AMR/docs/articles/resistance_predict.html

370 lines
21 KiB
HTML
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!DOCTYPE html>
<!-- Generated by pkgdown: do not edit by hand --><html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>How to predict antimicrobial resistance • AMR (for R)</title>
<!-- favicons --><link rel="icon" type="image/png" sizes="16x16" href="../favicon-16x16.png">
<link rel="icon" type="image/png" sizes="32x32" href="../favicon-32x32.png">
<link rel="apple-touch-icon" type="image/png" sizes="180x180" href="../apple-touch-icon.png">
<link rel="apple-touch-icon" type="image/png" sizes="120x120" href="../apple-touch-icon-120x120.png">
<link rel="apple-touch-icon" type="image/png" sizes="76x76" href="../apple-touch-icon-76x76.png">
<link rel="apple-touch-icon" type="image/png" sizes="60x60" href="../apple-touch-icon-60x60.png">
<!-- jquery --><script src="https://cdnjs.cloudflare.com/ajax/libs/jquery/3.4.1/jquery.min.js" integrity="sha256-CSXorXvZcTkaix6Yvo6HppcZGetbYMGWSFlBw8HfCJo=" crossorigin="anonymous"></script><!-- Bootstrap --><link href="https://cdnjs.cloudflare.com/ajax/libs/bootswatch/3.4.0/flatly/bootstrap.min.css" rel="stylesheet" crossorigin="anonymous">
<script src="https://cdnjs.cloudflare.com/ajax/libs/twitter-bootstrap/3.4.1/js/bootstrap.min.js" integrity="sha256-nuL8/2cJ5NDSSwnKD8VqreErSWHtnEP9E7AySL+1ev4=" crossorigin="anonymous"></script><!-- bootstrap-toc --><link rel="stylesheet" href="../bootstrap-toc.css">
<script src="../bootstrap-toc.js"></script><!-- Font Awesome icons --><link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/all.min.css" integrity="sha256-mmgLkCYLUQbXn0B1SRqzHar6dCnv9oZFPEC1g1cwlkk=" crossorigin="anonymous">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.12.1/css/v4-shims.min.css" integrity="sha256-wZjR52fzng1pJHwx4aV2AO3yyTOXrcDW7jBpJtTwVxw=" crossorigin="anonymous">
<!-- clipboard.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/clipboard.js/2.0.6/clipboard.min.js" integrity="sha256-inc5kl9MA1hkeYUt+EC3BhlIgyp/2jDIyBLS6k3UxPI=" crossorigin="anonymous"></script><!-- headroom.js --><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/headroom.min.js" integrity="sha256-AsUX4SJE1+yuDu5+mAVzJbuYNPHj/WroHuZ8Ir/CkE0=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/headroom/0.11.0/jQuery.headroom.min.js" integrity="sha256-ZX/yNShbjqsohH1k95liqY9Gd8uOiE1S4vZc+9KQ1K4=" crossorigin="anonymous"></script><!-- pkgdown --><link href="../pkgdown.css" rel="stylesheet">
<script src="../pkgdown.js"></script><link href="../extra.css" rel="stylesheet">
<script src="../extra.js"></script><meta property="og:title" content="How to predict antimicrobial resistance">
<meta property="og:description" content="AMR">
<meta property="og:image" content="https://msberends.gitlab.io/AMR/logo.svg">
<!-- mathjax --><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js" integrity="sha256-nvJJv9wWKEm88qvoQl9ekL2J+k/RWIsaSScxxlsrv8k=" crossorigin="anonymous"></script><script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/config/TeX-AMS-MML_HTMLorMML.js" integrity="sha256-84DKXVJXs0/F8OTMzX4UR909+jtl4G7SPypPavF+GfA=" crossorigin="anonymous"></script><!--[if lt IE 9]>
<script src="https://oss.maxcdn.com/html5shiv/3.7.3/html5shiv.min.js"></script>
<script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
<![endif]-->
</head>
<body data-spy="scroll" data-target="#toc">
<div class="container template-article">
<header><div class="navbar navbar-default navbar-fixed-top" role="navigation">
<div class="container">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar" aria-expanded="false">
<span class="sr-only">Toggle navigation</span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<span class="navbar-brand">
<a class="navbar-link" href="../index.html">AMR (for R)</a>
<span class="version label label-default" data-toggle="tooltip" data-placement="bottom" title="Latest development version">1.1.0</span>
</span>
</div>
<div id="navbar" class="navbar-collapse collapse">
<ul class="nav navbar-nav">
<li>
<a href="../index.html">
<span class="fa fa-home"></span>
Home
</a>
</li>
<li class="dropdown">
<a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-expanded="false">
<span class="fa fa-question-circle"></span>
How to
<span class="caret"></span>
</a>
<ul class="dropdown-menu" role="menu">
<li>
<a href="../articles/AMR.html">
<span class="fa fa-directions"></span>
Conduct AMR analysis
</a>
</li>
<li>
<a href="../articles/resistance_predict.html">
<span class="fa fa-dice"></span>
Predict antimicrobial resistance
</a>
</li>
<li>
<a href="../articles/PCA.html">
<span class="fa fa-compress"></span>
Conduct principal component analysis for AMR
</a>
</li>
<li>
<a href="../articles/MDR.html">
<span class="fa fa-skull-crossbones"></span>
Determine multi-drug resistance (MDR)
</a>
</li>
<li>
<a href="../articles/WHONET.html">
<span class="fa fa-globe-americas"></span>
Work with WHONET data
</a>
</li>
<li>
<a href="../articles/SPSS.html">
<span class="fa fa-file-upload"></span>
Import data from SPSS/SAS/Stata
</a>
</li>
<li>
<a href="../articles/EUCAST.html">
<span class="fa fa-exchange-alt"></span>
Apply EUCAST rules
</a>
</li>
<li>
<a href="../reference/mo_property.html">
<span class="fa fa-bug"></span>
Get properties of a microorganism
</a>
</li>
<li>
<a href="../reference/ab_property.html">
<span class="fa fa-capsules"></span>
Get properties of an antibiotic
</a>
</li>
<li>
<a href="../articles/benchmarks.html">
<span class="fa fa-shipping-fast"></span>
Other: benchmarks
</a>
</li>
</ul>
</li>
<li>
<a href="../reference/">
<span class="fa fa-book-open"></span>
Manual
</a>
</li>
<li>
<a href="../authors.html">
<span class="fa fa-users"></span>
Authors
</a>
</li>
<li>
<a href="../news/">
<span class="far fa far fa-newspaper"></span>
Changelog
</a>
</li>
</ul>
<ul class="nav navbar-nav navbar-right">
<li>
<a href="https://gitlab.com/msberends/AMR">
<span class="fab fa fab fa-gitlab"></span>
Source Code
</a>
</li>
<li>
<a href="../LICENSE-text.html">
<span class="fa fa-book"></span>
Licence
</a>
</li>
</ul>
</div>
<!--/.nav-collapse -->
</div>
<!--/.container -->
</div>
<!--/.navbar -->
</header><div class="row">
<div class="col-md-9 contents">
<div class="page-header toc-ignore">
<h1 data-toc-skip>How to predict antimicrobial resistance</h1>
<h4 class="author">Matthijs S. Berends</h4>
<h4 class="date">15 April 2020</h4>
<small class="dont-index">Source: <a href="https://gitlab.com/msberends/AMR/blob/master/vignettes/resistance_predict.Rmd"><code>vignettes/resistance_predict.Rmd</code></a></small>
<div class="hidden name"><code>resistance_predict.Rmd</code></div>
</div>
<div id="needed-r-packages" class="section level2">
<h2 class="hasAnchor">
<a href="#needed-r-packages" class="anchor"></a>Needed R packages</h2>
<p>As with many uses in R, we need some additional packages for AMR analysis. Our package works closely together with the <a href="https://www.tidyverse.org">tidyverse packages</a> <a href="https://dplyr.tidyverse.org/"><code>dplyr</code></a> and <a href="https://ggplot2.tidyverse.org"><code>ggplot2</code></a> by Dr Hadley Wickham. The tidyverse tremendously improves the way we conduct data science - it allows for a very natural way of writing syntaxes and creating beautiful plots in R.</p>
<p>Our <code>AMR</code> package depends on these packages and even extends their use and functions.</p>
<div class="sourceCode" id="cb1"><html><body><pre class="r"><span class="fu"><a href="https://rdrr.io/r/base/library.html">library</a></span>(<span class="no">dplyr</span>)
<span class="fu"><a href="https://rdrr.io/r/base/library.html">library</a></span>(<span class="no">ggplot2</span>)
<span class="fu"><a href="https://rdrr.io/r/base/library.html">library</a></span>(<span class="no">AMR</span>)
<span class="co"># (if not yet installed, install with:)</span>
<span class="co"># install.packages(c("tidyverse", "AMR"))</span></pre></body></html></div>
</div>
<div id="prediction-analysis" class="section level2">
<h2 class="hasAnchor">
<a href="#prediction-analysis" class="anchor"></a>Prediction analysis</h2>
<p>Our package contains a function <code><a href="../reference/resistance_predict.html">resistance_predict()</a></code>, which takes the same input as functions for <a href="./AMR.html">other AMR analysis</a>. Based on a date column, it calculates cases per year and uses a regression model to predict antimicrobial resistance.</p>
<p>It is basically as easy as:</p>
<div class="sourceCode" id="cb2"><html><body><pre class="r"># resistance prediction of piperacillin/tazobactam (TZP):
resistance_predict(tbl = example_isolates, col_date = "date", col_ab = "TZP", model = "binomial")
# or:
example_isolates %&gt;%
resistance_predict(col_ab = "TZP",
model "binomial")
# to bind it to object 'predict_TZP' for example:
predict_TZP %
resistance_predict(col_ab = "TZP",
model = "binomial")</pre></body></html></div>
<p>The function will look for a date column itself if <code>col_date</code> is not set.</p>
<p>When running any of these commands, a summary of the regression model will be printed unless using <code><a href="../reference/resistance_predict.html">resistance_predict(..., info = FALSE)</a></code>.</p>
<pre><code># NOTE: Using column `date` as input for `col_date`.</code></pre>
<p>This text is only a printed summary - the actual result (output) of the function is a <code>data.frame</code> containing for each year: the number of observations, the actual observed resistance, the estimated resistance and the standard error below and above the estimation:</p>
<div class="sourceCode" id="cb4"><html><body><pre class="r"><span class="no">predict_TZP</span>
<span class="co"># year value se_min se_max observations observed estimated</span>
<span class="co"># 1 2003 0.06250000 NA NA 32 0.06250000 0.05486389</span>
<span class="co"># 2 2004 0.08536585 NA NA 82 0.08536585 0.06089002</span>
<span class="co"># 3 2005 0.05000000 NA NA 60 0.05000000 0.06753075</span>
<span class="co"># 4 2006 0.05084746 NA NA 59 0.05084746 0.07483801</span>
<span class="co"># 5 2007 0.12121212 NA NA 66 0.12121212 0.08286570</span>
<span class="co"># 6 2008 0.04166667 NA NA 72 0.04166667 0.09166918</span>
<span class="co"># 7 2009 0.01639344 NA NA 61 0.01639344 0.10130461</span>
<span class="co"># 8 2010 0.05660377 NA NA 53 0.05660377 0.11182814</span>
<span class="co"># 9 2011 0.18279570 NA NA 93 0.18279570 0.12329488</span>
<span class="co"># 10 2012 0.30769231 NA NA 65 0.30769231 0.13575768</span>
<span class="co"># 11 2013 0.06896552 NA NA 58 0.06896552 0.14926576</span>
<span class="co"># 12 2014 0.10000000 NA NA 60 0.10000000 0.16386307</span>
<span class="co"># 13 2015 0.23636364 NA NA 55 0.23636364 0.17958657</span>
<span class="co"># 14 2016 0.22619048 NA NA 84 0.22619048 0.19646431</span>
<span class="co"># 15 2017 0.16279070 NA NA 86 0.16279070 0.21451350</span>
<span class="co"># 16 2018 0.23373852 0.2021578 0.2653193 NA NA 0.23373852</span>
<span class="co"># 17 2019 0.25412909 0.2168525 0.2914057 NA NA 0.25412909</span>
<span class="co"># 18 2020 0.27565854 0.2321869 0.3191302 NA NA 0.27565854</span>
<span class="co"># 19 2021 0.29828252 0.2481942 0.3483709 NA NA 0.29828252</span>
<span class="co"># 20 2022 0.32193804 0.2649008 0.3789753 NA NA 0.32193804</span>
<span class="co"># 21 2023 0.34654311 0.2823269 0.4107593 NA NA 0.34654311</span>
<span class="co"># 22 2024 0.37199700 0.3004860 0.4435080 NA NA 0.37199700</span>
<span class="co"># 23 2025 0.39818127 0.3193839 0.4769787 NA NA 0.39818127</span>
<span class="co"># 24 2026 0.42496142 0.3390173 0.5109056 NA NA 0.42496142</span>
<span class="co"># 25 2027 0.45218939 0.3593720 0.5450068 NA NA 0.45218939</span>
<span class="co"># 26 2028 0.47970658 0.3804212 0.5789920 NA NA 0.47970658</span>
<span class="co"># 27 2029 0.50734745 0.4021241 0.6125708 NA NA 0.50734745</span>
<span class="co"># 28 2030 0.53494347 0.4244247 0.6454622 NA NA 0.53494347</span></pre></body></html></div>
<p>The function <code>plot</code> is available in base R, and can be extended by other packages to depend the output based on the type of input. We extended its function to cope with resistance predictions:</p>
<div class="sourceCode" id="cb5"><html><body><pre class="r"><span class="fu"><a href="https://rdrr.io/r/graphics/plot.html">plot</a></span>(<span class="no">predict_TZP</span>)</pre></body></html></div>
<p><img src="resistance_predict_files/figure-html/unnamed-chunk-4-1.png" width="720"></p>
<p>This is the fastest way to plot the result. It automatically adds the right axes, error bars, titles, number of available observations and type of model.</p>
<p>We also support the <code>ggplot2</code> package with our custom function <code><a href="../reference/resistance_predict.html">ggplot_rsi_predict()</a></code> to create more appealing plots:</p>
<div class="sourceCode" id="cb6"><html><body><pre class="r"><span class="fu"><a href="../reference/resistance_predict.html">ggplot_rsi_predict</a></span>(<span class="no">predict_TZP</span>)</pre></body></html></div>
<p><img src="resistance_predict_files/figure-html/unnamed-chunk-5-1.png" width="720"></p>
<div class="sourceCode" id="cb7"><html><body><pre class="r">
<span class="co"># choose for error bars instead of a ribbon</span>
<span class="fu"><a href="../reference/resistance_predict.html">ggplot_rsi_predict</a></span>(<span class="no">predict_TZP</span>, <span class="kw">ribbon</span> <span class="kw">=</span> <span class="fl">FALSE</span>)</pre></body></html></div>
<p><img src="resistance_predict_files/figure-html/unnamed-chunk-5-2.png" width="720"></p>
<div id="choosing-the-right-model" class="section level3">
<h3 class="hasAnchor">
<a href="#choosing-the-right-model" class="anchor"></a>Choosing the right model</h3>
<p>Resistance is not easily predicted; if we look at vancomycin resistance in Gram positives, the spread (i.e. standard error) is enormous:</p>
<div class="sourceCode" id="cb8"><html><body><pre class="r"><span class="no">example_isolates</span> <span class="kw">%&gt;%</span>
<span class="fu"><a href="https://dplyr.tidyverse.org/reference/filter.html">filter</a></span>(<span class="fu"><a href="../reference/mo_property.html">mo_gramstain</a></span>(<span class="no">mo</span>, <span class="kw">language</span> <span class="kw">=</span> <span class="kw">NULL</span>) <span class="kw">==</span> <span class="st">"Gram-positive"</span>) <span class="kw">%&gt;%</span>
<span class="fu"><a href="../reference/resistance_predict.html">resistance_predict</a></span>(<span class="kw">col_ab</span> <span class="kw">=</span> <span class="st">"VAN"</span>, <span class="kw">year_min</span> <span class="kw">=</span> <span class="fl">2010</span>, <span class="kw">info</span> <span class="kw">=</span> <span class="fl">FALSE</span>, <span class="kw">model</span> <span class="kw">=</span> <span class="st">"binomial"</span>) <span class="kw">%&gt;%</span>
<span class="fu"><a href="../reference/resistance_predict.html">ggplot_rsi_predict</a></span>()
<span class="co"># NOTE: Using column `date` as input for `col_date`.</span></pre></body></html></div>
<p><img src="resistance_predict_files/figure-html/unnamed-chunk-6-1.png" width="720"></p>
<p>Vancomycin resistance could be 100% in ten years, but might also stay around 0%.</p>
<p>You can define the model with the <code>model</code> parameter. The model chosen above is a generalised linear regression model using a binomial distribution, assuming that a period of zero resistance was followed by a period of increasing resistance leading slowly to more and more resistance.</p>
<p>Valid values are:</p>
<table class="table">
<colgroup>
<col width="32%">
<col width="25%">
<col width="42%">
</colgroup>
<thead><tr class="header">
<th>Input values</th>
<th>Function used by R</th>
<th>Type of model</th>
</tr></thead>
<tbody>
<tr class="odd">
<td>
<code>"binomial"</code> or <code>"binom"</code> or <code>"logit"</code>
</td>
<td><code><a href="https://rdrr.io/r/stats/glm.html">glm(..., family = binomial)</a></code></td>
<td>Generalised linear model with binomial distribution</td>
</tr>
<tr class="even">
<td>
<code>"loglin"</code> or <code>"poisson"</code>
</td>
<td><code><a href="https://rdrr.io/r/stats/glm.html">glm(..., family = poisson)</a></code></td>
<td>Generalised linear model with poisson distribution</td>
</tr>
<tr class="odd">
<td>
<code>"lin"</code> or <code>"linear"</code>
</td>
<td><code><a href="https://rdrr.io/r/stats/lm.html">lm()</a></code></td>
<td>Linear model</td>
</tr>
</tbody>
</table>
<p>For the vancomycin resistance in Gram positive bacteria, a linear model might be more appropriate since no (left half of a) binomial distribution is to be expected based on the observed years:</p>
<div class="sourceCode" id="cb9"><html><body><pre class="r"><span class="no">example_isolates</span> <span class="kw">%&gt;%</span>
<span class="fu"><a href="https://dplyr.tidyverse.org/reference/filter.html">filter</a></span>(<span class="fu"><a href="../reference/mo_property.html">mo_gramstain</a></span>(<span class="no">mo</span>, <span class="kw">language</span> <span class="kw">=</span> <span class="kw">NULL</span>) <span class="kw">==</span> <span class="st">"Gram-positive"</span>) <span class="kw">%&gt;%</span>
<span class="fu"><a href="../reference/resistance_predict.html">resistance_predict</a></span>(<span class="kw">col_ab</span> <span class="kw">=</span> <span class="st">"VAN"</span>, <span class="kw">year_min</span> <span class="kw">=</span> <span class="fl">2010</span>, <span class="kw">info</span> <span class="kw">=</span> <span class="fl">FALSE</span>, <span class="kw">model</span> <span class="kw">=</span> <span class="st">"linear"</span>) <span class="kw">%&gt;%</span>
<span class="fu"><a href="../reference/resistance_predict.html">ggplot_rsi_predict</a></span>()
<span class="co"># NOTE: Using column `date` as input for `col_date`.</span></pre></body></html></div>
<p><img src="resistance_predict_files/figure-html/unnamed-chunk-7-1.png" width="720"></p>
<p>This seems more likely, doesnt it?</p>
<p>The model itself is also available from the object, as an <code>attribute</code>:</p>
<div class="sourceCode" id="cb10"><html><body><pre class="r"><span class="no">model</span> <span class="kw">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/attributes.html">attributes</a></span>(<span class="no">predict_TZP</span>)$<span class="no">model</span>
<span class="fu"><a href="https://rdrr.io/r/base/summary.html">summary</a></span>(<span class="no">model</span>)$<span class="no">family</span>
<span class="co"># </span>
<span class="co"># Family: binomial </span>
<span class="co"># Link function: logit</span>
<span class="fu"><a href="https://rdrr.io/r/base/summary.html">summary</a></span>(<span class="no">model</span>)$<span class="no">coefficients</span>
<span class="co"># Estimate Std. Error z value Pr(&gt;|z|)</span>
<span class="co"># (Intercept) -224.3987194 48.0335384 -4.671709 2.987038e-06</span>
<span class="co"># year 0.1106102 0.0238753 4.632831 3.606990e-06</span></pre></body></html></div>
</div>
</div>
</div>
<div class="col-md-3 hidden-xs hidden-sm" id="pkgdown-sidebar">
<nav id="toc" data-toggle="toc"><h2 data-toc-skip>Contents</h2>
</nav>
</div>
</div>
<footer><div class="copyright">
<p>Developed by <a href="https://www.rug.nl/staff/m.s.berends/">Matthijs S. Berends</a>, <a href="https://www.rug.nl/staff/c.f.luz/">Christian F. Luz</a>, <a href="https://www.rug.nl/staff/a.w.friedrich/">Alexander W. Friedrich</a>, <a href="https://www.rug.nl/staff/b.sinha/">Bhanu N. M. Sinha</a>, <a href="https://www.rug.nl/staff/c.j.albers/">Casper J. Albers</a>, <a href="https://www.rug.nl/staff/c.glasner/">Corinna Glasner</a>.</p>
</div>
<div class="pkgdown">
<p>Site built with <a href="https://pkgdown.r-lib.org/">pkgdown</a> 1.5.0.</p>
</div>
</footer>
</div>
</body>
</html>