Mercurial > repos > recetox > waveica

changeset 7:1a2aeb8137bf 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 013d7c85fa9d77b8a27d194b350cd6b2d127a80f
author recetox
date Thu, 06 Jun 2024 12:25:05 +0000
parents 071a424241ec
children bf32ae95a06f
files macros.xml test-data/test10_output1.tsv test-data/test10_output2.tsv test-data/test9_output1.parquet test-data/test9_output2.parquet waveica.xml waveica_wrapper.R
diffstat 7 files changed, 129 insertions(+), 35 deletions(-) [+]
line wrap: on
line diff
--- a/macros.xml	Thu May 30 14:54:02 2024 +0000
+++ b/macros.xml	Thu Jun 06 12:25:05 2024 +0000
@@ -96,10 +96,20 @@
             </when>
         </conditional>
     </xml>
+    <xml name="split_output">
+        <param name = "keep_two_output" label="Output metadata and data matrix as separate files" type="boolean" checked="false" 
+                truevalue="TRUE" falsevalue="FALSE" help="Keep two output files, one being the data matrix (feature table) and the second being the  metadata table." />
+    </xml>
 
     <xml name="outputs">
         <outputs>
-            <data name="normalized_data" format="tsv">
+            <data name="normalized_data" format="tabular" label="Normalized table of ${on_string}">
+                <change_format>
+                    <when input_dataset="data" attribute="ext" value="parquet" format="parquet" />
+                </change_format>
+            </data>
+            <data name="metadata" format="tabular" label="Metadata table of ${on_string}">
+                <filter>keep_two_output</filter>
                 <change_format>
                     <when input_dataset="data" attribute="ext" value="parquet" format="parquet" />
                 </change_format>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/test10_output1.tsv	Thu Jun 06 12:25:05 2024 +0000
@@ -0,0 +1,5 @@
+id	VT_160120_002	VT_160120_004	VT_160120_006	VT_160120_008	VT_160120_010
+M85T34	355200.506508035	216897.826587868	362337.195084504	143303.377379009	189065.516447239
+M86T41	75115889.9077485	75204863.1495248	76490295.1450204	83771659.9549148	84108898.7658797
+M86T518	6101488.54615418	6170882.26270475	12588041.969092	6181538.46316058	6103964.42378424
+M86T539	2007379.02604984	2069979.64992079	1818589.63912375	1975712.25920485	1935671.32085241
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/test10_output2.tsv	Thu Jun 06 12:25:05 2024 +0000
@@ -0,0 +1,6 @@
+sampleName	class	sampleType	injectionOrder	batch
+VT_160120_002	sample	sample	1	1
+VT_160120_004	sample	sample	2	1
+VT_160120_006	sample	sample	3	1
+VT_160120_008	sample	sample	4	1
+VT_160120_010	sample	sample	5	1
Binary file test-data/test9_output1.parquet has changed
Binary file test-data/test9_output2.parquet has changed
--- a/waveica.xml	Thu May 30 14:54:02 2024 +0000
+++ b/waveica.xml	Thu Jun 06 12:25:05 2024 +0000
@@ -1,4 +1,4 @@
-<tool id="waveica" name="WaveICA" version="@TOOL_VERSION@+galaxy6" profile="21.09">
+<tool id="waveica" name="WaveICA" version="@TOOL_VERSION@+galaxy7" profile="21.09">
     <description>removal of batch effects for untargeted metabolomics data</description>
     <macros>
         <import>macros.xml</import>
@@ -51,14 +51,14 @@
             )'
             #end if
 
-            -e 'store_data(normalized_data, "$normalized_data", "$input_num.data.ext")'
+            -e 'store_data(normalized_data, "$normalized_data", "$metadata", "$input_num.data.ext", $keep_two_output)'
     ]]></command>
 
     <inputs>
         <conditional name="input_num">
             <param name="input_choice" type="select" label="Choose input files:">
-                    <option value="1" selected="true">1</option>
-                    <option value="2">2</option>
+                    <option value="1" selected="true">1: intensity-by-feature table with metadata</option>
+                    <option value="2">2: intensity-by-feature table and metadata table separately</option>
             </param>
             <when value="1">
                 <expand macro="input_data"/>
@@ -89,12 +89,13 @@
             </when>
         </conditional>
         <expand macro="exclude_blanks"/>
+        <expand macro="split_output"/>
     </inputs>
 
     <expand macro="outputs"/>
 
     <tests>
-        <test><!-- TEST 1 -->
+        <test expect_num_outputs="1"><!-- TEST 1 -->
             <param name="data" value="input_data.csv" ftype="csv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
@@ -103,9 +104,9 @@
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
-            <output name="normalized_data" file="normalized_data.tsv" ftype="tsv"/>
+            <output name="normalized_data" file="normalized_data.tsv" ftype="tabular"/>
         </test>
-        <test><!-- TEST 2 -->
+        <test expect_num_outputs="1"><!-- TEST 2 -->
             <param name="data" value="input_data.tsv" ftype="tsv"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
@@ -114,9 +115,9 @@
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
-            <output name="normalized_data" file="normalized_data.tsv" ftype="tsv"/>
+            <output name="normalized_data" file="normalized_data.tsv" ftype="tabular"/>
         </test>
-        <test><!-- TEST 3 -->
+        <test expect_num_outputs="1"><!-- TEST 3 -->
             <param name="data" value="input_data.parquet" ftype="parquet"/>
             <param name="mode" value="batchwise"/>
             <param name="wavelet_filter" value="d"/>
@@ -127,7 +128,7 @@
             <param name="alpha" value="0"/>
             <output name="normalized_data" file="normalized_data.parquet" ftype="parquet"/>
         </test>
-        <test><!-- TEST 4 -->
+        <test expect_num_outputs="1"><!-- TEST 4 -->
             <param name="input_choice" value="2"/>
             <param name="data" value="feature_table.csv" ftype="csv"/>
             <param name="metadata" value="metadata.csv" ftype="csv"/>
@@ -138,9 +139,9 @@
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
-            <output name="normalized_data" file="normalized_data.tsv" ftype="tsv"/>
+            <output name="normalized_data" file="normalized_data.tsv" ftype="tabular"/>
         </test>
-        <test><!-- TEST 5 -->
+        <test expect_num_outputs="1"><!-- TEST 5 -->
             <param name="input_choice" value="2"/>
             <param name="data" value="feature_table.tsv" ftype="tsv"/>
             <param name="metadata" value="metadata.tsv" ftype="tsv"/>
@@ -151,9 +152,9 @@
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
-            <output name="normalized_data" file="normalized_data.tsv" ftype="tsv"/>
+            <output name="normalized_data" file="normalized_data.tsv" ftype="tabular"/>
         </test>
-        <test><!-- TEST 6 -->
+        <test expect_num_outputs="1"><!-- TEST 6 -->
             <param name="input_choice" value="2"/>
             <param name="data" value="feature_table.parquet" ftype="parquet"/>
             <param name="metadata" value="metadata.csv" ftype="csv"/>
@@ -166,7 +167,7 @@
             <param name="alpha" value="0"/>
             <output name="normalized_data" file="normalized_data.parquet" compare="sim_size" delta="200" ftype="parquet"/>
         </test>
-        <test><!-- TEST 7 -->
+        <test expect_num_outputs="1"><!-- 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"/>
@@ -180,7 +181,7 @@
             <param name="alpha" value="0"/>
             <output name="normalized_data" file="normalized_data.parquet" compare="sim_size" delta="200" ftype="parquet"/>
         </test>
-        <test><!-- TEST 8 -->
+        <test expect_num_outputs="1"><!-- 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"/>
@@ -192,7 +193,7 @@
             <param name="t" value="0.05"/>
             <param name="t2" value="0.05"/>
             <param name="alpha" value="0"/>
-            <output name="normalized_data" file="normalized_data.tsv" ftype="tsv"/>
+            <output name="normalized_data" file="normalized_data.tsv" ftype="tabular"/>
         </test>
         <!-- The following test has different results on three platform I've tried -->
         <!-- <test>
@@ -205,10 +206,42 @@
             <param name="cutoff" value="0"/>
             <output name="normalized_data" file="normalized_data_nobatch.tsv"/>
         </test> -->
-        <test expect_failure="true">
+        <test expect_num_outputs="2"><!-- TEST 9 -->
+            <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"/>
+            <param name="keep_two_output" value="TRUE"/>
+            <output name="normalized_data" file="test9_output1.parquet" ftype="parquet"/>
+            <output name="metadata" file="test9_output2.parquet" ftype="parquet"/>
+        </test>
+        <test expect_num_outputs="2"><!-- TEST 10 -->
+            <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"/>
+            <param name="keep_two_output" value="TRUE"/>
+            <output name="normalized_data" file="test10_output1.tsv" ftype="tabular"/>
+            <output name="metadata" file="test10_output2.tsv" ftype="tabular"/>
+        </test>
+        <test expect_failure="true"><!-- TEST 11 -->
             <param name="data" value="na_data.csv" ftype="csv"/>
         </test>
-        <test expect_failure="true">
+        <test expect_failure="true"><!-- TEST 12 -->
             <param name="data" value="incomplete_metadata_data.csv" ftype="csv"/>
         </test>
     </tests>
--- a/waveica_wrapper.R	Thu May 30 14:54:02 2024 +0000
+++ b/waveica_wrapper.R	Thu Jun 06 12:25:05 2024 +0000
@@ -3,10 +3,7 @@
 
     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
+        data <- tranpose_data(data, col_names)
     }
 
     if (!is.na(metadata)) {
@@ -133,7 +130,6 @@
     return(data)
 }
 
-
 sort_by_injection_order <- function(data) {
     if ("batch" %in% colnames(data)) {
         data <- data[order(data[, "batch"], data[, "injectionOrder"], decreasing = FALSE), ]
@@ -143,7 +139,6 @@
     return(data)
 }
 
-
 verify_input_dataframe <- function(data, required_columns) {
     if (anyNA(data)) {
         stop("Error: dataframe cannot contain NULL values!
@@ -194,7 +189,6 @@
     return(data)
 }
 
-
 # Match group labels with [blank/sample/qc] and enumerate them
 enumerate_groups <- function(group) {
     group[grepl("blank", tolower(group))] <- 0
@@ -204,7 +198,6 @@
     return(group)
 }
 
-
 # Create appropriate input for R wavelets function
 get_wf <- function(wavelet_filter, wavelet_length) {
     wf <- paste(wavelet_filter, wavelet_length, sep = "")
@@ -217,7 +210,6 @@
     return(wf)
 }
 
-
 # Exclude blanks from a dataframe
 exclude_group <- function(data, group) {
     row_idx_to_exclude <- which(group %in% 0)
@@ -230,14 +222,62 @@
     }
 }
 
-store_data <- function(data, output, ext) {
+store_data <- function(data, feature_output, metadata_output, ext, split_output = FALSE) {
     if (ext == "parquet") {
-        arrow::write_parquet(data, output)
+        if (split_output == TRUE) {
+            split_df <- split_output(data)
+            arrow::write_parquet(split_df$metadata, metadata_output)
+            arrow::write_parquet(split_df$feature_table, feature_output)
+        } else {
+            arrow::write_parquet(data, feature_output)
+        }
     } else {
-        write.table(data,
-            file = output, sep = "\t",
-            row.names = FALSE, quote = FALSE
-        )
+        if (split_output == TRUE) {
+            split_df <- split_output(data)
+            write.table(split_df$metadata,
+                file = metadata_output, sep = "\t",
+                row.names = FALSE, quote = FALSE
+            )
+            write.table(split_df$feature_table,
+                file = feature_output, sep = "\t",
+                row.names = FALSE, quote = FALSE
+            )
+        } else {
+            write.table(data,
+                file = feature_output, sep = "\t",
+                row.names = FALSE, quote = FALSE
+            )
+        }
     }
     cat("Normalization has been completed.\n")
 }
+
+split_output <- function(df) {
+    required_columns_set1 <- c("sampleName", "class", "sampleType", "injectionOrder", "batch")
+    required_columns_set2 <- c("sampleName", "class", "sampleType", "injectionOrder")
+
+    if (all(required_columns_set1 %in% colnames(df))) {
+        metadata_df <- df[, required_columns_set1, drop = FALSE]
+        df <- df[, -c(2:5)]
+    } else if (all(required_columns_set2 %in% colnames(df))) {
+        metadata_df <- df[, required_columns_set2, drop = FALSE]
+        df <- df[, -c(2:4)]
+    } else {
+        stop("Neither set of required columns is present in the dataframe.")
+    }
+
+    # Transpose the feature table
+    col_names <- c("id", as.vector(df[[1]]))
+    feature_table <- tranpose_data(df, col_names)
+
+    return(list(metadata = metadata_df, feature_table = feature_table))
+}
+
+tranpose_data <- function(data, column_names) {
+    t_data <- data[-1]
+    t_data <- t(t_data)
+    tranposed_data <- data.frame(rownames(t_data), t_data)
+    colnames(tranposed_data) <- column_names
+
+    return(tranposed_data)
+}