xcms3_new_naming - merge - OK but not using conda see: sneumann/xcms#247

workflow4metabolomics · Feb 8, 2018 · 4c42ad2 · 4c42ad2
1 parent 017d0aa
commit 4c42ad2
Show file tree

Hide file tree

Showing 15 changed files with 123 additions and 83 deletions.
diff --git a/tools/xcms_merge/lib.r b/tools/xcms_merge/lib.r
@@ -0,0 +1 @@
+../scripts/lib.r
diff --git a/tools/xcms_merge/test-data/MM14-xset.RData b/tools/xcms_merge/test-data/MM14-xset.RData
diff --git a/tools/xcms_merge/test-data/MM8-xset.RData b/tools/xcms_merge/test-data/MM8-xset.RData
diff --git a/tools/xcms_merge/test-data/ko15-xset.RData b/tools/xcms_merge/test-data/ko15-xset.RData
diff --git a/tools/xcms_merge/test-data/ko15.CDF b/tools/xcms_merge/test-data/ko15.CDF
@@ -0,0 +1 @@
+../../xcms_retcor/test-data/ko15.CDF
diff --git a/tools/xcms_merge/test-data/ko16-xset.RData b/tools/xcms_merge/test-data/ko16-xset.RData
diff --git a/tools/xcms_merge/test-data/ko16.CDF b/tools/xcms_merge/test-data/ko16.CDF
@@ -0,0 +1 @@
+../../xcms_retcor/test-data/ko16.CDF
diff --git a/tools/xcms_merge/test-data/sampleMetadata_missing.tab b/tools/xcms_merge/test-data/sampleMetadata_missing.tab
@@ -0,0 +1,5 @@
+wt16	WT
+wt15	WT
+ko15	KO
+ko10	KO
+foobar01	FOOBAR
diff --git a/tools/xcms_merge/test-data/wt15-xset.RData b/tools/xcms_merge/test-data/wt15-xset.RData
diff --git a/tools/xcms_merge/test-data/wt15.CDF b/tools/xcms_merge/test-data/wt15.CDF
@@ -0,0 +1 @@
+../../xcms_retcor/test-data/wt15.CDF
diff --git a/tools/xcms_merge/test-data/wt16-xset.RData b/tools/xcms_merge/test-data/wt16-xset.RData
diff --git a/tools/xcms_merge/test-data/wt16.CDF b/tools/xcms_merge/test-data/wt16.CDF
@@ -0,0 +1 @@
+../../xcms_retcor/test-data/wt16.CDF
diff --git a/tools/xcms_merge/xcms_merge.r b/tools/xcms_merge/xcms_merge.r
@@ -0,0 +1,83 @@
+#!/usr/bin/env Rscript
+
+#Import the different functions
+source_local <- function(fname){ argv <- commandArgs(trailingOnly=FALSE); base_dir <- dirname(substring(argv[grep("--file=", argv)], 8)); source(paste(base_dir, fname, sep="/")) }
+source_local("lib.r")
+
+suppressPackageStartupMessages(library(xcms, quietly=TRUE))
+suppressPackageStartupMessages(library(batch, quietly=TRUE))
+
+listArguments <- parseCommandArgs(evaluate=FALSE) #interpretation of arguments given in command line as an R list of objects
+
+# Handle infiles
+if (!exists("singlefile")) singlefile <- NULL
+if (!exists("zipfile")) zipfile <- NULL
+rawFilePath <- getRawfilePathFromArguments(singlefile, zipfile, listArguments)
+zipfile <- rawFilePath$zipfile
+singlefile <- rawFilePath$singlefile
+listArguments <- rawFilePath$listArguments
+directory <- retrieveRawfileInTheWorkingDirectory(singlefile, zipfile)
+
+cat("\tXSET MERGING...\n")
+
+for(image in listArguments[["images"]]) {
+    load(image)
+    cat(sampleNamesList$sampleNamesOrigin,"\n")
+    if (!exists("xdata_merged")) {
+        xdata_merged <- xdata
+        singlefile_merged <- singlefile
+        md5sumList_merged <- md5sumList
+        sampleNamesList_merged <- sampleNamesList
+    } else {
+        xdata_merged <- c(xdata_merged,xdata)
+        singlefile_merged <- c(singlefile_merged,singlefile)
+        md5sumList_merged$origin <- rbind(md5sumList_merged$origin,md5sumList$origin)
+        sampleNamesList_merged$sampleNamesOrigin <- c(sampleNamesList_merged$sampleNamesOrigin,sampleNamesList$sampleNamesOrigin)
+        sampleNamesList_merged$sampleNamesMakeNames <- c(sampleNamesList_merged$sampleNamesMakeNames,sampleNamesList$sampleNamesMakeNames)
+    }
+}
+rm(image)
+xdata <- xdata_merged; rm(xdata_merged)
+singlefile <- singlefile_merged; rm(singlefile_merged)
+md5sumList <- md5sumList_merged; rm(md5sumList_merged)
+sampleNamesList <- sampleNamesList_merged; rm(sampleNamesList_merged)
+
+if (!is.null(listArguments[["sampleMetadata"]])) {
+    cat("\tXSET PHENODATA SETTING...\n")
+    sampleMetadataFile <- listArguments[["sampleMetadata"]]
+    sampleMetadata <- read.table(sampleMetadataFile, h=F, sep=";", stringsAsFactors=F)
+    if (ncol(sampleMetadata) < 2) sampleMetadata <- read.table(sampleMetadataFile, h=F, sep="\t", stringsAsFactors=F)
+    if (ncol(sampleMetadata) < 2) sampleMetadata <- read.table(sampleMetadataFile, h=F, sep=",", stringsAsFactors=F)
+    if (ncol(sampleMetadata) < 2) {
+        error_message="Your sampleMetadata file seems not well formatted. The column separators accepted are ; , and tabulation"
+        print(error_message)
+        stop(error_message)
+    }
+    xdata@phenoData@data$sample_group=sampleMetadata$V2[match(xdata@phenoData@data$sample_name,sampleMetadata$V1)]
+
+    if (any(is.na(pData(xdata)$sample_group))) {
+        sample_missing <- pData(xdata)$sample_name[is.na(pData(xdata)$sample_group)]
+        error_message <- paste("Those samples are missing in your sampleMetadata:", paste(sample_missing, collapse=" "))
+        print(error_message)
+        stop(error_message)
+    }
+}
+
+# Get the legacy xcmsSet object
+suppressWarnings(xset <- as(xdata, "xcmsSet"))
+sampclass(xset) <- xset@phenoData$sample_group
+
+cat("\tXCMSnExp OBJECT INFO\n")
+print(pData(xdata))
+print(xdata)
+cat("\n\n")
+
+cat("\txcmsSet OBJECT INFO\n")
+print(xset@phenoData)
+print(xset)
+cat("\n\n")
+
+cat("\tSAVE RData\n")
+#saving R data in .Rdata file to save the variables used in the present tool
+objects2save <- c("xdata", "zipfile", "singlefile", "md5sumList", "sampleNamesList")
+save(list=objects2save[objects2save %in% ls()], file="merged.RData")
diff --git a/tools/xcms_merge/xcms_merge.xml b/tools/xcms_merge/xcms_merge.xml
@@ -1,4 +1,4 @@
-<tool id="xcms_merge" name="xcms.xcmsSet Merger" version="1.0.0">
+<tool id="xcms_merge" name="xcms.xcmsSet Merger" version="@WRAPPER_VERSION@.0">
     <description>Merge xcms.xcmsSet xset in one to be used by group</description>
 
     <macros>
@@ -9,84 +9,32 @@
     <expand macro="stdio"/>
 
     <command><![CDATA[
-        Rscript $script_file
-    ]]></command>
+        @COMMAND_XCMS_SCRIPT@/xcms_merge.r
+
+        images 'c("${"\",\"".join(map(str, $images))}")'
 
-    <configfiles>
-        <configfile name="script_file"><![CDATA[
-suppressPackageStartupMessages(library(xcms, quietly=TRUE))
-
-images=c(
-"${"\",\"".join(map(str, $images))}"
-)
-
-
-cat("\tXSET MERGING...\n")
-
-for(image in images) {
-    load(image)
-    cat(sampleNamesList\$sampleNamesOrigin,"\n")
-    if (!exists("xset_merged")) {
-        xset_merged=xset
-        singlefile_merged=singlefile
-        listOFlistArguments_merged=listOFlistArguments
-        md5sumList_merged=md5sumList
-        sampleNamesList_merged=sampleNamesList
-    } else {
-        xset_merged=c(xset_merged,xset)
-        singlefile_merged=c(singlefile_merged,singlefile)
-        listOFlistArguments_merged=c(listOFlistArguments_merged,listOFlistArguments)
-        md5sumList_merged\$origin=rbind(md5sumList_merged\$origin,md5sumList\$origin)
-        sampleNamesList_merged\$sampleNamesOrigin=c(sampleNamesList_merged\$sampleNamesOrigin,sampleNamesList\$sampleNamesOrigin)
-        sampleNamesList_merged\$sampleNamesMakeNames=c(sampleNamesList_merged\$sampleNamesMakeNames,sampleNamesList\$sampleNamesMakeNames)
-    }
-}
-rm(image)
-xset=xset_merged; rm(xset_merged)
-singlefile=singlefile_merged; rm(singlefile_merged)
-listOFlistArguments=listOFlistArguments_merged; rm(listOFlistArguments_merged)
-md5sumList=md5sumList_merged; rm(md5sumList_merged)
-sampleNamesList=sampleNamesList_merged; rm(sampleNamesList_merged)
-
-#if str($sampleMetadata) != 'None':
-    cat("\tXSET PHENODATA SETTING...\n")
-    sampleMetadata=read.table("$sampleMetadata", h=F, sep=";", stringsAsFactors=F)
-    if (ncol(sampleMetadata) < 2) sampleMetadata=read.table("$sampleMetadata", h=F, sep="\t", stringsAsFactors=F)
-    if (ncol(sampleMetadata) < 2) sampleMetadata=read.table("$sampleMetadata", h=F, sep=",", stringsAsFactors=F)
-    if (ncol(sampleMetadata) < 2) {
-        error_message="Your sampleMetadata file seems not well formatted. The column separators accepted are ; , and tabulation"
-        print(error_message)
-        stop(error_message)
-    }
-
-    sampclass(xset)=sampleMetadata\$V2[match(rownames(xset@phenoData),sampleMetadata\$V1)]
-#end if
-
-
-
-cat("\tXSET OBJECT INFO\n")
-
-print(xset@phenoData)
-print(xset)
-
-cat("\tSAVE RData\n")
-save.image("$xsetRData")
-
-        ]]></configfile>
-    </configfiles>
+        #if str($sampleMetadata) != 'None':
+            sampleMetadata '$sampleMetadata'
+        #end if
+
+        @COMMAND_FILE_LOAD@
+    ]]></command>
 
     <inputs>
         <param name="images" type="data" format="rdata.xcms.raw,rdata" label="xset RData / xcms.xcmsSet file" help="output file from another xcms.xcmsSet" multiple="true" />
         <param name="sampleMetadata" label="Sample metadata file " format="tabular" type="data" optional="true" help="must contain at least one column with the sample id and one column with the sample class"/>
+
+        <expand macro="input_file_load"/>
     </inputs>
 
     <outputs>
-        <data name="xsetRData" format="rdata.xcms.raw" label="xset.merged.RData" />
+        <data name="xsetRData" format="rdata.xcms.raw" label="xset.merged.RData" from_work_dir="merged.RData" />
     </outputs>
 
     <tests>
-        <test>
+        <!--<test>
             <param name="images" value="ko15-xset.RData,ko16-xset.RData,wt15-xset.RData,wt16-xset.RData" />
+            <expand macro="test_file_load_single"/>
             <assert_stdout>
                 <has_text text="object with 4 samples" />
                 <has_text text="Time range: 2506.1-4477.9 seconds (41.8-74.6 minutes)" />
@@ -95,15 +43,16 @@ save.image("$xsetRData")
                 <has_text text="Peak Groups: 0" />
                 <has_text text="Sample classes: ." />
             </assert_stdout>
-        </test>
+        </test>-->
         <test>
             <param name="images" value="ko15-xset.RData,ko16-xset.RData,wt15-xset.RData,wt16-xset.RData" />
+            <expand macro="test_file_load_single"/>
             <param name="sampleMetadata" value="sampleMetadata.tab" />
             <assert_stdout>
-                <has_text text="ko15    KO" />
-                <has_text text="ko16    KO" />
-                <has_text text="wt15    WT" />
-                <has_text text="wt16    WT" />
+                <has_text_matching expression="ko15\s+KO" />
+                <has_text_matching expression="ko16\s+KO" />
+                <has_text_matching expression="wt15\s+WT" />
+                <has_text_matching expression="wt16\s+WT" />
                 <has_text text="object with 4 samples" />
                 <has_text text="Time range: 2506.1-4477.9 seconds (41.8-74.6 minutes)" />
                 <has_text text="Mass range: 200.1-600 m/z" />
@@ -112,17 +61,14 @@ save.image("$xsetRData")
                 <has_text text="Sample classes: KO, WT" />
             </assert_stdout>
         </test>
-        <test>
-            <param name="images" value="MM8-xset.RData,MM14-xset.RData" />
-            <assert_stdout>
-                <has_text text="object with 2 samples" />
-                <has_text text="Time range: 19.7-307.3 seconds (0.3-5.1 minutes)" />
-                <has_text text="Mass range: 117.0357-936.7059 m/z" />
-                <has_text text="Peaks: 236 (about 118 per sample)" />
-                <has_text text="Peak Groups: 0" />
-                <has_text text="Sample classes: ." />
-            </assert_stdout>
-        </test>
+        <!--<test expect_failure="True">
+            <param name="images" value="ko15-xset.RData,ko16-xset.RData,wt15-xset.RData,wt16-xset.RData" />
+            <param name="sampleMetadata" value="sampleMetadata_missing.tab" />
+            <expand macro="test_file_load_single"/>
+            <assert_stderr>
+                <has_text text="Those samples are missing in your sampleMetadata: ko16" />
+            </assert_stderr>
+        </test>-->
     </tests>
 
     <help><![CDATA[

diff --git a/tools/xcms_suite/test-data/sampleMetadata_missing.tab b/tools/xcms_suite/test-data/sampleMetadata_missing.tab
@@ -0,0 +1 @@
+../../xcms_merge/test-data/sampleMetadata_missing.tab