUCAST). https://www.eucast.org/clinical_breakpoints. WHONET source machine-reading clinical breakpoints (read ), 1989-2024, Collaborating Centre Surveillance Antimicrobial Resistance. https://whonet.org/.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"x vector values (class mic: MIC values mg/L, class disk: disk diffusion radius millimetres) ... using data.frame: names columns apply .sir() (supports tidy selection column1:column4). Otherwise: arguments passed methods. threshold maximum fraction invalid antimicrobial interpretations x, see Examples S, , R, NI, SDD case-independent regular expression translate input result. regular expression run non-letters whitespaces removed input. mo vector (column name) characters can coerced valid microorganism codes .mo(), can left empty determine automatically ab vector (column name) characters can coerced valid antimicrobial drug code .ab() guideline defaults EUCAST 2024 (latest implemented EUCAST guideline clinical_breakpoints data set), can set package option AMR_guideline. Currently supports EUCAST (2011-2024) CLSI (2011-2024), see Details. uti (Urinary Tract Infection) vector (column name) logicals (TRUE FALSE) specify whether UTI specific interpretation guideline chosen. using .sir() data.frame, can also column containing logicals left blank, data set searched column 'specimen', rows within column containing 'urin' ('urine', 'urina') regarded isolates UTI. See Examples. conserve_capped_values logical indicate MIC values starting \">\" (\">=\") must always return \"R\" , MIC values starting \"<\" (\"<=\") must always return \"S\" add_intrinsic_resistance (useful using EUCAST guideline) logical indicate whether intrinsic antibiotic resistance must also considered applicable bug-drug combinations, meaning e.g. ampicillin always return \"R\" Klebsiella species. Determination based intrinsic_resistant data set, based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021). reference_data data.frame used interpretation, defaults clinical_breakpoints data set. Changing argument allows using interpretation guidelines. argument must contain data set equal structure clinical_breakpoints data set (column names column types). Please note guideline argument ignored reference_data manually set. include_screening logical indicate clinical breakpoints screening allowed - default FALSE. Can also set package option AMR_include_screening. include_PKPD logical indicate PK/PD clinical breakpoints must applied last resort - default TRUE. Can also set package option AMR_include_PKPD. breakpoint_type type breakpoints use, either \"ECOFF\", \"animal\", \"human\". ECOFF stands Epidemiological Cut-values. default \"human\", can also set package option AMR_breakpoint_type. host set values veterinary species, automatically set \"animal\". host vector (column name) characters indicate host. useful veterinary breakpoints, requires breakpoint_type = \"animal\". values can text resembling animal species, even 20 supported languages package. foreign languages, sure set language set_AMR_locale() (though automatically guessed based system language). verbose logical indicate notes printed interpretation MIC values disk diffusion values. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). clean logical indicate whether previously stored results forgotten returning 'logbook' results","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Ordered factor new class sir","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"Note: clinical breakpoints package validated , imported , WHONET. public use AMR package endorsed CLSI EUCAST. See clinical_breakpoints information. NA_sir_ missing value new sir class, analogous e.g. base R's NA_character_.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"how-it-works","dir":"Reference","previous_headings":"","what":"How it Works","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":".sir() function can work four ways: cleaning raw / untransformed data. data cleaned contain valid values, namely: S susceptible, intermediate 'susceptible, increased exposure', R resistant, NI non-interpretable, SDD susceptible dose-dependent. can set using regular expression. Furthermore, .sir() try best clean intelligence. example, mixed values SIR interpretations MIC values \"<0.25; S\" coerced \"S\". Combined interpretations multiple test methods (seen laboratory records) \"S; S\" coerced \"S\", value like \"S; \" return NA warning input invalid. interpreting minimum inhibitory concentration (MIC) values according EUCAST CLSI. must clean MIC values first using .mic(), also gives columns new data class mic. Also, sure column microorganism names codes. found automatically, can set manually using mo argument. Using dplyr, SIR interpretation can done easily either:   Operators like \"<=\" stripped interpretation. using conserve_capped_values = TRUE, MIC value e.g. \">2\" always return \"R\", even breakpoint according chosen guideline \">=4\". prevent capped values raw laboratory data treated conservatively. default behaviour (conserve_capped_values = FALSE) considers \">2\" lower \">=4\" might case return \"S\" \"\". interpreting disk diffusion diameters according EUCAST CLSI. must clean disk zones first using .disk(), also gives columns new data class disk. Also, sure column microorganism names codes. found automatically, can set manually using mo argument. Using dplyr, SIR interpretation can done easily either:   interpreting complete data set, automatic determination MIC values, disk diffusion diameters, microorganism names codes, antimicrobial test results. done simply running .sir(your_data). points 2, 3 4: Use sir_interpretation_history() retrieve data.frame (tibble tibble package installed) results last .sir() call.","code":"your_data %>% mutate_if(is.mic, as.sir) your_data %>% mutate(across(where(is.mic), as.sir)) your_data %>% mutate_if(is.mic, as.sir, ab = \"column_with_antibiotics\", mo = \"column_with_microorganisms\") your_data %>% mutate_if(is.mic, as.sir, ab = c(\"cipro\", \"ampicillin\", ...), mo = c(\"E. coli\", \"K. pneumoniae\", ...))  # for veterinary breakpoints, also set `host`: your_data %>% mutate_if(is.mic, as.sir, host = \"column_with_animal_species\", guideline = \"CLSI\") your_data %>% mutate_if(is.disk, as.sir) your_data %>% mutate(across(where(is.disk), as.sir)) your_data %>% mutate_if(is.disk, as.sir, ab = \"column_with_antibiotics\", mo = \"column_with_microorganisms\") your_data %>% mutate_if(is.disk, as.sir, ab = c(\"cipro\", \"ampicillin\", ...), mo = c(\"E. coli\", \"K. pneumoniae\", ...))  # for veterinary breakpoints, also set `host`: your_data %>% mutate_if(is.disk, as.sir, host = \"column_with_animal_species\", guideline = \"CLSI\")"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"supported-guidelines","dir":"Reference","previous_headings":"","what":"Supported Guidelines","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"interpreting MIC values well disk diffusion diameters, currently implemented guidelines clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024, veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024. Thus, guideline argument must set e.g., \"EUCAST 2024\" \"CLSI 2024\". simply using \"EUCAST\" (default) \"CLSI\" input, latest included version guideline automatically selected. can set data set using reference_data argument. guideline argument ignored. can set default guideline package option AMR_guideline (e.g. .Rprofile file), :   veterinary guidelines, might best options:   applying veterinary breakpoints (setting host setting breakpoint_type = \"animal\"), CLSI VET09 guideline applied cope missing animal species-specific breakpoints.","code":"options(AMR_guideline = \"CLSI\")   options(AMR_guideline = \"CLSI 2018\")   options(AMR_guideline = \"EUCAST 2020\")   # or to reset:   options(AMR_guideline = NULL) options(AMR_guideline = \"CLSI\")   options(AMR_breakpoint_type = \"animal\")"},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"after-interpretation","dir":"Reference","previous_headings":"","what":"After Interpretation","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"using .sir(), can use eucast_rules() defined EUCAST (1) apply inferred susceptibility resistance based results antimicrobials (2) apply intrinsic resistance based taxonomic properties microorganism. determine isolates multi-drug resistant, sure run mdro() (applies MDR/PDR/XDR guideline 2012 default) data set contains S//R values. Read interpreting multidrug-resistant organisms .","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"machine-readable-clinical-breakpoints","dir":"Reference","previous_headings":"","what":"Machine-Readable Clinical Breakpoints","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"repository package contains machine-readable version guidelines. CSV file consisting 34 063 rows 14 columns. file machine-readable, since contains one row every unique combination test method (MIC disk diffusion), antimicrobial drug microorganism. allows easy implementation rules laboratory information systems (LIS). Note contains interpretation guidelines humans - interpretation guidelines CLSI animals removed.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"other","dir":"Reference","previous_headings":"","what":"Other","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"function .sir() detects input contains class sir. input data.frame, iterates columns returns logical vector. base R function .double() can used retrieve quantitative values sir object: \"S\" = 1, \"\"/\"SDD\" = 2, \"R\" = 3. values rendered NA . Note: use .integer(), since (R works internally) return factor level indices, aforementioned quantitative values. function is_sir_eligible() returns TRUE column contains 5% invalid antimicrobial interpretations (S //R /NI /SDD), FALSE otherwise. threshold 5% can set threshold argument. input data.frame, iterates columns returns logical vector.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/as.sir.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Translate MIC and Disk Diffusion to SIR, or Clean Existing SIR Data — as.sir","text":"","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … summary(example_isolates) # see all SIR results at a glance #>       date              patient               age           gender          #>  Min.   :2002-01-02   Length:2000        Min.   : 0.00   Length:2000        #>  1st Qu.:2005-07-31   Class :character   1st Qu.:63.00   Class :character   #>  Median :2009-07-31   Mode  :character   Median :74.00   Mode  :character   #>  Mean   :2009-11-20                      Mean   :70.69                      #>  3rd Qu.:2014-05-30                      3rd Qu.:82.00                      #>  Max.   :2017-12-28                      Max.   :97.00                      #>      ward                mo                  PEN                 #>  Length:2000        Class :mo             Class:sir              #>  Class :character   <NA>  :0              %S   :25.6% (n=417)    #>  Mode  :character   Unique:90             %SDD : 0.0% (n=0)      #>                     #1    :B_ESCHR_COLI   %I   : 0.7% (n=11)     #>                     #2    :B_STPHY_CONS   %R   :73.7% (n=1201)   #>                     #3    :B_STPHY_AURS   %NI  : 0.0% (n=0)      #>     OXA                   FLC                   AMX                #>  Class:sir             Class:sir             Class:sir             #>  %S   :68.8% (n=251)   %S   :70.5% (n=665)   %S   :40.2% (n=543)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)     %I   : 0.2% (n=3)     #>  %R   :31.2% (n=114)   %R   :29.5% (n=278)   %R   :59.6% (n=804)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     AMC                    AMP                   TZP                #>  Class:sir              Class:sir             Class:sir             #>  %S   :71.4% (n=1342)   %S   :40.2% (n=543)   %S   :86.1% (n=862)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 4.8% (n=91)     %I   : 0.2% (n=3)     %I   : 1.3% (n=13)    #>  %R   :23.7% (n=446)    %R   :59.6% (n=804)   %R   :12.6% (n=126)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     CZO                   FEP                   CXM                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :54.9% (n=245)   %S   :85.6% (n=620)   %S   :72.5% (n=1297)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 0.4% (n=2)     %I   : 0.1% (n=1)     %I   : 1.2% (n=22)     #>  %R   :44.6% (n=199)   %R   :14.2% (n=103)   %R   :26.3% (n=470)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     FOX                   CTX                   CAZ                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :71.6% (n=586)   %S   :84.4% (n=796)   %S   :33.5% (n=607)    #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 1.0% (n=8)     %I   : 0.1% (n=1)     %I   : 0.0% (n=0)      #>  %R   :27.4% (n=224)   %R   :15.5% (n=146)   %R   :66.5% (n=1204)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     CRO                   GEN                    TOB                #>  Class:sir             Class:sir              Class:sir             #>  %S   :84.4% (n=796)   %S   :74.0% (n=1372)   %S   :65.1% (n=879)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.1% (n=1)     %I   : 1.5% (n=27)     %I   : 0.5% (n=7)     #>  %R   :15.5% (n=146)   %R   :24.6% (n=456)    %R   :34.4% (n=465)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     AMK                   KAN                    TMP                #>  Class:sir             Class:sir              Class:sir             #>  %S   :36.3% (n=251)   %S   : 0.0% (n=0)      %S   :61.2% (n=918)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)      %I   : 0.7% (n=10)    #>  %R   :63.7% (n=441)   %R   :100.0% (n=471)   %R   :38.1% (n=571)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     SXT                    NIT                   FOS                #>  Class:sir              Class:sir             Class:sir             #>  %S   :79.1% (n=1392)   %S   :76.0% (n=565)   %S   :57.8% (n=203)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.3% (n=6)      %I   : 6.9% (n=51)    %I   : 0.0% (n=0)     #>  %R   :20.5% (n=361)    %R   :17.1% (n=127)   %R   :42.2% (n=148)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     LNZ                   CIP                    MFX                #>  Class:sir             Class:sir              Class:sir             #>  %S   :30.7% (n=314)   %S   :78.9% (n=1112)   %S   :64.5% (n=136)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 4.9% (n=69)     %I   : 1.9% (n=4)     #>  %R   :69.3% (n=709)   %R   :16.2% (n=228)    %R   :33.6% (n=71)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     VAN                    TEC                   TCY                #>  Class:sir              Class:sir             Class:sir             #>  %S   :61.7% (n=1149)   %S   :24.3% (n=237)   %S   :68.3% (n=820)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)      %I   : 0.0% (n=0)     %I   : 1.9% (n=23)    #>  %R   :38.3% (n=712)    %R   :75.7% (n=739)   %R   :29.8% (n=357)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     #>     TGC                   DOX                   ERY                 #>  Class:sir             Class:sir             Class:sir              #>  %S   :87.3% (n=697)   %S   :71.7% (n=814)   %S   :42.3% (n=801)    #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      #>  %I   : 0.0% (n=0)     %I   : 0.6% (n=7)     %I   : 0.5% (n=9)      #>  %R   :12.7% (n=101)   %R   :27.7% (n=315)   %R   :57.2% (n=1084)   #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      #>     CLI                   AZM                    IPM                #>  Class:sir             Class:sir              Class:sir             #>  %S   :38.6% (n=586)   %S   :42.3% (n=801)    %S   :92.7% (n=824)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     #>  %I   : 0.3% (n=4)     %I   : 0.5% (n=9)      %I   : 1.1% (n=10)    #>  %R   :61.2% (n=930)   %R   :57.2% (n=1084)   %R   : 6.2% (n=55)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     #>     MEM                   MTR                  CHL                #>  Class:sir             Class:sir            Class:sir             #>  %S   :94.1% (n=780)   %S   :85.3% (n=29)   %S   :78.6% (n=121)   #>  %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)    %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)     %I   : 0.0% (n=0)    %I   : 0.0% (n=0)     #>  %R   : 5.9% (n=49)    %R   :14.7% (n=5)    %R   :21.4% (n=33)    #>  %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)    %NI  : 0.0% (n=0)     #>     COL                    MUP                   RIF                #>  Class:sir              Class:sir             Class:sir             #>  %S   :18.8% (n=309)    %S   :93.0% (n=251)   %S   :30.2% (n=303)   #>  %SDD : 0.0% (n=0)      %SDD : 0.0% (n=0)     %SDD : 0.0% (n=0)     #>  %I   : 0.0% (n=0)      %I   : 1.1% (n=3)     %I   : 0.2% (n=2)     #>  %R   :81.2% (n=1331)   %R   : 5.9% (n=16)    %R   :69.6% (n=698)   #>  %NI  : 0.0% (n=0)      %NI  : 0.0% (n=0)     %NI  : 0.0% (n=0)      # For INTERPRETING disk diffusion and MIC values -----------------------  # example data sets, with combined MIC values and disk zones df_wide <- data.frame(   microorganism = \"Escherichia coli\",   amoxicillin = as.mic(8),   cipro = as.mic(0.256),   tobra = as.disk(16),   genta = as.disk(18),   ERY = \"R\" ) df_long <- data.frame(   bacteria = rep(\"Escherichia coli\", 4),   antibiotic = c(\"amoxicillin\", \"cipro\", \"tobra\", \"genta\"),   mics = as.mic(c(0.01, 1, 4, 8)),   disks = as.disk(c(6, 10, 14, 18)) )  # \\donttest{ ## Using dplyr ------------------------------------------------- if (require(\"dplyr\")) {   # approaches that all work without additional arguments:   df_wide %>% mutate_if(is.mic, as.sir)   df_wide %>% mutate_if(function(x) is.mic(x) | is.disk(x), as.sir)   df_wide %>% mutate(across(where(is.mic), as.sir))   df_wide %>% mutate_at(vars(amoxicillin:tobra), as.sir)   df_wide %>% mutate(across(amoxicillin:tobra, as.sir))      # approaches that all work with additional arguments:   df_long %>%     # given a certain data type, e.g. MIC values     mutate_if(is.mic, as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               guideline = \"CLSI\")   df_long %>%     mutate(across(where(is.mic),                   function(x) as.sir(x,                                      mo = \"bacteria\",                                      ab = \"antibiotic\",                                      guideline = \"CLSI\")))   df_wide %>%     # given certain columns, e.g. from 'cipro' to 'genta'     mutate_at(vars(cipro:genta), as.sir,               mo = \"bacteria\",               guideline = \"CLSI\")   df_wide %>%     mutate(across(cipro:genta,                        function(x) as.sir(x,                                           mo = \"bacteria\",                                           guideline = \"CLSI\")))                           # for veterinary breakpoints, add 'host':   df_long$animal_species <- c(\"cats\", \"dogs\", \"horses\", \"cattle\")   df_long %>%     # given a certain data type, e.g. MIC values     mutate_if(is.mic, as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               host = \"animal_species\",               guideline = \"CLSI\")   df_long %>%     mutate(across(where(is.mic),                   function(x) as.sir(x,                                      mo = \"bacteria\",                                      ab = \"antibiotic\",                                      host = \"animal_species\",                                      guideline = \"CLSI\")))   df_wide %>%     mutate_at(vars(cipro:genta), as.sir,               mo = \"bacteria\",               ab = \"antibiotic\",               host = \"animal_species\",               guideline = \"CLSI\")   df_wide %>%     mutate(across(cipro:genta,                        function(x) as.sir(x,                                           mo = \"bacteria\",                                           host = \"animal_species\",                                           guideline = \"CLSI\")))      # to include information about urinary tract infections (UTI)   data.frame(mo = \"E. coli\",              nitrofuratoin = c(\"<= 2\", 32),              from_the_bladder = c(TRUE, FALSE)) %>%     as.sir(uti = \"from_the_bladder\")    data.frame(mo = \"E. coli\",              nitrofuratoin = c(\"<= 2\", 32),              specimen = c(\"urine\", \"blood\")) %>%     as.sir() # automatically determines urine isolates    df_wide %>%     mutate_at(vars(cipro:genta), as.sir, mo = \"E. coli\", uti = TRUE) } #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for gentamicin (GEN) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin) based on column #> 'microorganism', EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin) based on #> column 'microorganism', EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'amoxicillin' (AMX), 'cipro' (CIP, ciprofloxacin), #> 'tobra' (TOB, tobramycin), and 'genta' (GEN, gentamicin) based on column #> 'bacteria', CLSI 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'antibiotic' (TESTAB, test Antibiotic), CLSI #> 2024... #>  WARNING  #>   • No CLSI 2024 MIC breakpoints available for test Antibiotic (TESTAB). #> Interpreting disk diffusion zones: 'antibiotic' (TESTAB, test Antibiotic), #> CLSI 2024... #>  WARNING  #>   • No CLSI 2024 DISK breakpoints available for test Antibiotic (TESTAB). #> Interpreting disk diffusion zones: 'antibiotic' (TESTAB, test Antibiotic), #> CLSI 2024... #>  WARNING  #>   • No CLSI 2024 DISK breakpoints available for test Antibiotic (TESTAB). #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `cipro = (function (x, ...) ...`. #> Caused by warning: #> ! The following animal host(s) could not be coerced: \"animal_species\" #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> ℹ Assuming breakpoint_type = \"animal\", since host is set. #>  #>  #> ℹ Please note that in the absence of specific veterinary breakpoints for #>   certain animal hosts, the CLSI guideline VET09 will be applied where #>   possible. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), CLSI 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), CLSI 2024... #>  OK  #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `across(...)`. #> Caused by warning: #> ! The following animal host(s) could not be coerced: \"animal_species\" #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'nitrofuratoin' (NIT, #> nitrofurantoin), EUCAST 2024... #>  OK  #> ℹ Assuming value \"urine\" in column 'specimen' reflects a urinary tract #>   infection. #>   Use as.sir(uti = FALSE) to prevent this. #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: column 'nitrofuratoin' (NIT, #> nitrofurantoin), EUCAST 2024... #>  OK  #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'cipro' (CIP, ciprofloxacin), EUCAST 2024... #>  OK  #> Interpreting disk diffusion zones: 'tobra' (TOB, tobramycin), EUCAST #> 2024... #>  OK  #> Interpreting disk diffusion zones: 'genta' (GEN, gentamicin), EUCAST #> 2024... #>  OK  #>      microorganism amoxicillin cipro tobra genta ERY #> 1 Escherichia coli           8  <NA>     S     S   R   ## Using base R ------------------------------------------------  as.sir(df_wide) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: column 'amoxicillin' (AMX), EUCAST 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for amoxicillin (AMX) in Escherichia coli - assuming body site 'Intravenous'. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting MIC values: column 'cipro' (CIP, ciprofloxacin), EUCAST #> 2024... #>  NOTE  #>   • Multiple breakpoints available for ciprofloxacin (CIP) in Escherichia coli - assuming body site 'Non-meningitis'. #> Interpreting disk diffusion zones: column 'tobra' (TOB, tobramycin), EUCAST #> 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for tobramycin (TOB) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Interpreting disk diffusion zones: column 'genta' (GEN, gentamicin), EUCAST #> 2024... #>  NOTE  #>   • Breakpoints for UTI and non-UTI available for gentamicin (GEN) in Escherichia coli - assuming an unspecified body site. Use argument uti to set which isolates are from urine. See ?as.sir. #> Assigning class 'sir' to already clean column 'ERY' (erythromycin)... #>  OK  #>      microorganism amoxicillin cipro tobra genta ERY #> 1 Escherichia coli           S     I     S     S   R  # return a 'logbook' about the results: sir_interpretation_history() #> # A tibble: 57 × 16 #>    datetime            index ab_given mo_given host_given ab   mo               #>  * <dttm>              <int> <chr>    <chr>    <chr>      <ab> <mo>             #>  1 2025-02-22 21:12:34     4 AMX      B_STRPT… human      AMX  B_STRPT_PNMN     #>  2 2025-02-22 21:12:41     4 genta    Escheri… human      GEN  B_[ORD]_ENTRBCTR #>  3 2025-02-22 21:12:41     4 genta    Escheri… human      GEN  B_[ORD]_ENTRBCTR #>  4 2025-02-22 21:12:42     4 genta    Escheri… cattle     GEN  B_ESCHR_COLI     #>  5 2025-02-22 21:12:42     4 genta    Escheri… cattle     GEN  B_ESCHR_COLI     #>  6 2025-02-22 21:12:34     3 AMX      B_STRPT… human      AMX  B_STRPT_PNMN     #>  7 2025-02-22 21:12:41     3 tobra    Escheri… human      TOB  B_[ORD]_ENTRBCTR #>  8 2025-02-22 21:12:41     3 tobra    Escheri… human      TOB  B_[ORD]_ENTRBCTR #>  9 2025-02-22 21:12:42     3 tobra    Escheri… horses     TOB  B_ESCHR_COLI     #> 10 2025-02-22 21:12:42     3 tobra    Escheri… horses     TOB  B_ESCHR_COLI     #> # ℹ 47 more rows #> # ℹ 9 more variables: host <chr>, method <chr>, input <chr>, outcome <sir>, #> #   notes <chr>, guideline <chr>, ref_table <chr>, uti <lgl>, #> #   breakpoint_S_R <chr>  # for single values as.sir(   x = as.mic(2),   mo = as.mo(\"S. pneumoniae\"),   ab = \"AMP\",   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting MIC values: 'AMP' (ampicillin), EUCAST 2024... #>  NOTE  #>   • Multiple breakpoints available for ampicillin (AMP) in Streptococcus pneumoniae - assuming body site 'Non-meningitis'. #> Class 'sir' #> [1] R  as.sir(   x = as.disk(18),   mo = \"Strep pneu\", # `mo` will be coerced with as.mo()   ab = \"ampicillin\", # and `ab` with as.ab()   guideline = \"EUCAST\" ) #>  #> ℹ Run sir_interpretation_history() afterwards to retrieve a logbook with #>   all the details of the breakpoint interpretations. #>  #> Interpreting disk diffusion zones: 'ampicillin' (AMP), EUCAST 2024... #>  OK  #> Class 'sir' #> [1] R   # For CLEANING existing SIR values ------------------------------------  as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", \"A\", \"B\", \"C\")) #> Warning: in as.sir(): 3 results in column '20' truncated (38%) that were invalid #> antimicrobial interpretations: \"A\", \"B\", and \"C\" as.sir(\"<= 0.002; S\") # will return \"S\" sir_data <- as.sir(c(rep(\"S\", 474), rep(\"I\", 36), rep(\"R\", 370))) is.sir(sir_data) #> [1] TRUE plot(sir_data) # for percentages  barplot(sir_data) # for frequencies   # as common in R, you can use as.integer() to return factor indices: as.integer(as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", NA))) #> [1]  1  2  3  4  5 NA # but for computational use, as.double() will return 1 for S, 2 for I/SDD, and 3 for R: as.double(as.sir(c(\"S\", \"SDD\", \"I\", \"R\", \"NI\", NA))) #> [1]  1  2  2  3 NA NA  # the dplyr way if (require(\"dplyr\")) {   example_isolates %>%     mutate_at(vars(PEN:RIF), as.sir)   # same:   example_isolates %>%     as.sir(PEN:RIF)    # fastest way to transform all columns with already valid AMR results to class `sir`:   example_isolates %>%     mutate_if(is_sir_eligible, as.sir)    # since dplyr 1.0.0, this can also be:   # example_isolates %>%   #   mutate(across(where(is_sir_eligible), as.sir)) } #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":null,"dir":"Reference","previous_headings":"","what":"Get ATC Properties from WHOCC Website — atc_online_property","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"Gets data WHOCC website determine properties Anatomical Therapeutic Chemical (ATC) (e.g. antibiotic), name, defined daily dose (DDD) standard unit.","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"","code":"atc_online_property(atc_code, property, administration = \"O\",   url = \"https://atcddd.fhi.no/atc_ddd_index/?code=%s&showdescription=no\",   url_vet = \"https://atcddd.fhi.no/atcvet/atcvet_index/?code=%s&showdescription=no\")  atc_online_groups(atc_code, ...)  atc_online_ddd(atc_code, ...)  atc_online_ddd_units(atc_code, ...)"},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"https://atcddd.fhi./atc_ddd_alterations__cumulative/ddd_alterations/abbrevations/","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"atc_code character (vector) ATC code(s) antibiotics, coerced .ab() ab_atc() internally valid ATC code property property ATC code. Valid values \"ATC\", \"Name\", \"DDD\", \"U\" (\"unit\"), \"Adm.R\", \"Note\" groups. last option, hierarchical groups ATC code returned, see Examples. administration type administration using property = \"Adm.R\", see Details url url website WHOCC. sign %s can used placeholder ATC codes. url_vet url website WHOCC veterinary medicine. sign %s can used placeholder ATC_vet codes (start \"Q\"). ... arguments pass atc_property","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"Options argument administration: \"Implant\" = Implant \"Inhal\" = Inhalation \"Instill\" = Instillation \"N\" = nasal \"O\" = oral \"P\" = parenteral \"R\" = rectal \"SL\" = sublingual/buccal \"TD\" = transdermal \"V\" = vaginal Abbreviations return values using property = \"U\" (unit): \"g\" = gram \"mg\" = milligram \"mcg\" = microgram \"U\" = unit \"TU\" = thousand units \"MU\" = million units \"mmol\" = millimole \"ml\" = millilitre (e.g. eyedrops) N.B. function requires internet connection works following packages installed: curl, rvest, xml2.","code":""},{"path":"https://msberends.github.io/AMR/reference/atc_online.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get ATC Properties from WHOCC Website — atc_online_property","text":"","code":"# \\donttest{ if (requireNamespace(\"curl\") && requireNamespace(\"rvest\") && requireNamespace(\"xml2\")) {   # oral DDD (Defined Daily Dose) of amoxicillin   atc_online_property(\"J01CA04\", \"DDD\", \"O\")   atc_online_ddd(ab_atc(\"amox\"))    # parenteral DDD (Defined Daily Dose) of amoxicillin   atc_online_property(\"J01CA04\", \"DDD\", \"P\")    atc_online_property(\"J01CA04\", property = \"groups\") # search hierarchical groups of amoxicillin } #> Loading required namespace: rvest #> [1] \"ANTIINFECTIVES FOR SYSTEMIC USE\"         #> [2] \"ANTIBACTERIALS FOR SYSTEMIC USE\"         #> [3] \"BETA-LACTAM ANTIBACTERIALS, PENICILLINS\" #> [4] \"Penicillins with extended spectrum\"      # }"},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":null,"dir":"Reference","previous_headings":"","what":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"Use function e.g. clinical texts health care records. returns list antiviral drugs, doses forms administration found texts.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"","code":"av_from_text(text, type = c(\"drug\", \"dose\", \"administration\"),   collapse = NULL, translate_av = FALSE, thorough_search = NULL,   info = interactive(), ...)"},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"text text analyse type type property search , either \"drug\", \"dose\" \"administration\", see Examples collapse character pass paste(, collapse = ...) return one character per element text, see Examples translate_av type = \"drug\": column name antivirals data set translate antibiotic abbreviations , using av_property(). default FALSE. Using TRUE equal using \"name\". thorough_search logical indicate whether input must extensively searched misspelling faulty input values. Setting TRUE take considerably time using FALSE. default, turn TRUE input elements contain maximum three words. info logical indicate whether progress bar printed - default TRUE interactive mode ... arguments passed .av()","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"list,  character collapse NULL","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"function also internally used .av(), although searches first drug name throw note drug names returned. Note: .av() function may use long regular expression match brand names antiviral drugs. may fail systems.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"argument-type","dir":"Reference","previous_headings":"","what":"Argument type","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"default, function search antiviral drug names. text elements searched official names, ATC codes brand names. uses .av() internally, correct misspelling. type = \"dose\" (similar, like \"dosing\", \"doses\"), text elements searched numeric values higher 100 resemble years. output numeric. supports unit (g, mg, IE, etc.) multiple values one clinical text, see Examples. type = \"administration\" (abbreviations, like \"admin\", \"adm\"), text elements searched form drug administration. supports following forms (including common abbreviations): buccal, implant, inhalation, instillation, intravenous, nasal, oral, parenteral, rectal, sublingual, transdermal vaginal. Abbreviations oral ('po', 'per os') become \"oral\", values intravenous ('iv', 'intraven') become \"iv\". supports multiple values one clinical text, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"argument-collapse","dir":"Reference","previous_headings":"","what":"Argument collapse","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"Without using collapse, function return list. can convenient use e.g. inside mutate()):df %>% mutate(avx = av_from_text(clinical_text)) returned AV codes can transformed official names, groups, etc. av_* functions av_name() av_group(), using translate_av argument. using collapse, function return character:df %>% mutate(avx = av_from_text(clinical_text, collapse = \"|\"))","code":""},{"path":"https://msberends.github.io/AMR/reference/av_from_text.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Retrieve Antiviral Drug Names and Doses from Clinical Text — av_from_text","text":"","code":"av_from_text(\"28/03/2020 valaciclovir po tid\") #> [[1]] #> Class 'av' #> [1] VALA #>  av_from_text(\"28/03/2020 valaciclovir po tid\", type = \"admin\") #> [[1]] #> [1] \"oral\" #>"},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of an Antiviral Drug — av_property","title":"Get Properties of an Antiviral Drug — av_property","text":"Use functions return specific property antiviral drug antivirals data set. input values evaluated internally .av().","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of an Antiviral Drug — av_property","text":"","code":"av_name(x, language = get_AMR_locale(), tolower = FALSE, ...)  av_cid(x, ...)  av_synonyms(x, ...)  av_tradenames(x, ...)  av_group(x, language = get_AMR_locale(), ...)  av_atc(x, ...)  av_loinc(x, ...)  av_ddd(x, administration = \"oral\", ...)  av_ddd_units(x, administration = \"oral\", ...)  av_info(x, language = get_AMR_locale(), ...)  av_url(x, open = FALSE, ...)  av_property(x, property = \"name\", language = get_AMR_locale(), ...)"},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of an Antiviral Drug — av_property","text":"x (vector ) text can coerced valid antiviral drug code .av() language language returned text - default system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. tolower logical indicate whether first character every output transformed lower case character. ... arguments passed .av() administration way administration, either \"oral\" \"iv\" open browse URL using utils::browseURL() property one column names one antivirals data set: vector_or(colnames(antivirals), sort = FALSE).","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of an Antiviral Drug — av_property","text":"integer case av_cid() named list case av_info() multiple av_atc()/av_synonyms()/av_tradenames() double case av_ddd() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of an Antiviral Drug — av_property","text":"output translated possible. function av_url() return direct URL official website. warning returned required ATC code available.","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of an Antiviral Drug — av_property","text":"World Health Organization () Collaborating Centre Drug Statistics Methodology: https://atcddd.fhi./atc_ddd_index/ European Commission Public Health PHARMACEUTICALS - COMMUNITY REGISTER: https://ec.europa.eu/health/documents/community-register/html/reg_hum_atc.htm","code":""},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of an Antiviral Drug — av_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/av_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of an Antiviral Drug — av_property","text":"","code":"# all properties: av_name(\"ACI\") #> [1] \"Aciclovir\" av_atc(\"ACI\") #> [1] \"J05AB01\" av_cid(\"ACI\") #> [1] 135398513 av_synonyms(\"ACI\") #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            av_tradenames(\"ACI\") #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            av_group(\"ACI\") #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" av_url(\"ACI\") #>                                                              Aciclovir  #> \"https://atcddd.fhi.no/atc_ddd_index/?code=J05AB01&showdescription=no\"   # lowercase transformation av_name(x = c(\"ACI\", \"VALA\")) #> [1] \"Aciclovir\"    \"Valaciclovir\" av_name(x = c(\"ACI\", \"VALA\"), tolower = TRUE) #> [1] \"aciclovir\"    \"valaciclovir\"  # defined daily doses (DDD) av_ddd(\"ACI\", \"oral\") #> [1] 4 av_ddd_units(\"ACI\", \"oral\") #> [1] \"g\" av_ddd(\"ACI\", \"iv\") #> [1] 4 av_ddd_units(\"ACI\", \"iv\") #> [1] \"g\"  av_info(\"ACI\") # all properties as a list #> $av #> [1] \"ACI\" #>  #> $cid #> [1] 135398513 #>  #> $name #> [1] \"Aciclovir\" #>  #> $group #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" #>  #> $atc #> [1] \"J05AB01\" #>  #> $tradenames #>  [1] \"acicloftal\"        \"aciclovier\"        \"aciclovirum\"       #>  [4] \"activir\"           \"acyclofoam\"        \"acycloguanosine\"   #>  [7] \"acyclovir\"         \"acyclovir lauriad\" \"avaclyr\"           #> [10] \"cargosil\"          \"cyclovir\"          \"genvir\"            #> [13] \"gerpevir\"          \"hascovir\"          \"maynar\"            #> [16] \"novirus\"           \"poviral\"           \"sitavig\"           #> [19] \"sitavir\"           \"vipral\"            \"viropump\"          #> [22] \"virorax\"           \"zovirax\"           \"zyclir\"            #>  #> $loinc #> [1] \"\" #>  #> $ddd #> $ddd$oral #> $ddd$oral$amount #> [1] 4 #>  #> $ddd$oral$units #> [1] \"g\" #>  #>  #> $ddd$iv #> $ddd$iv$amount #> [1] 4 #>  #> $ddd$iv$units #> [1] \"g\" #>  #>  #>   # all av_* functions use as.av() internally, so you can go from 'any' to 'any': av_atc(\"ACI\") #> [1] \"J05AB01\" av_group(\"J05AB01\") #> [1] \"Nucleosides and nucleotides excl. reverse transcriptase inhibitors\" av_loinc(\"abacavir\") #> [1] \"29113-8\" \"30273-7\" \"30287-7\" \"30303-2\" \"78772-1\" \"78773-9\" \"79134-3\" #> [8] \"80118-3\" av_name(\"29113-8\") #> [1] \"Abacavir\" av_name(135398513) #> [1] \"Aciclovir\" av_name(\"J05AB01\") #> [1] \"Aciclovir\""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":null,"dir":"Reference","previous_headings":"","what":"Check Availability of Columns — availability","title":"Check Availability of Columns — availability","text":"Easy check data availability columns data set. makes easy get idea antimicrobial combinations can used calculation e.g. susceptibility() resistance().","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Check Availability of Columns — availability","text":"","code":"availability(tbl, width = NULL)"},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Check Availability of Columns — availability","text":"tbl data.frame list width number characters present visual availability - default filling width console","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Check Availability of Columns — availability","text":"data.frame column names tbl row names","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Check Availability of Columns — availability","text":"function returns data.frame columns \"resistant\" \"visual_resistance\". values columns calculated resistance().","code":""},{"path":"https://msberends.github.io/AMR/reference/availability.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Check Availability of Columns — availability","text":"","code":"availability(example_isolates) #>         count available     visual_availabilty resistant      visual_resistance #> date     2000    100.0% |####################|                                  #> patient  2000    100.0% |####################|                                  #> age      2000    100.0% |####################|                                  #> gender   2000    100.0% |####################|                                  #> ward     2000    100.0% |####################|                                  #> mo       2000    100.0% |####################|                                  #> PEN      1629     81.5% |################----|     73.7% |##############------| #> OXA       365     18.3% |###-----------------|     31.2% |######--------------| #> FLC       943     47.2% |#########-----------|     29.5% |#####---------------| #> AMX      1350     67.5% |#############-------|     59.6% |###########---------| #> AMC      1879     94.0% |##################--|     23.7% |####----------------| #> AMP      1350     67.5% |#############-------|     59.6% |###########---------| #> TZP      1001     50.0% |##########----------|     12.6% |##------------------| #> CZO       446     22.3% |####----------------|     44.6% |########------------| #> FEP       724     36.2% |#######-------------|     14.2% |##------------------| #> CXM      1789     89.5% |#################---|     26.3% |#####---------------| #> FOX       818     40.9% |########------------|     27.4% |#####---------------| #> CTX       943     47.2% |#########-----------|     15.5% |###-----------------| #> CAZ      1811     90.6% |##################--|     66.5% |#############-------| #> CRO       943     47.2% |#########-----------|     15.5% |###-----------------| #> GEN      1855     92.8% |##################--|     24.6% |####----------------| #> TOB      1351     67.6% |#############-------|     34.4% |######--------------| #> AMK       692     34.6% |######--------------|     63.7% |############--------| #> KAN       471     23.6% |####----------------|    100.0% |####################| #> TMP      1499     75.0% |###############-----|     38.1% |#######-------------| #> SXT      1759     88.0% |#################---|     20.5% |####----------------| #> NIT       743     37.2% |#######-------------|     17.1% |###-----------------| #> FOS       351     17.6% |###-----------------|     42.2% |########------------| #> LNZ      1023     51.2% |##########----------|     69.3% |#############-------| #> CIP      1409     70.5% |#############-------|     16.2% |###-----------------| #> MFX       211     10.6% |##------------------|     33.6% |######--------------| #> VAN      1861     93.1% |##################--|     38.3% |#######-------------| #> TEC       976     48.8% |#########-----------|     75.7% |###############-----| #> TCY      1200     60.0% |###########---------|     29.8% |#####---------------| #> TGC       798     39.9% |########------------|     12.7% |##------------------| #> DOX      1136     56.8% |###########---------|     27.7% |#####---------------| #> ERY      1894     94.7% |##################--|     57.2% |###########---------| #> CLI      1520     76.0% |###############-----|     61.2% |############--------| #> AZM      1894     94.7% |##################--|     57.2% |###########---------| #> IPM       889     44.5% |########------------|      6.2% |#-------------------| #> MEM       829     41.5% |########------------|      5.9% |#-------------------| #> MTR        34      1.7% |--------------------|     14.7% |##------------------| #> CHL       154      7.7% |#-------------------|     21.4% |####----------------| #> COL      1640     82.0% |################----|     81.2% |################----| #> MUP       270     13.5% |##------------------|      5.9% |#-------------------| #> RIF      1003     50.2% |##########----------|     69.6% |#############-------| # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     filter(mo == as.mo(\"Escherichia coli\")) %>%     select_if(is.sir) %>%     availability() } #>     count available       visual_availabilty resistant        visual_resistance #> PEN   467    100.0% |######################|    100.0% |######################| #> OXA     0      0.0% |----------------------|                                    #> FLC     0      0.0% |----------------------|                                    #> AMX   392     83.9% |##################----|     50.0% |###########-----------| #> AMC   467    100.0% |######################|     13.1% |##--------------------| #> AMP   392     83.9% |##################----|     50.0% |###########-----------| #> TZP   416     89.1% |###################---|      5.5% |#---------------------| #> CZO    82     17.6% |###-------------------|      2.4% |----------------------| #> FEP   317     67.9% |##############--------|      2.8% |----------------------| #> CXM   465     99.6% |######################|      5.4% |#---------------------| #> FOX   377     80.7% |#################-----|      6.9% |#---------------------| #> CTX   459     98.3% |#####################-|      2.4% |----------------------| #> CAZ   460     98.5% |#####################-|      2.4% |----------------------| #> CRO   459     98.3% |#####################-|      2.4% |----------------------| #> GEN   460     98.5% |#####################-|      2.0% |----------------------| #> TOB   462     98.9% |#####################-|      2.6% |----------------------| #> AMK   171     36.6% |########--------------|      0.0% |----------------------| #> KAN     0      0.0% |----------------------|                                    #> TMP   396     84.8% |##################----|     39.1% |########--------------| #> SXT   465     99.6% |######################|     31.6% |######----------------| #> NIT   458     98.1% |#####################-|      2.8% |----------------------| #> FOS    61     13.1% |##--------------------|      0.0% |----------------------| #> LNZ   467    100.0% |######################|    100.0% |######################| #> CIP   456     97.6% |#####################-|     12.5% |##--------------------| #> MFX    57     12.2% |##--------------------|    100.0% |######################| #> VAN   467    100.0% |######################|    100.0% |######################| #> TEC   467    100.0% |######################|    100.0% |######################| #> TCY     3      0.6% |----------------------|     66.7% |##############--------| #> TGC    68     14.6% |###-------------------|      0.0% |----------------------| #> DOX     0      0.0% |----------------------|                                    #> ERY   467    100.0% |######################|    100.0% |######################| #> CLI   467    100.0% |######################|    100.0% |######################| #> AZM   467    100.0% |######################|    100.0% |######################| #> IPM   422     90.4% |###################---|      0.0% |----------------------| #> MEM   418     89.5% |###################---|      0.0% |----------------------| #> MTR     2      0.4% |----------------------|      0.0% |----------------------| #> CHL     0      0.0% |----------------------|                                    #> COL   240     51.4% |###########-----------|      0.0% |----------------------| #> MUP     0      0.0% |----------------------|                                    #> RIF   467    100.0% |######################|    100.0% |######################| # }"},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Bug-Drug Combinations — bug_drug_combinations","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"Determine antimicrobial resistance (AMR) bug-drug combinations data set least 30 (default) isolates available per species. Use format() result prettify publishable/printable format, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"","code":"bug_drug_combinations(x, col_mo = NULL, FUN = mo_shortname,   include_n_rows = FALSE, ...)  # S3 method for class 'bug_drug_combinations' format(x, translate_ab = \"name (ab, atc)\",   language = get_AMR_locale(), minimum = 30, combine_SI = TRUE,   add_ab_group = TRUE, remove_intrinsic_resistant = FALSE,   decimal.mark = getOption(\"OutDec\"), big.mark = ifelse(decimal.mark ==   \",\", \".\", \",\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"x data set antibiotic columns, amox, AMX AMC col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). FUN function call mo column transform microorganism codes - default mo_shortname() include_n_rows logical indicate total number rows must included output ... arguments passed FUN translate_ab character length 1 containing column names antibiotics data set language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether values S, SDD, summed, resistance based R - default TRUE add_ab_group logical indicate group antimicrobials must included first column remove_intrinsic_resistant logical indicate rows columns 100% resistance tested antimicrobials must removed table decimal.mark character used indicate numeric     decimal point. big.mark character; empty used mark every     big.interval decimals (hence big)     decimal point.","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"function bug_drug_combinations() returns data.frame columns \"mo\", \"ab\", \"S\", \"SDD\", \"\", \"R\", \"total\".","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"function format() calculates resistance per bug-drug combination returns table ready reporting/publishing. Use combine_SI = TRUE (default) test R vs. S+combine_SI = FALSE test R+vs. S. table can also directly used R Markdown / Quarto without need e.g. knitr::kable().","code":""},{"path":"https://msberends.github.io/AMR/reference/bug_drug_combinations.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Bug-Drug Combinations — bug_drug_combinations","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # \\donttest{ x <- bug_drug_combinations(example_isolates) head(x) #> # A tibble: 6 × 8 #>   mo                ab        S   SDD     I     R    NI total #>   <chr>             <chr> <int> <int> <int> <int> <int> <int> #> 1 (unknown species) AMC      15     0     0     0     0    15 #> 2 (unknown species) AMK       0     0     0     0     0     0 #> 3 (unknown species) AMP      15     0     0     1     0    16 #> 4 (unknown species) AMX      15     0     0     1     0    16 #> 5 (unknown species) AZM       3     0     0     3     0     6 #> 6 (unknown species) CAZ       0     0     0     0     0     0 #> Use 'format()' on this result to get a publishable/printable format. format(x, translate_ab = \"name (atc)\") #> # A tibble: 39 × 12 #>    Group     Drug  CoNS  `E. coli` `E. faecalis` `K. pneumoniae` `P. aeruginosa` #>    <chr>     <chr> <chr> <chr>     <chr>         <chr>           <chr>           #>  1 \"Aminogl… Amik… \"100… \"  0.0% … \"100.0% (39/… \"\"              \"\"              #>  2 \"\"        Gent… \" 13… \"  2.0% … \"100.0% (39/… \" 10.3% (6/58)\" \"  0.0% (0/30)\" #>  3 \"\"        Kana… \"100… \"\"        \"100.0% (39/… \"\"              \"100.0% (30/30… #>  4 \"\"        Tobr… \" 78… \"  2.6% … \"100.0% (39/… \" 10.3% (6/58)\" \"  0.0% (0/30)\" #>  5 \"Antimyc… Rifa… \"\"    \"100.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  6 \"Beta-la… Amox… \" 93… \" 50.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  7 \"\"        Amox… \" 42… \" 13.1% … \"\"            \" 10.3% (6/58)\" \"100.0% (30/30… #>  8 \"\"        Ampi… \" 93… \" 50.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #>  9 \"\"        Benz… \" 77… \"100.0% … \"\"            \"100.0% (58/58… \"100.0% (30/30… #> 10 \"\"        Fluc… \" 42… \"\"        \"\"            \"\"              \"\"              #> # ℹ 29 more rows #> # ℹ 5 more variables: `P. mirabilis` <chr>, `S. aureus` <chr>, #> #   `S. epidermidis` <chr>, `S. hominis` <chr>, `S. pneumoniae` <chr>  # Use FUN to change to transformation of microorganism codes bug_drug_combinations(example_isolates,   FUN = mo_gramstain ) #> # A tibble: 80 × 8 #>    mo            ab        S   SDD     I     R    NI total #>  * <chr>         <chr> <int> <int> <int> <int> <int> <int> #>  1 Gram-negative AMC     463     0    89   174     0   726 #>  2 Gram-negative AMK     251     0     0     5     0   256 #>  3 Gram-negative AMP     226     0     0   405     0   631 #>  4 Gram-negative AMX     226     0     0   405     0   631 #>  5 Gram-negative AZM       1     0     2   696     0   699 #>  6 Gram-negative CAZ     607     0     0    27     0   634 #>  7 Gram-negative CHL       1     0     0    30     0    31 #>  8 Gram-negative CIP     610     0    11    63     0   684 #>  9 Gram-negative CLI      18     0     1   709     0   728 #> 10 Gram-negative COL     309     0     0    78     0   387 #> # ℹ 70 more rows #> Use 'format()' on this result to get a publishable/printable format.  bug_drug_combinations(example_isolates,   FUN = function(x) {     ifelse(x == as.mo(\"Escherichia coli\"),       \"E. coli\",       \"Others\"     )   } ) #> # A tibble: 80 × 8 #>    mo      ab        S   SDD     I     R    NI total #>  * <chr>   <chr> <int> <int> <int> <int> <int> <int> #>  1 E. coli AMC     332     0    74    61     0   467 #>  2 E. coli AMK     171     0     0     0     0   171 #>  3 E. coli AMP     196     0     0   196     0   392 #>  4 E. coli AMX     196     0     0   196     0   392 #>  5 E. coli AZM       0     0     0   467     0   467 #>  6 E. coli CAZ     449     0     0    11     0   460 #>  7 E. coli CHL       0     0     0     0     0     0 #>  8 E. coli CIP     398     0     1    57     0   456 #>  9 E. coli CLI       0     0     0   467     0   467 #> 10 E. coli COL     240     0     0     0     0   240 #> # ℹ 70 more rows #> Use 'format()' on this result to get a publishable/printable format. # }"},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Data set containing clinical breakpoints interpret MIC disk diffusion SIR values, according international guidelines. dataset contain breakpoints humans, 7 different animal groups, ECOFFs. breakpoints currently implemented: clinical microbiology: EUCAST 2011-2024 CLSI 2011-2024; veterinary microbiology: EUCAST 2021-2024 CLSI 2019-2024; ECOFFs (Epidemiological Cut-Values): EUCAST 2020-2024 CLSI 2022-2024. Use .sir() transform MICs disks measurements SIR values.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"","code":"clinical_breakpoints"},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"tibble 34 063 observations 14 variables: guideline Name guideline type Breakpoint type, either \"ECOFF\", \"animal\", \"human\" host Host infectious agent. mostly useful veterinary breakpoints either \"ECOFF\", \"aquatic\", \"cats\", \"cattle\", \"dogs\", \"horse\", \"human\", \"poultry\", \"swine\" method Testing method, either \"DISK\" \"MIC\" site Body site breakpoint must applied, e.g. \"Oral\" \"Respiratory\" mo Microbial ID, see .mo() rank_index Taxonomic rank index mo 1 (subspecies/infraspecies) 5 (unknown microorganism) ab Antibiotic code used package, EARS-Net WHONET, see .ab() ref_tbl Info guideline rule can found disk_dose Dose used disk diffusion method breakpoint_S Lowest MIC value highest number millimetres leads \"S\" breakpoint_R Highest MIC value lowest number millimetres leads \"R\" uti logical value (TRUE/FALSE) indicate whether rule applies urinary tract infection (UTI) is_SDD logical value (TRUE/FALSE) indicate whether intermediate range \"S\" \"R\" interpreted \"SDD\", instead \"\". currently applies 24 breakpoints.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"different-types-of-breakpoints","dir":"Reference","previous_headings":"","what":"Different types of breakpoints","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Supported types breakpoints ECOFF, animal, human. ECOFF (Epidemiological cut-) values used antimicrobial susceptibility testing differentiate wild-type non-wild-type strains bacteria fungi. default \"human\", can also set package option AMR_breakpoint_type. Use .sir(..., breakpoint_type = ...) interpret raw data using specific breakpoint type, e.g. .sir(..., breakpoint_type = \"ECOFF\") use ECOFFs.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"imported-from-whonet","dir":"Reference","previous_headings":"","what":"Imported from WHONET","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Clinical breakpoints package validated imported WHONET, free desktop Windows application developed supported Collaborating Centre Surveillance Antimicrobial Resistance. can read website. developers WHONET AMR package contact sharing work. highly appreciate great development WHONET software.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"response-from-clsi-and-eucast","dir":"Reference","previous_headings":"","what":"Response from CLSI and EUCAST","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"CEO CLSI chairman EUCAST endorsed work public use AMR package (consequently use breakpoints) June 2023, future development distributing clinical breakpoints discussed meeting CLSI, EUCAST, , developers WHONET software, developers AMR package.","code":""},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"download","dir":"Reference","previous_headings":"","what":"Download","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository. allow machine reading EUCAST CLSI guidelines, almost impossible MS Excel PDF files distributed EUCAST CLSI, though initiatives started overcome burdens. NOTE: AMR package (WHONET software well) contains rather complex internal methods apply guidelines. example, breakpoints must applied certain species groups (case package available microorganisms.groups data set). important considered using breakpoints use.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/clinical_breakpoints.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Clinical Breakpoints for SIR Interpretation — clinical_breakpoints","text":"","code":"clinical_breakpoints #> # A tibble: 34,063 × 14 #>    guideline   type  host  method site    mo            rank_index ab   ref_tbl  #>    <chr>       <chr> <chr> <chr>  <chr>   <mo>               <dbl> <ab> <chr>    #>  1 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 MEM  A. xylo… #>  2 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 MEM  A. xylo… #>  3 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 SXT  A. xylo… #>  4 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 SXT  A. xylo… #>  5 EUCAST 2024 human human DISK   NA      B_ACHRMB_XYLS          2 TZP  A. xylo… #>  6 EUCAST 2024 human human MIC    NA      B_ACHRMB_XYLS          2 TZP  A. xylo… #>  7 EUCAST 2024 human human DISK   NA      B_ACNTB                3 AMK  Acineto… #>  8 EUCAST 2024 human human DISK   Uncomp… B_ACNTB                3 AMK  Acineto… #>  9 EUCAST 2024 human human MIC    NA      B_ACNTB                3 AMK  Acineto… #> 10 EUCAST 2024 human human MIC    Uncomp… B_ACNTB                3 AMK  Acineto… #> # ℹ 34,053 more rows #> # ℹ 5 more variables: disk_dose <chr>, breakpoint_S <dbl>, breakpoint_R <dbl>, #> #   uti <lgl>, is_SDD <lgl>"},{"path":"https://msberends.github.io/AMR/reference/count.html","id":null,"dir":"Reference","previous_headings":"","what":"Count Available Isolates — count","title":"Count Available Isolates — count","text":"functions can used count resistant/susceptible microbial isolates. functions support quasiquotation pipes, can used summarise() dplyr package also support grouped variables, see Examples. count_resistant() used count resistant isolates, count_susceptible() used count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Count Available Isolates — count","text":"","code":"count_resistant(..., only_all_tested = FALSE)  count_susceptible(..., only_all_tested = FALSE)  count_S(..., only_all_tested = FALSE)  count_SI(..., only_all_tested = FALSE)  count_I(..., only_all_tested = FALSE)  count_IR(..., only_all_tested = FALSE)  count_R(..., only_all_tested = FALSE)  count_all(..., only_all_tested = FALSE)  n_sir(..., only_all_tested = FALSE)  count_df(data, translate_ab = \"name\", language = get_AMR_locale(),   combine_SI = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Count Available Isolates — count","text":"... one vectors (columns) antibiotic interpretations. transformed internally .sir() needed. only_all_tested (combination therapies, .e. using one variable ...): logical indicate isolates must tested antibiotics, see section Combination Therapy data data.frame containing columns class sir (see .sir()) translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Count Available Isolates — count","text":"integer","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Count Available Isolates — count","text":"functions meant count isolates. Use resistance()/susceptibility() functions calculate microbial resistance/susceptibility. function count_resistant() equal function count_R(). function count_susceptible() equal function count_SI(). function n_sir() alias count_all(). can used count available isolates, .e. input antibiotics available result (S, R). use equal n_distinct(). function equal count_susceptible(...) + count_resistant(...). function count_df() takes variable data sir class (created .sir()) counts number S's, 's R's. also supports grouped variables. function sir_df() works exactly like count_df(), adds percentage S, R.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Count Available Isolates — count","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"combination-therapy","dir":"Reference","previous_headings":"","what":"Combination Therapy","title":"Count Available Isolates — count","text":"using one variable ... (= combination therapy), use only_all_tested count isolates tested antibiotics/variables test . See example two antibiotics, Drug Drug B, susceptibility() works calculate %SI:   Please note , combination therapies, only_all_tested = TRUE applies :   , combination therapies, only_all_tested = FALSE applies :   Using only_all_tested impact using one antibiotic input.","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - -------------------------------------------------------------------- count_S()    +   count_I()    +   count_R()    = count_all()   proportion_S() + proportion_I() + proportion_R() = 1 count_S()    +   count_I()    +   count_R()    >= count_all()   proportion_S() + proportion_I() + proportion_R() >= 1"},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/count.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Count Available Isolates — count","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info.  # base R ------------------------------------------------------------ count_resistant(example_isolates$AMX) # counts \"R\" #> [1] 804 count_susceptible(example_isolates$AMX) # counts \"S\" and \"I\" #> [1] 546 count_all(example_isolates$AMX) # counts \"S\", \"I\" and \"R\" #> [1] 1350  # be more specific count_S(example_isolates$AMX) #> Using count_S() is discouraged; use count_susceptible() instead to also #> consider \"I\" and \"SDD\" being susceptible. This note will be shown once for #> this session. #> [1] 543 count_SI(example_isolates$AMX) #> Note that count_SI() will also count dose-dependent susceptibility, #> 'SDD'. This note will be shown once for this session. #> [1] 546 count_I(example_isolates$AMX) #> Note that count_I() will also count dose-dependent susceptibility, 'SDD'. #> This note will be shown once for this session. #> [1] 3 count_IR(example_isolates$AMX) #> Using count_IR() is discouraged; use count_resistant() instead to not #> consider \"I\" and \"SDD\" being resistant. This note will be shown once for #> this session. #> [1] 807 count_R(example_isolates$AMX) #> [1] 804  # Count all available isolates count_all(example_isolates$AMX) #> [1] 1350 n_sir(example_isolates$AMX) #> [1] 1350  # n_sir() is an alias of count_all(). # Since it counts all available isolates, you can # calculate back to count e.g. susceptible isolates. # These results are the same: count_susceptible(example_isolates$AMX) #> [1] 546 susceptibility(example_isolates$AMX) * n_sir(example_isolates$AMX) #> [1] 546  # dplyr ------------------------------------------------------------- # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       R = count_R(CIP),       I = count_I(CIP),       S = count_S(CIP),       n1 = count_all(CIP), # the actual total; sum of all three       n2 = n_sir(CIP), # same - analogous to n_distinct       total = n()     ) # NOT the number of tested isolates!    # Number of available isolates for a whole antibiotic class   # (i.e., in this data set columns GEN, TOB, AMK, KAN)   example_isolates %>%     group_by(ward) %>%     summarise(across(aminoglycosides(), n_sir))    # Count co-resistance between amoxicillin/clav acid and gentamicin,   # so we can see that combination therapy does a lot more than mono therapy.   # Please mind that `susceptibility()` calculates percentages right away instead.   example_isolates %>% count_susceptible(AMC) # 1433   example_isolates %>% count_all(AMC) # 1879    example_isolates %>% count_susceptible(GEN) # 1399   example_isolates %>% count_all(GEN) # 1855    example_isolates %>% count_susceptible(AMC, GEN) # 1764   example_isolates %>% count_all(AMC, GEN) # 1936    # Get number of S+I vs. R immediately of selected columns   example_isolates %>%     select(AMX, CIP) %>%     count_df(translate = FALSE)    # It also supports grouping variables   example_isolates %>%     select(ward, AMX, CIP) %>%     group_by(ward) %>%     count_df(translate = FALSE) } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> # A tibble: 12 × 4 #>    ward       antibiotic interpretation value #>  * <chr>      <chr>      <ord>          <int> #>  1 Clinical   AMX        SI               357 #>  2 Clinical   AMX        R                487 #>  3 Clinical   CIP        SI               741 #>  4 Clinical   CIP        R                128 #>  5 ICU        AMX        SI               158 #>  6 ICU        AMX        R                270 #>  7 ICU        CIP        SI               362 #>  8 ICU        CIP        R                 85 #>  9 Outpatient AMX        SI                31 #> 10 Outpatient AMX        R                 47 #> 11 Outpatient CIP        SI                78 #> 12 Outpatient CIP        R                 15 # }"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":null,"dir":"Reference","previous_headings":"","what":"Define Custom EUCAST Rules — custom_eucast_rules","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"Define custom EUCAST rules organisation specific analysis use output function eucast_rules().","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"","code":"custom_eucast_rules(...)"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"... rules formula notation, see instructions, Examples","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"list containing custom rules","code":""},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"organisations adoption EUCAST rules. function can used define custom EUCAST rules used eucast_rules() function.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"basics","dir":"Reference","previous_headings":"","what":"Basics","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"familiar case_when() function dplyr package, recognise input method set rules. Rules must set using R considers 'formula notation'. rule written tilde (~) consequence rule written tilde:   two custom EUCAST rules: TZP (piperacillin/tazobactam) \"S\", aminopenicillins (ampicillin amoxicillin) must made \"S\", TZP \"R\", aminopenicillins must made \"R\". rules can also printed console, immediately clear work:   rules (part tilde, example TZP == \"S\" TZP == \"R\") must evaluable data set: able run filter data set without errors. means example column TZP must exist. create sample data set test rules set:","code":"x <- custom_eucast_rules(TZP == \"S\" ~ aminopenicillins == \"S\",                          TZP == \"R\" ~ aminopenicillins == \"R\") x #> A set of custom EUCAST rules: #> #>   1. If TZP is \"S\" then set to  S : #>      amoxicillin (AMX), ampicillin (AMP) #> #>   2. If TZP is \"R\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) df <- data.frame(mo = c(\"Escherichia coli\", \"Klebsiella pneumoniae\"),                  TZP = as.sir(\"R\"),                  ampi = as.sir(\"S\"),                  cipro = as.sir(\"S\")) df #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    S     S #> 2 Klebsiella pneumoniae   R    S     S  eucast_rules(df, rules = \"custom\", custom_rules = x, info = FALSE) #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    R     S #> 2 Klebsiella pneumoniae   R    R     S"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"using-taxonomic-properties-in-rules","dir":"Reference","previous_headings":"","what":"Using taxonomic properties in rules","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"one exception columns used rules: column names microorganisms data set can also used, exist data set. column names : \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". Thus, next example work well, despite fact df data set contain column genus:","code":"y <- custom_eucast_rules(TZP == \"S\" & genus == \"Klebsiella\" ~ aminopenicillins == \"S\",                          TZP == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\")  eucast_rules(df, rules = \"custom\", custom_rules = y, info = FALSE) #>                      mo TZP ampi cipro #> 1      Escherichia coli   R    S     S #> 2 Klebsiella pneumoniae   R    R     S"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"usage-of-multiple-antibiotics-and-antibiotic-group-names","dir":"Reference","previous_headings":"","what":"Usage of multiple antibiotics and antibiotic group names","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"can define antibiotic groups instead single antibiotics rule consequence, part tilde (~). examples , antibiotic group aminopenicillins includes ampicillin amoxicillin. Rules can also applied multiple antibiotics antibiotic groups simultaneously. Use c() function combine multiple antibiotics. instance, following example sets aminopenicillins ureidopenicillins \"R\" column TZP (piperacillin/tazobactam) \"R\":   34 antibiotic groups allowed rules (case-insensitive) can used combination: aminoglycosides(amikacin, amikacin/fosfomycin, apramycin, arbekacin, astromicin, bekanamycin, dibekacin, framycetin, gentamicin, gentamicin-high, habekacin, hygromycin, isepamicin, kanamycin, kanamycin-high, kanamycin/cephalexin, micronomicin, neomycin, netilmicin, pentisomicin, plazomicin, propikacin, ribostamycin, sisomicin, streptoduocin, streptomycin, streptomycin-high, tobramycin, tobramycin-high) aminopenicillins(amoxicillin ampicillin) antifungals(amorolfine, amphotericin B, amphotericin B-high, anidulafungin, butoconazole, caspofungin, ciclopirox, clotrimazole, econazole, fluconazole, flucytosine, fosfluconazole, griseofulvin, hachimycin, ibrexafungerp, isavuconazole, isoconazole, itraconazole, ketoconazole, manogepix, micafungin, miconazole, nystatin, oteseconazole, pimaricin, posaconazole, rezafungin, ribociclib, sulconazole, terbinafine, terconazole, voriconazole) antimycobacterials(4-aminosalicylic acid, calcium aminosalicylate, capreomycin, clofazimine, delamanid, enviomycin, ethambutol, ethambutol/isoniazid, ethionamide, isoniazid, isoniazid/sulfamethoxazole/trimethoprim/pyridoxine, morinamide, p-aminosalicylic acid, pretomanid, protionamide, pyrazinamide, rifabutin, rifampicin, rifampicin/ethambutol/isoniazid, rifampicin/isoniazid, rifampicin/pyrazinamide/ethambutol/isoniazid, rifampicin/pyrazinamide/isoniazid, rifamycin, rifapentine, simvastatin/fenofibrate, sodium aminosalicylate, streptomycin/isoniazid, terizidone, thioacetazone, thioacetazone/isoniazid, tiocarlide, viomycin) betalactams(amoxicillin, amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin, ampicillin/sulbactam, apalcillin, aspoxicillin, azidocillin, azlocillin, aztreonam, aztreonam/avibactam, aztreonam/nacubactam, bacampicillin, benzathine benzylpenicillin, benzathine phenoxymethylpenicillin, benzylpenicillin, biapenem, carbenicillin, carindacillin, carumonam, cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, ciclacillin, clometocillin, cloxacillin, dicloxacillin, doripenem, epicillin, ertapenem, flucloxacillin, hetacillin, imipenem, imipenem/EDTA, imipenem/relebactam, latamoxef, lenampicillin, loracarbef, mecillinam, meropenem, meropenem/nacubactam, meropenem/vaborbactam, metampicillin, meticillin, mezlocillin, mezlocillin/sulbactam, nafcillin, oxacillin, panipenem, penamecillin, penicillin/novobiocin, penicillin/sulbactam, pheneticillin, phenoxymethylpenicillin, piperacillin, piperacillin/sulbactam, piperacillin/tazobactam, piridicillin, pivampicillin, pivmecillinam, procaine benzylpenicillin, propicillin, razupenem, ritipenem, ritipenem acoxil, sarmoxicillin, sulbenicillin, sultamicillin, talampicillin, tebipenem, temocillin, ticarcillin, ticarcillin/clavulanic acid, tigemonam) betalactams_with_inhibitor(amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin/sulbactam, aztreonam/avibactam, aztreonam/nacubactam, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefoperazone/sulbactam, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefpodoxime/clavulanic acid, ceftaroline/avibactam, ceftazidime/avibactam, ceftazidime/clavulanic acid, ceftolozane/tazobactam, ceftriaxone/beta-lactamase inhibitor, imipenem/relebactam, meropenem/nacubactam, meropenem/vaborbactam, mezlocillin/sulbactam, penicillin/novobiocin, penicillin/sulbactam, piperacillin/sulbactam, piperacillin/tazobactam, ticarcillin/clavulanic acid) carbapenems(biapenem, doripenem, ertapenem, imipenem, imipenem/EDTA, imipenem/relebactam, meropenem, meropenem/nacubactam, meropenem/vaborbactam, panipenem, razupenem, ritipenem, ritipenem acoxil, tebipenem) cephalosporins(cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, latamoxef, loracarbef) cephalosporins_1st(cefacetrile, cefadroxil, cefalexin, cefaloridine, cefalotin, cefapirin, cefatrizine, cefazedone, cefazolin, cefroxadine, ceftezole, cephradine) cephalosporins_2nd(cefaclor, cefamandole, cefmetazole, cefonicid, ceforanide, cefotetan, cefotiam, cefoxitin, cefoxitin screening, cefprozil, cefuroxime, cefuroxime axetil, loracarbef) cephalosporins_3rd(cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefetamet, cefetamet pivoxil, cefixime, cefmenoxime, cefodizime, cefoperazone, cefoperazone/sulbactam, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotiam hexetil, cefovecin, cefpimizole, cefpiramide, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefsulodin, ceftazidime, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, latamoxef) cephalosporins_4th(cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetecol, cefoselis, cefozopran, cefpirome, cefquinome) cephalosporins_5th(ceftaroline, ceftaroline/avibactam, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam) cephalosporins_except_caz(cefacetrile, cefaclor, cefadroxil, cefalexin, cefaloridine, cefalotin, cefamandole, cefapirin, cefatrizine, cefazedone, cefazolin, cefcapene, cefcapene pivoxil, cefdinir, cefditoren, cefditoren pivoxil, cefepime, cefepime/amikacin, cefepime/clavulanic acid, cefepime/nacubactam, cefepime/tazobactam, cefepime/zidebactam, cefetamet, cefetamet pivoxil, cefetecol, cefetrizole, cefiderocol, cefixime, cefmenoxime, cefmetazole, cefodizime, cefonicid, cefoperazone, cefoperazone/sulbactam, ceforanide, cefoselis, cefotaxime, cefotaxime/clavulanic acid, cefotaxime/sulbactam, cefotetan, cefotiam, cefotiam hexetil, cefovecin, cefoxitin, cefoxitin screening, cefozopran, cefpimizole, cefpiramide, cefpirome, cefpodoxime, cefpodoxime proxetil, cefpodoxime/clavulanic acid, cefprozil, cefquinome, cefroxadine, cefsulodin, cefsumide, ceftaroline, ceftaroline/avibactam, ceftazidime/avibactam, ceftazidime/clavulanic acid, cefteram, cefteram pivoxil, ceftezole, ceftibuten, ceftiofur, ceftizoxime, ceftizoxime alapivoxil, ceftobiprole, ceftobiprole medocaril, ceftolozane/tazobactam, ceftriaxone, ceftriaxone/beta-lactamase inhibitor, cefuroxime, cefuroxime axetil, cephradine, latamoxef, loracarbef) fluoroquinolones(besifloxacin, ciprofloxacin, clinafloxacin, danofloxacin, delafloxacin, difloxacin, enoxacin, enrofloxacin, finafloxacin, fleroxacin, garenoxacin, gatifloxacin, gemifloxacin, grepafloxacin, lascufloxacin, levofloxacin, levonadifloxacin, lomefloxacin, marbofloxacin, metioxate, miloxacin, moxifloxacin, nadifloxacin, nifuroquine, norfloxacin, ofloxacin, orbifloxacin, pazufloxacin, pefloxacin, pradofloxacin, premafloxacin, prulifloxacin, rufloxacin, sarafloxacin, sitafloxacin, sparfloxacin, temafloxacin, tilbroquinol, tioxacin, tosufloxacin, trovafloxacin) glycopeptides(avoparcin, dalbavancin, norvancomycin, oritavancin, ramoplanin, teicoplanin, teicoplanin-macromethod, telavancin, vancomycin, vancomycin-macromethod) glycopeptides_except_lipo(avoparcin, norvancomycin, ramoplanin, teicoplanin, teicoplanin-macromethod, vancomycin, vancomycin-macromethod) isoxazolylpenicillins(cloxacillin, dicloxacillin, flucloxacillin, meticillin, oxacillin) lincosamides(clindamycin, lincomycin, pirlimycin) lipoglycopeptides(dalbavancin, oritavancin, telavancin) macrolides(acetylmidecamycin, acetylspiramycin, azithromycin, clarithromycin, dirithromycin, erythromycin, flurithromycin, gamithromycin, josamycin, kitasamycin, meleumycin, midecamycin, miocamycin, nafithromycin, oleandomycin, rokitamycin, roxithromycin, solithromycin, spiramycin, telithromycin, tildipirosin, tilmicosin, troleandomycin, tulathromycin, tylosin, tylvalosin) monobactams(aztreonam, aztreonam/avibactam, aztreonam/nacubactam, carumonam, tigemonam) nitrofurans(furazidin, furazolidone, nifurtoinol, nitrofurantoin, nitrofurazone) oxazolidinones(cadazolid, cycloserine, linezolid, tedizolid, thiacetazone) penicillins(amoxicillin, amoxicillin/clavulanic acid, amoxicillin/sulbactam, ampicillin, ampicillin/sulbactam, apalcillin, aspoxicillin, azidocillin, azlocillin, bacampicillin, benzathine benzylpenicillin, benzathine phenoxymethylpenicillin, benzylpenicillin, carbenicillin, carindacillin, ciclacillin, clometocillin, cloxacillin, dicloxacillin, epicillin, flucloxacillin, hetacillin, lenampicillin, mecillinam, metampicillin, meticillin, mezlocillin, mezlocillin/sulbactam, nafcillin, oxacillin, penamecillin, penicillin/novobiocin, penicillin/sulbactam, pheneticillin, phenoxymethylpenicillin, piperacillin, piperacillin/sulbactam, piperacillin/tazobactam, piridicillin, pivampicillin, pivmecillinam, procaine benzylpenicillin, propicillin, sarmoxicillin, sulbenicillin, sultamicillin, talampicillin, temocillin, ticarcillin, ticarcillin/clavulanic acid) phenicols(chloramphenicol, florfenicol, thiamphenicol) polymyxins(colistin, polymyxin B, polymyxin B/polysorbate 80) quinolones(besifloxacin, cinoxacin, ciprofloxacin, ciprofloxacin/metronidazole, ciprofloxacin/ornidazole, ciprofloxacin/tinidazole, clinafloxacin, danofloxacin, delafloxacin, difloxacin, enoxacin, enrofloxacin, finafloxacin, fleroxacin, flumequine, garenoxacin, gatifloxacin, gemifloxacin, grepafloxacin, lascufloxacin, levofloxacin, levonadifloxacin, lomefloxacin, marbofloxacin, metioxate, miloxacin, moxifloxacin, nadifloxacin, nalidixic acid, nemonoxacin, nifuroquine, nitroxoline, norfloxacin, ofloxacin, orbifloxacin, oxolinic acid, pazufloxacin, pefloxacin, pipemidic acid, piromidic acid, pradofloxacin, premafloxacin, prulifloxacin, rosoxacin, rufloxacin, sarafloxacin, sitafloxacin, sparfloxacin, temafloxacin, tilbroquinol, tioxacin, tosufloxacin, trovafloxacin) rifamycins(rifabutin, rifampicin, rifampicin/ethambutol/isoniazid, rifampicin/isoniazid, rifampicin/pyrazinamide/ethambutol/isoniazid, rifampicin/pyrazinamide/isoniazid, rifamycin, rifapentine) streptogramins(pristinamycin quinupristin/dalfopristin) tetracyclines(cetocycline, chlortetracycline, clomocycline, demeclocycline, doxycycline, eravacycline, lymecycline, metacycline, minocycline, omadacycline, oxytetracycline, penimepicycline, rolitetracycline, sarecycline, tetracycline, tigecycline) tetracyclines_except_tgc(cetocycline, chlortetracycline, clomocycline, demeclocycline, doxycycline, eravacycline, lymecycline, metacycline, minocycline, omadacycline, oxytetracycline, penimepicycline, rolitetracycline, sarecycline, tetracycline) trimethoprims(brodimoprim, sulfadiazine, sulfadiazine/tetroxoprim, sulfadiazine/trimethoprim, sulfadimethoxine, sulfadimidine, sulfadimidine/trimethoprim, sulfafurazole, sulfaisodimidine, sulfalene, sulfamazone, sulfamerazine, sulfamerazine/trimethoprim, sulfamethizole, sulfamethoxazole, sulfamethoxypyridazine, sulfametomidine, sulfametoxydiazine, sulfametrole/trimethoprim, sulfamoxole, sulfamoxole/trimethoprim, sulfanilamide, sulfaperin, sulfaphenazole, sulfapyridine, sulfathiazole, sulfathiourea, trimethoprim, trimethoprim/sulfamethoxazole) ureidopenicillins(azlocillin, mezlocillin, piperacillin, piperacillin/tazobactam)","code":"x <- custom_eucast_rules(TZP == \"R\" ~ c(aminopenicillins, ureidopenicillins) == \"R\") x #> A set of custom EUCAST rules: #> #>   1. If TZP is \"R\" then set to \"R\": #>      amoxicillin (AMX), ampicillin (AMP), azlocillin (AZL), mezlocillin (MEZ), piperacillin (PIP), piperacillin/tazobactam (TZP)"},{"path":"https://msberends.github.io/AMR/reference/custom_eucast_rules.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Define Custom EUCAST Rules — custom_eucast_rules","text":"","code":"x <- custom_eucast_rules(   AMC == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\",   AMC == \"I\" & genus == \"Klebsiella\" ~ aminopenicillins == \"I\" ) x #> A set of custom EUCAST rules: #>  #>   1. If AMC is  R  and genus is \"Klebsiella\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   2. If AMC is  I  and genus is \"Klebsiella\" then set to  I : #>      amoxicillin (AMX), ampicillin (AMP)  # run the custom rule set (verbose = TRUE will return a logbook instead of the data set): eucast_rules(example_isolates,   rules = \"custom\",   custom_rules = x,   info = FALSE,   verbose = TRUE ) #> # A tibble: 8 × 9 #>     row col   mo_fullname     old   new   rule  rule_group rule_name rule_source #>   <int> <chr> <chr>           <ord> <chr> <chr> <chr>      <chr>     <chr>       #> 1    33 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 2    33 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 3    34 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 4    34 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 5   531 AMP   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 6   531 AMX   Klebsiella pne… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 7  1485 AMP   Klebsiella oxy… R     I     \"rep… Custom EU… Custom E… Object 'x'… #> 8  1485 AMX   Klebsiella oxy… R     I     \"rep… Custom EU… Custom E… Object 'x'…  # combine rule sets x2 <- c(   x,   custom_eucast_rules(TZP == \"R\" ~ carbapenems == \"R\") ) x2 #> A set of custom EUCAST rules: #>  #>   1. If AMC is  R  and genus is \"Klebsiella\" then set to  R : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   2. If AMC is  I  and genus is \"Klebsiella\" then set to  I : #>      amoxicillin (AMX), ampicillin (AMP) #>  #>   3. If TZP is  R  then set to  R : #>      biapenem (BIA), doripenem (DOR), ertapenem (ETP), imipenem (IPM), #>      imipenem/EDTA (IPE), imipenem/relebactam (IMR), meropenem (MEM), #>      meropenem/nacubactam (MNC), meropenem/vaborbactam (MEV), panipenem (PAN), #>      razupenem (RZM), ritipenem (RIT), ritipenem acoxil (RIA), tebipenem (TBP)"},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"EUCAST breakpoints used package based dosages data set. can retrieved eucast_dosage().","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"","code":"dosage"},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"tibble 503 observations 9 variables: ab Antibiotic ID used package (AMC), using official EARS-Net (European Antimicrobial Resistance Surveillance Network) codes available name Official name antimicrobial drug used WHONET/EARS-Net type Type dosage, either \"high_dosage\", \"standard_dosage\", \"uncomplicated_uti\" dose Dose, \"2 g\" \"25 mg/kg\" dose_times Number times dose must administered administration Route administration, either \"im\", \"iv\", \"oral\" notes Additional dosage notes original_txt Original text PDF file EUCAST eucast_version Version number EUCAST Clinical Breakpoints guideline dosages apply, either 13, 12, 11","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/dosage.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Treatment Dosages as Defined by EUCAST — dosage","text":"","code":"dosage #> # A tibble: 503 × 9 #>    ab   name            type  dose  dose_times administration notes original_txt #>    <ab> <chr>           <chr> <chr>      <int> <chr>          <chr> <chr>        #>  1 AMK  Amikacin        stan… 25-3…          1 iv             \"\"    25-30 mg/kg… #>  2 AMX  Amoxicillin     high… 2 g            6 iv             \"\"    2 g x 6 iv   #>  3 AMX  Amoxicillin     stan… 1 g            3 iv             \"\"    1 g x 3-4 iv #>  4 AMX  Amoxicillin     high… 0.75…          3 oral           \"\"    0.75-1 g x … #>  5 AMX  Amoxicillin     stan… 0.5 g          3 oral           \"\"    0.5 g x 3 o… #>  6 AMX  Amoxicillin     unco… 0.5 g          3 oral           \"\"    0.5 g x 3 o… #>  7 AMC  Amoxicillin/cl… high… 2 g …          3 iv             \"\"    (2 g amoxic… #>  8 AMC  Amoxicillin/cl… stan… 1 g …          3 iv             \"\"    (1 g amoxic… #>  9 AMC  Amoxicillin/cl… high… 0.87…          3 oral           \"\"    (0.875 g am… #> 10 AMC  Amoxicillin/cl… stan… 0.5 …          3 oral           \"\"    (0.5 g amox… #> # ℹ 493 more rows #> # ℹ 1 more variable: eucast_version <dbl>"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":null,"dir":"Reference","previous_headings":"","what":"Apply EUCAST Rules — eucast_rules","title":"Apply EUCAST Rules — eucast_rules","text":"Apply rules clinical breakpoints notes expected resistant phenotypes defined European Committee Antimicrobial Susceptibility Testing (EUCAST, https://www.eucast.org), see Source. Use eucast_dosage() get data.frame advised dosages certain bug-drug combination, based dosage data set. improve interpretation antibiogram EUCAST rules applied, non-EUCAST rules can applied default, see Details.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Apply EUCAST Rules — eucast_rules","text":"","code":"eucast_rules(x, col_mo = NULL, info = interactive(),   rules = getOption(\"AMR_eucastrules\", default = c(\"breakpoints\", \"expert\")),   verbose = FALSE, version_breakpoints = 12, version_expertrules = 3.3,   ampc_cephalosporin_resistance = NA, only_sir_columns = FALSE,   custom_rules = NULL, overwrite = TRUE, ...)  eucast_dosage(ab, administration = \"iv\", version_breakpoints = 12)"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Apply EUCAST Rules — eucast_rules","text":"EUCAST Expert Rules. Version 2.0, 2012. Leclercq et al. EUCAST expert rules antimicrobial susceptibility testing. Clin Microbiol Infect. 2013;19(2):141-60; doi:10.1111/j.1469-0691.2011.03703.x EUCAST Expert Rules, Intrinsic Resistance Exceptional Phenotypes Tables. Version 3.1, 2016. (link) EUCAST Intrinsic Resistance Unusual Phenotypes. Version 3.2, 2020. (link) EUCAST Intrinsic Resistance Unusual Phenotypes. Version 3.3, 2021. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 9.0, 2019. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 10.0, 2020. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 11.0, 2021. (link) EUCAST Breakpoint tables interpretation MICs zone diameters. Version 12.0, 2022. (link)","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Apply EUCAST Rules — eucast_rules","text":"x data set antibiotic columns, amox, AMX AMC col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). info logical indicate whether progress printed console - default print interactive sessions rules character vector specifies rules applied. Must one \"breakpoints\", \"expert\", \"\", \"custom\", \"\", defaults c(\"breakpoints\", \"expert\"). default value can set another value using package option AMR_eucastrules: options(AMR_eucastrules = \"\"). using \"custom\", sure fill argument custom_rules . Custom rules can created custom_eucast_rules(). verbose logical turn Verbose mode (default ). Verbose mode, function apply rules data, instead returns data set logbook form extensive info rows columns effected way. Using Verbose mode takes lot time. version_breakpoints version number use EUCAST Clinical Breakpoints guideline. Can \"12.0\", \"11.0\", \"10.0\". version_expertrules version number use EUCAST Expert Rules Intrinsic Resistance guideline. Can \"3.3\", \"3.2\", \"3.1\". ampc_cephalosporin_resistance character value applied cefotaxime, ceftriaxone ceftazidime AmpC de-repressed cephalosporin-resistant mutants - default NA. Currently works version_expertrules 3.2 higher; version 'EUCAST Expert Rules Enterobacterales' state results cefotaxime, ceftriaxone ceftazidime reported note, results suppressed (emptied) three drugs. value NA (default) argument remove results three drugs, e.g. value \"R\" make results drugs resistant. Use NULL FALSE alter results three drugs AmpC de-repressed cephalosporin-resistant mutants. Using TRUE equal using \"R\".  EUCAST Expert Rules v3.2, rule applies : Citrobacter braakii, Citrobacter freundii, Citrobacter gillenii, Citrobacter murliniae, Citrobacter rodenticum, Citrobacter sedlakii, Citrobacter werkmanii, Citrobacter youngae, Enterobacter, Hafnia alvei, Klebsiella aerogenes, Morganella morganii, Providencia, Serratia. only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE) custom_rules custom rules apply, created custom_eucast_rules() overwrite logical indicate whether non-NA values must overwritten (defaults TRUE). FALSE, NA values changed. ... column name antibiotic, see section Antibiotics ab (vector ) text can coerced valid antibiotic drug code .ab() administration route administration, either \"im\", \"iv\", \"oral\"","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Apply EUCAST Rules — eucast_rules","text":"input x, possibly edited values antibiotics. , verbose = TRUE, data.frame original new values affected bug-drug combinations.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Apply EUCAST Rules — eucast_rules","text":"Note: function translate MIC values SIR values. Use .sir() . Note: ampicillin (AMP, J01CA01) available amoxicillin (AMX, J01CA04) , latter used rules dependency ampicillin. drugs interchangeable comes expression antimicrobial resistance. file containing EUCAST rules located : https://github.com/msberends/AMR/blob/main/data-raw/eucast_rules.tsv.  Note: Old taxonomic names replaced current taxonomy applicable. example, Ochrobactrum anthropi renamed Brucella anthropi 2020; original EUCAST rules v3.1 v3.2 yet contain new taxonomic name. AMR package contains full microbial taxonomy updated June 24th, 2024, see microorganisms.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"custom-rules","dir":"Reference","previous_headings":"","what":"Custom Rules","title":"Apply EUCAST Rules — eucast_rules","text":"Custom rules can created using custom_eucast_rules(), e.g.:","code":"x <- custom_eucast_rules(AMC == \"R\" & genus == \"Klebsiella\" ~ aminopenicillins == \"R\",                          AMC == \"I\" & genus == \"Klebsiella\" ~ aminopenicillins == \"I\")  eucast_rules(example_isolates, rules = \"custom\", custom_rules = x)"},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"-other-rules","dir":"Reference","previous_headings":"","what":"'Other' Rules","title":"Apply EUCAST Rules — eucast_rules","text":"processing, two non-EUCAST rules drug combinations can applied improve efficacy EUCAST rules, reliability data (analysis). rules : drug enzyme inhibitor set S drug without enzyme inhibitor S drug without enzyme inhibitor set R drug enzyme inhibitor R Important examples include amoxicillin amoxicillin/clavulanic acid, trimethoprim trimethoprim/sulfamethoxazole. Needless say, rules work, drugs must available data set. Since rules officially approved EUCAST, applied default. use rules, include \"\" rules argument, use eucast_rules(..., rules = \"\"). can also set package option AMR_eucastrules, .e. run options(AMR_eucastrules = \"\").","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"antibiotics","dir":"Reference","previous_headings":"","what":"Antibiotics","title":"Apply EUCAST Rules — eucast_rules","text":"define antibiotics column names, leave determine automatically guess_ab_col() input text (case-insensitive), use NULL skip column (e.g. TIC = NULL skip ticarcillin). Manually defined non-existing columns skipped warning. following antibiotics eligible functions eucast_rules() mdro(). shown format 'name (antimicrobial ID, ATC code)', sorted alphabetically: Amikacin (AMK, J01GB06), amoxicillin (AMX, J01CA04), amoxicillin/clavulanic acid (AMC, J01CR02), ampicillin (AMP, J01CA01), ampicillin/sulbactam (SAM, J01CR01), apramycin (APR, QA07AA92), arbekacin (ARB, J01GB12), aspoxicillin (APX, J01CA19), azidocillin (AZD, J01CE04), azithromycin (AZM, J01FA10), azlocillin (AZL, J01CA09), aztreonam (ATM, J01DF01), bacampicillin (BAM, J01CA06), bekanamycin (BEK, J01GB13), benzathine benzylpenicillin (BNB, J01CE08), benzathine phenoxymethylpenicillin (BNP, J01CE10), benzylpenicillin (PEN, J01CE01), besifloxacin (BES, S01AE08), biapenem (BIA, J01DH05), carbenicillin (CRB, J01CA03), carindacillin (CRN, J01CA05), carumonam (CAR, J01DF02), cefacetrile (CAC, J01DB10), cefaclor (CEC, J01DC04), cefadroxil (CFR, J01DB05), cefalexin (LEX, J01DB01), cefaloridine (RID, J01DB02), cefalotin (CEP, J01DB03), cefamandole (MAN, J01DC03), cefapirin (HAP, J01DB08), cefatrizine (CTZ, J01DB07), cefazedone (CZD, J01DB06), cefazolin (CZO, J01DB04), cefcapene (CCP, J01DD17), cefdinir (CDR, J01DD15), cefditoren (DIT, J01DD16), cefepime (FEP, J01DE01), cefepime/amikacin (CFA, J01RA06), cefetamet (CAT, J01DD10), cefiderocol (FDC, J01DI04), cefixime (CFM, J01DD08), cefmenoxime (CMX, J01DD05), cefmetazole (CMZ, J01DC09), cefodizime (DIZ, J01DD09), cefonicid (CID, J01DC06), cefoperazone (CFP, J01DD12), cefoperazone/sulbactam (CSL, J01DD62), ceforanide (CND, J01DC11), cefotaxime (CTX, J01DD01), cefotaxime/clavulanic acid (CTC, J01DD51), cefotetan (CTT, J01DC05), cefotiam (CTF, J01DC07), cefovecin (FOV, QJ01DD91), cefoxitin (FOX, J01DC01), cefozopran (ZOP, J01DE03), cefpiramide (CPM, J01DD11), cefpirome (CPO, J01DE02), cefpodoxime (CPD, J01DD13), cefprozil (CPR, J01DC10), cefquinome (CEQ, QG51AA07), cefroxadine (CRD, J01DB11), cefsulodin (CFS, J01DD03), ceftaroline (CPT, J01DI02), ceftazidime (CAZ, J01DD02), ceftazidime/clavulanic acid (CCV, J01DD52), cefteram (CEM, J01DD18), ceftezole (CTL, J01DB12), ceftibuten (CTB, J01DD14), ceftiofur (TIO, QJ01DD90), ceftizoxime (CZX, J01DD07), ceftobiprole medocaril (CFM1, J01DI01), ceftolozane/tazobactam (CZT, J01DI54), ceftriaxone (CRO, J01DD04), ceftriaxone/beta-lactamase inhibitor (CEB, J01DD63), cefuroxime (CXM, J01DC02), cephradine (CED, J01DB09), chloramphenicol (CHL, J01BA01), ciprofloxacin (CIP, J01MA02), clarithromycin (CLR, J01FA09), clindamycin (CLI, J01FF01), clometocillin (CLM, J01CE07), cloxacillin (CLO, J01CF02), colistin (COL, J01XB01), cycloserine (CYC, J04AB01), dalbavancin (DAL, J01XA04), danofloxacin (DAN, QJ01MA92), daptomycin (DAP, J01XX09), delafloxacin (DFX, J01MA23), dibekacin (DKB, J01GB09), dicloxacillin (DIC, J01CF01), difloxacin (DIF, QJ01MA94), dirithromycin (DIR, J01FA13), doripenem (DOR, J01DH04), doxycycline (DOX, J01AA02), enoxacin (ENX, J01MA04), enrofloxacin (ENR, QJ01MA90), epicillin (EPC, J01CA07), ertapenem (ETP, J01DH03), erythromycin (ERY, J01FA01), fleroxacin (FLE, J01MA08), flucloxacillin (FLC, J01CF05), flurithromycin (FLR1, J01FA14), fosfomycin (FOS, J01XX01), framycetin (FRM, D09AA01), fusidic acid (FUS, J01XC01), gamithromycin (GAM, QJ01FA95), garenoxacin (GRN, J01MA19), gatifloxacin (GAT, J01MA16), gemifloxacin (GEM, J01MA15), gentamicin (GEN, J01GB03), grepafloxacin (GRX, J01MA11), hetacillin (HET, J01CA18), imipenem (IPM, J01DH51), imipenem/relebactam (IMR, J01DH56), isepamicin (ISE, J01GB11), josamycin (JOS, J01FA07), kanamycin (KAN, J01GB04), kitasamycin (KIT, QJ01FA93), lascufloxacin (LSC, J01MA25), latamoxef (LTM, J01DD06), levofloxacin (LVX, J01MA12), levonadifloxacin (LND, J01MA24), lincomycin (LIN, J01FF02), linezolid (LNZ, J01XX08), lomefloxacin (LOM, J01MA07), loracarbef (LOR, J01DC08), marbofloxacin (MAR, QJ01MA93), mecillinam (MEC, J01CA11), meropenem (MEM, J01DH02), meropenem/vaborbactam (MEV, J01DH52), metampicillin (MTM, J01CA14), meticillin (MET, J01CF03), mezlocillin (MEZ, J01CA10), micronomicin (MCR, S01AA22), midecamycin (MID, J01FA03), minocycline (MNO, J01AA08), miocamycin (MCM, J01FA11), moxifloxacin (MFX, J01MA14), nadifloxacin (NAD, D10AF05), nafcillin (NAF, J01CF06), nalidixic acid (NAL, J01MB02), neomycin (NEO, J01GB05), netilmicin (NET, J01GB07), nitrofurantoin (NIT, J01XE01), norfloxacin (, J01MA06), novobiocin (NOV, QJ01XX95), ofloxacin (OFX, J01MA01), oleandomycin (OLE, J01FA05), orbifloxacin (ORB, QJ01MA95), oritavancin (ORI, J01XA05), oxacillin (OXA, J01CF04), panipenem (PAN, J01DH55), pazufloxacin (PAZ, J01MA18), pefloxacin (PEF, J01MA03), penamecillin (PNM, J01CE06), pheneticillin (PHE, J01CE05), phenoxymethylpenicillin (PHN, J01CE02), piperacillin (PIP, J01CA12), piperacillin/tazobactam (TZP, J01CR05), pirlimycin (PRL, QJ51FF90), pivampicillin (PVM, J01CA02), pivmecillinam (PME, J01CA08), plazomicin (PLZ, J01GB14), polymyxin B (PLB, J01XB02), pradofloxacin (PRA, QJ01MA97), pristinamycin (PRI, J01FG01), procaine benzylpenicillin (PRB, J01CE09), propicillin (PRP, J01CE03), prulifloxacin (PRU, J01MA17), quinupristin/dalfopristin (QDA, QJ01FG02), ribostamycin (RST, J01GB10), rifampicin (RIF, J04AB02), rokitamycin (ROK, J01FA12), roxithromycin (RXT, J01FA06), rufloxacin (RFL, J01MA10), sarafloxacin (SAR, QJ01MA98), sisomicin (SIS, J01GB08), sitafloxacin (SIT, J01MA21), solithromycin (SOL, J01FA16), sparfloxacin (SPX, J01MA09), spiramycin (SPI, J01FA02), streptoduocin (STR, J01GA02), streptomycin (STR1, J01GA01), sulbenicillin (SBC, J01CA16), sulfadiazine (SDI, J01EC02), sulfadiazine/trimethoprim (SLT1, J01EE02), sulfadimethoxine (SUD, J01ED01), sulfadimidine (SDM, J01EB03), sulfadimidine/trimethoprim (SLT2, J01EE05), sulfafurazole (SLF, J01EB05), sulfaisodimidine (SLF1, J01EB01), sulfalene (SLF2, J01ED02), sulfamazone (SZO, J01ED09), sulfamerazine (SLF3, J01ED07), sulfamerazine/trimethoprim (SLT3, J01EE07), sulfamethizole (SLF4, J01EB02), sulfamethoxazole (SMX, J01EC01), sulfamethoxypyridazine (SLF5, J01ED05), sulfametomidine (SLF6, J01ED03), sulfametoxydiazine (SLF7, J01ED04), sulfametrole/trimethoprim (SLT4, J01EE03), sulfamoxole (SLF8, J01EC03), sulfamoxole/trimethoprim (SLT5, J01EE04), sulfanilamide (SLF9, J01EB06), sulfaperin (SLF10, J01ED06), sulfaphenazole (SLF11, J01ED08), sulfapyridine (SLF12, J01EB04), sulfathiazole (SUT, J01EB07), sulfathiourea (SLF13, J01EB08), sultamicillin (SLT6, J01CR04), talampicillin (TAL, J01CA15), tebipenem (TBP, J01DH06), tedizolid (TZD, J01XX11), teicoplanin (TEC, J01XA02), telavancin (TLV, J01XA03), telithromycin (TLT, J01FA15), temafloxacin (TMX, J01MA05), temocillin (TEM, J01CA17), tetracycline (TCY, J01AA07), ticarcillin (TIC, J01CA13), ticarcillin/clavulanic acid (TCC, J01CR03), tigecycline (TGC, J01AA12), tilbroquinol (TBQ, P01AA05), tildipirosin (TIP, QJ01FA96), tilmicosin (TIL, QJ01FA91), tobramycin (TOB, J01GB01), tosufloxacin (TFX, J01MA22), trimethoprim (TMP, J01EA01), trimethoprim/sulfamethoxazole (SXT, J01EE01), troleandomycin (TRL, J01FA08), trovafloxacin (TVA, J01MA13), tulathromycin (TUL, QJ01FA94), tylosin (TYL, QJ01FA90), tylvalosin (TYL1, QJ01FA92), vancomycin (VAN, J01XA01)","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Apply EUCAST Rules — eucast_rules","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/eucast_rules.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Apply EUCAST Rules — eucast_rules","text":"","code":"# \\donttest{ a <- data.frame(   mo = c(     \"Staphylococcus aureus\",     \"Enterococcus faecalis\",     \"Escherichia coli\",     \"Klebsiella pneumoniae\",     \"Pseudomonas aeruginosa\"   ),   VAN = \"-\", # Vancomycin   AMX = \"-\", # Amoxicillin   COL = \"-\", # Colistin   CAZ = \"-\", # Ceftazidime   CXM = \"-\", # Cefuroxime   PEN = \"S\", # Benzylpenicillin   FOX = \"S\", # Cefoxitin   stringsAsFactors = FALSE )  head(a) #>                       mo VAN AMX COL CAZ CXM PEN FOX #> 1  Staphylococcus aureus   -   -   -   -   -   S   S #> 2  Enterococcus faecalis   -   -   -   -   -   S   S #> 3       Escherichia coli   -   -   -   -   -   S   S #> 4  Klebsiella pneumoniae   -   -   -   -   -   S   S #> 5 Pseudomonas aeruginosa   -   -   -   -   -   S   S   # apply EUCAST rules: some results wil be changed b <- eucast_rules(a) #> Warning: in eucast_rules(): not all columns with antimicrobial results are of #> class 'sir'. Transform them on beforehand, with e.g.: #>   - a %>% as.sir(PEN:FOX) #>   - a %>% mutate_if(is_sir_eligible, as.sir) #>   - a %>% mutate(across(where(is_sir_eligible), as.sir))  head(b) #>                       mo VAN AMX COL CAZ CXM PEN FOX #> 1  Staphylococcus aureus   -   -   R   R   S   S   S #> 2  Enterococcus faecalis   -   -   R   R   R   S   R #> 3       Escherichia coli   R   -   -   -   -   R   S #> 4  Klebsiella pneumoniae   R   R   -   -   -   R   S #> 5 Pseudomonas aeruginosa   R   R   -   -   R   R   R   # do not apply EUCAST rules, but rather get a data.frame # containing all details about the transformations: c <- eucast_rules(a, verbose = TRUE) #> Warning: in eucast_rules(): not all columns with antimicrobial results are of #> class 'sir'. Transform them on beforehand, with e.g.: #>   - a %>% as.sir(PEN:FOX) #>   - a %>% mutate_if(is_sir_eligible, as.sir) #>   - a %>% mutate(across(where(is_sir_eligible), as.sir)) head(c) #>   row col           mo_fullname old new rule   rule_group #> 1   1 CXM Staphylococcus aureus   -   S       Breakpoints #> 2   1 CAZ Staphylococcus aureus   -   R      Expert Rules #> 3   1 COL Staphylococcus aureus   -   R      Expert Rules #> 4   2 CAZ Enterococcus faecalis   -   R      Expert Rules #> 5   2 COL Enterococcus faecalis   -   R      Expert Rules #> 6   2 CXM Enterococcus faecalis   -   R      Expert Rules #>                                                 rule_name #> 1                                          Staphylococcus #> 2 Table 4: Intrinsic resistance in gram-positive bacteria #> 3 Table 4: Intrinsic resistance in gram-positive bacteria #> 4 Table 4: Intrinsic resistance in gram-positive bacteria #> 5 Table 4: Intrinsic resistance in gram-positive bacteria #> 6 Table 4: Intrinsic resistance in gram-positive bacteria #>                                                                                 rule_source #> 1                                           'EUCAST Clinical Breakpoint Tables' v12.0, 2022 #> 2 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 3 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 4 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 5 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 #> 6 'EUCAST Expert Rules' and 'EUCAST Intrinsic Resistance and Unusual Phenotypes' v3.3, 2021 # }  # Dosage guidelines:  eucast_dosage(c(\"tobra\", \"genta\", \"cipro\"), \"iv\") #> ℹ Dosages for antimicrobial drugs, as meant for 'EUCAST Clinical Breakpoint #>   Tables' v12.0 (2022). This note will be shown once per session. #> # A tibble: 3 × 4 #>   ab   name          standard_dosage  high_dosage  #>   <ab> <chr>         <chr>            <chr>        #> 1 TOB  Tobramycin    6-7 mg/kg x 1 iv NA           #> 2 GEN  Gentamicin    6-7 mg/kg x 1 iv NA           #> 3 CIP  Ciprofloxacin 0.4 g x 2 iv     0.4 g x 3 iv  eucast_dosage(c(\"tobra\", \"genta\", \"cipro\"), \"iv\", version_breakpoints = 10) #> ℹ Dosages for antimicrobial drugs, as meant for 'EUCAST Clinical Breakpoint #>   Tables' v10.0 (2020). This note will be shown once per session. #> # A tibble: 3 × 4 #>   ab   name          standard_dosage  high_dosage  #>   <ab> <chr>         <chr>            <chr>        #> 1 TOB  Tobramycin    6-7 mg/kg x 1 iv NA           #> 2 GEN  Gentamicin    6-7 mg/kg x 1 iv NA           #> 3 CIP  Ciprofloxacin 0.4 g x 2 iv     0.4 g x 3 iv"},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 2 000 Example Isolates — example_isolates","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"data set containing 2 000 microbial isolates full antibiograms. data set contains randomised fictitious data, reflects reality can used practise AMR data analysis. examples, please read tutorial website.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"","code":"example_isolates"},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"tibble 2 000 observations 46 variables: date Date receipt laboratory patient ID patient age Age patient gender Gender patient, either \"F\" \"M\" ward Ward type patient admitted, either \"Clinical\", \"ICU\", \"Outpatient\" mo ID microorganism created .mo(), see also microorganisms data set PEN:RIF 40 different antibiotics class sir (see .sir()); column names occur antibiotics data set can translated set_ab_names() ab_name()","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 2 000 Example Isolates — example_isolates","text":"","code":"example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Unclean Data — example_isolates_unclean","title":"Data Set with Unclean Data — example_isolates_unclean","text":"data set containing 3 000 microbial isolates cleaned consequently ready AMR data analysis. data set can used practice.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Unclean Data — example_isolates_unclean","text":"","code":"example_isolates_unclean"},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Unclean Data — example_isolates_unclean","text":"tibble 3 000 observations 8 variables: patient_id ID patient date date receipt laboratory hospital ID hospital, C bacteria info microorganism can transformed .mo(), see also microorganisms AMX:GEN 4 different antibiotics transformed .sir()","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Unclean Data — example_isolates_unclean","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/example_isolates_unclean.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Unclean Data — example_isolates_unclean","text":"","code":"example_isolates_unclean #> # A tibble: 3,000 × 8 #>    patient_id hospital date       bacteria      AMX   AMC   CIP   GEN   #>    <chr>      <chr>    <date>     <chr>         <chr> <chr> <chr> <chr> #>  1 J3         A        2012-11-21 E. coli       R     I     S     S     #>  2 R7         A        2018-04-03 K. pneumoniae R     I     S     S     #>  3 P3         A        2014-09-19 E. coli       R     S     S     S     #>  4 P10        A        2015-12-10 E. coli       S     I     S     S     #>  5 B7         A        2015-03-02 E. coli       S     S     S     S     #>  6 W3         A        2018-03-31 S. aureus     R     S     R     S     #>  7 J8         A        2016-06-14 E. coli       R     S     S     S     #>  8 M3         A        2015-10-25 E. coli       R     S     S     S     #>  9 J3         A        2019-06-19 E. coli       S     S     S     S     #> 10 G6         A        2015-04-27 S. aureus     S     S     S     S     #> # ℹ 2,990 more rows"},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":null,"dir":"Reference","previous_headings":"","what":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"Export Data Set NCBI BioSample Antibiogram","code":""},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"","code":"export_ncbi_biosample(x, filename = paste0(\"biosample_\", format(Sys.time(),   \"%Y-%m-%d-%H%M%S\"), \".xlsx\"), type = \"pathogen MIC\",   columns = where(is.mic), save_as_xlsx = TRUE)"},{"path":"https://msberends.github.io/AMR/reference/export_ncbi_biosample.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Export Data Set as NCBI BioSample Antibiogram — export_ncbi_biosample","text":"x data set filename character string specifying file name type character string specifying type data set, either \"pathogen MIC\" \"beta-lactamase MIC\", see https://www.ncbi.nlm.nih.gov/biosample/docs/","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine First Isolates — first_isolate","title":"Determine First Isolates — first_isolate","text":"Determine first isolates microorganisms every patient per episode (needed) per specimen type. functions support four methods summarised Hindler et al. 2007 (doi:10.1086/511864 ). determine patient episodes necessarily based microorganisms, use is_new_episode() also supports grouping dplyr package.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine First Isolates — first_isolate","text":"","code":"first_isolate(x = NULL, col_date = NULL, col_patient_id = NULL,   col_mo = NULL, col_testcode = NULL, col_specimen = NULL,   col_icu = NULL, col_keyantimicrobials = NULL, episode_days = 365,   testcodes_exclude = NULL, icu_exclude = FALSE, specimen_group = NULL,   type = \"points\", method = c(\"phenotype-based\", \"episode-based\",   \"patient-based\", \"isolate-based\"), ignore_I = TRUE, points_threshold = 2,   info = interactive(), include_unknown = FALSE,   include_untested_sir = TRUE, ...)  filter_first_isolate(x = NULL, col_date = NULL, col_patient_id = NULL,   col_mo = NULL, episode_days = 365, method = c(\"phenotype-based\",   \"episode-based\", \"patient-based\", \"isolate-based\"), ...)"},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Determine First Isolates — first_isolate","text":"Methodology functions strictly based : M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/. Hindler JF Stelling J (2007). Analysis Presentation Cumulative Antibiograms: New Consensus Guideline Clinical Laboratory Standards Institute. Clinical Infectious Diseases, 44(6), 867-873. doi:10.1086/511864","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine First Isolates — first_isolate","text":"x data.frame containing isolates. Can left blank automatic determination, see Examples. col_date column name result date (date received lab) - default first column date class col_patient_id column name unique IDs patients - default first column starts 'patient' 'patid' (case insensitive) col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). col_testcode column name test codes. Use col_testcode = NULL exclude certain test codes (test codes screening). case testcodes_exclude ignored. col_specimen column name specimen type group col_icu column name logicals (TRUE/FALSE) whether ward department Intensive Care Unit (ICU). can also logical vector length rows x. col_keyantimicrobials (useful method = \"phenotype-based\") column name key antimicrobials determine first isolates, see key_antimicrobials(). default first column starts 'key' followed 'ab' 'antibiotics' 'antimicrobials' (case insensitive). Use col_keyantimicrobials = FALSE prevent . Can also output key_antimicrobials(). episode_days episode days genus/species combination determined 'first isolate' . default 365 days based guideline CLSI, see Source. testcodes_exclude character vector test codes excluded (case-insensitive) icu_exclude logical indicate whether ICU isolates excluded (rows value TRUE column set col_icu) specimen_group value column set col_specimen filter type type determine weighed isolates; can \"keyantimicrobials\" \"points\", see Details method method apply, either \"phenotype-based\", \"episode-based\", \"patient-based\" \"isolate-based\" (can abbreviated), see Details. default \"phenotype-based\" antimicrobial test results present data, \"episode-based\" otherwise. ignore_I logical indicate whether antibiotic interpretations \"\" ignored type = \"keyantimicrobials\", see Details points_threshold minimum number points require differences antibiogram lead inclusion isolate type = \"points\", see Details info logical indicate info printed - default TRUE interactive mode include_unknown logical indicate whether 'unknown' microorganisms included , .e. microbial code \"UNKNOWN\", defaults FALSE. WHONET users, means records organism code \"con\" (contamination) excluded default. Isolates microbial ID NA always excluded first isolate. include_untested_sir logical indicate whether also rows without antibiotic results still eligible becoming first isolate. Use include_untested_sir = FALSE always return FALSE rows. checks data set columns class sir consequently requires transforming columns antibiotic results using .sir() first. ... arguments passed first_isolate() using filter_first_isolate(), otherwise arguments passed key_antimicrobials() (universal, gram_negative, gram_positive)","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine First Isolates — first_isolate","text":"logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine First Isolates — first_isolate","text":"methodology implemented functions strictly based recommendations outlined CLSI Guideline M39 research overview Hindler et al. (2007, doi:10.1086/511864 ). conduct epidemiological analyses antimicrobial resistance data, -called first isolates included prevent overestimation underestimation antimicrobial resistance. Different methods can used , see . functions context-aware. means x argument can left blank used inside data.frame call, see Examples. first_isolate() function wrapper around is_new_episode() function, efficient data sets containing microorganism codes names. isolates microbial ID NA excluded first isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"different-methods","dir":"Reference","previous_headings":"","what":"Different methods","title":"Determine First Isolates — first_isolate","text":"According previously-mentioned sources, different methods (algorithms) select first isolates increasing reliability: isolate-based, patient-based, episode-based phenotype-based. methods select combination taxonomic genus species (subspecies). mentioned methods covered first_isolate() function:","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"isolate-based","dir":"Reference","previous_headings":"","what":"Isolate-based","title":"Determine First Isolates — first_isolate","text":"method require selection, isolates included. , however, respect arguments set first_isolate() function. example, default setting include_unknown (FALSE) omit selection rows without microbial ID.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"patient-based","dir":"Reference","previous_headings":"","what":"Patient-based","title":"Determine First Isolates — first_isolate","text":"include every genus-species combination per patient , set episode_days Inf. method makes sure duplicate isolates selected patient. method preferred e.g. identify first MRSA finding patient determine incidence. Conversely, large longitudinal data set, mean isolates excluded found years initial isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"episode-based","dir":"Reference","previous_headings":"","what":"Episode-based","title":"Determine First Isolates — first_isolate","text":"include every genus-species combination per patient episode , set episode_days sensible number days. Depending type analysis, 14, 30, 60 365. Short episodes common analysing specific hospital ward data ICU cases, long episodes common analysing regional national data. common method correct duplicate isolates. Patients categorised episodes based ID dates (e.g., date specimen receipt laboratory result). common method, take account antimicrobial test results. means e.g. methicillin-resistant Staphylococcus aureus (MRSA) isolate differentiated wildtype Staphylococcus aureus isolate.","code":""},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"phenotype-based","dir":"Reference","previous_headings":"","what":"Phenotype-based","title":"Determine First Isolates — first_isolate","text":"reliable method, since also weighs antibiogram (antimicrobial test results) yielding -called 'first weighted isolates'. two different methods weigh antibiogram: Using type = \"points\" argument points_threshold (default) method weighs antimicrobial drugs available data set. difference S R (vice versa) counts 0.5 points, difference S R (vice versa) counts 1 point. sum points exceeds points_threshold, defaults 2, isolate selected first weighted isolate. antimicrobials internally selected using all_antimicrobials() function. output function need passed first_isolate() function. Using type = \"keyantimicrobials\" argument ignore_I method weighs specific antimicrobial drugs, called key antimicrobials. difference S R (vice versa) key antimicrobials select isolate first weighted isolate. ignore_I = FALSE, also differences S R (vice versa) lead . Key antimicrobials internally selected using key_antimicrobials() function, can also added manually variable data set col_keyantimicrobials argument. Another option pass output key_antimicrobials() function directly col_keyantimicrobials argument. default method phenotype-based (using type = \"points\") episode-based (using episode_days = 365). makes sure every genus-species combination selected per patient per year, taking account antimicrobial test results. antimicrobial test results available data set, episode-based method applied default.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/first_isolate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine First Isolates — first_isolate","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  example_isolates[first_isolate(info = TRUE), ] #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Using column 'date' as input for col_date. #> ℹ Using column 'patient' as input for col_patient_id. #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> ℹ Excluding 16 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 1,387 'phenotype-based' first isolates (69.4% of total where a #>    microbial ID was available) #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # \\donttest{ # get all first Gram-negatives example_isolates[which(first_isolate(info = FALSE) & mo_is_gram_negative()), ] #> ℹ Using column 'mo' as input for mo_is_gram_negative() #> # A tibble: 441 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  4 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #>  5 2002-02-05 067927     45 F      ICU      B_SERRT_MRCS   R     NA    NA    R   #>  6 2002-02-27 066895     85 F      Clinical B_KLBSL_PNMN   R     NA    NA    R   #>  7 2002-03-08 4FC193     69 M      Clinical B_ESCHR_COLI   R     NA    NA    R   #>  8 2002-03-16 4FC193     69 M      Clinical B_PSDMN_AERG   R     NA    NA    R   #>  9 2002-04-01 496896     46 F      ICU      B_ESCHR_COLI   R     NA    NA    NA  #> 10 2002-04-23 EE2510     69 F      ICU      B_ESCHR_COLI   R     NA    NA    NA  #> # ℹ 431 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  if (require(\"dplyr\")) {   # filter on first isolates using dplyr:   example_isolates %>%     filter(first_isolate(info = TRUE)) } #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #> ℹ Excluding 16 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 1,387 'phenotype-based' first isolates (69.4% of total where a #>    microbial ID was available) #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … if (require(\"dplyr\")) {   # short-hand version:   example_isolates %>%     filter_first_isolate(info = FALSE) } #> # A tibble: 1,387 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #>  6 2002-01-17 495616     67 M      Clinical B_STPHY_EPDR   R     NA    S     NA  #>  7 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  8 2002-01-21 462081     75 F      Clinical B_CTRBC_FRND   R     NA    NA    R   #>  9 2002-01-22 F35553     50 M      ICU      B_PROTS_MRBL   R     NA    NA    NA  #> 10 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #> # ℹ 1,377 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … if (require(\"dplyr\")) {   # flag the first isolates per group:   example_isolates %>%     group_by(ward) %>%     mutate(first = first_isolate(info = TRUE)) %>%     select(ward, date, patient, mo, first) } #> ℹ Determining first isolates using an episode length of 365 days #> ℹ Basing inclusion on all antimicrobial results, using a points threshold #>   of 2 #>  #> Group: ward = \"Clinical\" #> ℹ Excluding 9 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 865 'phenotype-based' first isolates (70.1% of total where a #>    microbial ID was available) #>  #> Group: ward = \"ICU\" #> ℹ Excluding 6 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 452 'phenotype-based' first isolates (70.0% of total where a #>    microbial ID was available) #>  #> Group: ward = \"Outpatient\" #> ℹ Excluding 1 isolates with a microbial ID 'UNKNOWN' (in column 'mo') #> => Found 99 'phenotype-based' first isolates (82.5% of total where a #>    microbial ID was available) #> # A tibble: 2,000 × 5 #> # Groups:   ward [3] #>    ward     date       patient mo           first #>    <chr>    <date>     <chr>   <mo>         <lgl> #>  1 Clinical 2002-01-02 A77334  B_ESCHR_COLI TRUE  #>  2 Clinical 2002-01-03 A77334  B_ESCHR_COLI FALSE #>  3 ICU      2002-01-07 067927  B_STPHY_EPDR TRUE  #>  4 ICU      2002-01-07 067927  B_STPHY_EPDR FALSE #>  5 ICU      2002-01-13 067927  B_STPHY_EPDR FALSE #>  6 ICU      2002-01-13 067927  B_STPHY_EPDR FALSE #>  7 Clinical 2002-01-14 462729  B_STPHY_AURS TRUE  #>  8 Clinical 2002-01-14 462729  B_STPHY_AURS FALSE #>  9 ICU      2002-01-16 067927  B_STPHY_EPDR TRUE  #> 10 ICU      2002-01-17 858515  B_STPHY_EPDR TRUE  #> # ℹ 1,990 more rows # }"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":null,"dir":"Reference","previous_headings":"","what":"G-test for Count Data — g.test","title":"G-test for Count Data — g.test","text":"g.test() performs chi-squared contingency table tests goodness--fit tests, just like chisq.test() reliable (1). G-test can used see whether number observations category fits theoretical expectation (called G-test goodness--fit), see whether proportions one variable different different values variable (called G-test independence).","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"G-test for Count Data — g.test","text":"","code":"g.test(x, y = NULL, p = rep(1/length(x), length(x)), rescale.p = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"G-test for Count Data — g.test","text":"code function identical chisq.test(), except : calculation statistic changed \\(2 * sum(x * log(x / E))\\) Yates' continuity correction removed apply G-test possibility simulate p values simulate.p.value removed","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"G-test for Count Data — g.test","text":"x numeric vector matrix. x y can also     factors. y numeric vector; ignored x matrix.      x factor, y factor length. p vector probabilities length x.     error given entry p negative. rescale.p logical scalar; TRUE p rescaled     (necessary) sum 1.  rescale.p FALSE,     p sum 1, error given.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"G-test for Count Data — g.test","text":"list class \"htest\" containing following   components: statistic value chi-squared test statistic. parameter degrees freedom approximate     chi-squared distribution test statistic, NA     p-value computed Monte Carlo simulation. p.value p-value test. method character string indicating type test     performed, whether Monte Carlo simulation continuity     correction used. data.name character string giving name(s) data. observed observed counts. expected expected counts null hypothesis. residuals Pearson residuals,     (observed - expected) / sqrt(expected). stdres standardized residuals,     (observed - expected) / sqrt(V), V     residual cell variance (Agresti, 2007, section 2.4.5     case x matrix, n * p * (1 - p) otherwise).","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"G-test for Count Data — g.test","text":"x matrix one row column, x vector y given, goodness--fit test performed (x treated one-dimensional contingency table). entries x must non-negative integers. case, hypothesis tested whether population probabilities equal p, equal p given. x matrix least two rows columns, taken two-dimensional contingency table: entries x must non-negative integers.  Otherwise, x y must vectors factors length; cases missing values removed, objects coerced factors, contingency table computed .  Pearson's chi-squared test performed null hypothesis joint distribution cell counts 2-dimensional contingency table product row column marginals. p-value computed asymptotic chi-squared distribution test statistic. contingency table case simulation done random sampling set contingency tables given marginals, works marginals strictly positive. Note usual sampling situation assumed chi-squared test (G-test) rather Fisher's exact test. goodness--fit case simulation done random sampling discrete distribution specified p, sample size n = sum(x). simulation done R may slow.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"g-test-of-goodness-of-fit-likelihood-ratio-test-","dir":"Reference","previous_headings":"","what":"G-test Of Goodness-of-Fit (Likelihood Ratio Test)","title":"G-test for Count Data — g.test","text":"Use G-test goodness--fit one nominal variable two values (male female, red, pink white flowers). compare observed counts numbers observations category expected counts, calculate using kind theoretical expectation (1:1 sex ratio 1:2:1 ratio genetic cross). expected number observations category small, G-test may give inaccurate results, use exact test instead (fisher.test()). G-test goodness--fit alternative chi-square test goodness--fit (chisq.test()); tests advantages disadvantages, results two tests usually similar.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"g-test-of-independence","dir":"Reference","previous_headings":"","what":"G-test of Independence","title":"G-test for Count Data — g.test","text":"Use G-test independence two nominal variables, two possible values. want know whether proportions one variable different among values variable. also possible G-test independence two nominal variables. example, Jackson et al. (2013) also data children 3, analysis old vs. young, thigh vs. arm, reaction vs. reaction, analyzed together. Fisher's exact test (fisher.test()) exact test, G-test still approximation. 2x2 table, Fisher's Exact test may slower still run seconds, even sum observations multiple millions. G-test independence alternative chi-square test independence (chisq.test()), give approximately results.","code":""},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"how-the-test-works","dir":"Reference","previous_headings":"","what":"How the Test Works","title":"G-test for Count Data — g.test","text":"Unlike exact test goodness--fit (fisher.test()), G-test directly calculate probability obtaining observed results something extreme. Instead, like almost statistical tests, G-test intermediate step; uses data calculate test statistic measures far observed data null expectation. use mathematical relationship, case chi-square distribution, estimate probability obtaining value test statistic. G-test uses log ratio two likelihoods test statistic, also called likelihood ratio test log-likelihood ratio test. formula calculate G-statistic : \\(G = 2 * sum(x * log(x / E))\\) E expected values. Since chi-square distributed, p value can calculated R :   df degrees freedom. two categories want find ones significantly different null expectation, can use method testing category vs. sum categories, Bonferroni correction. use G-tests category, course.","code":"p <- stats::pchisq(G, df, lower.tail = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"references","dir":"Reference","previous_headings":"","what":"References","title":"G-test for Count Data — g.test","text":"McDonald, J.H. 2014. Handbook Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, Maryland. http://www.biostathandbook.com/gtestgof.html.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/g.test.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"G-test for Count Data — g.test","text":"","code":"# = EXAMPLE 1 = # Shivrain et al. (2006) crossed clearfield rice (which are resistant # to the herbicide imazethapyr) with red rice (which are susceptible to # imazethapyr). They then crossed the hybrid offspring and examined the # F2 generation, where they found 772 resistant plants, 1611 moderately # resistant plants, and 737 susceptible plants. If resistance is controlled # by a single gene with two co-dominant alleles, you would expect a 1:2:1 # ratio.  x <- c(772, 1611, 737) g.test(x, p = c(1, 2, 1) / 4) #>  #> \tG-test of goodness-of-fit (likelihood ratio test) #>  #> data:  x #> X-squared = 4.1471, p-value = 0.1257 #>   # There is no significant difference from a 1:2:1 ratio. # Meaning: resistance controlled by a single gene with two co-dominant # alleles, is plausible.   # = EXAMPLE 2 = # Red crossbills (Loxia curvirostra) have the tip of the upper bill either # right or left of the lower bill, which helps them extract seeds from pine # cones. Some have hypothesized that frequency-dependent selection would # keep the number of right and left-billed birds at a 1:1 ratio. Groth (1992) # observed 1752 right-billed and 1895 left-billed crossbills.  x <- c(1752, 1895) g.test(x) #>  #> \tG-test of goodness-of-fit (likelihood ratio test) #>  #> data:  x #> X-squared = 5.6085, p-value = 0.01787 #>   # There is a significant difference from a 1:1 ratio. # Meaning: there are significantly more left-billed birds."},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Clinical or Epidemic Episodes — get_episode","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"functions determine items vector can considered (start ) new episode. can used determine clinical episodes epidemiological analysis. get_episode() function returns index number episode per group, is_new_episode() function returns TRUE every new get_episode() index. absolute relative episode determination supported.","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"","code":"get_episode(x, episode_days = NULL, case_free_days = NULL, ...)  is_new_episode(x, episode_days = NULL, case_free_days = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"x vector dates (class Date POSIXt), sorted internally determine episodes episode_days episode length days specify time period new episode begins, can also less day Inf, see Details case_free_days (inter-epidemic) interval length days new episode start, can also less day Inf, see Details ... ignored, place allow future extensions","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"get_episode(): integer vector is_new_episode(): logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"Episodes can determined two ways: absolute relative. Absolute method uses episode_days define episode length days, new episode start. common use case AMR data analysis microbial epidemiology: episodes S. aureus bacteraemia ICU patients example. episode length 30 days, new S. aureus isolates ICU episode 30 days considered different (new) episode. Thus, method counts since start previous episode. Relative method uses case_free_days quantify duration case-free days (inter-epidemic interval), new episode start. common use case infectious disease epidemiology: episodes norovirus outbreaks hospital example. case-free period 14 days, new norovirus cases time considered different (new) episode. Thus, methods counts since last case previous episode. table: ** marks start new episode, 8 January 2023 7 days since start previous episode (1 January 2023).  *** marks start new episode, 21 January 2023 7 days since last case previous episode (8 January 2023). Either episode_days case_free_days must provided function.","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"difference-between-get-episode-and-is-new-episode-","dir":"Reference","previous_headings":"","what":"Difference between get_episode() and is_new_episode()","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"get_episode() function returns index number episode, cases/patients/isolates first episode number 1, cases/patients/isolates second episode number 2, etc. is_new_episode() function hand, returns TRUE every new get_episode() index. specify, setting episode_days = 365 (using method 1 explained ), two functions differ:","code":""},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"other","dir":"Reference","previous_headings":"","what":"Other","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"first_isolate() function wrapper around is_new_episode() function, efficient data sets containing microorganism codes names allows different isolate selection methods. dplyr package required functions work, episode functions support variable grouping work conveniently inside dplyr verbs filter(), mutate() summarise().","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/get_episode.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Clinical or Epidemic Episodes — get_episode","text":"","code":"# difference between absolute and relative determination of episodes: x <- data.frame(dates = as.Date(c(   \"2021-01-01\",   \"2021-01-02\",   \"2021-01-05\",   \"2021-01-08\",   \"2021-02-21\",   \"2021-02-22\",   \"2021-02-23\",   \"2021-02-24\",   \"2021-03-01\",   \"2021-03-01\" ))) x$absolute <- get_episode(x$dates, episode_days = 7) x$relative <- get_episode(x$dates, case_free_days = 7) x #>         dates absolute relative #> 1  2021-01-01        1        1 #> 2  2021-01-02        1        1 #> 3  2021-01-05        1        1 #> 4  2021-01-08        2        1 #> 5  2021-02-21        3        2 #> 6  2021-02-22        3        2 #> 7  2021-02-23        3        2 #> 8  2021-02-24        3        2 #> 9  2021-03-01        4        2 #> 10 2021-03-01        4        2   # `example_isolates` is a data set available in the AMR package. # See ?example_isolates df <- example_isolates[sample(seq_len(2000), size = 100), ]  get_episode(df$date, episode_days = 60) # indices #>   [1] 47 28 47 47 48  1 48 13  7 38 28  5 14 49 13 22 31 18 28 21  6  8  6 39 13 #>  [26]  5 11 15  9 42 48 43 28 18  4 45 26  2 31 43 10  4 23 47 25 35  1 38 24 15 #>  [51] 19  4 16  5 16  1 44 39 41 25  4 11 30  1 17 30  3 48  8 39  6 38  5 20 37 #>  [76] 36 44 31 48 40 42 43 44 33 12  4 10 34 11  8 12 32 10 33 29 46 34 27 22 23 is_new_episode(df$date, episode_days = 60) # TRUE/FALSE #>   [1]  TRUE  TRUE FALSE FALSE  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE #>  [13]  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE  TRUE FALSE  TRUE #>  [25] FALSE FALSE  TRUE  TRUE  TRUE  TRUE FALSE  TRUE FALSE FALSE  TRUE  TRUE #>  [37]  TRUE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE  TRUE FALSE FALSE #>  [49]  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE #>  [61] FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE #>  [73] FALSE  TRUE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE #>  [85]  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE #>  [97] FALSE  TRUE FALSE FALSE  # filter on results from the third 60-day episode only, using base R df[which(get_episode(df$date, 60) == 3), ] #> # A tibble: 1 × 46 #>   date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>   <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #> 1 2003-01-31 151041     85 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, RIF <sir>  # the functions also work for less than a day, e.g. to include one per hour: get_episode(   c(     Sys.time(),     Sys.time() + 60 * 60   ),   episode_days = 1 / 24 ) #> [1] 1 2  # \\donttest{ if (require(\"dplyr\")) {   # is_new_episode() can also be used in dplyr verbs to determine patient   # episodes based on any (combination of) grouping variables:   df %>%     mutate(condition = sample(       x = c(\"A\", \"B\", \"C\"),       size = 100,       replace = TRUE     )) %>%     group_by(patient, condition) %>%     mutate(new_episode = is_new_episode(date, 365)) %>%     select(patient, date, condition, new_episode) %>%     arrange(patient, condition, date) } #> # A tibble: 100 × 4 #> # Groups:   patient, condition [99] #>    patient date       condition new_episode #>    <chr>   <date>     <chr>     <lgl>       #>  1 067927  2002-01-29 A         TRUE        #>  2 069276  2015-06-18 A         TRUE        #>  3 078381  2014-08-17 A         TRUE        #>  4 09B453  2010-03-21 C         TRUE        #>  5 0DBF93  2015-10-12 A         TRUE        #>  6 118928  2004-02-01 B         TRUE        #>  7 119392  2010-11-01 A         TRUE        #>  8 122506  2007-08-10 B         TRUE        #>  9 146120  2011-09-23 B         TRUE        #> 10 151041  2003-01-31 B         TRUE        #> # ℹ 90 more rows  if (require(\"dplyr\")) {   df %>%     group_by(ward, patient) %>%     transmute(date,       patient,       new_index = get_episode(date, 60),       new_logical = is_new_episode(date, 60)     ) %>%     arrange(patient, ward, date) } #> # A tibble: 100 × 5 #> # Groups:   ward, patient [94] #>    ward     date       patient new_index new_logical #>    <chr>    <date>     <chr>       <int> <lgl>       #>  1 ICU      2002-01-29 067927          1 TRUE        #>  2 Clinical 2015-06-18 069276          1 TRUE        #>  3 ICU      2014-08-17 078381          1 TRUE        #>  4 Clinical 2010-03-21 09B453          1 TRUE        #>  5 Clinical 2015-10-12 0DBF93          1 TRUE        #>  6 Clinical 2004-02-01 118928          1 TRUE        #>  7 Clinical 2010-11-01 119392          1 TRUE        #>  8 Clinical 2007-08-10 122506          1 TRUE        #>  9 Clinical 2011-09-23 146120          1 TRUE        #> 10 Clinical 2003-01-31 151041          1 TRUE        #> # ℹ 90 more rows  if (require(\"dplyr\")) {   df %>%     group_by(ward) %>%     summarise(       n_patients = n_distinct(patient),       n_episodes_365 = sum(is_new_episode(date, episode_days = 365)),       n_episodes_60 = sum(is_new_episode(date, episode_days = 60)),       n_episodes_30 = sum(is_new_episode(date, episode_days = 30))     ) } #> # A tibble: 3 × 5 #>   ward       n_patients n_episodes_365 n_episodes_60 n_episodes_30 #>   <chr>           <int>          <int>         <int>         <int> #> 1 Clinical           57             14            38            48 #> 2 ICU                31             12            25            28 #> 3 Outpatient          6              5             6             6  # grouping on patients and microorganisms leads to the same # results as first_isolate() when using 'episode-based': if (require(\"dplyr\")) {   x <- df %>%     filter_first_isolate(       include_unknown = TRUE,       method = \"episode-based\"     )    y <- df %>%     group_by(patient, mo) %>%     filter(is_new_episode(date, 365)) %>%     ungroup()    identical(x, y) } #> [1] TRUE  # but is_new_episode() has a lot more flexibility than first_isolate(), # since you can now group on anything that seems relevant: if (require(\"dplyr\")) {   df %>%     group_by(patient, mo, ward) %>%     mutate(flag_episode = is_new_episode(date, 365)) %>%     select(group_vars(.), flag_episode) } #> # A tibble: 100 × 4 #> # Groups:   patient, mo, ward [98] #>    patient mo           ward     flag_episode #>    <chr>   <mo>         <chr>    <lgl>        #>  1 F5F794  B_STPHY_AURS Clinical TRUE         #>  2 146120  B_STPHY_AURS Clinical TRUE         #>  3 501607  B_STRPT_AGLC ICU      TRUE         #>  4 402950  B_ENTRC_FCLS ICU      TRUE         #>  5 560323  B_ESCHR_COLI Clinical TRUE         #>  6 4FC193  B_ESCHR_COLI Clinical TRUE         #>  7 D80438  B_CRYNB_STRT Clinical TRUE         #>  8 F81262  B_STRPT_PNMN Clinical TRUE         #>  9 619705  B_STRPT_ANGN ICU      TRUE         #> 10 280834  B_STPHY_CONS Clinical TRUE         #> # ℹ 90 more rows # }"},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":null,"dir":"Reference","previous_headings":"","what":"PCA Biplot with ggplot2 — ggplot_pca","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"Produces ggplot2 variant -called biplot PCA (principal component analysis), flexible appealing base R biplot() function.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"","code":"ggplot_pca(x, choices = 1:2, scale = 1, pc.biplot = TRUE,   labels = NULL, labels_textsize = 3, labels_text_placement = 1.5,   groups = NULL, ellipse = TRUE, ellipse_prob = 0.68,   ellipse_size = 0.5, ellipse_alpha = 0.5, points_size = 2,   points_alpha = 0.25, arrows = TRUE, arrows_colour = \"darkblue\",   arrows_size = 0.5, arrows_textsize = 3, arrows_textangled = TRUE,   arrows_alpha = 0.75, base_textsize = 10, ...)"},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"ggplot_pca() function based ggbiplot() function ggbiplot package Vince Vu, found GitHub: https://github.com/vqv/ggbiplot (retrieved: 2 March 2020, latest commit: 7325e88; 12 February 2015). per GPL-2 licence demands documentation code changes, changes made based source code : Rewritten code remove dependency packages plyr, scales grid Parametrised options, like arrow ellipse settings Hardened input possibilities defining exact type user input every argument Added total amount explained variance caption plot Cleaned syntax based lintr package, fixed grammatical errors added integrity checks Updated documentation","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"x object returned pca(), prcomp() princomp() choices length 2 vector specifying components plot. default     biplot strict sense. scale variables scaled lambda ^ scale     observations scaled lambda ^ (1-scale)     lambda singular values computed     princomp. Normally 0 <= scale <= 1, warning     issued specified scale outside range. pc.biplot true, use Gabriel (1971) refers \"principal component     biplot\", lambda = 1 observations scaled sqrt(n)     variables scaled sqrt(n).  inner products     variables approximate covariances distances observations     approximate Mahalanobis distance. labels optional vector labels observations. set, labels placed respective points. using pca() function input x, determined automatically based attribute non_numeric_cols, see pca(). labels_textsize size text used labels labels_text_placement adjustment factor placement variable names (>=1 means away arrow head) groups optional vector groups labels, length labels. set, points labels coloured according groups. using pca() function input x, determined automatically based attribute non_numeric_cols, see pca(). ellipse logical indicate whether normal data ellipse drawn group (set groups) ellipse_prob statistical size ellipse normal probability ellipse_size size ellipse line ellipse_alpha alpha (transparency) ellipse line points_size size points points_alpha alpha (transparency) points arrows logical indicate whether arrows drawn arrows_colour colour arrow text arrows_size size (thickness) arrow lines arrows_textsize size text end arrows arrows_textangled logical whether text end arrows angled arrows_alpha alpha (transparency) arrows text base_textsize text size plot elements except labels arrows ... arguments passed functions","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"colours labels points can changed adding another scale layer colour, scale_colour_viridis_d() scale_colour_brewer().","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_pca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"PCA Biplot with ggplot2 — ggplot_pca","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # \\donttest{ if (require(\"dplyr\")) {   # calculate the resistance per group first   resistance_data <- example_isolates %>%     group_by(       order = mo_order(mo), # group on anything, like order       genus = mo_genus(mo)     ) %>% #   and genus as we do here;     filter(n() >= 30) %>% # filter on only 30 results per group     summarise_if(is.sir, resistance) # then get resistance of all drugs    # now conduct PCA for certain antimicrobial drugs   pca_result <- resistance_data %>%     pca(AMC, CXM, CTX, CAZ, GEN, TOB, TMP, SXT)    summary(pca_result)    # old base R plotting method:   biplot(pca_result, main = \"Base R biplot\")    # new ggplot2 plotting method using this package:   if (require(\"ggplot2\")) {     ggplot_pca(pca_result) +       labs(title = \"ggplot2 biplot\")   }   if (require(\"ggplot2\")) {     # still extendible with any ggplot2 function     ggplot_pca(pca_result) +       scale_colour_viridis_d() +       labs(title = \"ggplot2 biplot\")   } } #> Warning: There were 73 warnings in `summarise()`. #> The first warning was: #> ℹ In argument: `PEN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 5: `order = \"Lactobacillales\"` `genus = \"Enterococcus\"`. #> Caused by warning: #> ! Introducing NA: only 14 results available for PEN in group: order = #> \"Lactobacillales\", genus = \"Enterococcus\" (minimum = 30). #> ℹ Run `dplyr::last_dplyr_warnings()` to see the 72 remaining warnings. #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\"  #>    # }"},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":null,"dir":"Reference","previous_headings":"","what":"AMR Plots with ggplot2 — ggplot_sir","title":"AMR Plots with ggplot2 — ggplot_sir","text":"Use functions create bar plots AMR data analysis. functions rely ggplot2 functions.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"AMR Plots with ggplot2 — ggplot_sir","text":"","code":"ggplot_sir(data, position = NULL, x = \"antibiotic\",   fill = \"interpretation\", facet = NULL, breaks = seq(0, 1, 0.1),   limits = NULL, translate_ab = \"name\", combine_SI = TRUE,   minimum = 30, language = get_AMR_locale(), nrow = NULL, colours = c(S   = \"#3CAEA3\", SI = \"#3CAEA3\", I = \"#F6D55C\", IR = \"#ED553B\", R = \"#ED553B\"),   datalabels = TRUE, datalabels.size = 2.5, datalabels.colour = \"grey15\",   title = NULL, subtitle = NULL, caption = NULL,   x.title = \"Antimicrobial\", y.title = \"Proportion\", ...)  geom_sir(position = NULL, x = c(\"antibiotic\", \"interpretation\"),   fill = \"interpretation\", translate_ab = \"name\", minimum = 30,   language = get_AMR_locale(), combine_SI = TRUE, ...)"},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"AMR Plots with ggplot2 — ggplot_sir","text":"data data.frame column(s) class sir (see .sir()) position position adjustment bars, either \"fill\", \"stack\" \"dodge\" x variable show x axis, either \"antibiotic\" (default) \"interpretation\" grouping variable fill variable categorise using plots legend, either \"antibiotic\" (default) \"interpretation\" grouping variable facet variable split plots , either \"interpretation\" (default) \"antibiotic\" grouping variable breaks numeric vector positions limits numeric vector length two providing limits scale, use NA refer existing minimum maximum translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. nrow (using facet) number rows colours named vactor colour used filling. default colours colour-blind friendly. datalabels show datalabels using labels_sir_count() datalabels.size size datalabels datalabels.colour colour datalabels title text show title plot subtitle text show subtitle plot caption text show caption plot x.title text show x axis description y.title text show y axis description ... arguments passed geom_sir() , case scale_sir_colours(), named values set colours. default colours colour-blind friendly, maintaining convention e.g. 'susceptible' green 'resistant' red. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"AMR Plots with ggplot2 — ggplot_sir","text":"default, names antibiotics shown plots using ab_name(). can set translate_ab argument. See count_df(). geom_sir() take variable data sir class (created .sir()) using sir_df() plot bars percentage S, , R. default behaviour bars stacked different antibiotics x axis. Additional functions include: facet_sir() creates 2d plots (default based S//R) using ggplot2::facet_wrap(). scale_y_percent() transforms y axis 0 100% range using ggplot2::scale_y_continuous(). scale_sir_colours() sets colours bars (green S, yellow , red R). multilingual support. default colours colour-blind friendly, maintaining convention e.g. 'susceptible' green 'resistant' red. theme_sir() [ggplot2 theme][ggplot2::theme() minimal distraction. labels_sir_count() print datalabels bars percentage amount isolates using ggplot2::geom_text(). ggplot_sir() wrapper around functions uses data first input. makes possible use function pipe (%>%). See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/ggplot_sir.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"AMR Plots with ggplot2 — ggplot_sir","text":"","code":"# \\donttest{ if (require(\"ggplot2\") && require(\"dplyr\")) {   # get antimicrobial results for drugs against a UTI:   ggplot(example_isolates %>% select(AMX, NIT, FOS, TMP, CIP)) +     geom_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # prettify the plot using some additional functions:   df <- example_isolates %>% select(AMX, NIT, FOS, TMP, CIP)   ggplot(df) +     geom_sir() +     scale_y_percent() +     scale_sir_colours(aesthetics = \"fill\") +     labels_sir_count() +     theme_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # or better yet, simplify this using the wrapper function - a single command:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir() }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # get only proportions and no counts:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(datalabels = FALSE) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # add other ggplot2 arguments as you like:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(       width = 0.5,       colour = \"black\",       size = 1,       linetype = 2,       alpha = 0.25     ) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # you can alter the colours with colour names:   example_isolates %>%     select(AMX) %>%     ggplot_sir(colours = c(SI = \"yellow\")) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # but you can also use the built-in colour-blind friendly colours for   # your plots, where \"S\" is green, \"I\" is yellow and \"R\" is red:   data.frame(     x = c(\"Value1\", \"Value2\", \"Value3\"),     y = c(1, 2, 3),     z = c(\"Value4\", \"Value5\", \"Value6\")   ) %>%     ggplot() +     geom_col(aes(x = x, y = y, fill = z)) +     scale_sir_colours(aesthetics = \"fill\",                       Value4 = \"S\", Value5 = \"I\", Value6 = \"R\") }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # resistance of ciprofloxacine per age group   example_isolates %>%     mutate(first_isolate = first_isolate()) %>%     filter(       first_isolate == TRUE,       mo == as.mo(\"Escherichia coli\")     ) %>%     # age_groups() is also a function in this AMR package:     group_by(age_group = age_groups(age)) %>%     select(age_group, CIP) %>%     ggplot_sir(x = \"age_group\") } #> Warning: Removed 6 rows containing missing values or values outside the scale range #> (`geom_col()`). #> Warning: Removed 6 rows containing missing values or values outside the scale range #> (`geom_text()`).  if (require(\"ggplot2\") && require(\"dplyr\")) {   # a shorter version which also adjusts data label colours:   example_isolates %>%     select(AMX, NIT, FOS, TMP, CIP) %>%     ggplot_sir(colours = FALSE) }  if (require(\"ggplot2\") && require(\"dplyr\")) {   # it also supports groups (don't forget to use the group var on `x` or `facet`):   example_isolates %>%     filter(mo_is_gram_negative(), ward != \"Outpatient\") %>%     # select only UTI-specific drugs     select(ward, AMX, NIT, FOS, TMP, CIP) %>%     group_by(ward) %>%     ggplot_sir(       x = \"ward\",       facet = \"antibiotic\",       nrow = 1,       title = \"AMR of Anti-UTI Drugs Per Ward\",       x.title = \"Ward\",       datalabels = FALSE     ) } #> ℹ Using column 'mo' as input for mo_is_gram_negative()  # }"},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":null,"dir":"Reference","previous_headings":"","what":"Guess Antibiotic Column — guess_ab_col","title":"Guess Antibiotic Column — guess_ab_col","text":"tries find column name data set based information antibiotics data set. Also supports WHONET abbreviations.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Guess Antibiotic Column — guess_ab_col","text":"","code":"guess_ab_col(x = NULL, search_string = NULL, verbose = FALSE,   only_sir_columns = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Guess Antibiotic Column — guess_ab_col","text":"x data.frame search_string text search x , checked .ab() value column x verbose logical indicate whether additional info printed only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE)","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Guess Antibiotic Column — guess_ab_col","text":"column name x, NULL result found.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Guess Antibiotic Column — guess_ab_col","text":"can look antibiotic (trade) name abbreviation search x antibiotics data set column containing name code antibiotic.","code":""},{"path":"https://msberends.github.io/AMR/reference/guess_ab_col.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Guess Antibiotic Column — guess_ab_col","text":"","code":"df <- data.frame(   amox = \"S\",   tetr = \"R\" )  guess_ab_col(df, \"amoxicillin\") #> [1] \"amox\" guess_ab_col(df, \"J01AA07\") # ATC code of tetracycline #> [1] \"tetr\"  guess_ab_col(df, \"J01AA07\", verbose = TRUE) #> Auto-guessing columns suitable for analysis #> ... #>  OK. #> ℹ Using column 'amox' as input for AMX (amoxicillin). #> ℹ Using column 'tetr' as input for TCY (tetracycline). #> ℹ Using column 'tetr' as input for J01AA07 (tetracycline). #> [1] \"tetr\" # NOTE: Using column 'tetr' as input for J01AA07 (tetracycline).  # WHONET codes df <- data.frame(   AMP_ND10 = \"R\",   AMC_ED20 = \"S\" ) guess_ab_col(df, \"ampicillin\") #> [1] \"AMP_ND10\" guess_ab_col(df, \"J01CR02\") #> [1] \"AMC_ED20\" guess_ab_col(df, as.ab(\"augmentin\")) #> [1] \"AMC_ED20\""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"Data set containing defined intrinsic resistance EUCAST bug-drug combinations.","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"","code":"intrinsic_resistant"},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"tibble 301 583 observations 2 variables: mo Microorganism ID ab Antibiotic ID","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"data set based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021).","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository. allow machine reading EUCAST CLSI guidelines, almost impossible MS Excel PDF files distributed EUCAST CLSI.","code":""},{"path":"https://msberends.github.io/AMR/reference/intrinsic_resistant.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with Bacterial Intrinsic Resistance — intrinsic_resistant","text":"","code":"intrinsic_resistant #> # A tibble: 301,583 × 2 #>    mo          ab   #>    <mo>        <ab> #>  1 B_GRAMP     ATM  #>  2 B_GRAMP     COL  #>  3 B_GRAMP     NAL  #>  4 B_GRAMP     PLB  #>  5 B_GRAMP     TEM  #>  6 B_ANAER-POS ATM  #>  7 B_ANAER-POS COL  #>  8 B_ANAER-POS NAL  #>  9 B_ANAER-POS PLB  #> 10 B_ANAER-POS TEM  #> # ℹ 301,573 more rows"},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":null,"dir":"Reference","previous_headings":"","what":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"According binomial nomenclature, lowest four taxonomic levels (family, genus, species, subspecies) printed italics. function finds taxonomic names within strings makes italic.","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"","code":"italicise_taxonomy(string, type = c(\"markdown\", \"ansi\", \"html\"))  italicize_taxonomy(string, type = c(\"markdown\", \"ansi\", \"html\"))"},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"string character (vector) type type conversion taxonomic names, either \"markdown\", \"html\" \"ansi\", see Details","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"function finds taxonomic names makes italic based microorganisms data set. taxonomic names can italicised using markdown (default) adding * taxonomic names, <> <\/> using html. using 'ansi', ANSI colours added using \\033[3m \\033[23m taxonomic names. multiple ANSI colours available, conversion occur. function also supports abbreviation genus followed species, \"E. coli\" \"K. pneumoniae ozaenae\".","code":""},{"path":"https://msberends.github.io/AMR/reference/italicise_taxonomy.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Italicise Taxonomic Families, Genera, Species, Subspecies — italicise_taxonomy","text":"","code":"italicise_taxonomy(\"An overview of Staphylococcus aureus isolates\") #> [1] \"An overview of *Staphylococcus aureus* isolates\" italicise_taxonomy(\"An overview of S. aureus isolates\") #> [1] \"An overview of *S. aureus* isolates\"  cat(italicise_taxonomy(\"An overview of S. aureus isolates\", type = \"ansi\")) #> An overview of S. aureus isolates"},{"path":"https://msberends.github.io/AMR/reference/join.html","id":null,"dir":"Reference","previous_headings":"","what":"Join microorganisms to a Data Set — join","title":"Join microorganisms to a Data Set — join","text":"Join data set microorganisms easily existing data set character vector.","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Join microorganisms to a Data Set — join","text":"","code":"inner_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  left_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  right_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  full_join_microorganisms(x, by = NULL, suffix = c(\"2\", \"\"), ...)  semi_join_microorganisms(x, by = NULL, ...)  anti_join_microorganisms(x, by = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Join microorganisms to a Data Set — join","text":"x existing data set join, character vector. case character vector, resulting data.frame contain column 'x' values. variable join - left empty search column class mo (created .mo()) \"mo\" column name exists x, otherwise column name x values exist microorganisms$mo (= \"bacteria_id\"), another column microorganisms (named, like = c(\"bacteria_id\" = \"fullname\")) suffix non-joined duplicate variables x y, suffixes added output disambiguate . character vector length 2. ... ignored, place allow future extensions","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Join microorganisms to a Data Set — join","text":"data.frame","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Join microorganisms to a Data Set — join","text":"Note: opposed join() functions dplyr, character vectors supported default existing columns get suffix \"2\" newly joined columns get suffix. dplyr package installed, join functions used. Otherwise, much slower merge() interaction() functions base R used.","code":""},{"path":"https://msberends.github.io/AMR/reference/join.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Join microorganisms to a Data Set — join","text":"","code":"left_join_microorganisms(as.mo(\"K. pneumoniae\")) #> # A tibble: 1 × 26 #>   mo           fullname   status kingdom phylum class order family genus species #>   <mo>         <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #> 1 B_KLBSL_PNMN Klebsiell… accep… Bacter… Pseud… Gamm… Ente… Enter… Kleb… pneumo… #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list> left_join_microorganisms(\"B_KLBSL_PNMN\") #> # A tibble: 1 × 26 #>   mo           fullname   status kingdom phylum class order family genus species #>   <mo>         <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #> 1 B_KLBSL_PNMN Klebsiell… accep… Bacter… Pseud… Gamm… Ente… Enter… Kleb… pneumo… #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list>  df <- data.frame(   date = seq(     from = as.Date(\"2018-01-01\"),     to = as.Date(\"2018-01-07\"),     by = 1   ),   bacteria = as.mo(c(     \"S. aureus\", \"MRSA\", \"MSSA\", \"STAAUR\",     \"E. coli\", \"E. coli\", \"E. coli\"   )),   stringsAsFactors = FALSE ) colnames(df) #> [1] \"date\"     \"bacteria\"  df_joined <- left_join_microorganisms(df, \"bacteria\") colnames(df_joined) #>  [1] \"date\"                \"bacteria\"            \"fullname\"            #>  [4] \"status\"              \"kingdom\"             \"phylum\"              #>  [7] \"class\"               \"order\"               \"family\"              #> [10] \"genus\"               \"species\"             \"subspecies\"          #> [13] \"rank\"                \"ref\"                 \"oxygen_tolerance\"    #> [16] \"source\"              \"lpsn\"                \"lpsn_parent\"         #> [19] \"lpsn_renamed_to\"     \"mycobank\"            \"mycobank_parent\"     #> [22] \"mycobank_renamed_to\" \"gbif\"                \"gbif_parent\"         #> [25] \"gbif_renamed_to\"     \"prevalence\"          \"snomed\"               # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     left_join_microorganisms() %>%     colnames() } #> Joining, by = \"mo\" #>  [1] \"date\"                \"patient\"             \"age\"                 #>  [4] \"gender\"              \"ward\"                \"mo\"                  #>  [7] \"PEN\"                 \"OXA\"                 \"FLC\"                 #> [10] \"AMX\"                 \"AMC\"                 \"AMP\"                 #> [13] \"TZP\"                 \"CZO\"                 \"FEP\"                 #> [16] \"CXM\"                 \"FOX\"                 \"CTX\"                 #> [19] \"CAZ\"                 \"CRO\"                 \"GEN\"                 #> [22] \"TOB\"                 \"AMK\"                 \"KAN\"                 #> [25] \"TMP\"                 \"SXT\"                 \"NIT\"                 #> [28] \"FOS\"                 \"LNZ\"                 \"CIP\"                 #> [31] \"MFX\"                 \"VAN\"                 \"TEC\"                 #> [34] \"TCY\"                 \"TGC\"                 \"DOX\"                 #> [37] \"ERY\"                 \"CLI\"                 \"AZM\"                 #> [40] \"IPM\"                 \"MEM\"                 \"MTR\"                 #> [43] \"CHL\"                 \"COL\"                 \"MUP\"                 #> [46] \"RIF\"                 \"fullname\"            \"status\"              #> [49] \"kingdom\"             \"phylum\"              \"class\"               #> [52] \"order\"               \"family\"              \"genus\"               #> [55] \"species\"             \"subspecies\"          \"rank\"                #> [58] \"ref\"                 \"oxygen_tolerance\"    \"source\"              #> [61] \"lpsn\"                \"lpsn_parent\"         \"lpsn_renamed_to\"     #> [64] \"mycobank\"            \"mycobank_parent\"     \"mycobank_renamed_to\" #> [67] \"gbif\"                \"gbif_parent\"         \"gbif_renamed_to\"     #> [70] \"prevalence\"          \"snomed\"              # }"},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":null,"dir":"Reference","previous_headings":"","what":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"functions can used determine first weighted isolates considering phenotype isolate selection (see first_isolate()). Using phenotype-based method determine first isolates reliable methods disregard phenotypes.","code":""},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"","code":"key_antimicrobials(x = NULL, col_mo = NULL, universal = c(\"ampicillin\",   \"amoxicillin/clavulanic acid\", \"cefuroxime\", \"piperacillin/tazobactam\",   \"ciprofloxacin\", \"trimethoprim/sulfamethoxazole\"),   gram_negative = c(\"gentamicin\", \"tobramycin\", \"colistin\", \"cefotaxime\",   \"ceftazidime\", \"meropenem\"), gram_positive = c(\"vancomycin\", \"teicoplanin\",   \"tetracycline\", \"erythromycin\", \"oxacillin\", \"rifampin\"),   antifungal = c(\"anidulafungin\", \"caspofungin\", \"fluconazole\", \"miconazole\",   \"nystatin\", \"voriconazole\"), only_sir_columns = FALSE, ...)  all_antimicrobials(x = NULL, only_sir_columns = FALSE, ...)  antimicrobials_equal(y, z, type = c(\"points\", \"keyantimicrobials\"),   ignore_I = TRUE, points_threshold = 2, ...)"},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"x data.frame antibiotics columns, like AMX amox. Can left blank determine automatically col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). universal names broad-spectrum antimicrobial drugs, case-insensitive. Set NULL ignore. See Details default antimicrobial drugs gram_negative names antibiotic drugs Gram-positives, case-insensitive. Set NULL ignore. See Details default antibiotic drugs gram_positive names antibiotic drugs Gram-negatives, case-insensitive. Set NULL ignore. See Details default antibiotic drugs antifungal names antifungal drugs fungi, case-insensitive. Set NULL ignore. See Details default antifungal drugs only_sir_columns logical indicate whether columns must included transformed class sir (see .sir()) beforehand (default FALSE) ... ignored, place allow future extensions y, z character vectors compare type type determine weighed isolates; can \"keyantimicrobials\" \"points\", see Details ignore_I logical indicate whether antibiotic interpretations \"\" ignored type = \"keyantimicrobials\", see Details points_threshold minimum number points require differences antibiogram lead inclusion isolate type = \"points\", see Details","code":""},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"key_antimicrobials() all_antimicrobials() functions context-aware. means x argument can left blank used inside data.frame call, see Examples. function key_antimicrobials() returns character vector 12 antimicrobial results every isolate. function all_antimicrobials() returns character vector antimicrobial drug results every isolate. vectors can compared using antimicrobials_equal(), check two isolates generally antibiogram. Missing invalid values replaced dot (\".\") key_antimicrobials() ignored antimicrobials_equal(). Please see first_isolate() function important functions enable 'phenotype-based' method determination first isolates. default antimicrobial drugs used rows (set universal) : Ampicillin Amoxicillin/clavulanic acid Cefuroxime Ciprofloxacin Piperacillin/tazobactam Trimethoprim/sulfamethoxazole default antimicrobial drugs used Gram-negative bacteria (set gram_negative) : Cefotaxime Ceftazidime Colistin Gentamicin Meropenem Tobramycin default antimicrobial drugs used Gram-positive bacteria (set gram_positive) : Erythromycin Oxacillin Rifampin Teicoplanin Tetracycline Vancomycin default antimicrobial drugs used fungi (set antifungal) : Anidulafungin Caspofungin Fluconazole Miconazole Nystatin Voriconazole","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/key_antimicrobials.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"(Key) Antimicrobials for First Weighted Isolates — key_antimicrobials","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # output of the `key_antimicrobials()` function could be like this: strainA <- \"SSSRR.S.R..S\" strainB <- \"SSSIRSSSRSSS\"  # those strings can be compared with: antimicrobials_equal(strainA, strainB, type = \"keyantimicrobials\") #> [1] TRUE # TRUE, because I is ignored (as well as missing values)  antimicrobials_equal(strainA, strainB, type = \"keyantimicrobials\", ignore_I = FALSE) #> [1] FALSE # FALSE, because I is not ignored and so the 4th [character] differs  # \\donttest{ if (require(\"dplyr\")) {   # set key antibiotics to a new variable   my_patients <- example_isolates %>%     mutate(keyab = key_antimicrobials(antifungal = NULL)) %>% # no need to define `x`     mutate(       # now calculate first isolates       first_regular = first_isolate(col_keyantimicrobials = FALSE),       # and first WEIGHTED isolates       first_weighted = first_isolate(col_keyantimicrobials = \"keyab\")     )    # Check the difference in this data set, 'weighted' results in more isolates:   sum(my_patients$first_regular, na.rm = TRUE)   sum(my_patients$first_weighted, na.rm = TRUE) } #> [1] 1383 # }"},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":null,"dir":"Reference","previous_headings":"","what":"Kurtosis of the Sample — kurtosis","title":"Kurtosis of the Sample — kurtosis","text":"Kurtosis measure \"tailedness\" probability distribution real-valued random variable. normal distribution kurtosis 3 excess kurtosis 0.","code":""},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Kurtosis of the Sample — kurtosis","text":"","code":"kurtosis(x, na.rm = FALSE, excess = FALSE)  # Default S3 method kurtosis(x, na.rm = FALSE, excess = FALSE)  # S3 method for class 'matrix' kurtosis(x, na.rm = FALSE, excess = FALSE)  # S3 method for class 'data.frame' kurtosis(x, na.rm = FALSE, excess = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Kurtosis of the Sample — kurtosis","text":"x vector values, matrix data.frame na.rm logical indicate whether NA values stripped computation proceeds excess logical indicate whether excess kurtosis returned, defined kurtosis minus 3.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/kurtosis.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Kurtosis of the Sample — kurtosis","text":"","code":"kurtosis(rnorm(10000)) #> [1] 2.971316 kurtosis(rnorm(10000), excess = TRUE) #> [1] -0.05342143"},{"path":"https://msberends.github.io/AMR/reference/like.html","id":null,"dir":"Reference","previous_headings":"","what":"Vectorised Pattern Matching with Keyboard Shortcut — like","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"Convenient wrapper around grepl() match pattern: x %like% pattern. always returns logical vector always case-insensitive (use x %like_case% pattern case-sensitive matching). Also, pattern can long x compare items index vectors, can length iterate cases.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"","code":"like(x, pattern, ignore.case = TRUE)  x %like% pattern  x %unlike% pattern  x %like_case% pattern  x %unlike_case% pattern"},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"Idea like function data.table package, although altered explained Details.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"x character vector matches sought, object can coerced .character() character vector. pattern character vector containing regular expressions (character string fixed = TRUE) matched given character vector. Coerced .character() character string possible. ignore.case FALSE, pattern matching case sensitive TRUE, case ignored matching.","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"logical vector","code":""},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"like() %like%/%unlike% functions: case-insensitive (use %like_case%/%unlike_case% case-sensitive matching) Support multiple patterns Check pattern valid regular expression sets fixed = TRUE , greatly improve speed (vectorised pattern) Always use compatibility Perl unless fixed = TRUE, greatly improve speed Using RStudio? %like%/%unlike% functions can also directly inserted code Addins menu can keyboard shortcut like Shift+Ctrl+L Shift+Cmd+L (see menu Tools > Modify Keyboard Shortcuts...). keep pressing shortcut, inserted text iterated %like% -> %unlike% -> %like_case% -> %unlike_case%.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/like.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Vectorised Pattern Matching with Keyboard Shortcut — like","text":"","code":"# data.table has a more limited version of %like%, so unload it: try(detach(\"package:data.table\", unload = TRUE), silent = TRUE)  a <- \"This is a test\" b <- \"TEST\" a %like% b #> [1] TRUE b %like% a #> [1] FALSE  # also supports multiple patterns a <- c(\"Test case\", \"Something different\", \"Yet another thing\") b <- c(\"case\", \"diff\", \"yet\") a %like% b #> [1] TRUE TRUE TRUE a %unlike% b #> [1] FALSE FALSE FALSE  a[1] %like% b #> [1]  TRUE FALSE FALSE a %like% b[1] #> [1]  TRUE FALSE FALSE  # \\donttest{ # get isolates whose name start with 'Entero' (case-insensitive) example_isolates[which(mo_name() %like% \"^entero\"), ] #> ℹ Using column 'mo' as input for mo_name() #> # A tibble: 106 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinic… B_ENTRC_FACM    NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-08-13 F35553     52 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU     B_ENTRC         NA    NA    NA    NA  #> 10 2003-09-05 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #> # ℹ 96 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_name() %like% \"^ent\") } #> ℹ Using column 'mo' as input for mo_name() #> # A tibble: 106 × 46 #>    date       patient   age gender ward    mo            PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>   <mo>          <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinic… B_ENTRC_FACM    NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinic… B_ENTRC_FCLS    NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU     B_ENTRC         NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #>  8 2003-08-13 F35553     52 M      ICU     B_ENTRC_FCLS    NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU     B_ENTRC         NA    NA    NA    NA  #> 10 2003-09-05 F35553     52 M      ICU     B_ENTRBC_CLOC   R     NA    NA    R   #> # ℹ 96 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, … # }"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":null,"dir":"Reference","previous_headings":"","what":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Determine isolates multidrug-resistant organisms (MDRO) according international, national, custom guidelines.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"","code":"mdro(x = NULL, guideline = \"CMI2012\", col_mo = NULL, esbl = NA,   carbapenemase = NA, mecA = NA, mecC = NA, vanA = NA, vanB = NA,   info = interactive(), pct_required_classes = 0.5, combine_SI = TRUE,   verbose = FALSE, only_sir_columns = FALSE, ...)  custom_mdro_guideline(..., as_factor = TRUE)  brmo(x = NULL, only_sir_columns = FALSE, ...)  mrgn(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  mdr_tb(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  mdr_cmi2012(x = NULL, only_sir_columns = FALSE, verbose = FALSE, ...)  eucast_exceptional_phenotypes(x = NULL, only_sir_columns = FALSE,   verbose = FALSE, ...)"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"See supported guidelines list publications used function.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"x data.frame antibiotics columns, like AMX amox. Can left blank automatic determination. guideline specific guideline follow, see sections Supported international / national guidelines Using Custom Guidelines . left empty, publication Magiorakos et al. (see ) followed. col_mo column name names codes microorganisms (see .mo()) - default first column class mo. Values coerced using .mo(). esbl logical values, column name containing logical values, indicating presence ESBL gene (production proteins) carbapenemase logical values, column name containing logical values, indicating presence carbapenemase gene (production proteins) mecA logical values, column name containing logical values, indicating presence mecA gene (production proteins) mecC logical values, column name containing logical values, indicating presence mecC gene (production proteins) vanA logical values, column name containing logical values, indicating presence vanA gene (production proteins) vanB logical values, column name containing logical values, indicating presence vanB gene (production proteins) info logical indicate whether progress printed console - default print interactive sessions pct_required_classes minimal required percentage antimicrobial classes must available per isolate, rounded . example, default guideline, 17 antimicrobial classes must available S. aureus. Setting pct_required_classes argument 0.5 (default) means every S. aureus isolate least 8 different classes must available. lower number available classes return NA isolate. combine_SI logical indicate whether values S must merged one, resistance considered isolates R, . default behaviour mdro() function, follows redefinition EUCAST interpretation (increased exposure) 2019, see section 'Interpretation S, R' . using combine_SI = FALSE, resistance considered isolates R . verbose logical turn Verbose mode (default ). Verbose mode, function return MDRO results, instead returns data set logbook form extensive info isolates MDRO-positive, . only_sir_columns logical indicate whether antibiotic columns must detected transformed class sir (see .sir()) beforehand (default FALSE) ... case custom_mdro_guideline(): set rules, see section Using Custom Guidelines . Otherwise: column name antibiotic, see section Antibiotics . as_factor logical indicate whether returned value ordered factor (TRUE, default), otherwise character vector","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"CMI 2012 paper - function mdr_cmi2012() mdro(): Ordered factor levels Negative < Multi-drug-resistant (MDR) < Extensively drug-resistant (XDR) < Pandrug-resistant (PDR) TB guideline - function mdr_tb() mdro(..., guideline = \"TB\"): Ordered factor levels Negative < Mono-resistant < Poly-resistant < Multi-drug-resistant < Extensively drug-resistant German guideline - function mrgn() mdro(..., guideline = \"MRGN\"): Ordered factor levels Negative < 3MRGN < 4MRGN Everything else, except custom guidelines: Ordered factor levels Negative < Positive, unconfirmed < Positive. value \"Positive, unconfirmed\" means , according guideline, entirely sure isolate multi-drug resistant confirmed additional (e.g. molecular) tests","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"functions context-aware. means x argument can left blank used inside data.frame call, see Examples. pct_required_classes argument, values 1 divided 100. support fractions (0.75 3/4) percentages (75). Note: Every test involves Enterobacteriaceae family, internally performed using newly named order Enterobacterales, since Enterobacteriaceae family taxonomically reclassified Adeolu et al. 2016. , Enterobacteriaceae family Enterobacteriales () order. species old Enterobacteriaceae family still new Enterobacterales (without ) order, divided multiple families. way tests performed now mdro() function makes sure results 2016 2016 identical.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"supported-international-national-guidelines","dir":"Reference","previous_headings":"","what":"Supported International / National Guidelines","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Currently supported guidelines (case-insensitive): guideline = \"CMI2012\" (default) Magiorakos AP, Srinivasan et al. \"Multidrug-resistant, extensively drug-resistant pandrug-resistant bacteria: international expert proposal interim standard definitions acquired resistance.\" Clinical Microbiology Infection (2012) (doi:10.1111/j.1469-0691.2011.03570.x ) guideline = \"EUCAST3.3\" (simply guideline = \"EUCAST\") European international guideline - EUCAST Expert Rules Version 3.3 \"Intrinsic Resistance Unusual Phenotypes\" (link) guideline = \"EUCAST3.2\" European international guideline - EUCAST Expert Rules Version 3.2 \"Intrinsic Resistance Unusual Phenotypes\" (link) guideline = \"EUCAST3.1\" European international guideline - EUCAST Expert Rules Version 3.1 \"Intrinsic Resistance Exceptional Phenotypes Tables\" (link) guideline = \"TB\" international guideline multi-drug resistant tuberculosis - World Health Organization \"Companion handbook guidelines programmatic management drug-resistant tuberculosis\" (link) guideline = \"MRGN\" German national guideline - Mueller et al. (2015) Antimicrobial Resistance Infection Control 4:7; doi:10.1186/s13756-015-0047-6 guideline = \"BRMO\" Dutch national guideline - Samenwerkingverband Richtlijnen Infectiepreventie (SRI) (2024) \"Bijzonder Resistente Micro-Organismen (BRMO)\" (link) Also: guideline = \"BRMO 2017\" former Dutch national guideline - Werkgroep Infectiepreventie (WIP), RIVM, last revision 2017: \"Bijzonder Resistente Micro-Organismen (BRMO)\" Please suggest implement guidelines letting us know: https://github.com/msberends/AMR/issues/new.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"using-custom-guidelines","dir":"Reference","previous_headings":"","what":"Using Custom Guidelines","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"Custom guidelines can set custom_mdro_guideline() function. great importance custom rules determine MDROs hospital, e.g., rules dependent ward, state contact isolation variables data. familiar case_when() function dplyr package, recognise input method set rules. Rules must set using R considers 'formula notation'. rule written tilde (~) consequence rule written tilde:   row/isolate matches first rule, value first ~ (case 'Elderly Type ') set MDRO value. Otherwise, second rule tried . number rules unlimited. can print rules set console overview. Colours help reading console supports colours.   outcome function can used guideline argument mdro() function:   Rules can also combined custom rules using c():   rules set (custom object case) exported shared file location using saveRDS() collaborate multiple users. custom rules set imported using readRDS().","code":"custom <- custom_mdro_guideline(CIP == \"R\" & age > 60 ~ \"Elderly Type A\",                                 ERY == \"R\" & age > 60 ~ \"Elderly Type B\") custom #> A set of custom MDRO rules: #>   1. CIP is \"R\" and age is higher than 60 -> Elderly Type A #>   2. ERY is \"R\" and age is higher than 60 -> Elderly Type B #>   3. Otherwise -> Negative #> #> Unmatched rows will return NA. x <- mdro(example_isolates,           guideline = custom) table(x) #>       Negative Elderly Type A Elderly Type B #>           1070            198            732 x <- mdro(example_isolates,           guideline = c(custom,                         custom_mdro_guideline(ERY == \"R\" & age > 50 ~ \"Elderly Type C\"))) table(x) #>       Negative Elderly Type A Elderly Type B Elderly Type C #>            961            198            732            109"},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"antibiotics","dir":"Reference","previous_headings":"","what":"Antibiotics","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"define antibiotics column names, leave determine automatically guess_ab_col() input text (case-insensitive), use NULL skip column (e.g. TIC = NULL skip ticarcillin). Manually defined non-existing columns skipped warning. following antibiotics eligible functions eucast_rules() mdro(). shown format 'name (antimicrobial ID, ATC code)', sorted alphabetically: Amikacin (AMK, J01GB06), amoxicillin (AMX, J01CA04), amoxicillin/clavulanic acid (AMC, J01CR02), ampicillin (AMP, J01CA01), ampicillin/sulbactam (SAM, J01CR01), apramycin (APR, QA07AA92), arbekacin (ARB, J01GB12), aspoxicillin (APX, J01CA19), azidocillin (AZD, J01CE04), azithromycin (AZM, J01FA10), azlocillin (AZL, J01CA09), aztreonam (ATM, J01DF01), bacampicillin (BAM, J01CA06), bekanamycin (BEK, J01GB13), benzathine benzylpenicillin (BNB, J01CE08), benzathine phenoxymethylpenicillin (BNP, J01CE10), benzylpenicillin (PEN, J01CE01), besifloxacin (BES, S01AE08), biapenem (BIA, J01DH05), carbenicillin (CRB, J01CA03), carindacillin (CRN, J01CA05), carumonam (CAR, J01DF02), cefacetrile (CAC, J01DB10), cefaclor (CEC, J01DC04), cefadroxil (CFR, J01DB05), cefalexin (LEX, J01DB01), cefaloridine (RID, J01DB02), cefalotin (CEP, J01DB03), cefamandole (MAN, J01DC03), cefapirin (HAP, J01DB08), cefatrizine (CTZ, J01DB07), cefazedone (CZD, J01DB06), cefazolin (CZO, J01DB04), cefcapene (CCP, J01DD17), cefdinir (CDR, J01DD15), cefditoren (DIT, J01DD16), cefepime (FEP, J01DE01), cefepime/amikacin (CFA, J01RA06), cefetamet (CAT, J01DD10), cefiderocol (FDC, J01DI04), cefixime (CFM, J01DD08), cefmenoxime (CMX, J01DD05), cefmetazole (CMZ, J01DC09), cefodizime (DIZ, J01DD09), cefonicid (CID, J01DC06), cefoperazone (CFP, J01DD12), cefoperazone/sulbactam (CSL, J01DD62), ceforanide (CND, J01DC11), cefotaxime (CTX, J01DD01), cefotaxime/clavulanic acid (CTC, J01DD51), cefotetan (CTT, J01DC05), cefotiam (CTF, J01DC07), cefovecin (FOV, QJ01DD91), cefoxitin (FOX, J01DC01), cefozopran (ZOP, J01DE03), cefpiramide (CPM, J01DD11), cefpirome (CPO, J01DE02), cefpodoxime (CPD, J01DD13), cefprozil (CPR, J01DC10), cefquinome (CEQ, QG51AA07), cefroxadine (CRD, J01DB11), cefsulodin (CFS, J01DD03), ceftaroline (CPT, J01DI02), ceftazidime (CAZ, J01DD02), ceftazidime/clavulanic acid (CCV, J01DD52), cefteram (CEM, J01DD18), ceftezole (CTL, J01DB12), ceftibuten (CTB, J01DD14), ceftiofur (TIO, QJ01DD90), ceftizoxime (CZX, J01DD07), ceftobiprole medocaril (CFM1, J01DI01), ceftolozane/tazobactam (CZT, J01DI54), ceftriaxone (CRO, J01DD04), ceftriaxone/beta-lactamase inhibitor (CEB, J01DD63), cefuroxime (CXM, J01DC02), cephradine (CED, J01DB09), chloramphenicol (CHL, J01BA01), ciprofloxacin (CIP, J01MA02), clarithromycin (CLR, J01FA09), clindamycin (CLI, J01FF01), clometocillin (CLM, J01CE07), cloxacillin (CLO, J01CF02), colistin (COL, J01XB01), cycloserine (CYC, J04AB01), dalbavancin (DAL, J01XA04), danofloxacin (DAN, QJ01MA92), daptomycin (DAP, J01XX09), delafloxacin (DFX, J01MA23), dibekacin (DKB, J01GB09), dicloxacillin (DIC, J01CF01), difloxacin (DIF, QJ01MA94), dirithromycin (DIR, J01FA13), doripenem (DOR, J01DH04), doxycycline (DOX, J01AA02), enoxacin (ENX, J01MA04), enrofloxacin (ENR, QJ01MA90), epicillin (EPC, J01CA07), ertapenem (ETP, J01DH03), erythromycin (ERY, J01FA01), fleroxacin (FLE, J01MA08), flucloxacillin (FLC, J01CF05), flurithromycin (FLR1, J01FA14), fosfomycin (FOS, J01XX01), framycetin (FRM, D09AA01), fusidic acid (FUS, J01XC01), gamithromycin (GAM, QJ01FA95), garenoxacin (GRN, J01MA19), gatifloxacin (GAT, J01MA16), gemifloxacin (GEM, J01MA15), gentamicin (GEN, J01GB03), grepafloxacin (GRX, J01MA11), hetacillin (HET, J01CA18), imipenem (IPM, J01DH51), imipenem/relebactam (IMR, J01DH56), isepamicin (ISE, J01GB11), josamycin (JOS, J01FA07), kanamycin (KAN, J01GB04), kitasamycin (KIT, QJ01FA93), lascufloxacin (LSC, J01MA25), latamoxef (LTM, J01DD06), levofloxacin (LVX, J01MA12), levonadifloxacin (LND, J01MA24), lincomycin (LIN, J01FF02), linezolid (LNZ, J01XX08), lomefloxacin (LOM, J01MA07), loracarbef (LOR, J01DC08), marbofloxacin (MAR, QJ01MA93), mecillinam (MEC, J01CA11), meropenem (MEM, J01DH02), meropenem/vaborbactam (MEV, J01DH52), metampicillin (MTM, J01CA14), meticillin (MET, J01CF03), mezlocillin (MEZ, J01CA10), micronomicin (MCR, S01AA22), midecamycin (MID, J01FA03), minocycline (MNO, J01AA08), miocamycin (MCM, J01FA11), moxifloxacin (MFX, J01MA14), nadifloxacin (NAD, D10AF05), nafcillin (NAF, J01CF06), nalidixic acid (NAL, J01MB02), neomycin (NEO, J01GB05), netilmicin (NET, J01GB07), nitrofurantoin (NIT, J01XE01), norfloxacin (, J01MA06), novobiocin (NOV, QJ01XX95), ofloxacin (OFX, J01MA01), oleandomycin (OLE, J01FA05), orbifloxacin (ORB, QJ01MA95), oritavancin (ORI, J01XA05), oxacillin (OXA, J01CF04), panipenem (PAN, J01DH55), pazufloxacin (PAZ, J01MA18), pefloxacin (PEF, J01MA03), penamecillin (PNM, J01CE06), pheneticillin (PHE, J01CE05), phenoxymethylpenicillin (PHN, J01CE02), piperacillin (PIP, J01CA12), piperacillin/tazobactam (TZP, J01CR05), pirlimycin (PRL, QJ51FF90), pivampicillin (PVM, J01CA02), pivmecillinam (PME, J01CA08), plazomicin (PLZ, J01GB14), polymyxin B (PLB, J01XB02), pradofloxacin (PRA, QJ01MA97), pristinamycin (PRI, J01FG01), procaine benzylpenicillin (PRB, J01CE09), propicillin (PRP, J01CE03), prulifloxacin (PRU, J01MA17), quinupristin/dalfopristin (QDA, QJ01FG02), ribostamycin (RST, J01GB10), rifampicin (RIF, J04AB02), rokitamycin (ROK, J01FA12), roxithromycin (RXT, J01FA06), rufloxacin (RFL, J01MA10), sarafloxacin (SAR, QJ01MA98), sisomicin (SIS, J01GB08), sitafloxacin (SIT, J01MA21), solithromycin (SOL, J01FA16), sparfloxacin (SPX, J01MA09), spiramycin (SPI, J01FA02), streptoduocin (STR, J01GA02), streptomycin (STR1, J01GA01), sulbenicillin (SBC, J01CA16), sulfadiazine (SDI, J01EC02), sulfadiazine/trimethoprim (SLT1, J01EE02), sulfadimethoxine (SUD, J01ED01), sulfadimidine (SDM, J01EB03), sulfadimidine/trimethoprim (SLT2, J01EE05), sulfafurazole (SLF, J01EB05), sulfaisodimidine (SLF1, J01EB01), sulfalene (SLF2, J01ED02), sulfamazone (SZO, J01ED09), sulfamerazine (SLF3, J01ED07), sulfamerazine/trimethoprim (SLT3, J01EE07), sulfamethizole (SLF4, J01EB02), sulfamethoxazole (SMX, J01EC01), sulfamethoxypyridazine (SLF5, J01ED05), sulfametomidine (SLF6, J01ED03), sulfametoxydiazine (SLF7, J01ED04), sulfametrole/trimethoprim (SLT4, J01EE03), sulfamoxole (SLF8, J01EC03), sulfamoxole/trimethoprim (SLT5, J01EE04), sulfanilamide (SLF9, J01EB06), sulfaperin (SLF10, J01ED06), sulfaphenazole (SLF11, J01ED08), sulfapyridine (SLF12, J01EB04), sulfathiazole (SUT, J01EB07), sulfathiourea (SLF13, J01EB08), sultamicillin (SLT6, J01CR04), talampicillin (TAL, J01CA15), tebipenem (TBP, J01DH06), tedizolid (TZD, J01XX11), teicoplanin (TEC, J01XA02), telavancin (TLV, J01XA03), telithromycin (TLT, J01FA15), temafloxacin (TMX, J01MA05), temocillin (TEM, J01CA17), tetracycline (TCY, J01AA07), ticarcillin (TIC, J01CA13), ticarcillin/clavulanic acid (TCC, J01CR03), tigecycline (TGC, J01AA12), tilbroquinol (TBQ, P01AA05), tildipirosin (TIP, QJ01FA96), tilmicosin (TIL, QJ01FA91), tobramycin (TOB, J01GB01), tosufloxacin (TFX, J01MA22), trimethoprim (TMP, J01EA01), trimethoprim/sulfamethoxazole (SXT, J01EE01), troleandomycin (TRL, J01FA08), trovafloxacin (TVA, J01MA13), tulathromycin (TUL, QJ01FA94), tylosin (TYL, QJ01FA90), tylvalosin (TYL1, QJ01FA92), vancomycin (VAN, J01XA01)","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/mdro.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Determine Multidrug-Resistant Organisms (MDRO) — mdro","text":"","code":"out <- mdro(example_isolates, guideline = \"EUCAST\") str(out) #>  Ord.factor w/ 3 levels \"Negative\"<\"Positive, unconfirmed\"<..: 1 1 1 1 1 1 1 1 1 1 ... table(out) #> out #>              Negative Positive, unconfirmed              Positive  #>                  1994                     0                     6   out <- mdro(example_isolates,   guideline = custom_mdro_guideline(     AMX == \"R\" ~ \"Custom MDRO 1\",     VAN == \"R\" ~ \"Custom MDRO 2\"   ) ) table(out) #> out #>      Negative Custom MDRO 1 Custom MDRO 2  #>           870           804           326   # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     mdro() %>%     table()    # no need to define `x` when used inside dplyr verbs:   example_isolates %>%     mutate(MDRO = mdro()) %>%     pull(MDRO) %>%     table() } #> Warning: in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) #> Warning: There was 1 warning in `mutate()`. #> ℹ In argument: `MDRO = mdro()`. #> Caused by warning: #> ! in mdro(): NA introduced for isolates where the available percentage of #> antimicrobial classes was below 50% (set with pct_required_classes) #> . #>                         Negative       Multi-drug-resistant (MDR)  #>                             1617                              128  #> Extensively drug-resistant (XDR)          Pandrug-resistant (PDR)  #>                                0                                0  # }"},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate the Mean AMR Distance — mean_amr_distance","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"Calculates normalised mean antimicrobial resistance multiple observations, help identify similar isolates without comparing antibiograms hand.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"","code":"mean_amr_distance(x, ...)  # S3 method for class 'sir' mean_amr_distance(x, ..., combine_SI = TRUE)  # S3 method for class 'data.frame' mean_amr_distance(x, ..., combine_SI = TRUE)  amr_distance_from_row(amr_distance, row)"},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"x vector class sir, mic disk, data.frame containing columns classes ... variables select (supports tidyselect language column1:column4 (.mic), can thus also antimicrobial selectors combine_SI logical indicate whether values S, SDD, must merged one, input consists S+vs. R (susceptible vs. resistant) - default TRUE amr_distance outcome mean_amr_distance() row index, row number","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"mean AMR distance effectively Z-score; normalised numeric value compare AMR test results can help identify similar isolates, without comparing antibiograms hand. MIC values (see .mic()) transformed log2() first; distance thus calculated (log2(x) - mean(log2(x))) / sd(log2(x)). SIR values (see .sir()) transformed using \"S\" = 1, \"\" = 2, \"R\" = 3. combine_SI TRUE (default), \"\" considered 1. data sets, mean AMR distance calculated per column, mean per row returned, see Examples. Use amr_distance_from_row() subtract distances distance one row, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"interpretation","dir":"Reference","previous_headings":"","what":"Interpretation","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"Isolates distances less 0.01 difference considered similar. Differences lower 0.025 considered suspicious.","code":""},{"path":"https://msberends.github.io/AMR/reference/mean_amr_distance.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate the Mean AMR Distance — mean_amr_distance","text":"","code":"sir <- random_sir(10) sir #> Class 'sir' #>  [1] S R S S S R R R R R mean_amr_distance(sir) #>  [1] -1.1618950  0.7745967 -1.1618950 -1.1618950 -1.1618950  0.7745967 #>  [7]  0.7745967  0.7745967  0.7745967  0.7745967  mic <- random_mic(10) mic #> Class 'mic' #>  [1] 0.005 0.25  1     2     >=16  8     1     1     0.025 4     mean_amr_distance(mic) #>  [1] -1.9714967 -0.4147020  0.1369756  0.4128144  1.2403309  0.9644921 #>  [7]  0.1369756  0.1369756 -1.3310188  0.6886533 # equal to the Z-score of their log2: (log2(mic) - mean(log2(mic))) / sd(log2(mic)) #>  [1] -1.9714967 -0.4147020  0.1369756  0.4128144  1.2403309  0.9644921 #>  [7]  0.1369756  0.1369756 -1.3310188  0.6886533  disk <- random_disk(10) disk #> Class 'disk' #>  [1] 15 15 50 41 10 21 44 42 10 41 mean_amr_distance(disk) #>  [1] -0.8693350 -0.8693350  1.3196381  0.7567593 -1.1820455 -0.4940825 #>  [7]  0.9443855  0.8193014 -1.1820455  0.7567593  y <- data.frame(   id = LETTERS[1:10],   amox = random_sir(10, ab = \"amox\", mo = \"Escherichia coli\"),   cipr = random_disk(10, ab = \"cipr\", mo = \"Escherichia coli\"),   gent = random_mic(10, ab = \"gent\", mo = \"Escherichia coli\"),   tobr = random_mic(10, ab = \"tobr\", mo = \"Escherichia coli\") ) y #>    id amox cipr   gent tobr #> 1   A    S   26    >=4 0.25 #> 2   B    S   20      1  >=4 #> 3   C    S   28      2 0.25 #> 4   D    S   26 <=0.25  0.5 #> 5   E    S   23      2 0.25 #> 6   F    I   31    0.5    1 #> 7   G    R   31      2  >=4 #> 8   H    R   29      1    2 #> 9   I    R   27 <=0.25  >=4 #> 10  J    I   22    0.5 0.25 mean_amr_distance(y) #> ℹ Calculating mean AMR distance based on columns \"amox\", \"cipr\", \"gent\", #>   and \"tobr\" #>  [1] -0.04047504 -0.25789134 -0.08826444 -0.63237868 -0.42488867  0.02453739 #>  [7]  1.18267023  0.72772870  0.36605356 -0.85709171 y$amr_distance <- mean_amr_distance(y, where(is.mic)) #> Error in .subset(x, j): invalid subscript type 'list' y[order(y$amr_distance), ] #> Error in order(y$amr_distance): argument 1 is not a vector  if (require(\"dplyr\")) {   y %>%     mutate(       amr_distance = mean_amr_distance(y),       check_id_C = amr_distance_from_row(amr_distance, id == \"C\")     ) %>%     arrange(check_id_C) } #> ℹ Calculating mean AMR distance based on columns \"amox\", \"cipr\", \"gent\", #>   and \"tobr\" #>    id amox cipr   gent tobr amr_distance check_id_C #> 1   C    S   28      2 0.25  -0.08826444  0.0000000 #> 2   A    S   26    >=4 0.25  -0.04047504  0.0477894 #> 3   F    I   31    0.5    1   0.02453739  0.1128018 #> 4   B    S   20      1  >=4  -0.25789134  0.1696269 #> 5   E    S   23      2 0.25  -0.42488867  0.3366242 #> 6   I    R   27 <=0.25  >=4   0.36605356  0.4543180 #> 7   D    S   26 <=0.25  0.5  -0.63237868  0.5441142 #> 8   J    I   22    0.5 0.25  -0.85709171  0.7688273 #> 9   H    R   29      1    2   0.72772870  0.8159931 #> 10  G    R   31      2  >=4   1.18267023  1.2709347 if (require(\"dplyr\")) {   # support for groups   example_isolates %>%     filter(mo_genus() == \"Enterococcus\" & mo_species() != \"\") %>%     select(mo, TCY, carbapenems()) %>%     group_by(mo) %>%     mutate(dist = mean_amr_distance(.)) %>%     arrange(mo, dist) } #> ℹ Using column 'mo' as input for mo_genus() #> ℹ Using column 'mo' as input for mo_species() #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> ℹ Calculating mean AMR distance based on columns \"TCY\", \"IPM\", and \"MEM\" #> # A tibble: 63 × 5 #> # Groups:   mo [4] #>    mo           TCY   IPM   MEM     dist #>    <mo>         <sir> <sir> <sir>  <dbl> #>  1 B_ENTRC_AVIM   S     S     NA   0     #>  2 B_ENTRC_AVIM   S     S     NA   0     #>  3 B_ENTRC_CSSL   NA    S     NA  NA     #>  4 B_ENTRC_FACM   S     S     NA  -2.66  #>  5 B_ENTRC_FACM   S     R     R   -0.423 #>  6 B_ENTRC_FACM   S     R     R   -0.423 #>  7 B_ENTRC_FACM   NA    R     R    0.224 #>  8 B_ENTRC_FACM   NA    R     R    0.224 #>  9 B_ENTRC_FACM   NA    R     R    0.224 #> 10 B_ENTRC_FACM   NA    R     R    0.224 #> # ℹ 53 more rows"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"data set containing commonly used codes microorganisms, laboratory systems WHONET. Define set_mo_source(). searched using .mo() consequently mo_* functions.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"","code":"microorganisms.codes"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"tibble 4 971 observations 2 variables: code Commonly used code microorganism. unique identifier. mo ID microorganism microorganisms data set","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.codes.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 4 971 Common Microorganism Codes — microorganisms.codes","text":"","code":"microorganisms.codes #> # A tibble: 4,971 × 2 #>    code  mo                #>    <chr> <mo>              #>  1 1011  B_GRAMP           #>  2 1012  B_GRAMP           #>  3 1013  B_GRAMN           #>  4 1014  B_GRAMN           #>  5 1015  F_YEAST           #>  6 103   B_ESCHR_COLI      #>  7 104   B_SLMNL_ENTR_ENTR #>  8 1100  B_STRPT           #>  9 1101  B_STRPT_VIRI      #> 10 1102  B_STRPT_HAEM      #> # ℹ 4,961 more rows  # 'ECO' or 'eco' is the WHONET code for E. coli: microorganisms.codes[microorganisms.codes$code == \"ECO\", ] #> # A tibble: 1 × 2 #>   code  mo           #>   <chr> <mo>         #> 1 ECO   B_ESCHR_COLI  # and therefore, 'eco' will be understood as E. coli in this package: mo_info(\"eco\") #> $mo #> [1] \"B_ESCHR_COLI\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Escherichia\" #>  #> $species #> [1] \"coli\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] \"facultative anaerobe\" #>  #> $url #> [1] \"https://lpsn.dsmz.de/species/escherichia-coli\" #>  #> $ref #> [1] \"Castellani et al., 1919\" #>  #> $snomed #>  [1] \"1095001000112106\" \"715307006\"        \"737528008\"        \"416989002\"        #>  [5] \"116397003\"        \"414097009\"        \"414098004\"        \"414099007\"        #>  [9] \"414100004\"        \"116395006\"        \"735270003\"        \"116396007\"        #> [13] \"83285000\"         \"116394005\"        \"112283007\"        \"710886005\"        #> [17] \"710887001\"        \"710888006\"        \"710889003\"        \"414132004\"        #> [21] \"721892009\"        \"416812001\"        \"416740004\"        \"417216001\"        #> [25] \"457541006\"        \"710253004\"        \"416530004\"        \"417189006\"        #> [29] \"409800005\"        \"713925008\"        \"444771000124108\"  \"838549008\"        #>  #> $lpsn #> [1] \"776057\" #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] \"11286021\" #>  #> $group_members #> character(0) #>   # works for all AMR functions: mo_is_intrinsic_resistant(\"eco\", ab = \"vancomycin\") #> [1] TRUE"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"data set containing species groups microbiological complexes, used clinical breakpoints table.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"","code":"microorganisms.groups"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"tibble 521 observations 4 variables: mo_group ID species group / microbiological complex mo ID microorganism belonging species group / microbiological complex mo_group_name Name species group / microbiological complex, retrieved mo_name() mo_name Name microorganism belonging species group / microbiological complex, retrieved mo_name()","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.groups.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 521 Microorganisms In Species Groups — microorganisms.groups","text":"","code":"microorganisms.groups #> # A tibble: 521 × 4 #>    mo_group       mo           mo_group_name                   mo_name           #>    <mo>           <mo>         <chr>                           <chr>             #>  1 B_ACNTB_BMNN-C B_ACNTB_BMNN Acinetobacter baumannii complex Acinetobacter ba… #>  2 B_ACNTB_BMNN-C B_ACNTB_CLCC Acinetobacter baumannii complex Acinetobacter ca… #>  3 B_ACNTB_BMNN-C B_ACNTB_LCTC Acinetobacter baumannii complex Acinetobacter di… #>  4 B_ACNTB_BMNN-C B_ACNTB_NSCM Acinetobacter baumannii complex Acinetobacter no… #>  5 B_ACNTB_BMNN-C B_ACNTB_PITT Acinetobacter baumannii complex Acinetobacter pi… #>  6 B_ACNTB_BMNN-C B_ACNTB_SFRT Acinetobacter baumannii complex Acinetobacter se… #>  7 B_BCTRD_FRGL-C B_BCTRD_FRGL Bacteroides fragilis complex    Bacteroides frag… #>  8 B_BCTRD_FRGL-C B_BCTRD_OVTS Bacteroides fragilis complex    Bacteroides ovat… #>  9 B_BCTRD_FRGL-C B_BCTRD_THTT Bacteroides fragilis complex    Bacteroides thet… #> 10 B_BCTRD_FRGL-C B_PHCCL_VLGT Bacteroides fragilis complex    Bacteroides vulg… #> # ℹ 511 more rows  # these are all species in the Bacteroides fragilis group, as per WHONET: microorganisms.groups[microorganisms.groups$mo_group == \"B_BCTRD_FRGL-C\", ] #> # A tibble: 5 × 4 #>   mo_group       mo           mo_group_name                mo_name               #>   <mo>           <mo>         <chr>                        <chr>                 #> 1 B_BCTRD_FRGL-C B_BCTRD_FRGL Bacteroides fragilis complex Bacteroides fragilis  #> 2 B_BCTRD_FRGL-C B_BCTRD_OVTS Bacteroides fragilis complex Bacteroides ovatus    #> 3 B_BCTRD_FRGL-C B_BCTRD_THTT Bacteroides fragilis complex Bacteroides thetaiot… #> 4 B_BCTRD_FRGL-C B_PHCCL_VLGT Bacteroides fragilis complex Bacteroides vulgatus  #> 5 B_BCTRD_FRGL-C B_PRBCT_DSTS Bacteroides fragilis complex Parabacteroides dist…"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"data set containing full microbial taxonomy (last updated: June 24th, 2024) six kingdoms. data set backbone AMR package. MO codes can looked using .mo() microorganism properties can looked using mo_* functions. data set carefully crafted, yet made 100% reproducible public authoritative taxonomic sources (using script), namely: List Prokaryotic names Standing Nomenclature (LPSN) bacteria, MycoBank fungi, Global Biodiversity Information Facility (GBIF) others taxons.","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"","code":"microorganisms"},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"format","dir":"Reference","previous_headings":"","what":"Format","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"tibble 78 678 observations 26 variables: mo ID microorganism used package. unique identifier. fullname Full name, like \"Escherichia coli\". taxonomic ranks genus, species subspecies, 'pasted' text genus, species, subspecies. taxonomic ranks higher genus, name taxon. unique identifier. status  Status taxon, either \"accepted\", \"validly published\", \"synonym\", \"unknown\" kingdom, phylum, class, order, family, genus, species, subspecies Taxonomic rank microorganism. Note fungi, phylum equal taxonomic division. Also, fungi, subkingdom subdivision left since occur bacterial taxonomy. rank Text taxonomic rank microorganism, \"species\" \"genus\" ref Author(s) year related scientific publication. contains first surname year latest authors, e.g. \"Wallis et al. 2006 emend. Smith Jones 2018\" becomes \"Smith et al., 2018\". field directly retrieved source specified column source. Moreover, accents removed comply CRAN allows ASCII characters. oxygen_tolerance  Oxygen tolerance, either \"aerobe\", \"anaerobe\", \"anaerobe/microaerophile\", \"facultative anaerobe\", \"likely facultative anaerobe\", \"microaerophile\". data retrieved BacDive (see Source). Items contain \"likely\" missing BacDive extrapolated species within genus guess oxygen tolerance. Currently 68.3% ~39 000 bacteria data set contain oxygen tolerance. source Either \"GBIF\", \"LPSN\", \"MycoBank\", \"manually added\" (see Source) lpsn Identifier ('Record number') List Prokaryotic names Standing Nomenclature (LPSN). first/highest LPSN identifier keep one identifier per row. example, Acetobacter ascendens LPSN Record number 7864 11011. first available microorganisms data set. unique identifier, though available ~33 000 records. lpsn_parent LPSN identifier parent taxon lpsn_renamed_to LPSN identifier currently valid taxon mycobank Identifier ('MycoBank #') MycoBank. unique identifier, though available ~18 000 records. mycobank_parent MycoBank identifier parent taxon mycobank_renamed_to MycoBank identifier currently valid taxon gbif Identifier ('taxonID') Global Biodiversity Information Facility (GBIF). unique identifier, though available ~49 000 records. gbif_parent GBIF identifier parent taxon gbif_renamed_to GBIF identifier currently valid taxon prevalence Prevalence microorganism based Bartlett et al. (2022, doi:10.1099/mic.0.001269 ), see mo_matching_score() full explanation snomed Systematized Nomenclature Medicine (SNOMED) code microorganism, version July 16th, 2024 (see Source). Use mo_snomed() retrieve quickly, see mo_property().","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Taxonomic entries imported order importance: List Prokaryotic names Standing Nomenclature (LPSN): Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. MycoBank: Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. Global Biodiversity Information Facility (GBIF): GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Furthermore, sources used additional details: BacDive: Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Systematized Nomenclature Medicine - Clinical Terms (SNOMED-CT): Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). Accessed https://www.cdc.gov/phin/php/phinvads/ July 16th, 2024. Grimont et al. (2007). Antigenic Formulae Salmonella Serovars, 9th Edition. Collaborating Centre Reference Research Salmonella (WHOCC-SALM). Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Please note entries based LPSN, MycoBank, GBIF (see ). Since sources incorporate entries based (recent) publications International Journal Systematic Evolutionary Microbiology (IJSEM), can happen year publication sometimes later one might expect. example, Staphylococcus pettenkoferi described first time Diagnostic Microbiology Infectious Disease 2002 (doi:10.1016/s0732-8893(02)00399-1 ), 2007 publication IJSEM followed (doi:10.1099/ijs.0.64381-0 ). Consequently, AMR package returns 2007 mo_year(\"S. pettenkoferi\").","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"included-taxa","dir":"Reference","previous_headings":"","what":"Included Taxa","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Included taxonomic data LPSN, MycoBank, GBIF : ~39 000 (sub)species kingdoms Archaea Bacteria ~28 000 species kingdom Fungi. kingdom Fungi large taxon almost 300,000 different (sub)species, microbial (rather macroscopic, like mushrooms). , fungi fit scope package. relevant fungi covered (species Aspergillus, Candida, Cryptococcus, Histoplasma, Pneumocystis, Saccharomyces Trichophyton). ~8 100 (sub)species kingdom Protozoa ~1 600 (sub)species 39 relevant genera kingdom Animalia (Strongyloides Taenia) ~22 000 previously accepted names included (sub)species (taxonomically renamed) complete taxonomic tree included (sub)species: kingdom subspecies identifier parent taxons year first author related scientific publication","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"manual-additions","dir":"Reference","previous_headings":"","what":"Manual additions","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"convenience, entries added manually: ~1 500 entries Salmonella, city-like serovars groups H 36 species groups (beta-haemolytic Streptococcus groups K, coagulase-negative Staphylococcus (CoNS), Mycobacterium tuberculosis complex, etc.), group compositions stored microorganisms.groups data set 1 entry Blastocystis (B.  hominis), although officially exist (Noel et al. 2005, PMID 15634993) 1 entry Moraxella (M. catarrhalis), formally named Branhamella catarrhalis (Catlin, 1970) though change never accepted within field clinical microbiology 8 'undefined' entries (unknown, unknown Gram-negatives, unknown Gram-positives, unknown yeast, unknown fungus, unknown anaerobic Gram-pos/Gram-neg bacteria) syntax used transform original data cleansed R format, can found .","code":""},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"direct-download","dir":"Reference","previous_headings":"","what":"Direct download","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"Like data sets package, data set publicly available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Data Set with 78 678 Taxonomic Records of Microorganisms — microorganisms","text":"","code":"microorganisms #> # A tibble: 78,678 × 26 #>    mo          fullname   status kingdom phylum class order family genus species #>    <mo>        <chr>      <chr>  <chr>   <chr>  <chr> <chr> <chr>  <chr> <chr>   #>  1 B_GRAMN     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  2 B_GRAMP     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  3 B_ANAER-NEG (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  4 B_ANAER-POS (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  5 B_ANAER     (unknown … unkno… Bacter… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  6 F_FUNGUS    (unknown … unkno… Fungi   (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  7   UNKNOWN   (unknown … unkno… (unkno… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  8 P_PROTOZOAN (unknown … unkno… Protoz… (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #>  9 F_YEAST     (unknown … unkno… Fungi   (unkn… (unk… (unk… \"(unk… (unk… \"(unkn… #> 10 F_AABRN     Aabaarnia  unkno… Fungi   Ascom… Leca… Ostr… \"\"     Aaba… \"\"      #> # ℹ 78,668 more rows #> # ℹ 16 more variables: subspecies <chr>, rank <chr>, ref <chr>, #> #   oxygen_tolerance <chr>, source <chr>, lpsn <chr>, lpsn_parent <chr>, #> #   lpsn_renamed_to <chr>, mycobank <chr>, mycobank_parent <chr>, #> #   mycobank_renamed_to <chr>, gbif <chr>, gbif_parent <chr>, #> #   gbif_renamed_to <chr>, prevalence <dbl>, snomed <list>"},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate the Matching Score for Microorganisms — mo_matching_score","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"algorithm used .mo() mo_* functions determine probable match taxonomic records based user input.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"","code":"mo_matching_score(x, n)"},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"x user input value(s) n full taxonomic name, exists microorganisms$fullname","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"note","dir":"Reference","previous_headings":"","what":"Note","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"algorithm originally developed 2018 subsequently described : Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 . Later, work Bartlett et al. bacterial pathogens infecting humans (2022, doi:10.1099/mic.0.001269 ) incorporated, optimalisations algorithm made.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"ambiguous user input .mo() mo_* functions, returned results chosen based matching score using mo_matching_score(). matching score \\(m\\), calculated : $$m_{(x, n)} = \\frac{l_{n} - 0.5 \\cdot \\min \\begin{cases}l_{n} \\\\ \\textrm{lev}(x, n)\\end{cases}}{l_{n} \\cdot p_{n} \\cdot k_{n}}$$ : \\(x\\) user input; \\(n\\) taxonomic name (genus, species, subspecies); \\(l_n\\) length \\(n\\); \\(lev\\) Levenshtein distance function (counting insertion 1, deletion substitution 2) needed change \\(x\\) \\(n\\); \\(p_n\\) human pathogenic prevalence group \\(n\\), described ; \\(k_n\\) taxonomic kingdom \\(n\\), set Bacteria = 1, Fungi = 1.25, Protozoa = 1.5, Chromista = 1.75, Archaea = 2, others = 3. grouping human pathogenic prevalence \\(p\\) based recent work Bartlett et al. (2022, doi:10.1099/mic.0.001269 ) extensively studied medical-scientific literature categorise bacterial species groups: Established, taxonomic species infected least three persons three references. records prevalence = 1.15 microorganisms data set; Putative, taxonomic species fewer three known cases. records prevalence = 1.25 microorganisms data set. Furthermore, Genera World Health Organization's () Priority Pathogen List prevalence = 1.0 microorganisms data set; genus present established list also prevalence = 1.15 microorganisms data set; genus present putative list prevalence = 1.25 microorganisms data set; species subspecies genus present two aforementioned groups, prevalence = 1.5 microorganisms data set; non-bacterial genus, species subspecies genus present following list, prevalence = 1.25 microorganisms data set: Absidia, Acanthamoeba, Acremonium, Actinomucor, Aedes, Alternaria, Amoeba, Ancylostoma, Angiostrongylus, Anisakis, Anopheles, Apophysomyces, Arthroderma, Aspergillus, Aureobasidium, Basidiobolus, Beauveria, Bipolaris, Blastobotrys, Blastocystis, Blastomyces, Candida, Capillaria, Chaetomium, Chilomastix, Chrysonilia, Chrysosporium, Cladophialophora, Cladosporium, Clavispora, Coccidioides, Cokeromyces, Conidiobolus, Coniochaeta, Contracaecum, Cordylobia, Cryptococcus, Cryptosporidium, Cunninghamella, Curvularia, Cyberlindnera, Debaryozyma, Demodex, Dermatobia, Dientamoeba, Diphyllobothrium, Dirofilaria, Echinostoma, Entamoeba, Enterobius, Epidermophyton, Exidia, Exophiala, Exserohilum, Fasciola, Fonsecaea, Fusarium, Geotrichum, Giardia, Graphium, Haloarcula, Halobacterium, Halococcus, Hansenula, Hendersonula, Heterophyes, Histomonas, Histoplasma, Hortaea, Hymenolepis, Hypomyces, Hysterothylacium, Kloeckera, Kluyveromyces, Kodamaea, Lacazia, Leishmania, Lichtheimia, Lodderomyces, Lomentospora, Madurella, Malassezia, Malbranchea, Metagonimus, Meyerozyma, Microsporidium, Microsporum, Millerozyma, Mortierella, Mucor, Mycocentrospora, Nannizzia, Necator, Nectria, Ochroconis, Oesophagostomum, Oidiodendron, Opisthorchis, Paecilomyces, Paracoccidioides, Pediculus, Penicillium, Phaeoacremonium, Phaeomoniella, Phialophora, Phlebotomus, Phoma, Pichia, Piedraia, Pithomyces, Pityrosporum, Pneumocystis, Pseudallescheria, Pseudoscopulariopsis, Pseudoterranova, Pulex, Purpureocillium, Quambalaria, Rhinocladiella, Rhizomucor, Rhizopus, Rhodotorula, Saccharomyces, Saksenaea, Saprochaete, Sarcoptes, Scedosporium, Schistosoma, Schizosaccharomyces, Scolecobasidium, Scopulariopsis, Scytalidium, Spirometra, Sporobolomyces, Sporopachydermia, Sporothrix, Sporotrichum, Stachybotrys, Strongyloides, Syncephalastrum, Syngamus, Taenia, Talaromyces, Teleomorph, Toxocara, Trichinella, Trichobilharzia, Trichoderma, Trichomonas, Trichophyton, Trichosporon, Trichostrongylus, Trichuris, Tritirachium, Trombicula, Trypanosoma, Tunga, Ulocladium, Ustilago, Verticillium, Wallemia, Wangiella, Wickerhamomyces, Wuchereria, Yarrowia, Zygosaccharomyces; records prevalence = 2.0 microorganisms data set. calculating matching score, characters \\(x\\) \\(n\\) ignored -Z, -z, 0-9, spaces parentheses. matches sorted descending matching score user input values, top match returned. lead effect e.g., \"E. coli\" return microbial ID Escherichia coli (\\(m = 0.688\\), highly prevalent microorganism found humans) Entamoeba coli (\\(m = 0.381\\), less prevalent microorganism humans), although latter alphabetically come first.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_matching_score.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate the Matching Score for Microorganisms — mo_matching_score","text":"","code":"mo_reset_session() #> ℹ Reset 17 previously matched input values.  as.mo(\"E. coli\") #> Class 'mo' #> [1] B_ESCHR_COLI mo_uncertainties() #> Matching scores are based on the resemblance between the input and the full #> taxonomic name, and the pathogenicity in humans. See ?mo_matching_score. #> Colour keys:  0.000-0.549  0.550-0.649  0.650-0.749  0.750-1.000  #>  #> -------------------------------------------------------------------------------- #> \"E. coli\" -> Escherichia coli (B_ESCHR_COLI, 0.688) #> Also matched: Enterococcus crotali (0.650), Escherichia coli coli #>               (0.643), Escherichia coli expressing (0.611), Enterobacter cowanii #>               (0.600), Enterococcus columbae (0.595), Enterococcus camelliae (0.591), #>               Enterococcus casseliflavus (0.577), Enterobacter cloacae cloacae #>               (0.571), Enterobacter cloacae complex (0.571), and Enterobacter cloacae #>               dissolvens (0.565) #>  #> Only the first 10 other matches of each record are shown. Run #> print(mo_uncertainties(), n = ...) to view more entries, or save #> mo_uncertainties() to an object.  mo_matching_score(   x = \"E. coli\",   n = c(\"Escherichia coli\", \"Entamoeba coli\") ) #> [1] 0.6875000 0.3809524"},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":null,"dir":"Reference","previous_headings":"","what":"Get Properties of a Microorganism — mo_property","title":"Get Properties of a Microorganism — mo_property","text":"Use functions return specific property microorganism based latest accepted taxonomy. input values evaluated internally .mo(), makes possible use microbial abbreviations, codes names input. See Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get Properties of a Microorganism — mo_property","text":"","code":"mo_name(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_fullname(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_shortname(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_subspecies(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_species(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_genus(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_family(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_order(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_class(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_phylum(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_kingdom(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_domain(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_type(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_status(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_pathogenicity(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_gramstain(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_gram_negative(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_gram_positive(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_yeast(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_intrinsic_resistant(x, ab, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_oxygen_tolerance(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_is_anaerobic(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_snomed(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_ref(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_authors(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_year(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_lpsn(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_mycobank(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_gbif(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_rank(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_taxonomy(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_synonyms(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_current(x, language = get_AMR_locale(), ...)  mo_group_members(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_info(x, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_url(x, open = FALSE, language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)  mo_property(x, property = \"fullname\", language = get_AMR_locale(),   keep_synonyms = getOption(\"AMR_keep_synonyms\", FALSE), ...)"},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get Properties of a Microorganism — mo_property","text":"x character (vector) can coerced valid microorganism code .mo(). Can left blank auto-guessing column containing microorganism codes used data set, see Examples. language language translate text like \"growth\", defaults system language (see get_AMR_locale()) keep_synonyms logical indicate old, previously valid taxonomic names must preserved corrected currently accepted names. default FALSE, return note old taxonomic names processed. default can set package option AMR_keep_synonyms, .e. options(AMR_keep_synonyms = TRUE) options(AMR_keep_synonyms = FALSE). ... arguments passed .mo(), 'minimum_matching_score', 'ignore_pattern', 'remove_from_input' ab (vector ) text can coerced valid antibiotic drug code .ab() open browse URL using browseURL() property one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\", must \"shortname\"","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get Properties of a Microorganism — mo_property","text":"integer case mo_year() ordered factor case mo_pathogenicity() list case mo_taxonomy(), mo_synonyms(), mo_snomed(), mo_info() logical case mo_is_anaerobic(), mo_is_gram_negative(), mo_is_gram_positive(), mo_is_intrinsic_resistant(), mo_is_yeast() named character case mo_synonyms() mo_url() character cases","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get Properties of a Microorganism — mo_property","text":"functions , default, keep old taxonomic properties, synonyms automatically replaced current taxonomy. Take example Enterobacter aerogenes, initially named 1960 renamed Klebsiella aerogenes 2017: mo_genus(\"Enterobacter aerogenes\") return \"Klebsiella\" (note renaming) mo_genus(\"Enterobacter aerogenes\", keep_synonyms = TRUE) return \"Enterobacter\" (-per-session warning name outdated) mo_ref(\"Enterobacter aerogenes\") return \"Tindall et al., 2017\" (note renaming) mo_ref(\"Enterobacter aerogenes\", keep_synonyms = TRUE) return \"Hormaeche et al., 1960\" (-per-session warning name outdated) short name (mo_shortname()) returns first character genus full species, \"E. coli\", species subspecies. Exceptions abbreviations staphylococci (\"CoNS\", Coagulase-Negative Staphylococci) beta-haemolytic streptococci (\"GBS\", Group B Streptococci). Please bear mind e.g. E. coli mean Escherichia coli (kingdom Bacteria) well Entamoeba coli (kingdom Protozoa). Returning full name done using .mo() internally, giving priority bacteria human pathogens, .e. \"E. coli\" considered Escherichia coli. result, mo_fullname(mo_shortname(\"Entamoeba coli\")) returns \"Escherichia coli\". Since top-level taxonomy sometimes referred 'kingdom' sometimes 'domain', functions mo_kingdom() mo_domain() return exact results. Determination human pathogenicity (mo_pathogenicity()) strongly based Bartlett et al. (2022, doi:10.1099/mic.0.001269 ). function returns factor levels Pathogenic, Potentially pathogenic, Non-pathogenic, Unknown. Determination Gram stain (mo_gramstain()) based taxonomic kingdom phylum. Originally, Cavalier-Smith defined -called subkingdoms Negibacteria Posibacteria (2002, PMID 11837318), considered phyla Posibacteria: Actinobacteria, Chloroflexi, Firmicutes, Tenericutes. phyla later renamed Actinomycetota, Chloroflexota, Bacillota, Mycoplasmatota (2021, PMID 34694987). Bacteria phyla considered Gram-positive AMR package, except members class Negativicutes (within phylum Bacillota) Gram-negative. bacteria considered Gram-negative. Species outside kingdom Bacteria return value NA. Functions mo_is_gram_negative() mo_is_gram_positive() always return TRUE FALSE (NA input NA MO code UNKNOWN), thus always return FALSE species outside taxonomic kingdom Bacteria. Determination yeasts (mo_is_yeast()) based taxonomic kingdom class. Budding yeasts yeasts reproduce asexually process called budding, new cell develops small protrusion parent cell. Taxonomically, members phylum Ascomycota, class Saccharomycetes (also called Hemiascomycetes) Pichiomycetes. True yeasts quite specifically refers yeasts underlying order Saccharomycetales (Saccharomyces cerevisiae). Thus, microorganisms member taxonomic class Saccharomycetes Pichiomycetes, function return TRUE. returns FALSE otherwise (NA input NA MO code UNKNOWN). Determination intrinsic resistance (mo_is_intrinsic_resistant()) based intrinsic_resistant data set, based 'EUCAST Expert Rules' 'EUCAST Intrinsic Resistance Unusual Phenotypes' v3.3 (2021). mo_is_intrinsic_resistant() function can vectorised argument x (input microorganisms) ab (input antibiotics). Determination bacterial oxygen tolerance (mo_oxygen_tolerance()) based BacDive, see Source. function mo_is_anaerobic() returns TRUE oxygen tolerance \"anaerobe\", indicting obligate anaerobic species genus. always returns FALSE species outside taxonomic kingdom Bacteria. function mo_url() return direct URL online database entry, also shows scientific reference concerned species. MycoBank URL used fungi wherever available , LPSN URL bacteria wherever available, GBIF link otherwise. SNOMED codes (mo_snomed()) last updated July 16th, 2024. See Source microorganisms data set info. Old taxonomic names (-called 'synonyms') can retrieved mo_synonyms() (scientific reference name), current taxonomic name can retrieved mo_current(). functions return full names. output translated possible.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"matching-score-for-microorganisms","dir":"Reference","previous_headings":"","what":"Matching Score for Microorganisms","title":"Get Properties of a Microorganism — mo_property","text":"function uses .mo() internally, uses advanced algorithm translate arbitrary user input valid taxonomy using -called matching score. can read public algorithm MO matching score page.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Get Properties of a Microorganism — mo_property","text":"Berends MS et al. (2022). AMR: R Package Working Antimicrobial Resistance Data. Journal Statistical Software, 104(3), 1-31; doi:10.18637/jss.v104.i03 Parte, AC et al. (2020). List Prokaryotic names Standing Nomenclature (LPSN) moves DSMZ. International Journal Systematic Evolutionary Microbiology, 70, 5607-5612; doi:10.1099/ijsem.0.004332 . Accessed https://lpsn.dsmz.de June 24th, 2024. Vincent, R et al (2013). MycoBank gearing new horizons. IMA Fungus, 4(2), 371-9; doi:10.5598/imafungus.2013.04.02.16 . Accessed https://www.mycobank.org June 24th, 2024. GBIF Secretariat (2023). GBIF Backbone Taxonomy. Checklist dataset doi:10.15468/39omei . Accessed https://www.gbif.org June 24th, 2024. Reimer, LC et al. (2022). BacDive 2022: knowledge base standardized bacterial archaeal data. Nucleic Acids Res., 50(D1):D741-D74; doi:10.1093/nar/gkab961 . Accessed https://bacdive.dsmz.de July 16th, 2024. Public Health Information Network Vocabulary Access Distribution System (PHIN VADS). US Edition SNOMED CT 1 September 2020. Value Set Name 'Microorganism', OID 2.16.840.1.114222.4.11.1009 (v12). URL: https://www.cdc.gov/phin/php/phinvads/ Bartlett et al. (2022). comprehensive list bacterial pathogens infecting humans Microbiology 168:001269; doi:10.1099/mic.0.001269","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"reference-data-publicly-available","dir":"Reference","previous_headings":"","what":"Reference Data Publicly Available","title":"Get Properties of a Microorganism — mo_property","text":"data sets AMR package (microorganisms, antibiotics, SIR interpretation, EUCAST rules, etc.) publicly freely available download following formats: R, MS Excel, Apache Feather, Apache Parquet, SPSS, Stata. also provide tab-separated plain text files machine-readable suitable input software program, laboratory information systems. Please visit website download links. actual files course available GitHub repository.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/mo_property.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get Properties of a Microorganism — mo_property","text":"","code":"# taxonomic tree -----------------------------------------------------------  mo_kingdom(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_phylum(\"Klebsiella pneumoniae\") #> [1] \"Pseudomonadota\" mo_class(\"Klebsiella pneumoniae\") #> [1] \"Gammaproteobacteria\" mo_order(\"Klebsiella pneumoniae\") #> [1] \"Enterobacterales\" mo_family(\"Klebsiella pneumoniae\") #> [1] \"Enterobacteriaceae\" mo_genus(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella\" mo_species(\"Klebsiella pneumoniae\") #> [1] \"pneumoniae\" mo_subspecies(\"Klebsiella pneumoniae\") #> [1] \"\"   # full names and short names -----------------------------------------------  mo_name(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella pneumoniae\" mo_fullname(\"Klebsiella pneumoniae\") #> [1] \"Klebsiella pneumoniae\" mo_shortname(\"Klebsiella pneumoniae\") #> [1] \"K. pneumoniae\"   # other properties ---------------------------------------------------------  mo_pathogenicity(\"Klebsiella pneumoniae\") #> [1] Pathogenic #> Levels: Pathogenic < Potentially pathogenic < Non-pathogenic < Unknown mo_gramstain(\"Klebsiella pneumoniae\") #> [1] \"Gram-negative\" mo_snomed(\"Klebsiella pneumoniae\") #> [[1]] #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\"        #>  mo_type(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_rank(\"Klebsiella pneumoniae\") #> [1] \"species\" mo_url(\"Klebsiella pneumoniae\") #>                                Klebsiella pneumoniae  #> \"https://lpsn.dsmz.de/species/klebsiella-pneumoniae\"  mo_is_yeast(c(\"Candida\", \"Trichophyton\", \"Klebsiella\")) #> [1]  TRUE FALSE FALSE  mo_group_members(c(\"Streptococcus group A\",                    \"Streptococcus group C\",                    \"Streptococcus group G\",                    \"Streptococcus group L\")) #> $`Streptococcus Group A` #> [1] \"Streptococcus pyogenes\" #>  #> $`Streptococcus Group C` #> [1] \"Streptococcus dysgalactiae\"              #> [2] \"Streptococcus dysgalactiae dysgalactiae\" #> [3] \"Streptococcus dysgalactiae equisimilis\"  #> [4] \"Streptococcus equi\"                      #> [5] \"Streptococcus equi equi\"                 #> [6] \"Streptococcus equi ruminatorum\"          #> [7] \"Streptococcus equi zooepidemicus\"        #>  #> $`Streptococcus Group G` #> [1] \"Streptococcus canis\"                     #> [2] \"Streptococcus dysgalactiae\"              #> [3] \"Streptococcus dysgalactiae dysgalactiae\" #> [4] \"Streptococcus dysgalactiae equisimilis\"  #>  #> $`Streptococcus Group L` #> [1] \"Streptococcus dysgalactiae\"              #> [2] \"Streptococcus dysgalactiae dysgalactiae\" #> [3] \"Streptococcus dysgalactiae equisimilis\"  #>    # scientific reference -----------------------------------------------------  mo_ref(\"Klebsiella aerogenes\") #> [1] \"Tindall et al., 2017\" mo_authors(\"Klebsiella aerogenes\") #> [1] \"Tindall et al.\" mo_year(\"Klebsiella aerogenes\") #> [1] 2017 mo_synonyms(\"Klebsiella aerogenes\") #>   Hormaeche et al., 1960     Bascomb et al., 1971  #> \"Enterobacter aerogenes\"     \"Klebsiella mobilis\"  mo_lpsn(\"Klebsiella aerogenes\") #> [1] \"777146\" mo_gbif(\"Klebsiella aerogenes\") #> [1] \"9281703\" mo_mycobank(\"Candida albicans\") #> [1] \"256187\" mo_mycobank(\"Candida krusei\") #> [1] \"337013\" mo_mycobank(\"Candida krusei\", keep_synonyms = TRUE) #> Warning: Function as.mo() returned one old taxonomic name. Use as.mo(..., #> keep_synonyms = FALSE) to clean the input to currently accepted taxonomic #> names, or set the R option AMR_keep_synonyms to FALSE. This warning #> will be shown once per session. #> [1] \"268707\"   # abbreviations known in the field -----------------------------------------  mo_genus(\"MRSA\") #> [1] \"Staphylococcus\" mo_species(\"MRSA\") #> [1] \"aureus\" mo_shortname(\"VISA\") #> [1] \"S. aureus\" mo_gramstain(\"VISA\") #> [1] \"Gram-positive\"  mo_genus(\"EHEC\") #> [1] \"Escherichia\" mo_species(\"EIEC\") #> [1] \"coli\" mo_name(\"UPEC\") #> [1] \"Escherichia coli\"   # known subspecies ---------------------------------------------------------  mo_fullname(\"K. pneu rh\") #> [1] \"Klebsiella pneumoniae rhinoscleromatis\" mo_shortname(\"K. pneu rh\") #> [1] \"K. pneumoniae\"  # \\donttest{ # Becker classification, see ?as.mo ----------------------------------------  mo_fullname(\"Staph epidermidis\") #> [1] \"Staphylococcus epidermidis\" mo_fullname(\"Staph epidermidis\", Becker = TRUE) #> [1] \"Coagulase-negative Staphylococcus (CoNS)\" mo_shortname(\"Staph epidermidis\") #> [1] \"S. epidermidis\" mo_shortname(\"Staph epidermidis\", Becker = TRUE) #> [1] \"CoNS\"   # Lancefield classification, see ?as.mo ------------------------------------  mo_fullname(\"Strep agalactiae\") #> [1] \"Streptococcus agalactiae\" mo_fullname(\"Strep agalactiae\", Lancefield = TRUE) #> [1] \"Streptococcus Group B\" mo_shortname(\"Strep agalactiae\") #> [1] \"S. agalactiae\" mo_shortname(\"Strep agalactiae\", Lancefield = TRUE) #> [1] \"GBS\"   # language support  --------------------------------------------------------  mo_gramstain(\"Klebsiella pneumoniae\", language = \"de\") # German #> [1] \"Gramnegativ\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"nl\") # Dutch #> [1] \"Gram-negatief\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"es\") # Spanish #> [1] \"Gram negativo\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"el\") # Greek #> [1] \"Αρνητικό κατά Gram\" mo_gramstain(\"Klebsiella pneumoniae\", language = \"uk\") # Ukrainian #> [1] \"Грамнегативні\"  # mo_type is equal to mo_kingdom, but mo_kingdom will remain untranslated mo_kingdom(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_type(\"Klebsiella pneumoniae\") #> [1] \"Bacteria\" mo_kingdom(\"Klebsiella pneumoniae\", language = \"zh\") # Chinese, no effect #> [1] \"Bacteria\" mo_type(\"Klebsiella pneumoniae\", language = \"zh\") # Chinese, translated #> [1] \"细菌\"  mo_fullname(\"S. pyogenes\", Lancefield = TRUE, language = \"de\") #> [1] \"Streptococcus Gruppe A\" mo_fullname(\"S. pyogenes\", Lancefield = TRUE, language = \"uk\") #> [1] \"Streptococcus Група A\"   # other --------------------------------------------------------------------  # gram stains and intrinsic resistance can be used as a filter in dplyr verbs if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_is_gram_positive()) %>%     count(mo_genus(), sort = TRUE) } #> ℹ Using column 'mo' as input for mo_is_gram_positive() #> ℹ Using column 'mo' as input for mo_genus() #> # A tibble: 18 × 2 #>    `mo_genus()`        n #>    <chr>           <int> #>  1 Staphylococcus    840 #>  2 Streptococcus     275 #>  3 Enterococcus       83 #>  4 Corynebacterium    17 #>  5 Micrococcus         6 #>  6 Gemella             3 #>  7 Aerococcus          2 #>  8 Cutibacterium       1 #>  9 Dermabacter         1 #> 10 Fusibacter          1 #> 11 Globicatella        1 #> 12 Granulicatella      1 #> 13 Lactobacillus       1 #> 14 Leuconostoc         1 #> 15 Listeria            1 #> 16 Paenibacillus       1 #> 17 Rothia              1 #> 18 Schaalia            1 if (require(\"dplyr\")) {   example_isolates %>%     filter(mo_is_intrinsic_resistant(ab = \"vanco\")) %>%     count(mo_genus(), sort = TRUE) } #> ℹ Using column 'mo' as input for mo_is_intrinsic_resistant() #> ℹ Using column 'mo' as input for mo_genus() #> # A tibble: 20 × 2 #>    `mo_genus()`         n #>    <chr>            <int> #>  1 Escherichia        467 #>  2 Klebsiella          77 #>  3 Proteus             39 #>  4 Pseudomonas         30 #>  5 Serratia            25 #>  6 Enterobacter        23 #>  7 Citrobacter         11 #>  8 Haemophilus          8 #>  9 Acinetobacter        6 #> 10 Morganella           6 #> 11 Pantoea              4 #> 12 Salmonella           3 #> 13 Neisseria            2 #> 14 Stenotrophomonas     2 #> 15 Campylobacter        1 #> 16 Enterococcus         1 #> 17 Hafnia               1 #> 18 Lactobacillus        1 #> 19 Leuconostoc          1 #> 20 Pseudescherichia     1  # get a list with the complete taxonomy (from kingdom to subspecies) mo_taxonomy(\"Klebsiella pneumoniae\") #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Klebsiella\" #>  #> $species #> [1] \"pneumoniae\" #>  #> $subspecies #> [1] \"\" #>   # get a list with the taxonomy, the authors, Gram-stain, # SNOMED codes, and URL to the online database mo_info(\"Klebsiella pneumoniae\") #> $mo #> [1] \"B_KLBSL_PNMN\" #>  #> $rank #> [1] \"species\" #>  #> $kingdom #> [1] \"Bacteria\" #>  #> $phylum #> [1] \"Pseudomonadota\" #>  #> $class #> [1] \"Gammaproteobacteria\" #>  #> $order #> [1] \"Enterobacterales\" #>  #> $family #> [1] \"Enterobacteriaceae\" #>  #> $genus #> [1] \"Klebsiella\" #>  #> $species #> [1] \"pneumoniae\" #>  #> $subspecies #> [1] \"\" #>  #> $status #> [1] \"accepted\" #>  #> $synonyms #> NULL #>  #> $gramstain #> [1] \"Gram-negative\" #>  #> $oxygen_tolerance #> [1] \"facultative anaerobe\" #>  #> $url #> [1] \"https://lpsn.dsmz.de/species/klebsiella-pneumoniae\" #>  #> $ref #> [1] \"Trevisan, 1887\" #>  #> $snomed #> [1] \"1098101000112102\" \"446870005\"        \"1098201000112108\" \"409801009\"        #> [5] \"56415008\"         \"714315002\"        \"713926009\"        #>  #> $lpsn #> [1] \"777151\" #>  #> $mycobank #> [1] NA #>  #> $gbif #> [1] \"3221874\" #>  #> $group_members #> character(0) #>  # }"},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":null,"dir":"Reference","previous_headings":"","what":"User-Defined Reference Data Set for Microorganisms — mo_source","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"functions can used predefine reference used .mo() consequently mo_* functions (mo_genus() mo_gramstain()). fastest way organisation (analysis) specific codes picked translated package, since bother setting .","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"","code":"set_mo_source(path, destination = getOption(\"AMR_mo_source\",   \"~/mo_source.rds\"))  get_mo_source(destination = getOption(\"AMR_mo_source\", \"~/mo_source.rds\"))"},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"path location reference file, can text file (comma-, tab- pipe-separated) Excel file (see Details). Can also \"\", NULL FALSE delete reference file. destination destination compressed data file - default user's home directory.","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"reference file can text file separated commas (CSV) tabs pipes, Excel file (either 'xls' 'xlsx' format) R object file (extension '.rds'). use Excel file, need readxl package installed. set_mo_source() check file validity: must data.frame, must column named \"mo\" contains values microorganisms$mo microorganisms$fullname must reference column defined values. tests pass, set_mo_source() read file R ask export \"~/mo_source.rds\". CRAN policy disallows packages write file system, although 'exceptions may allowed interactive sessions package obtains confirmation user'. reason, function works interactive sessions user can specifically confirm allow file created. destination file can set destination argument defaults user's home directory. can also set package option AMR_mo_source, e.g. options(AMR_mo_source = \"/location/file.rds\"). created compressed data file \"mo_source.rds\" used default MO determination (function .mo() consequently mo_* functions like mo_genus() mo_gramstain()). location timestamp original file saved attribute compressed data file. function get_mo_source() return data set reading \"mo_source.rds\" readRDS(). original file changed (checking location timestamp original file), call set_mo_source() update data file automatically used interactive session. Reading Excel file (.xlsx) one row size 8-9 kB. compressed file created set_mo_source() size 0.1 kB can read get_mo_source() couple microseconds (millionths second).","code":""},{"path":"https://msberends.github.io/AMR/reference/mo_source.html","id":"how-to-setup","dir":"Reference","previous_headings":"","what":"How to Setup","title":"User-Defined Reference Data Set for Microorganisms — mo_source","text":"Imagine data sheet Excel file. first column contains organisation specific codes, second column contains valid taxonomic names:   save \"/Users//Documents/ourcodes.xlsx\". Now set source:   now created file \"~/mo_source.rds\" contents Excel file. first column foreign values 'mo' column kept creating RDS file. now can use functions:   edit Excel file , say, adding row 4 like :   ...new usage MO function package update data file:   delete reference data file, just use \"\", NULL FALSE input set_mo_source():   original file (previous case Excel file) moved deleted, mo_source.rds file removed upon next use .mo() mo_* function.","code":"|         A          |            B          | --|--------------------|-----------------------| 1 | Organisation XYZ   | mo                    | 2 | lab_mo_ecoli       | Escherichia coli      | 3 | lab_mo_kpneumoniae | Klebsiella pneumoniae | 4 |                    |                       | set_mo_source(\"/Users/me/Documents/ourcodes.xlsx\") #> NOTE: Created mo_source file '/Users/me/mo_source.rds' (0.3 kB) from #>       '/Users/me/Documents/ourcodes.xlsx' (9 kB), columns #>       \"Organisation XYZ\" and \"mo\" as.mo(\"lab_mo_ecoli\") #> Class 'mo' #> [1] B_ESCHR_COLI  mo_genus(\"lab_mo_kpneumoniae\") #> [1] \"Klebsiella\"  # other input values still work too as.mo(c(\"Escherichia coli\", \"E. coli\", \"lab_mo_ecoli\")) #> NOTE: Translation to one microorganism was guessed with uncertainty. #>       Use mo_uncertainties() to review it. #> Class 'mo' #> [1] B_ESCHR_COLI B_ESCHR_COLI B_ESCHR_COLI |         A          |            B          | --|--------------------|-----------------------| 1 | Organisation XYZ   | mo                    | 2 | lab_mo_ecoli       | Escherichia coli      | 3 | lab_mo_kpneumoniae | Klebsiella pneumoniae | 4 | lab_Staph_aureus   | Staphylococcus aureus | 5 |                    |                       | as.mo(\"lab_mo_ecoli\") #> NOTE: Updated mo_source file '/Users/me/mo_source.rds' (0.3 kB) from #>       '/Users/me/Documents/ourcodes.xlsx' (9 kB), columns #>        \"Organisation XYZ\" and \"mo\" #> Class 'mo' #> [1] B_ESCHR_COLI  mo_genus(\"lab_Staph_aureus\") #> [1] \"Staphylococcus\" set_mo_source(NULL) #> Removed mo_source file '/Users/me/mo_source.rds'"},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":null,"dir":"Reference","previous_headings":"","what":"Principal Component Analysis (for AMR) — pca","title":"Principal Component Analysis (for AMR) — pca","text":"Performs principal component analysis (PCA) based data set automatic determination afterwards plotting groups labels, automatic filtering suitable (.e. non-empty numeric) variables.","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Principal Component Analysis (for AMR) — pca","text":"","code":"pca(x, ..., retx = TRUE, center = TRUE, scale. = TRUE, tol = NULL,   rank. = NULL)"},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Principal Component Analysis (for AMR) — pca","text":"x data.frame containing numeric columns ... columns x selected PCA, can unquoted since supports quasiquotation. retx logical value indicating whether rotated variables     returned. center logical value indicating whether variables     shifted zero centered. Alternately, vector     length equal number columns x can supplied.     value passed scale. scale. logical value indicating whether variables     scaled unit variance analysis takes     place.  default FALSE consistency S,     general scaling advisable.  Alternatively, vector length     equal number columns x can supplied.      value passed scale. tol value indicating magnitude components     omitted. (Components omitted     standard deviations less equal tol times     standard deviation first component.)  default null     setting, components omitted (unless rank. specified     less min(dim(x)).).  settings tol     tol = 0 tol = sqrt(.Machine$double.eps),     omit essentially constant components. rank. optionally, number specifying maximal rank, .e.,     maximal number principal components used.  Can set     alternative addition tol, useful notably     desired rank considerably smaller dimensions matrix.","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Principal Component Analysis (for AMR) — pca","text":"object classes pca prcomp","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Principal Component Analysis (for AMR) — pca","text":"pca() function takes data.frame input performs actual PCA R function prcomp(). result pca() function prcomp object, additional attribute non_numeric_cols vector column names columns contain numeric values. probably groups labels, used ggplot_pca().","code":""},{"path":"https://msberends.github.io/AMR/reference/pca.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Principal Component Analysis (for AMR) — pca","text":"","code":"# `example_isolates` is a data set available in the AMR package. # See ?example_isolates.  # \\donttest{ if (require(\"dplyr\")) {   # calculate the resistance per group first   resistance_data <- example_isolates %>%     group_by(       order = mo_order(mo), # group on anything, like order       genus = mo_genus(mo)     ) %>% #   and genus as we do here;     filter(n() >= 30) %>% # filter on only 30 results per group     summarise_if(is.sir, resistance) # then get resistance of all drugs    # now conduct PCA for certain antimicrobial drugs   pca_result <- resistance_data %>%     pca(AMC, CXM, CTX, CAZ, GEN, TOB, TMP, SXT)    pca_result   summary(pca_result)    # old base R plotting method:   biplot(pca_result)   # new ggplot2 plotting method using this package:   if (require(\"ggplot2\")) {     ggplot_pca(pca_result)      ggplot_pca(pca_result) +       scale_colour_viridis_d() +       labs(title = \"Title here\")   } } #> Warning: There were 73 warnings in `summarise()`. #> The first warning was: #> ℹ In argument: `PEN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 5: `order = \"Lactobacillales\"` `genus = \"Enterococcus\"`. #> Caused by warning: #> ! Introducing NA: only 14 results available for PEN in group: order = #> \"Lactobacillales\", genus = \"Enterococcus\" (minimum = 30). #> ℹ Run `dplyr::last_dplyr_warnings()` to see the 72 remaining warnings. #> ℹ Columns selected for PCA: \"AMC\", \"CAZ\", \"CTX\", \"CXM\", \"GEN\", \"SXT\", #>   \"TMP\", and \"TOB\". Total observations available: 7. #> Groups (n=4, named as 'order'): #> [1] \"Caryophanales\"    \"Enterobacterales\" \"Lactobacillales\"  \"Pseudomonadales\"  #>    # }"},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":null,"dir":"Reference","previous_headings":"","what":"Plotting Helpers for AMR Data Analysis — plot","title":"Plotting Helpers for AMR Data Analysis — plot","text":"Functions plot classes sir, mic disk, support base R ggplot2. Especially scale_*_mic() functions relevant wrappers plot MIC values ggplot2. allows custom MIC ranges plot intermediate log2 levels missing MIC values.","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Plotting Helpers for AMR Data Analysis — plot","text":"","code":"scale_x_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_y_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_colour_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_fill_mic(keep_operators = \"edges\", mic_range = NULL, ...)  scale_x_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  scale_colour_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  scale_fill_sir(colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), eucast_I = getOption(\"AMR_guideline\",   \"EUCAST\") == \"EUCAST\", ...)  # S3 method for class 'mic' plot(x, mo = NULL, ab = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   main = deparse(substitute(x)), ylab = translate_AMR(\"Frequency\", language   = language),   xlab = translate_AMR(\"Minimum Inhibitory Concentration (mg/L)\", language =   language), colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'mic' autoplot(object, mo = NULL, ab = NULL,   guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   title = deparse(substitute(object)), ylab = translate_AMR(\"Frequency\",   language = language),   xlab = translate_AMR(\"Minimum Inhibitory Concentration (mg/L)\", language =   language), colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'disk' plot(x, main = deparse(substitute(x)),   ylab = translate_AMR(\"Frequency\", language = language),   xlab = translate_AMR(\"Disk diffusion diameter (mm)\", language = language),   mo = NULL, ab = NULL, guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'disk' autoplot(object, mo = NULL, ab = NULL,   title = deparse(substitute(object)), ylab = translate_AMR(\"Frequency\",   language = language), xlab = translate_AMR(\"Disk diffusion diameter (mm)\",   language = language), guideline = getOption(\"AMR_guideline\", \"EUCAST\"),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), expand = TRUE,   include_PKPD = getOption(\"AMR_include_PKPD\", TRUE),   breakpoint_type = getOption(\"AMR_breakpoint_type\", \"human\"), ...)  # S3 method for class 'sir' plot(x, ylab = translate_AMR(\"Percentage\", language =   language), xlab = translate_AMR(\"Antimicrobial Interpretation\", language =   language), main = deparse(substitute(x)), language = get_AMR_locale(),   ...)  # S3 method for class 'sir' autoplot(object, title = deparse(substitute(object)),   xlab = translate_AMR(\"Antimicrobial Interpretation\", language = language),   ylab = translate_AMR(\"Frequency\", language = language),   colours_SIR = c(\"#3CAEA3\", \"#F6D55C\", \"#ED553B\"),   language = get_AMR_locale(), ...)  facet_sir(facet = c(\"interpretation\", \"antibiotic\"), nrow = NULL)  scale_y_percent(breaks = function(x) seq(0, max(x, na.rm = TRUE), 0.1),   limits = c(0, NA))  scale_sir_colours(..., aesthetics, colours_SIR = c(\"#3CAEA3\", \"#F6D55C\",   \"#ED553B\"))  theme_sir()  labels_sir_count(position = NULL, x = \"antibiotic\",   translate_ab = \"name\", minimum = 30, language = get_AMR_locale(),   combine_SI = TRUE, datalabels.size = 3, datalabels.colour = \"grey15\")"},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Plotting Helpers for AMR Data Analysis — plot","text":"keep_operators character specifying handle operators (> <=) input. Accepts one three values: \"\" (TRUE) keep operators, \"none\" (FALSE) remove operators, \"edges\" keep operators ends range. mic_range manual range rescale MIC values (using rescale_mic()), e.g., mic_range = c(0.001, 32). Use NA prevent rescaling one side, e.g., mic_range = c(NA, 32). Note: rescales values filter - use ggplot2 limits argument separately exclude values plot. ... arguments passed methods colours_SIR colours use filling bars, must vector three values (order S, R). default colours colour-blind friendly. language language used translate 'Susceptible', 'Increased exposure'/'Intermediate' 'Resistant' - default system language (see get_AMR_locale()) can overwritten setting package option AMR_locale, e.g. options(AMR_locale = \"de\"), see translate. Use language = NULL prevent translation. eucast_I logical indicate whether '' must interpreted \"Susceptible, increased exposure\". TRUE default AMR interpretation guideline set EUCAST (default). FALSE, interpreted \"Intermediate\". x, object values created .mic(), .disk() .sir() (random_* variants, random_mic()) mo (vector ) text can coerced valid microorganism code .mo() ab (vector ) text can coerced valid antimicrobial drug code .ab() guideline interpretation guideline use - default latest included EUCAST guideline, see Details main, title title plot xlab, ylab axis title expand logical indicate whether range x axis expanded lowest highest value. MIC values, intermediate values factors 2 starting highest MIC value. disk diameters, whole diameter range filled. include_PKPD logical indicate PK/PD clinical breakpoints must applied last resort - default TRUE. Can also set package option AMR_include_PKPD. breakpoint_type type breakpoints use, either \"ECOFF\", \"animal\", \"human\". ECOFF stands Epidemiological Cut-values. default \"human\", can also set package option AMR_breakpoint_type. host set values veterinary species, automatically set \"animal\". facet variable split plots , either \"interpretation\" (default) \"antibiotic\" grouping variable nrow (using facet) number rows breaks numeric vector positions limits numeric vector length two providing limits scale, use NA refer existing minimum maximum aesthetics aesthetics apply colours - default \"fill\" can also (combination ) \"alpha\", \"colour\", \"fill\", \"linetype\", \"shape\" \"size\" position position adjustment bars, either \"fill\", \"stack\" \"dodge\" translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE datalabels.size size datalabels datalabels.colour colour datalabels","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Plotting Helpers for AMR Data Analysis — plot","text":"autoplot() functions return ggplot model extendible ggplot2 function.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"the-scale-mic-functions","dir":"Reference","previous_headings":"","what":"The scale_*_mic() Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"functions scale_x_mic(), scale_y_mic(), scale_colour_mic(), scale_fill_mic() functions allow plot mic class (MIC values) continuous, logarithmic scale. also allow rescale MIC range 'inside' 'outside' range required, retain signs MIC values desired. Missing intermediate log2 levels plotted .","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"the-scale-sir-functions","dir":"Reference","previous_headings":"","what":"The scale_*_sir() Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"functions scale_x_sir(), scale_colour_sir(), scale_fill_sir() functions allow plot sir class right order (S < SDD < < R < NI). default, translate S//R values interpretative text (\"Susceptible\", \"Resistant\", etc.) 20 supported languages (use language = NULL keep S//R). Also, except scale_x_sir(), set colour-blind friendly colours colour fill aesthetics.","code":""},{"path":"https://msberends.github.io/AMR/reference/plot.html","id":"additional-ggplot-functions","dir":"Reference","previous_headings":"","what":"Additional ggplot2 Functions","title":"Plotting Helpers for AMR Data Analysis — plot","text":"package contains functions extend ggplot2 package, help visualising AMR data results. functions internally used ggplot_sir() . facet_sir() creates 2d plots (default based S//R) using ggplot2::facet_wrap(). scale_y_percent() transforms y axis 0 100% range using ggplot2::scale_y_continuous(). scale_sir_colours() allows set colours aesthetic, even shape linetype. theme_sir() [ggplot2 theme][ggplot2::theme() minimal distraction. labels_sir_count() print datalabels bars percentage number isolates, using ggplot2::geom_text(). interpretation \"\" named \"Increased exposure\" EUCAST guidelines since 2019, named \"Intermediate\" cases. interpreting MIC values well disk diffusion diameters, default guideline EUCAST 2024, unless package option AMR_guideline set. See .sir() information.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":null,"dir":"Reference","previous_headings":"","what":"Calculate Antimicrobial Resistance — proportion","title":"Calculate Antimicrobial Resistance — proportion","text":"functions can used calculate (co-)resistance susceptibility microbial isolates (.e. percentage S, SI, , IR R). functions support quasiquotation pipes, can used summarise() dplyr package also support grouped variables, see Examples. resistance() used calculate resistance, susceptibility() used calculate susceptibility.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Calculate Antimicrobial Resistance — proportion","text":"","code":"resistance(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  susceptibility(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  sir_confidence_interval(..., ab_result = \"R\", minimum = 30,   as_percent = FALSE, only_all_tested = FALSE, confidence_level = 0.95,   side = \"both\", collapse = FALSE)  proportion_R(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_IR(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_I(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_SI(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_S(..., minimum = 30, as_percent = FALSE,   only_all_tested = FALSE)  proportion_df(data, translate_ab = \"name\", language = get_AMR_locale(),   minimum = 30, as_percent = FALSE, combine_SI = TRUE,   confidence_level = 0.95)  sir_df(data, translate_ab = \"name\", language = get_AMR_locale(),   minimum = 30, as_percent = FALSE, combine_SI = TRUE,   confidence_level = 0.95)"},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"source","dir":"Reference","previous_headings":"","what":"Source","title":"Calculate Antimicrobial Resistance — proportion","text":"M39 Analysis Presentation Cumulative Antimicrobial Susceptibility Test Data, 5th Edition, 2022, Clinical Laboratory Standards Institute (CLSI). https://clsi.org/standards/products/microbiology/documents/m39/.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Calculate Antimicrobial Resistance — proportion","text":"... one vectors (columns) antibiotic interpretations. transformed internally .sir() needed. Use multiple columns calculate (lack ) co-resistance: probability one two drugs resistant susceptible result. See Examples. minimum minimum allowed number available (tested) isolates. isolate count lower minimum return NA warning. default number 30 isolates advised Clinical Laboratory Standards Institute (CLSI) best practice, see Source. as_percent logical indicate whether output must returned hundred fold % sign (character). value 0.123456 returned \"12.3%\". only_all_tested (combination therapies, .e. using one variable ...): logical indicate isolates must tested antibiotics, see section Combination Therapy ab_result antibiotic results test , must one values \"S\", \"SDD\", \"\", \"R\" confidence_level confidence level returned confidence interval. calculation, number S SI isolates, R isolates compared total number available isolates R, S, using binom.test(), .e., Clopper-Pearson method. side side confidence interval return. default \"\" length 2 vector, can also (abbreviated ) \"min\"/\"left\"/\"lower\"/\"less\" \"max\"/\"right\"/\"higher\"/\"greater\". collapse logical indicate whether output values 'collapsed', .e. merged together one value, character value use collapsing data data.frame containing columns class sir (see .sir()) translate_ab column name antibiotics data set translate antibiotic abbreviations , using ab_property() language language returned text - default current system language (see get_AMR_locale()) can also set package option AMR_locale. Use language = NULL language = \"\" prevent translation. combine_SI logical indicate whether values S, SDD, must merged one, output consists S+SDD+vs. R (susceptible vs. resistant) - default TRUE","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Calculate Antimicrobial Resistance — proportion","text":"double , as_percent = TRUE, character.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Calculate Antimicrobial Resistance — proportion","text":"automated comprehensive analysis, consider using antibiogram() wisca(), streamline many aspects susceptibility reporting , importantly, also support WISCA. functions described offer hands-, manual approach greater customisation. Remember filter data let contain first isolates! needed exclude duplicates reduce selection bias. Use first_isolate() determine data set one four available algorithms. function resistance() equal function proportion_R(). function susceptibility() equal function proportion_SI(). Since AMR v3.0, proportion_SI() proportion_I() include dose-dependent susceptibility ('SDD'). Use sir_confidence_interval() calculate confidence interval, relies binom.test(), .e., Clopper-Pearson method. function returns vector length 2 default antimicrobial resistance. Change side argument \"left\"/\"min\" \"right\"/\"max\" return single value, change ab_result argument e.g. c(\"S\", \"\") test antimicrobial susceptibility, see Examples. functions meant count isolates, calculate proportion resistance/susceptibility. Use count_*() functions count isolates. function susceptibility() essentially equal count_susceptible()/count_all(). Low counts can influence outcome - proportion_*() functions may camouflage , since return proportion (albeit dependent minimum argument). function proportion_df() takes variable data sir class (created .sir()) calculates proportions S, , R. also supports grouped variables. function sir_df() works exactly like proportion_df(), adds number isolates.","code":""},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"combination-therapy","dir":"Reference","previous_headings":"","what":"Combination Therapy","title":"Calculate Antimicrobial Resistance — proportion","text":"using one variable ... (= combination therapy), use only_all_tested count isolates tested antibiotics/variables test . See example two antibiotics, Drug Drug B, susceptibility() works calculate %SI:   Please note , combination therapies, only_all_tested = TRUE applies :   , combination therapies, only_all_tested = FALSE applies :   Using only_all_tested impact using one antibiotic input.","code":"--------------------------------------------------------------------                     only_all_tested = FALSE  only_all_tested = TRUE                     -----------------------  -----------------------  Drug A    Drug B   considered   considered  considered   considered                     susceptible    tested    susceptible    tested --------  --------  -----------  ----------  -----------  ----------  S or I    S or I        X            X           X            X    R       S or I        X            X           X            X   <NA>     S or I        X            X           -            -  S or I      R           X            X           X            X    R         R           -            X           -            X   <NA>       R           -            -           -            -  S or I     <NA>         X            X           -            -    R        <NA>         -            -           -            -   <NA>      <NA>         -            -           -            - -------------------------------------------------------------------- count_S()    +   count_I()    +   count_R()    = count_all()   proportion_S() + proportion_I() + proportion_R() = 1 count_S()    +   count_I()    +   count_R()    >= count_all()   proportion_S() + proportion_I() + proportion_R() >= 1"},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Calculate Antimicrobial Resistance — proportion","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/proportion.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Calculate Antimicrobial Resistance — proportion","text":"","code":"# example_isolates is a data set available in the AMR package. # run ?example_isolates for more info. example_isolates #> # A tibble: 2,000 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,990 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …   # base R ------------------------------------------------------------ # determines %R resistance(example_isolates$AMX) #> [1] 0.5955556 sir_confidence_interval(example_isolates$AMX) #> [1] 0.5688204 0.6218738 sir_confidence_interval(example_isolates$AMX,   confidence_level = 0.975 ) #> [1] 0.5650148 0.6255670 sir_confidence_interval(example_isolates$AMX,   confidence_level = 0.975,   collapse = \", \" ) #> [1] \"0.565, 0.626\"  # determines %S+I: susceptibility(example_isolates$AMX) #> [1] 0.4044444 sir_confidence_interval(example_isolates$AMX,   ab_result = c(\"S\", \"I\") ) #> [1] 0.3781262 0.4311796  # be more specific proportion_S(example_isolates$AMX) #> [1] 0.4022222 proportion_SI(example_isolates$AMX) #> Note that proportion_SI() will also include dose-dependent #> susceptibility, 'SDD'. This note will be shown once for this session. #> [1] 0.4044444 proportion_I(example_isolates$AMX) #> Note that proportion_I() will also include dose-dependent susceptibility, #> 'SDD'. This note will be shown once for this session. #> [1] 0.002222222 proportion_IR(example_isolates$AMX) #> Note that proportion_IR() will also include dose-dependent #> susceptibility, 'SDD'. This note will be shown once for this session. #> [1] 0.5977778 proportion_R(example_isolates$AMX) #> [1] 0.5955556  # dplyr ------------------------------------------------------------- # \\donttest{ if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       r = resistance(CIP),       n = n_sir(CIP)     ) # n_sir works like n_distinct in dplyr, see ?n_sir } #> # A tibble: 3 × 3 #>   ward           r     n #>   <chr>      <dbl> <int> #> 1 Clinical   0.147   869 #> 2 ICU        0.190   447 #> 3 Outpatient 0.161    93 if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       cipro_R = resistance(CIP),       ci_min = sir_confidence_interval(CIP, side = \"min\"),       ci_max = sir_confidence_interval(CIP, side = \"max\"),     ) } #> # A tibble: 3 × 4 #>   ward       cipro_R ci_min ci_max #>   <chr>        <dbl>  <dbl>  <dbl> #> 1 Clinical     0.147 0.124   0.173 #> 2 ICU          0.190 0.155   0.230 #> 3 Outpatient   0.161 0.0932  0.252 if (require(\"dplyr\")) {   # scoped dplyr verbs with antimicrobial selectors   # (you could also use across() of course)   example_isolates %>%     group_by(ward) %>%     summarise_at(       c(aminoglycosides(), carbapenems()),       resistance     ) } #> ℹ For aminoglycosides() using columns 'GEN' (gentamicin), 'TOB' #>   (tobramycin), 'AMK' (amikacin), and 'KAN' (kanamycin) #> ℹ For carbapenems() using columns 'IPM' (imipenem) and 'MEM' (meropenem) #> Warning: There was 1 warning in `summarise()`. #> ℹ In argument: `KAN = (function (..., minimum = 30, as_percent = FALSE, #>   only_all_tested = FALSE) ...`. #> ℹ In group 3: `ward = \"Outpatient\"`. #> Caused by warning: #> ! Introducing NA: only 23 results available for KAN in group: ward = #> \"Outpatient\" (minimum = 30). #> # A tibble: 3 × 7 #>   ward         GEN   TOB   AMK   KAN    IPM    MEM #>   <chr>      <dbl> <dbl> <dbl> <dbl>  <dbl>  <dbl> #> 1 Clinical   0.229 0.315 0.626     1 0.0498 0.0458 #> 2 ICU        0.290 0.400 0.662     1 0.0862 0.0894 #> 3 Outpatient 0.2   0.368 0.605    NA 0.0541 0.0541 if (require(\"dplyr\")) {   example_isolates %>%     group_by(ward) %>%     summarise(       R = resistance(CIP, as_percent = TRUE),       SI = susceptibility(CIP, as_percent = TRUE),       n1 = count_all(CIP), # the actual total; sum of all three       n2 = n_sir(CIP), # same - analogous to n_distinct       total = n()     ) # NOT the number of tested isolates!    # Calculate co-resistance between amoxicillin/clav acid and gentamicin,   # so we can see that combination therapy does a lot more than mono therapy:   example_isolates %>% susceptibility(AMC) # %SI = 76.3%   example_isolates %>% count_all(AMC) #   n = 1879    example_isolates %>% susceptibility(GEN) # %SI = 75.4%   example_isolates %>% count_all(GEN) #   n = 1855    example_isolates %>% susceptibility(AMC, GEN) # %SI = 94.1%   example_isolates %>% count_all(AMC, GEN) #   n = 1939     # See Details on how `only_all_tested` works. Example:   example_isolates %>%     summarise(       numerator = count_susceptible(AMC, GEN),       denominator = count_all(AMC, GEN),       proportion = susceptibility(AMC, GEN)     )    example_isolates %>%     summarise(       numerator = count_susceptible(AMC, GEN, only_all_tested = TRUE),       denominator = count_all(AMC, GEN, only_all_tested = TRUE),       proportion = susceptibility(AMC, GEN, only_all_tested = TRUE)     )     example_isolates %>%     group_by(ward) %>%     summarise(       cipro_p = susceptibility(CIP, as_percent = TRUE),       cipro_n = count_all(CIP),       genta_p = susceptibility(GEN, as_percent = TRUE),       genta_n = count_all(GEN),       combination_p = susceptibility(CIP, GEN, as_percent = TRUE),       combination_n = count_all(CIP, GEN)     )    # Get proportions S/I/R immediately of all sir columns   example_isolates %>%     select(AMX, CIP) %>%     proportion_df(translate = FALSE)    # It also supports grouping variables   # (use sir_df to also include the count)   example_isolates %>%     select(ward, AMX, CIP) %>%     group_by(ward) %>%     sir_df(translate = FALSE) } #> # A tibble: 12 × 7 #>    ward       antibiotic interpretation value ci_min ci_max isolates #>  * <chr>      <chr>      <ord>          <dbl>  <dbl>  <dbl>    <int> #>  1 Clinical   AMX        SI             0.423 0.389   0.457      357 #>  2 Clinical   AMX        R              0.577 0.543   0.611      487 #>  3 Clinical   CIP        SI             0.853 0.827   0.876      741 #>  4 Clinical   CIP        R              0.147 0.124   0.173      128 #>  5 ICU        AMX        SI             0.369 0.323   0.417      158 #>  6 ICU        AMX        R              0.631 0.583   0.677      270 #>  7 ICU        CIP        SI             0.810 0.770   0.845      362 #>  8 ICU        CIP        R              0.190 0.155   0.230       85 #>  9 Outpatient AMX        SI             0.397 0.288   0.515       31 #> 10 Outpatient AMX        R              0.603 0.485   0.712       47 #> 11 Outpatient CIP        SI             0.839 0.748   0.907       78 #> 12 Outpatient CIP        R              0.161 0.0932  0.252       15 # }"},{"path":"https://msberends.github.io/AMR/reference/random.html","id":null,"dir":"Reference","previous_headings":"","what":"Random MIC Values/Disk Zones/SIR Generation — random","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"functions can used generating random MIC values disk diffusion diameters, AMR data analysis practice. providing microorganism antimicrobial drug, generated results reflect reality much possible.","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"","code":"random_mic(size = NULL, mo = NULL, ab = NULL, ...)  random_disk(size = NULL, mo = NULL, ab = NULL, ...)  random_sir(size = NULL, prob_SIR = c(0.33, 0.33, 0.33), ...)"},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"size desired size returned vector. used data.frame call dplyr verb, get current (group) size left blank. mo character can coerced valid microorganism code .mo() ab character can coerced valid antimicrobial drug code .ab() ... ignored, place allow future extensions prob_SIR vector length 3: probabilities \"S\" (1st value), \"\" (2nd value) \"R\" (3rd value)","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"class mic random_mic() (see .mic()) class disk random_disk() (see .disk())","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"base R function sample() used generating values. Generated values based EUCAST 2024 guideline implemented clinical_breakpoints data set. create specific generated values per bug drug, set mo /ab argument.","code":""},{"path":"https://msberends.github.io/AMR/reference/random.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Random MIC Values/Disk Zones/SIR Generation — random","text":"","code":"random_mic(25) #> Class 'mic' #>  [1] 0.001  256    0.025  0.0625 0.125  0.25   2      0.001  32     0.01   #> [11] 0.0625 0.001  0.01   256    0.0625 32     0.001  0.25   128    256    #> [21] 1      0.002  0.5    0.01   0.002  random_disk(25) #> Class 'disk' #>  [1] 41 43 23 17 19 28 14 38 18 30 42 11 37 40 39 16 41 43 49 45 44 28 36 14 10 random_sir(25) #> Class 'sir' #>  [1] R S R I R R R I R S R I I S S R I I I S I R I I S  # \\donttest{ # make the random generation more realistic by setting a bug and/or drug: random_mic(25, \"Klebsiella pneumoniae\") # range 0.0625-64 #> Class 'mic' #>  [1] 0.25  0.5   32    0.002 0.005 0.25  0.25  4     0.001 0.005 0.025 0.01  #> [13] 0.25  0.005 0.005 2     0.01  1     4     8     0.5   16    1     >=64  #> [25] 0.25  random_mic(25, \"Klebsiella pneumoniae\", \"meropenem\") # range 0.0625-16 #> Class 'mic' #>  [1] 1  1  1  4  16 8  1  16 4  16 8  16 4  2  16 4  4  1  2  4  4  1  16 16 2  random_mic(25, \"Streptococcus pneumoniae\", \"meropenem\") # range 0.0625-4 #> Class 'mic' #>  [1] 2      4      0.25   0.0625 2      4      0.25   4      1      0.0625 #> [11] 0.25   0.25   4      1      0.5    0.0625 4      0.5    2      0.5    #> [21] 0.5    0.0625 1      0.0625 0.125   random_disk(25, \"Klebsiella pneumoniae\") # range 8-50 #> Class 'disk' #>  [1] 25 11 13 40 46 37 44 42 10 41 50 39 39 40 33 19 48 39 20 11 32 20 50 38 46 random_disk(25, \"Klebsiella pneumoniae\", \"ampicillin\") # range 11-17 #> Class 'disk' #>  [1] 14 15 14 14 17 13 16 17 12 15 17 11 17 17 17 12 17 11 11 14 13 17 13 15 15 random_disk(25, \"Streptococcus pneumoniae\", \"ampicillin\") # range 12-27 #> Class 'disk' #>  [1] 26 16 22 23 19 22 19 18 19 23 23 24 22 16 17 15 17 16 18 17 26 16 16 20 22 # }"},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":null,"dir":"Reference","previous_headings":"","what":"Predict Antimicrobial Resistance — resistance_predict","title":"Predict Antimicrobial Resistance — resistance_predict","text":"Create prediction model predict antimicrobial resistance next years statistical solid ground. Standard errors (SE) returned columns se_min se_max. See Examples real live example.","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Predict Antimicrobial Resistance — resistance_predict","text":"","code":"resistance_predict(x, col_ab, col_date = NULL, year_min = NULL,   year_max = NULL, year_every = 1, minimum = 30, model = NULL,   I_as_S = TRUE, preserve_measurements = TRUE, info = interactive(), ...)  sir_predict(x, col_ab, col_date = NULL, year_min = NULL, year_max = NULL,   year_every = 1, minimum = 30, model = NULL, I_as_S = TRUE,   preserve_measurements = TRUE, info = interactive(), ...)  # S3 method for class 'resistance_predict' plot(x, main = paste(\"Resistance Prediction of\",   x_name), ...)  ggplot_sir_predict(x, main = paste(\"Resistance Prediction of\", x_name),   ribbon = TRUE, ...)  # S3 method for class 'resistance_predict' autoplot(object,   main = paste(\"Resistance Prediction of\", x_name), ribbon = TRUE, ...)"},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Predict Antimicrobial Resistance — resistance_predict","text":"x data.frame containing isolates. Can left blank automatic determination, see Examples. col_ab column name x containing antimicrobial interpretations (\"R\", \"\" \"S\") col_date column name date, used calculate years column consist years already - default first column date class year_min lowest year use prediction model, dafaults lowest year col_date year_max highest year use prediction model - default 10 years today year_every unit sequence lowest year found data year_max minimum minimal amount available isolates per year include. Years containing less observations estimated model. model statistical model choice. generalised linear regression model binomial distribution (.e. using glm(..., family = binomial), assuming period zero resistance followed period increasing resistance leading slowly resistance. See Details valid options. I_as_S logical indicate whether values \"\" treated \"S\" (otherwise treated \"R\"). default, TRUE, follows redefinition EUCAST interpretation (increased exposure) 2019, see section Interpretation S, R . preserve_measurements logical indicate whether predictions years actually available data overwritten original data. standard errors years NA. info logical indicate whether textual analysis printed name summary() statistical model. ... arguments passed functions main title plot ribbon logical indicate whether ribbon shown (default) error bars object model data plotted","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Predict Antimicrobial Resistance — resistance_predict","text":"data.frame extra class resistance_predict columns: year value, estimated preserve_measurements = FALSE, combination observed estimated otherwise se_min, lower bound standard error minimum 0 (standard error never go 0%) se_max upper bound standard error maximum 1 (standard error never go 100%) observations, total number available observations year, .e. \\(S + + R\\) observed, original observed resistant percentages estimated, estimated resistant percentages, calculated model Furthermore, model available attribute: attributes(x)$model, see Examples.","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Predict Antimicrobial Resistance — resistance_predict","text":"Valid options statistical model (argument model) : \"binomial\" \"binom\" \"logit\": generalised linear regression model binomial distribution \"loglin\" \"poisson\": generalised log-linear regression model poisson distribution \"lin\" \"linear\": linear regression model","code":""},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"interpretation-of-sir","dir":"Reference","previous_headings":"","what":"Interpretation of SIR","title":"Predict Antimicrobial Resistance — resistance_predict","text":"2019, European Committee Antimicrobial Susceptibility Testing (EUCAST) decided change definitions susceptibility testing categories S, , R shown (https://www.eucast.org/newsiandr): S - Susceptible, standard dosing regimen microorganism categorised \"Susceptible, standard dosing regimen\", high likelihood therapeutic success using standard dosing regimen agent. - Susceptible, increased exposure  microorganism categorised \"Susceptible, Increased exposure\" high likelihood therapeutic success exposure agent increased adjusting dosing regimen concentration site infection. R = Resistant microorganism categorised \"Resistant\" high likelihood therapeutic failure even increased exposure. Exposure function mode administration, dose, dosing interval, infusion time, well distribution excretion antimicrobial agent influence infecting organism site infection. AMR package honours insight. Use susceptibility() (equal proportion_SI()) determine antimicrobial susceptibility count_susceptible() (equal count_SI()) count susceptible isolates.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/resistance_predict.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Predict Antimicrobial Resistance — resistance_predict","text":"","code":"x <- resistance_predict(example_isolates,   col_ab = \"AMX\",   year_min = 2010,   model = \"binomial\" ) plot(x)  # \\donttest{ if (require(\"ggplot2\")) {   ggplot_sir_predict(x) }   # using dplyr: if (require(\"dplyr\")) {   x <- example_isolates %>%     filter_first_isolate() %>%     filter(mo_genus(mo) == \"Staphylococcus\") %>%     resistance_predict(\"PEN\", model = \"binomial\")   print(plot(x))    # get the model from the object   mymodel <- attributes(x)$model   summary(mymodel) }  #> NULL #>  #> Call: #> glm(formula = df_matrix ~ year, family = binomial) #>  #> Coefficients: #>             Estimate Std. Error z value Pr(>|z|) #> (Intercept) 35.76101   72.29172   0.495    0.621 #> year        -0.01720    0.03603  -0.477    0.633 #>  #> (Dispersion parameter for binomial family taken to be 1) #>  #>     Null deviance: 5.3681  on 11  degrees of freedom #> Residual deviance: 5.1408  on 10  degrees of freedom #> AIC: 50.271 #>  #> Number of Fisher Scoring iterations: 4 #>   # create nice plots with ggplot2 yourself if (require(\"dplyr\") && require(\"ggplot2\")) {   data <- example_isolates %>%     filter(mo == as.mo(\"E. coli\")) %>%     resistance_predict(       col_ab = \"AMX\",       col_date = \"date\",       model = \"binomial\",       info = FALSE,       minimum = 15     )   head(data)   autoplot(data) }  # }"},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":null,"dir":"Reference","previous_headings":"","what":"Skewness of the Sample — skewness","title":"Skewness of the Sample — skewness","text":"Skewness measure asymmetry probability distribution real-valued random variable mean. negative ('left-skewed'): left tail longer; mass distribution concentrated right histogram. positive ('right-skewed'): right tail longer; mass distribution concentrated left histogram. normal distribution skewness 0.","code":""},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Skewness of the Sample — skewness","text":"","code":"skewness(x, na.rm = FALSE)  # Default S3 method skewness(x, na.rm = FALSE)  # S3 method for class 'matrix' skewness(x, na.rm = FALSE)  # S3 method for class 'data.frame' skewness(x, na.rm = FALSE)"},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Skewness of the Sample — skewness","text":"x vector values, matrix data.frame na.rm logical value indicating whether NA values stripped computation proceeds","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/skewness.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Skewness of the Sample — skewness","text":"","code":"skewness(runif(1000)) #> [1] 0.01041273"},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":null,"dir":"Reference","previous_headings":"","what":"Filter Top n Microorganisms — top_n_microorganisms","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"function filters data set include top n microorganisms based specified property, taxonomic family genus. example, can filter data set top 3 species, species top 5 genera, top 3 species top 5 genera.","code":""},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"","code":"top_n_microorganisms(x, n, property = \"fullname\", n_for_each = NULL,   col_mo = NULL, ...)"},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"x data frame containing microbial data n integer specifying maximum number unique values property include output property character string indicating microorganism property use filtering. Must one column names microorganisms data set: \"mo\", \"fullname\", \"status\", \"kingdom\", \"phylum\", \"class\", \"order\", \"family\", \"genus\", \"species\", \"subspecies\", \"rank\", \"ref\", \"oxygen_tolerance\", \"source\", \"lpsn\", \"lpsn_parent\", \"lpsn_renamed_to\", \"mycobank\", \"mycobank_parent\", \"mycobank_renamed_to\", \"gbif\", \"gbif_parent\", \"gbif_renamed_to\", \"prevalence\", \"snomed\". NULL, raw values col_mo used without transformation. n_for_each optional integer specifying maximum number rows retain value selected property. NULL, rows within top n groups included. col_mo character string indicating column x contains microorganism names codes. Defaults first column class mo. Values coerced using .mo(). ... Additional arguments passed mo_property() property NULL.","code":""},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"function useful preprocessing data creating antibiograms analyses require focused subsets microbial data. example, can filter data set include isolates top 10 species.","code":""},{"path":[]},{"path":"https://msberends.github.io/AMR/reference/top_n_microorganisms.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Filter Top n Microorganisms — top_n_microorganisms","text":"","code":"# filter to the top 3 species: top_n_microorganisms(example_isolates,                      n = 3) #> # A tibble: 1,015 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  4 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  5 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  6 2002-01-19 738003     71 M      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  7 2002-02-03 481442     76 M      ICU      B_STPHY_CONS   R     NA    S     NA  #>  8 2002-02-14 067927     45 F      ICU      B_STPHY_CONS   R     NA    R     NA  #>  9 2002-02-14 067927     45 F      ICU      B_STPHY_CONS   S     NA    S     NA  #> 10 2002-02-21 A56499     64 M      Clinical B_STPHY_CONS   S     NA    S     NA  #> # ℹ 1,005 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter to any species in the top 5 genera: top_n_microorganisms(example_isolates,                      n = 5, property = \"genus\") #> # A tibble: 1,742 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-01-02 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  2 2002-01-03 A77334     65 F      Clinical B_ESCHR_COLI   R     NA    NA    NA  #>  3 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  4 2002-01-07 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  5 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  6 2002-01-13 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #>  7 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  8 2002-01-14 462729     78 M      Clinical B_STPHY_AURS   R     NA    S     R   #>  9 2002-01-16 067927     45 F      ICU      B_STPHY_EPDR   R     NA    R     NA  #> 10 2002-01-17 858515     79 F      ICU      B_STPHY_EPDR   R     NA    S     NA  #> # ℹ 1,732 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …  # filter to the top 3 species in each of the top 5 genera: top_n_microorganisms(example_isolates,                      n = 5, property = \"genus\", n_for_each = 3) #> # A tibble: 1,497 × 46 #>    date       patient   age gender ward     mo           PEN   OXA   FLC   AMX   #>    <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <sir> <sir> <sir> <sir> #>  1 2002-02-21 4FC193     69 M      Clinical B_ENTRC_FACM   NA    NA    NA    NA  #>  2 2002-04-08 130252     78 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #>  3 2002-06-23 798871     82 M      Clinical B_ENTRC_FCLS   NA    NA    NA    NA  #>  4 2002-06-23 798871     82 M      Clinical B_ENTRC_FCLS   NA    NA    NA    NA  #>  5 2003-04-20 6BC362     62 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  6 2003-04-21 6BC362     62 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  7 2003-08-13 F35553     52 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #>  8 2003-09-05 F35553     52 M      ICU      B_ENTRC        NA    NA    NA    NA  #>  9 2003-09-05 F35553     52 M      ICU      B_ENTRC_FCLS   NA    NA    NA    NA  #> 10 2003-09-28 1B0933     80 M      Clinical B_ENTRC        NA    NA    NA    NA  #> # ℹ 1,487 more rows #> # ℹ 36 more variables: AMC <sir>, AMP <sir>, TZP <sir>, CZO <sir>, FEP <sir>, #> #   CXM <sir>, FOX <sir>, CTX <sir>, CAZ <sir>, CRO <sir>, GEN <sir>, #> #   TOB <sir>, AMK <sir>, KAN <sir>, TMP <sir>, SXT <sir>, NIT <sir>, #> #   FOS <sir>, LNZ <sir>, CIP <sir>, MFX <sir>, VAN <sir>, TEC <sir>, #> #   TCY <sir>, TGC <sir>, DOX <sir>, ERY <sir>, CLI <sir>, AZM <sir>, #> #   IPM <sir>, MEM <sir>, MTR <sir>, CHL <sir>, COL <sir>, MUP <sir>, …"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":null,"dir":"Reference","previous_headings":"","what":"Translate Strings from the AMR Package — translate","title":"Translate Strings from the AMR Package — translate","text":"language-dependent output AMR functions, mo_name(), mo_gramstain(), mo_type() ab_name().","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Translate Strings from the AMR Package — translate","text":"","code":"get_AMR_locale()  set_AMR_locale(language)  reset_AMR_locale()  translate_AMR(x, language = get_AMR_locale())"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Translate Strings from the AMR Package — translate","text":"language language choose. Use one supported language names ISO-639-1 codes: English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). x text translate","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Translate Strings from the AMR Package — translate","text":"currently 20 supported languages English (en), Chinese (zh), Czech (cs), Danish (da), Dutch (nl), Finnish (fi), French (fr), German (de), Greek (el), Italian (), Japanese (ja), Norwegian (), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Spanish (es), Swedish (sv), Turkish (tr), Ukrainian (uk). languages translations available antimicrobial drugs colloquial microorganism names. permanently silence -per-session language note non-English operating system, can set package option AMR_locale .Rprofile file like :   save file. Please read adding updating language Wiki.","code":"# Open .Rprofile file utils::file.edit(\"~/.Rprofile\")  # Then add e.g. Italian support to that file using: options(AMR_locale = \"Italian\")"},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"changing-the-default-language","dir":"Reference","previous_headings":"","what":"Changing the Default Language","title":"Translate Strings from the AMR Package — translate","text":"system language used default (returned Sys.getenv(\"LANG\") , LANG set, Sys.getlocale(\"LC_COLLATE\")), language supported. language used can overwritten two ways checked order: Setting package option AMR_locale, either using e.g. set_AMR_locale(\"German\") running e.g. options(AMR_locale = \"German\"). Note setting R option works session. Save command options(AMR_locale = \"(language)\") .Rprofile file apply every session. Run utils::file.edit(\"~/.Rprofile\") edit .Rprofile file. Setting system variable LANGUAGE LANG, e.g. adding LANGUAGE=\"de_DE.utf8\" .Renviron file home directory. Thus, package option AMR_locale set, system variables LANGUAGE LANG ignored.","code":""},{"path":"https://msberends.github.io/AMR/reference/translate.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Translate Strings from the AMR Package — translate","text":"","code":"# Current settings (based on system language) ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacin\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Coagulase-negative Staphylococcus (CoNS)\"  # setting another language set_AMR_locale(\"Dutch\") #> ℹ Using Dutch (Nederlands) for the AMR package for this session. ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacine\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Coagulase-negatieve Staphylococcus (CNS)\"  # setting yet another language set_AMR_locale(\"German\") #> ℹ Using German (Deutsch) for the AMR package for this session. ab_name(\"Ciprofloxacin\") #> [1] \"Ciprofloxacin\" mo_name(\"Coagulase-negative Staphylococcus (CoNS)\") #> [1] \"Koagulase-negative Staphylococcus (KNS)\"  # set_AMR_locale() understands endonyms, English exonyms, and ISO-639-1: set_AMR_locale(\"Deutsch\") #> ℹ Using German (Deutsch) for the AMR package for this session. set_AMR_locale(\"German\") #> ℹ Using German (Deutsch) for the AMR package for this session. set_AMR_locale(\"de\") #> ℹ Using German (Deutsch) for the AMR package for this session. ab_name(\"amox/clav\") #> ℹ More than one result was found for item 1: amoxicillin and clavulansäure #> [1] \"Amoxicillin/Clavulansäure\"  # reset to system default reset_AMR_locale() #> ℹ Using the English language (English) for the AMR package for this #>   session. ab_name(\"amox/clav\") #> [1] \"Amoxicillin/clavulanic acid\""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"amr-2119154","dir":"Changelog","previous_headings":"","what":"AMR 2.1.1.9154","title":"AMR 2.1.1.9154","text":"(beta version eventually become v3.0. ’re happy reach new major milestone soon, new One Health support! Install beta using instructions .)","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"a-new-milestone-amr-v30-with-one-health-support--human--veterinary--environmental-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"A New Milestone: AMR v3.0 with One Health Support (= Human + Veterinary + Environmental)","title":"AMR 2.1.1.9154","text":"package now supports tools AMR data analysis clinical settings, also veterinary environmental microbiology. made possible collaboration University Prince Edward Island’s Atlantic Veterinary College, Canada. celebrate great improvement package, also updated package logo reflect change.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"breaking-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"Breaking","title":"AMR 2.1.1.9154","text":"Removed functions references used deprecated rsi class, replaced sir equivalents two years ago","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"new-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"New","title":"AMR 2.1.1.9154","text":"Function .sir() now extensive support veterinary breakpoints CLSI. Use breakpoint_type = \"animal\" set host argument variable contains animal species names. CLSI VET09 guideline implemented address cases veterinary breakpoints missing (applies guideline set CLSI) clinical_breakpoints data set contains breakpoints, can downloaded download page. antibiotics data set contains veterinary antibiotics, pradofloxacin enrofloxacin. WHOCC codes veterinary use added well. ab_atc() now supports ATC codes veterinary antibiotics (start “Q”) ab_url() now supports retrieving WHOCC url ATCvet pages antibiogram() function now supports creating true Weighted-Incidence Syndromic Combination Antibiograms (WISCA), powerful Bayesian method estimating regimen coverage probabilities using pathogen incidence antimicrobial susceptibility data. WISCA offers improved precision syndrome-specific treatment, even datasets sparse data. dedicated wisca() function also available easy usage. MycoBank now integrated primary taxonomic source fungi. microorganisms data set enriched new columns (mycobank, mycobank_parent, mycobank_renamed_to) provide detailed information fungal species. remarkable addition 20,000 new fungal records New function mo_mycobank() retrieve MycoBank record number, analogous existing functions mo_lpsn() mo_gbif(). .mo() function mo_*() functions now include only_fungi argument, allowing users restrict results solely fungal species. ensures fungi prioritised bacteria microorganism identification. can also set globally new AMR_only_fungi option. Also updated kingdoms, welcoming total 2,149 new records 2023 927 2024. EUCAST 2024 CLSI 2024 now supported, adding 4,000 new clinical breakpoints clinical_breakpoints data set usage .sir(). EUCAST 2024 now new default guideline MIC disk diffusion interpretations. .sir() now brings additional factor levels: “NI” non-interpretable “SDD” susceptible dose-dependent. Currently, clinical_breakpoints data set contains 24 breakpoints can return value “SDD” instead “”. New function group scale_*_mic(), namely: scale_x_mic(), scale_y_mic(), scale_colour_mic() scale_fill_mic(). allow easy plotting MIC values. allow manual range definition plotting missing intermediate log2 levels. New function group scale_*_sir(), namely: scale_x_sir(), scale_colour_sir() scale_fill_sir(). allow plot sir class, translates system language default. also set colourblind-safe colours plots. New function rescale_mic(), allows users rescale MIC values manually set range. powerhouse behind scale_*_mic() functions, can used independently , instance, compare equality MIC distributions rescaling range first. using R heavy lifting, ‘AMR’ Python Package developed run AMR R package natively Python. Python package always version number R package, built automatically every code change. antimicrobial selectors (aminoglycosides() betalactams()) now supported tidymodels packages recipe parsnip. See info tutorial using AMR functions predictive modelling. New function top_n_microorganisms() filter data set top n taxonomic property, e.g., filter top 3 species, filter species top 5 genera, filter top 3 species top 5 genera New function mo_group_members() retrieve member microorganisms microorganism group. example, mo_group_members(\"Strep group C\") returns vector microorganisms belong group.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"changed-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"Changed","title":"AMR 2.1.1.9154","text":"now possible use column names argument ab, mo, uti: .sir(..., ab = \"column1\", mo = \"column2\", uti = \"column3\"). greatly improves flexibility users. Users can now set criteria (using regular expressions) considered S, , R, SDD, NI. get quantitative values, .double() sir object return 1 S, 2 SDD/, 3 R (NI become NA). functions using sir classes (e.g., summary()) updated reflect change contain NI SDD. Fix conserve_capped_values, now works expected: MIC values, <x always S, >x always R New argument formatting_type set 22 options formatting ‘cells’. defaults 10 non-WISCA 14 WISCA, changing output antibiograms cells info. reason, add_total_n now FALSE default since denominators added cells ab_transform argument now defaults \"name\", displaying antibiotic column names instead codes ‘Antibiotic selectors’ now called ‘antimicrobial selectors’ since scope broader just antibiotics. documentation updated, ab_class() ab_selector() replaced amr_class() amr_selector(). old functions now deprecated removed future version. Added selectors isoxazolylpenicillins(), monobactams(), nitrofurans(), phenicols(), rifamycins() using antimicrobial selectors exclude non-treatable drugs (gentamicin-high using aminoglycosides()), function now always returns warning can included using only_treatable = FALSE Added new argument return_all selectors, defaults TRUE include match. FALSE, old behaviour, first hit unique antimicrobial returned. selectors can now run separate command retrieve vector possible antimicrobials selector can select selectors lincosamides() macrolides() overlap anymore - antibiotic now classified either Added “clindamycin inducible screening” CLI1. Since clindamycin lincosamide, antimicrobial selector lincosamides() now contains argument only_treatable = TRUE (similar antibiotic selectors contain non-treatable drugs) Added Amorolfine (AMO, D01AE16), now also part antifungals() selector Added Efflux (EFF), allow mapping AMRFinderPlus Added Tigemonam (TNM), monobactam Added valid levels: 4096, 6 powers 0.0625, 5 powers 192 (192, 384, 576, 768, 960) Fixed bug .mic() failed translation scientifically formatted numbers Added new argument keep_operators .mic(). can \"\" (default), \"none\", \"edges\". argument also available new rescale_mic() scale_*_mic() functions. Comparisons MIC values now strict. example, >32 higher (never equal ) 32. Thus, .mic(\">32\") == .mic(32) now returns FALSE, .mic(\">32\") > .mic(32) now returns TRUE. Sorting MIC values (using sort()) fixed manner; <0.001 now gets sorted 0.001, >0.001 gets sorted 0.001. Intermediate log2 levels used MIC plotting now common values instead following strict dilution range eucast_rules() now argument overwrite (default: TRUE) indicate whether non-NA values overwritten Disks 0 5 mm now allowed, newly allowed range disk diffusion (.disk()) now 0 50 mm Updated italicise_taxonomy() support HTML output custom_eucast_rules() now supports multiple antibiotics antibiotic groups affected single rule mo_info() now contains extra element rank group_members (contents new mo_group_members() function) Updated ATC codes WHOCC Updated antibiotic DDDs WHOCC Added 1,500 trade names antibiotics Fix using manual value mo_transform antibiogram() Fixed bug antibiogram() returns empty data set Fix mapping ‘high level’ antibiotics .ab() (amphotericin B-high, gentamicin-high, kanamycin-high, streptomycin-high, tobramycin-high) Improved overall algorithm .ab() better performance accuracy weight given genus species combinations cases subspecies miswritten, result correct genus species Genera World Health Organization’s () Priority Pathogen List now highest prevalence Fixed bug sir_confidence_interval() isolates available Updated prevalence calculation include genera World Health Organization’s () Priority Pathogen List Improved algorithm first_isolate() using phenotype-based method, prioritise records highest availability SIR values scale_y_percent() can now cope ranges outside 0-100% range Implemented new Dutch national MDRO guideline (SRI-richtlijn BRMO, Nov 2024) Added arguments esbl, carbapenemase, mecA, mecC, vanA, vanB denote column names logical values indicating presence genes (production proteins) Added console colours support sir class Positron","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"other-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"Other","title":"AMR 2.1.1.9154","text":"Added Dr. Larisse Bolton contributor fantastic implementation WISCA mathematically solid way Added Matthew Saab, Dr. Jordan Stull, Prof. Javier Sanchez contributors tremendous input veterinary breakpoints interpretations Greatly improved vctrs integration, Tidyverse package working background many Tidyverse functions. users, means functions dplyr’s bind_rows(), rowwise() c_across() now supported e.g. columns class mic. Despite , AMR package still zero-dependent package, including dplyr vctrs. Greatly updated expanded documentation Stopped support SAS (.xpt) files, since file structure extremely inefficient requires disk space GitHub allows single commit.","code":""},{"path":"https://msberends.github.io/AMR/news/index.html","id":"older-versions-2-1-1-9154","dir":"Changelog","previous_headings":"","what":"Older Versions","title":"AMR 2.1.1.9154","text":"changelog contains changes AMR v3.0 (February 2025) later. prior v2 versions, please see v2 archive. prior v1 versions, please see v1 archive.","code":""}]