(v2.1.1.9157) improved as.ab(), fixed knit_print of antibiogram

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: AMR
-Version: 2.1.1.9156
-Date: 2025-02-23
+Version: 2.1.1.9157
+Date: 2025-02-26
 Title: Antimicrobial Resistance Data Analysis
 Description: Functions to simplify and standardise antimicrobial resistance (AMR)
   data analysis and to work with microbial and antimicrobial properties by

NAMESPACE

@@ -140,6 +140,7 @@ export(ab_info)
 export(ab_loinc)
 export(ab_name)
 export(ab_property)
+export(ab_reset_session)
 export(ab_selector)
 export(ab_synonyms)
 export(ab_tradenames)

NEWS.md

@@ -1,4 +1,4 @@
-# AMR 2.1.1.9156
+# AMR 2.1.1.9157
 
 *(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 [the instructions here](https://msberends.github.io/AMR/#latest-development-version).)*
 

PythonPackage/AMR/AMR.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: AMR
-Version: 2.1.1.9156
+Version: 2.1.1.9157
 Summary: A Python wrapper for the AMR R package
 Home-page: https://github.com/msberends/AMR
 Author: Matthijs Berends

PythonPackage/AMR/AMR/__init__.py

@@ -66,6 +66,7 @@ from .functions import administrable_iv
 from .functions import not_intrinsic_resistant
 from .functions import as_ab
 from .functions import is_ab
+from .functions import ab_reset_session
 from .functions import as_av
 from .functions import is_av
 from .functions import as_disk

PythonPackage/AMR/AMR/functions.py

@@ -228,6 +228,9 @@ def as_ab(x, *args, **kwargs):
 def is_ab(x):
     """See our website of the R package for the manual: https://msberends.github.io/AMR/index.html"""
     return convert_to_python(amr_r.is_ab(x))
+def ab_reset_session(*args, **kwargs):
+    """See our website of the R package for the manual: https://msberends.github.io/AMR/index.html"""
+    return convert_to_python(amr_r.ab_reset_session(*args, **kwargs))
 def as_av(x, *args, **kwargs):
     """See our website of the R package for the manual: https://msberends.github.io/AMR/index.html"""
     return convert_to_python(amr_r.as_av(x, *args, **kwargs))

PythonPackage/AMR/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name='AMR',
-    version='2.1.1.9156',
+    version='2.1.1.9157',
     packages=find_packages(),
     install_requires=[
         'rpy2',

R/ab.R

@@ -97,11 +97,9 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
   meet_criteria(flag_multiple_results, allow_class = "logical", has_length = 1)
   meet_criteria(info, allow_class = "logical", has_length = 1)
 
-  if (is.ab(x)) {
-    return(x)
-  }
-  if (all(x %in% c(AMR_env$AB_lookup$ab, NA))) {
-    # all valid AB codes, but not yet right class
+  if (is.ab(x) || all(x %in% c(AMR_env$AB_lookup$ab, NA))) {
+    # all valid AB codes, but not yet right class or might have additional attributes as AMR selector
+    attributes(x) <- NULL
     return(set_clean_class(x,
       new_class = c("ab", "character")
     ))
@@ -130,6 +128,7 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
 
   x <- unique(x_bak_clean) # this means that every x is in fact generalise_antibiotic_name(x)
   x_new <- rep(NA_character_, length(x))
+  x_uncertain <- character(0)
   x_unknown <- character(0)
   x_unknown_ATCs <- character(0)
   
@@ -176,6 +175,14 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
   x_new[known_codes_cid] <- AMR_env$AB_lookup$ab[match(x[known_codes_cid], AMR_env$AB_lookup$cid)]
   previously_coerced <- x %in% AMR_env$ab_previously_coerced$x
   x_new[previously_coerced & is.na(x_new)] <- AMR_env$ab_previously_coerced$ab[match(x[is.na(x_new) & x %in% AMR_env$ab_previously_coerced$x], AMR_env$ab_previously_coerced$x)]
+  prev <- x_bak[which(x[which(previously_coerced)] %in% x_bak_clean)]
+  if (any(previously_coerced) && isTRUE(info) && message_not_thrown_before("as.ab", prev, entire_session = TRUE)) {
+    message_(
+      "Returning previously coerced value", ifelse(length(unique(prev)) > 1, "s", ""),
+      " for ", vector_and(prev), ". Run `ab_reset_session()` to reset this. This note will be shown once per session for this input."
+    )
+  }
+
   already_known <- known_names | known_codes_ab | known_codes_atc | known_codes_cid | previously_coerced
 
   # fix for NAs
@@ -325,6 +332,18 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
     if (loop_time <= 2 && fast_mode == FALSE) {
       # only run on first and second try
       
+      # base on the Levensthein distance function if length >= 6
+      if (nchar(x[i]) >= 6) {
+        l_dist <- as.double(utils::adist(x[i], AMR_env$AB_lookup$generalised_name,
+                                         ignore.case = FALSE,
+                                         fixed = TRUE,
+                                         costs = c(insertions = 1, deletions = 2, substitutions = 2),
+                                         counts = FALSE))
+        x_new[i] <- AMR_env$AB_lookup$ab[order(l_dist)][1]
+        x_uncertain <- c(x_uncertain, x_bak[x[i] == x_bak_clean][1])
+        next
+      }
+      
       # try by removing all spaces
       if (x[i] %like% " ") {
         found <- suppressWarnings(as.ab(gsub(" +", "", x[i], perl = TRUE), loop_time = loop_time + 2))
@@ -554,6 +573,8 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
       vector_and(x_unknown_ATCs), "."
     )
   }
+  
+  # Throw note about uncertainties
   x_unknown <- x_unknown[!x_unknown %in% x_unknown_ATCs]
   x_unknown <- c(
     x_unknown,
@@ -567,6 +588,28 @@ as.ab <- function(x, flag_multiple_results = TRUE, info = interactive(), ...) {
     )
   }
   
+  # Throw note about uncertainties
+  if (isTRUE(info) && length(x_uncertain) > 0 && fast_mode == FALSE) {
+    if (message_not_thrown_before("as.ab", "uncertainties", x_bak)) {
+      plural <- c("", "this")
+      if (length(x_uncertain) > 1) {
+        plural <- c("s", "these uncertainties")
+      }
+      if (length(x_uncertain) <= 3) {
+        examples <- vector_and(
+          paste0(
+            '"', x_uncertain, '" (assumed ', 
+            ab_name(AMR_env$ab_previously_coerced$ab[which(AMR_env$ab_previously_coerced$x_bak %in% x_uncertain)], language = NULL, tolower = TRUE),
+            ", ", AMR_env$ab_previously_coerced$ab[which(AMR_env$ab_previously_coerced$x_bak %in% x_uncertain)], ")"),
+          quotes = FALSE)
+      } else {
+        examples <- paste0(nr2char(length(x_uncertain)), " antimicrobial", plural[1])
+      }
+      message_("Antimicrobial translation was uncertain for ", examples,
+               ". If required, use `add_custom_antimicrobials()` to add custom entries.")
+    }
+  }
+
   x_result <- x_new[match(x_bak_clean, x)]
   if (length(x_result) == 0) {
     x_result <- NA_character_
@@ -583,6 +626,18 @@ is.ab <- function(x) {
   inherits(x, "ab")
 }
 
+#' @rdname as.ab
+#' @export
+ab_reset_session <- function() {
+  if (NROW(AMR_env$ab_previously_coerced) > 0) {
+    message_("Reset ", nr2char(NROW(AMR_env$ab_previously_coerced)), " previously matched input value", ifelse(NROW(AMR_env$ab_previously_coerced) > 1, "s", ""), ".")
+    AMR_env$ab_previously_coerced <- AMR_env$ab_previously_coerced[0, , drop = FALSE]
+    AMR_env$mo_uncertainties <- AMR_env$mo_uncertainties[0, , drop = FALSE]
+  } else {
+    message_("No previously matched input values to reset.")
+  }
+}
+
 # will be exported using s3_register() in R/zzz.R
 pillar_shaft.ab <- function(x, ...) {
   out <- trimws(format(x))
@@ -606,6 +661,15 @@ type_sum.ab <- function(x, ...) {
 #' @export
 #' @noRd
 print.ab <- function(x, ...) {
+  if (!is.null(attributes(x)$amr_selector)) {
+    function_name <- attributes(x)$amr_selector
+    message_("This 'ab' vector was retrieved using `" , function_name, "()`, which should normally be used inside a `dplyr` verb or `data.frame` call, e.g.:\n",
+             "  ", AMR_env$bullet_icon, " your_data %>% select(", function_name, "())\n",
+             "  ", AMR_env$bullet_icon, " your_data %>% select(column_a, column_b, ", function_name, "())\n",
+             "  ", AMR_env$bullet_icon, " your_data %>% filter(any(", function_name, "() == \"R\"))\n",
+             "  ", AMR_env$bullet_icon, " your_data[, ", function_name, "()]\n",
+             "  ", AMR_env$bullet_icon, " your_data[, c(\"column_a\", \"column_b\", ", function_name, "())]")
+  }
   cat("Class 'ab'\n")
   print(as.character(x), quote = FALSE)
 }
@@ -692,7 +756,8 @@ generalise_antibiotic_name <- function(x) {
   # non-character, space or number should be a slash
   x <- gsub("[^A-Z0-9 -)(]", "/", x, perl = TRUE)
   # correct for 'high level' antibiotics
-  x <- gsub("([^A-Z0-9/ -]+)?(HIGH(.?LE?VE?L)?|[^A-Z0-9/]H[^A-Z0-9]?L)([^A-Z0-9 -]+)?", "-HIGH", x, perl = TRUE)
+  x <- trimws(gsub("([^A-Z0-9/ -]+)?(HIGH(.?LE?VE?L)?|[^A-Z0-9/]H[^A-Z0-9]?L)([^A-Z0-9 -]+)?", "-HIGH", x, perl = TRUE))
+  x <- trimws(gsub("^(-HIGH)(.*)", "\\2\\1", x))
   # remove part between brackets if that's followed by another string
   x <- gsub("(.*)+ [(].*[)]", "\\1", x)
   # spaces around non-characters must be removed: amox + clav -> amox/clav

R/amr_selectors.R

@@ -47,7 +47,7 @@
 #' @details
 #' These functions can be used in data set calls for selecting columns and filtering rows. They work with base \R, the Tidyverse, and `data.table`. They are heavily inspired by the [Tidyverse selection helpers][tidyselect::language] such as [`everything()`][tidyselect::everything()], but are not limited to `dplyr` verbs. Nonetheless, they are very convenient to use with `dplyr` functions such as [`select()`][dplyr::select()], [`filter()`][dplyr::filter()] and [`summarise()`][dplyr::summarise()], see *Examples*.
 #' 
-#' All selectors can also be used in `tidymodels` packages such as `recipe` and `parsnip`. See for more info [our tutorial](https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html) on using these AMR functions for predictive modelling.
+#' All selectors can also be used in `tidymodels` packages such as `recipe` and `parsnip`. See for more info [our tutorial](https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html) on using antimicrobial selectors for predictive modelling.
 #'
 #' All columns in the data in which these functions are called will be searched for known antimicrobial names, abbreviations, brand names, and codes (ATC, EARS-Net, WHO, etc.) according to the [antibiotics] data set. This means that a selector such as [aminoglycosides()] will pick up column names like 'gen', 'genta', 'J01GB03', 'tobra', 'Tobracin', etc.
 #'
@@ -747,16 +747,8 @@ amr_select_exec <- function(function_name,
   
   if (is.null(vars_df)) {
     # no data found, no antimicrobials, so no input. Happens if users run e.g. `aminoglycosides()` as a separate command.
-    examples <- paste0(
-      "  ", AMR_env$bullet_icon, " your_data %>% select(", function_name, "())\n",
-      "  ", AMR_env$bullet_icon, " your_data %>% select(column_a, column_b, ", function_name, "())\n",
-      "  ", AMR_env$bullet_icon, " your_data %>% filter(any(", function_name, "() == \"R\"))\n",
-      "  ", AMR_env$bullet_icon, " your_data[, ", function_name, "()]\n",
-      "  ", AMR_env$bullet_icon, " your_data[, c(\"column_a\", \"column_b\", ", function_name, "())]")
-    message_("The function `" , function_name, "()` should be used inside a `dplyr` verb or `data.frame` call, e.g.:\n",
-             examples,
-             "\n\nNow returning a vector of all possible antimicrobials that `" , function_name, "()` can select.")
-    return(sort(abx))
+    # print.ab will cover the additional printing text
+    return(structure(sort(abx), amr_selector = function_name))
   }
   
   # get the columns with a group names in the chosen ab class

R/antibiogram.R

@@ -441,7 +441,7 @@ antibiogram.default <- function(x,
   x <- ascertain_sir_classes(x, "x")
   meet_criteria(wisca, allow_class = "logical", has_length = 1)
   if (isTRUE(wisca)) {
-    if (!is.null(mo_transform)) {
+    if (!is.null(mo_transform) && !missing(mo_transform)) {
       warning_("WISCA must be based on the species level as WISCA parameters are based on this. For that reason, `mo_transform` will be ignored.")
     }
     mo_transform <- function(x) suppressMessages(suppressWarnings(paste(mo_genus(x, keep_synonyms = TRUE, language = NULL), mo_species(x, keep_synonyms = TRUE, language = NULL))))
@@ -1245,10 +1245,14 @@ knit_print.antibiogram <- function(x, italicise = TRUE, na = getOption("knitr.ka
   meet_criteria(italicise, allow_class = "logical", has_length = 1)
   meet_criteria(na, allow_class = "character", has_length = 1, allow_NA = TRUE)
   
-  if (!isTRUE(attributes(x)$wisca) && isTRUE(italicise) && "mo" %in% colnames(attributes(x)$long_numeric)) {
+  add_MO_lookup_to_AMR_env()
+  
+  cols_with_mo_names <- vapply(FUN.VALUE = logical(1), x, function(x) any(x %in% AMR_env$MO_lookup$fullname, na.rm = TRUE))
+  if (any(cols_with_mo_names)) {
+    for (i in which(cols_with_mo_names)) {
       # make all microorganism names italic, according to nomenclature
-    names_col <- ifelse(isTRUE(attributes(x)$has_syndromic_group), 2, 1)
-    x[[names_col]] <- italicise_taxonomy(x[[names_col]], type = "markdown")
+      x[[i]] <- italicise_taxonomy(x[[i]], type = "markdown")
+    }
   }
   
   old_option <- getOption("knitr.kable.NA")

R/plotting.R

@@ -218,12 +218,10 @@ create_scale_mic <- function(aest, keep_operators, mic_range = NULL, ...) {
     as.double(rescale_mic(x = as.double(as.mic(x)), keep_operators = keep_operators, mic_range = mic_range, as.mic = TRUE))
   }
   scale$transform_df <- function(self, df) {
-    stop_if(all(is.na(df[[aest]])),
-            "`scale_", aest, "_mic()`: All MIC values are `NA`. Check your input data.", call = FALSE)
     self$mic_values_rescaled <- rescale_mic(x = as.double(as.mic(df[[aest]])), keep_operators = keep_operators, mic_range = mic_range, as.mic = TRUE)
     # create new breaks and labels here
     lims <- range(self$mic_values_rescaled, na.rm = TRUE)
-    # support inner and outer mic_range settings (e.g., data ranges 0.5-8 and mic_range is set to 0.025-64)
+    # support inner and outer 'mic_range' settings (e.g., the data ranges 0.5-8 and 'mic_range' is set to 0.025-32)
     if (!is.null(mic_range) && !is.na(mic_range[1]) && !is.na(lims[1]) && mic_range[1] < lims[1]) {
       lims[1] <- mic_range[1]
     }

data-raw/gpt_training_text_v2.1.1.9157.txt

@@ -1,6 +1,6 @@
 This knowledge base contains all context you must know about the AMR package for R. You are a GPT trained to be an assistant for the AMR package in R. You are an incredible R specialist, especially trained in this package and in the tidyverse.
 
-First and foremost, you are trained on version 2.1.1.9156. Remember this whenever someone asks which AMR package version you’re at.
+First and foremost, you are trained on version 2.1.1.9157. Remember this whenever someone asks which AMR package version you’re at.
 
 Below are the contents of the  file, the  file, and all the  files (documentation) in the package. Every file content is split using 100 hypens.
 ----------------------------------------------------------------------------------------------------
@@ -151,6 +151,7 @@ export(ab_info)
 export(ab_loinc)
 export(ab_name)
 export(ab_property)
+export(ab_reset_session)
 export(ab_selector)
 export(ab_synonyms)
 export(ab_tradenames)
@@ -2334,7 +2335,7 @@ my_data_with_all_these_columns \%>\%
 \details{
 These functions can be used in data set calls for selecting columns and filtering rows. They work with base \R, the Tidyverse, and \code{data.table}. They are heavily inspired by the \link[tidyselect:language]{Tidyverse selection helpers} such as \code{\link[tidyselect:everything]{everything()}}, but are not limited to \code{dplyr} verbs. Nonetheless, they are very convenient to use with \code{dplyr} functions such as \code{\link[dplyr:select]{select()}}, \code{\link[dplyr:filter]{filter()}} and \code{\link[dplyr:summarise]{summarise()}}, see \emph{Examples}.
 
-All selectors can also be used in \code{tidymodels} packages such as \code{recipe} and \code{parsnip}. See for more info \href{https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html}{our tutorial} on using these AMR functions for predictive modelling.
+All selectors can also be used in \code{tidymodels} packages such as \code{recipe} and \code{parsnip}. See for more info \href{https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html}{our tutorial} on using antimicrobial selectors for predictive modelling.
 
 All columns in the data in which these functions are called will be searched for known antimicrobial names, abbreviations, brand names, and codes (ATC, EARS-Net, WHO, etc.) according to the \link{antibiotics} data set. This means that a selector such as \code{\link[=aminoglycosides]{aminoglycosides()}} will pick up column names like 'gen', 'genta', 'J01GB03', 'tobra', 'Tobracin', etc.
 
@@ -2573,11 +2574,14 @@ THE PART HEREAFTER CONTAINS CONTENTS FROM FILE 'man/as.ab.Rd':
 \alias{as.ab}
 \alias{ab}
 \alias{is.ab}
+\alias{ab_reset_session}
 \title{Transform Input to an Antibiotic ID}
 \usage{
 as.ab(x, flag_multiple_results = TRUE, info = interactive(), ...)
 
 is.ab(x)
+
+ab_reset_session()
 }
 \arguments{
 \item{x}{a \link{character} vector to determine to antibiotic ID}
@@ -9013,9 +9017,6 @@ We begin by loading the required libraries and preparing the `example_isolates`
 library(tidymodels)   # For machine learning workflows, and data manipulation (dplyr, tidyr, ...)
 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
@@ -9136,7 +9137,7 @@ metrics
 - `predict()` generates predictions on the testing set.
 - `metrics()` computes evaluation metrics like accuracy and kappa.
 
-It appears we can predict the Gram based on AMR results with a `r round(metrics$.estimate[1], 3)` accuracy based on AMR results of aminoglycosides and beta-lactam antibiotics. The ROC curve looks like this:
+It appears we can predict the Gram based on AMR results with a `r round(metrics$.estimate[1], 3) * 100`% accuracy based on AMR results of aminoglycosides and beta-lactam antibiotics. The ROC curve looks like this:
 
 ```{r}
 predictions %>%

man/antimicrobial_selectors.Rd

@@ -154,7 +154,7 @@ my_data_with_all_these_columns \%>\%
 \details{
 These functions can be used in data set calls for selecting columns and filtering rows. They work with base \R, the Tidyverse, and \code{data.table}. They are heavily inspired by the \link[tidyselect:language]{Tidyverse selection helpers} such as \code{\link[tidyselect:everything]{everything()}}, but are not limited to \code{dplyr} verbs. Nonetheless, they are very convenient to use with \code{dplyr} functions such as \code{\link[dplyr:select]{select()}}, \code{\link[dplyr:filter]{filter()}} and \code{\link[dplyr:summarise]{summarise()}}, see \emph{Examples}.
 
-All selectors can also be used in \code{tidymodels} packages such as \code{recipe} and \code{parsnip}. See for more info \href{https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html}{our tutorial} on using these AMR functions for predictive modelling.
+All selectors can also be used in \code{tidymodels} packages such as \code{recipe} and \code{parsnip}. See for more info \href{https://msberends.github.io/AMR/articles/AMR_with_tidymodels.html}{our tutorial} on using antimicrobial selectors for predictive modelling.
 
 All columns in the data in which these functions are called will be searched for known antimicrobial names, abbreviations, brand names, and codes (ATC, EARS-Net, WHO, etc.) according to the \link{antibiotics} data set. This means that a selector such as \code{\link[=aminoglycosides]{aminoglycosides()}} will pick up column names like 'gen', 'genta', 'J01GB03', 'tobra', 'Tobracin', etc.
 

man/as.ab.Rd

@@ -4,11 +4,14 @@
 \alias{as.ab}
 \alias{ab}
 \alias{is.ab}
+\alias{ab_reset_session}
 \title{Transform Input to an Antibiotic ID}
 \usage{
 as.ab(x, flag_multiple_results = TRUE, info = interactive(), ...)
 
 is.ab(x)
+
+ab_reset_session()
 }
 \arguments{
 \item{x}{a \link{character} vector to determine to antibiotic ID}

vignettes/AMR_with_tidymodels.Rmd

@@ -45,9 +45,6 @@ We begin by loading the required libraries and preparing the `example_isolates`
 library(tidymodels)   # For machine learning workflows, and data manipulation (dplyr, tidyr, ...)
 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
@@ -168,7 +165,7 @@ metrics
 - `predict()` generates predictions on the testing set.
 - `metrics()` computes evaluation metrics like accuracy and kappa.
 
-It appears we can predict the Gram based on AMR results with a `r round(metrics$.estimate[1], 3)` accuracy based on AMR results of aminoglycosides and beta-lactam antibiotics. The ROC curve looks like this:
+It appears we can predict the Gram based on AMR results with a `r round(metrics$.estimate[1], 3) * 100`% accuracy based on AMR results of aminoglycosides and beta-lactam antibiotics. The ROC curve looks like this:
 
 ```{r}
 predictions %>%