Mercurial > repos > recetox > waveica

changeset 2:6480c6d5fa36 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/RECETOX/galaxytools/tree/master/tools/waveica commit 77bea6c84c5a932fa203723d4d5fea6147f06505
author recetox
date Fri, 23 Jun 2023 07:27:29 +0000
parents b77023c41c76
children dbbedb14b44c
files macros.xml test-data/feature_table_transpose_version.csv test-data/feature_table_transpose_version.parquet waveica.xml waveica_wrapper.R
diffstat 5 files changed, 122 insertions(+), 22 deletions(-) [+]
line wrap: on
line diff
--- a/macros.xml	Thu Sep 29 15:21:04 2022 +0000
+++ b/macros.xml	Fri Jun 23 07:27:29 2023 +0000
@@ -20,6 +20,12 @@
         </creator>
     </xml>
 
+    <xml name="annotation">
+        <xrefs>
+               <xref type="bio.tools">waveica</xref>
+        </xrefs>
+     </xml>
+
     <xml name="input_data">
         <param type="data" name="data" label="Feature table" format="csv,tsv,parquet" help=""/>
     </xml>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/feature_table_transpose_version.csv	Fri Jun 23 07:27:29 2023 +0000
@@ -0,0 +1,5 @@
+id,VT_160120_002,VT_160120_004,VT_160120_006,VT_160120_008,VT_160120_010
+M85T34,228520.06430737,90217.384387202,235656.75288383896,16622.9351783435,62385.0742465736
+M86T41,35646729.21543971,35735702.457215995,37021134.452711605,44302499.262606,44639738.0735709
+M86T518,2386896.97966461,2456290.69621518,8873450.40260241,2466946.89667101,2389372.85729467
+M86T539,1026645.83653468,1089246.46040563,837856.449608585,994979.069689685,954938.131337246
Binary file test-data/feature_table_transpose_version.parquet has changed
--- a/waveica.xml	Thu Sep 29 15:21:04 2022 +0000
+++ b/waveica.xml	Fri Jun 23 07:27:29 2023 +0000
@@ -1,8 +1,9 @@
-<tool id="waveica" name="WaveICA" version="@TOOL_VERSION@+galaxy2" python_template_version="3.5">
+<tool id="waveica" name="WaveICA" version="@TOOL_VERSION@+galaxy3" python_template_version="3.5">
     <description>removal of batch effects for untargeted metabolomics data</description>
     <macros>
         <import>macros.xml</import>
     </macros>
+    <expand macro="annotation"/>
     <expand macro="creator"/>
 
     <requirements>
@@ -19,6 +20,7 @@
                 #if $input_num.input_choice == "2":
                     metadata = "$input_num.input_metadata.metadata",
                     ext = "$input_num.data.ext,$input_num.input_metadata.metadata.ext",
+                    transpose = $input_num.transpose_feature_table,
                 #else:
                     ext = "$input_num.data.ext",
                 #end if
@@ -36,6 +38,7 @@
                 #if $input_num.input_choice == "2":
                     metadata = "$input_num.input_metadata.metadata",
                     ext = "$input_num.data.ext,$input_num.input_metadata.metadata.ext",
+                    transpose = $input_num.transpose_feature_table,
                 #else:
                     ext = "$input_num.data.ext",
                 #end if
@@ -61,10 +64,12 @@
                 <expand macro="input_data"/>
             </when>
             <when value="2">
+                <expand macro="input_data"/>
                 <section name="input_metadata" title="Input metadata table" expanded="true">
                     <param name="metadata" label="Input metadata" type="data" format="csv,tsv,parquet" help="" />
-                </section> 
-                <expand macro="input_data"/>
+                </section>
+                <param name = "transpose_feature_table" label="Transpose feature table" type="boolean" checked="false" 
+                truevalue="TRUE" falsevalue="FALSE" help="Swap sample names with feature names as column headers (to fit recetox-aplcms outputs)." />
             </when>
         </conditional>
         <expand macro="general_parameters"/>
@@ -93,7 +98,7 @@
             <param name="data" value="input_data.csv" ftype="csv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
@@ -104,7 +109,7 @@
             <param name="data" value="input_data.tsv" ftype="tsv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
@@ -115,7 +120,7 @@
             <param name="data" value="input_data.parquet" ftype="parquet"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
@@ -128,7 +133,7 @@
             <param name="metadata" value="metadata.csv" ftype="csv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
@@ -141,7 +146,7 @@
             <param name="metadata" value="metadata.tsv" ftype="tsv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
@@ -154,13 +159,41 @@
             <param name="metadata" value="metadata.csv" ftype="csv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
-            <param name="filter_length" value="2"/>
+            <param name="wavelet_length" value="2"/>
             <param name="k" value="20"/>
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
             <output name="normalized_data" file="normalized_data.parquet" compare="sim_size" delta="200"/>
         </test>
+        <test><!-- TEST 7 -->
+            <param name="input_choice" value="2"/>
+            <param name="data" value="feature_table_transpose_version.parquet" ftype="parquet"/>
+            <param name="metadata" value="metadata.parquet" ftype="parquet"/>
+            <param name="transpose_feature_table" value="TRUE"/>
+            <param name="mode" value="batchwise"/>
+            <param name="wavelet_filter" value="d"/>
+            <param name="wavelet_length" value="2"/>
+            <param name="k" value="20"/>
+            <param name="t" value="0.05"/>
+            <param name="t2" value="0.05"/>
+            <param name="alpha" value="0"/>
+            <output name="normalized_data" file="normalized_data.parquet" compare="sim_size" delta="200"/>
+        </test>
+        <test><!-- TEST 8 -->
+            <param name="input_choice" value="2"/>
+            <param name="data" value="feature_table_transpose_version.csv" ftype="csv"/>
+            <param name="metadata" value="metadata.csv" ftype="csv"/>
+            <param name="transpose_feature_table" value="TRUE"/>
+            <param name="mode" value="batchwise"/>
+            <param name="wavelet_filter" value="d"/>
+            <param name="wavelet_length" value="2"/>
+            <param name="k" value="20"/>
+            <param name="t" value="0.05"/>
+            <param name="t2" value="0.05"/>
+            <param name="alpha" value="0"/>
+            <output name="normalized_data" file="normalized_data.csv"/>
+        </test>
         <!-- The following test has different results on three platform I've tried -->
         <!-- <test>
             <param name="data" value="input_data_nobatch.csv" ftype="csv"/>
--- a/waveica_wrapper.R	Thu Sep 29 15:21:04 2022 +0000
+++ b/waveica_wrapper.R	Fri Jun 23 07:27:29 2023 +0000
@@ -1,6 +1,14 @@
-read_file <- function(file, metadata, ft_ext, mt_ext) {
+read_file <- function(file, metadata, ft_ext, mt_ext, transpose) {
   data <- read_data(file, ft_ext)
 
+  if (transpose) {
+    col_names <- c("sampleName", data[[1]])
+    t_data <- data[-1]
+    t_data <- t(t_data)
+    data <- data.frame(rownames(t_data), t_data)
+    colnames(data) <- col_names
+  }
+
   if (!is.na(metadata)) {
     mt_data <- read_data(metadata, mt_ext)
     data <- merge(mt_data, data, by = "sampleName")
@@ -24,6 +32,7 @@
 waveica <- function(file,
                     metadata = NA,
                     ext,
+                    transpose = FALSE,
                     wavelet_filter,
                     wavelet_length,
                     k,
@@ -31,17 +40,19 @@
                     t2,
                     alpha,
                     exclude_blanks) {
-
   # get input from the Galaxy, preprocess data
   ext <- strsplit(x = ext, split = "\\,")[[1]]
 
   ft_ext <- ext[1]
   mt_ext <- ext[2]
 
-  data <- read_file(file, metadata, ft_ext, mt_ext)
+  data <- read_file(file, metadata, ft_ext, mt_ext, transpose)
 
-  required_columns <- c("sampleName", "class", "sampleType", "injectionOrder", "batch")
-  verify_input_dataframe(data, required_columns)
+  required_columns <- c(
+    "sampleName", "class", "sampleType",
+    "injectionOrder", "batch"
+  )
+  data <- verify_input_dataframe(data, required_columns)
 
   data <- sort_by_injection_order(data)
 
@@ -76,28 +87,30 @@
 waveica_singlebatch <- function(file,
                                 metadata = NA,
                                 ext,
+                                transpose = FALSE,
                                 wavelet_filter,
                                 wavelet_length,
                                 k,
                                 alpha,
                                 cutoff,
                                 exclude_blanks) {
-
   # get input from the Galaxy, preprocess data
   ext <- strsplit(x = ext, split = "\\,")[[1]]
 
   ft_ext <- ext[1]
   mt_ext <- ext[2]
 
-  data <- read_file(file, metadata, ft_ext, mt_ext)
+  data <- read_file(file, metadata, ft_ext, mt_ext, transpose)
 
   required_columns <- c("sampleName", "class", "sampleType", "injectionOrder")
   optional_columns <- c("batch")
-  verify_input_dataframe(data, required_columns)
+
+  data <- verify_input_dataframe(data, required_columns)
 
   data <- sort_by_injection_order(data)
 
-  feature_columns <- colnames(data)[!colnames(data) %in% c(required_columns, optional_columns)]
+  feature_columns <- colnames(data)[!colnames(data) %in%
+    c(required_columns, optional_columns)]
   features <- data[, feature_columns]
   injection_order <- data$injectionOrder
 
@@ -112,7 +125,7 @@
   )
 
   data[, feature_columns] <- features
-
+  group <- enumerate_groups(as.character(data$sampleType))
   # remove blanks from dataset
   if (exclude_blanks) {
     data <- exclude_group(data, group)
@@ -142,9 +155,49 @@
     stop("Error: dataframe cannot contain NULL values!
 Make sure that your dataframe does not contain empty cells")
   } else if (!all(required_columns %in% colnames(data))) {
-    stop("Error: missing metadata!
-Make sure that the following columns are present in your dataframe: ", paste(required_columns, collapse = ", "))
+    stop(
+      "Error: missing metadata!
+Make sure that the following columns are present in your dataframe: ",
+      paste(required_columns, collapse = ", ")
+    )
   }
+
+  data <- verify_column_types(data, required_columns)
+
+  return(data)
+}
+
+verify_column_types <- function(data, required_columns) {
+  # Specify the column names and their expected types
+  column_types <- list(
+    "sampleName" = c("character", "factor"),
+    "class" = c("character", "factor"),
+    "sampleType" = c("character", "factor"),
+    "injectionOrder" = "integer",
+    "batch" = "integer"
+  )
+
+  column_types <- column_types[required_columns]
+
+  for (col_name in names(data)) {
+    actual_type <- class(data[[col_name]])
+    if (col_name %in% names(column_types)) {
+      expected_types <- column_types[[col_name]]
+
+      if (!actual_type %in% expected_types) {
+        stop(
+          "Column ", col_name, " is of type ", actual_type,
+          " but expected type is ",
+          paste(expected_types, collapse = " or "), "\n"
+        )
+      }
+    } else {
+      if (actual_type != "numeric") {
+        data[[col_name]] <- as.numeric(as.character(data[[col_name]]))
+      }
+    }
+  }
+  return(data)
 }
 
 
@@ -187,7 +240,10 @@
   if (ext == "csv") {
     write.csv(data, file = output, row.names = FALSE, quote = FALSE)
   } else if (ext == "tsv") {
-    write.table(data, file = output, sep = "\t", row.names = FALSE, quote = FALSE)
+    write.table(data,
+      file = output, sep = "\t",
+      row.names = FALSE, quote = FALSE
+    )
   } else {
     arrow::write_parquet(data, sink = output)
   }