Mercurial > repos > md-anderson-bioinformatics > heat_map_creation

changeset 38:605ec840a06b draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author insilico-bob
date Thu, 20 Jun 2019 11:30:12 -0400
parents f2272e907f1b
children 436f03b71cf6
files CHM.R GalaxyMapGen.jar heatmap.sh mda_heatmap_gen.xml mda_heatmap_viz.zip
diffstat 5 files changed, 146 insertions(+), 58 deletions(-) [+]
line wrap: on
line diff
--- a/CHM.R	Thu Nov 08 14:59:04 2018 -0500
+++ b/CHM.R	Thu Jun 20 11:30:12 2019 -0400
@@ -17,7 +17,7 @@
 
 performDataOrdering<-function(dataFile, rowOrderMethod, rowDistanceMeasure, rowAgglomerationMethod, colOrderMethod, colDistanceMeasure, colAgglomerationMethod,rowOrderFile, colOrderFile, rowDendroFile, colDendroFile, rowCut, colCut)
 { 
-   dataMatrix = read.table(dataFile, header=TRUE, sep = "\t", row.names = 1, as.is=TRUE, na.strings=c("NA","N/A","-","?"))
+   dataMatrix = read.table(dataFile, header=TRUE, sep = "\t", check.names = FALSE, row.names = 1, as.is=TRUE, na.strings=c("NA","N/A","-","?"))
    rowOrder <-  createOrdering(dataMatrix, rowOrderMethod, "row", rowDistanceMeasure, rowAgglomerationMethod)  
    if (rowOrderMethod == "Hierarchical") {
       writeHCDataTSVs(rowOrder, rowDendroFile, rowOrderFile)
@@ -39,6 +39,7 @@
  
    data=matrix(,length(uDend$labels),2);
    for (i in 1:length(uDend$labels)) {
+   	  print(uDend$labels[i])
       data[i,1] = uDend$labels[i];
       data[i,2] = which(uDend$order==i);
    }
Binary file GalaxyMapGen.jar has changed
--- a/heatmap.sh	Thu Nov 08 14:59:04 2018 -0500
+++ b/heatmap.sh	Thu Jun 20 11:30:12 2019 -0400
@@ -1,16 +1,133 @@
-#echo $1 $2 $3 $4 $5 $6 $7 $8 $9 ${10} ${11} ${12} ${13} ${14} ${15} ${16} ${17} ${18}
+#echo "1: " $1" 2: " $2" 3: " $3" 4: "$4" 5: "$5 " 6: "$6 "7: "$7" 8: "$8 " 9: "$9" 10: "${10}" 11: "${11} "12: "${12} 
+#echo " 13: "${13}" 14: "${14}" 15: "${15}" 16: "${16} "17: "${17}" 18: "${18}" 19: "${19}" 20: "${20}" 21: "${21} " 22: "${22}" 23:" ${23} 
 
-#get tool data and tool install directories
-tooldir=$(cut -d';' -f1 <<< ${12})
-tooldata=$(cut -d';' -f2 <<< ${12})
+#Count total number of parameters and classification parameters
+parmSize=0
+classSize=0
+matrixSize=0
+for i in "$@"; do
+	currParm=$(cut -d'|' -f1 <<< $i)
+	parmSize=$((parmSize+1))
+	if [ $currParm = "classification" ]
+	then
+		classSize=$((classSize+1))
+  	fi
+done
 
+#Get tool data and tool install directories
+tooldir=$1
+tooldata=$2
 #create temp directory for row and col order and dendro files.
 tdir=$tooldata/$(date +%y%m%d%M%S)
-echo $tdir
 mkdir $tdir
+#echo "tdir: "$tdir
+
+#Extract parameters for row and column order and dendro files
+rowOrderFile=$tdir/ROfile.txt
+rowDendroFile=$tdir/RDfile.txt
+colOrderFile=$tdir/COfile.txt
+colDendroFile=$tdir/CDfile.txt
+rowOrderJson='"order_file": "'$rowOrderFile'",'
+rowDendroJson='"dendro_file": "'$rowDendroFile'",'
+colOrderJson='"order_file": "'$colOrderFile'",'
+colDendroJson='"dendro_file": "'$colDendroFile'",'
+
+#BEGIN: Construct JSON for all non-repeating parameters
+parmJson='{'
+rowConfigJson='"row_configuration": {'
+colConfigJson='"col_configuration": {'
+
+ctr=0
+for i in "$@"; do
+	if [ $ctr -gt 1 ]
+	then
+		currParm=$(cut -d'|' -f1 <<< $i)
+		if [ $currParm != "matrix_files" ] && [ $currParm != "row_configuration" ] && [ $currParm != "col_configuration" ] && [ $currParm != "classification" ]
+		then
+			#Parse pipe-delimited parameter parameter
+			parmJson=$parmJson' "'$(cut -d'|' -f1 <<< $i)'":"'$(cut -d'|' -f2 <<< $i)'",'
+	  	fi
+		if [ $currParm = "row_configuration" ]
+		then
+			rowOrder=$(cut -d'|' -f3 <<< $i)
+			rowDistance=$(cut -d'|' -f5 <<< $i)
+			rowAgglomeration=$(cut -d'|' -f7 <<< $i)
+			rowCuts=$(cut -d'|' -f9 <<< $i)
+			rowLabels=$(cut -d'|' -f11 <<< $i)
+			dataTypeJson='"'$(cut -d'|' -f10 <<< $i)'":["'$rowLabels'"]'
+			if [ $rowOrder = 'Hierarchical' ]
+			then
+				rowConfigJson=$rowConfigJson$rowOrderJson$rowDendroJson
+			fi
+			rowConfigJson=$rowConfigJson'"'$(cut -d'|' -f2 <<< $i)'":"'$(cut -d'|' -f3 <<< $i)'","'$(cut -d'|' -f4 <<< $i)'":"'$(cut -d'|' -f5 <<< $i)'","'$(cut -d'|' -f6 <<< $i)'":"'$(cut -d'|' -f7 <<< $i)'",'$dataTypeJson'},'
+	  	fi
+		if [ $currParm = "col_configuration" ]
+		then
+			colOrder=$(cut -d'|' -f3 <<< $i)
+			colDistance=$(cut -d'|' -f5 <<< $i)
+			colAgglomeration=$(cut -d'|' -f7 <<< $i)
+			colCuts=$(cut -d'|' -f9 <<< $i)
+			colLabels=$(cut -d'|' -f11 <<< $i)
+			dataTypeJson='"'$(cut -d'|' -f10 <<< $i)'":["'$colLabels'"]'
+			if [ $colOrder = 'Hierarchical' ]
+			then
+				colConfigJson=$colConfigJson$colOrderJson$colDendroJson
+			fi
+			colConfigJson=$colConfigJson'"'$(cut -d'|' -f2 <<< $i)'":"'$(cut -d'|' -f3 <<< $i)'","'$(cut -d'|' -f4 <<< $i)'":"'$(cut -d'|' -f5 <<< $i)'","'$(cut -d'|' -f6 <<< $i)'":"'$(cut -d'|' -f7 <<< $i)'",'$dataTypeJson'},'
+	  	fi
+	 fi
+	 ctr=$((ctr+1))
+done
+#END: Construct JSON for all non-repeating parameters
+#echo "rowCuts: "$rowCuts
+#echo "colCuts: "$colCuts
+#echo "ROW CONFIG JSON: "$rowConfigJson
+#echo "COL CONFIG JSON: "$colConfigJson
+
+#BEGIN: Construct JSON for data layers
+matrixJson='"matrix_files": [ '
+inputMatrix=''
+for i in "$@"; do
+	currParm=$(cut -d'|' -f1 <<< $i)
+	if [ $currParm = "matrix_files" ]
+	then
+		#Parse pipe-delimited parameter parameter
+		matrixJson=$matrixJson' {"'$(cut -d'|' -f2 <<< $i)'":"'$(cut -d'|' -f3 <<< $i)'","'$(cut -d'|' -f4 <<< $i)'":"'$(cut -d'|' -f5 <<< $i)'","'$(cut -d'|' -f6 <<< $i)'":"'$(cut -d'|' -f7 <<< $i)'"}'
+		inputMatrix=$(cut -d'|' -f3 <<< $i)
+  	fi
+done
+matrixJson=$matrixJson"],"
+#END: Construct JSON for data layers
+
+#BEGIN: Construct JSON for classification files
+classJson='"classification_files": [ '
+classIter=0
+for i in "$@"; do
+	currParm=$(cut -d'|' -f1 <<< $i)
+	if [ $currParm = "classification" ]
+	then
+		classIter=$((classIter+1))
+		#Parse pipe-delimited 3-part classification bar parameter
+		classJson=$classJson' {"'$(cut -d'|' -f2 <<< $i)'":"'$(cut -d'|' -f3 <<< $i)'","'$(cut -d'|' -f4 <<< $i)'":"'$(cut -d'|' -f5 <<< $i)'"'
+		classCat=$(cut -d'|' -f7 <<< $i)
+		classColorType=$(cut -d'_' -f2 <<< $classCat)
+		classJson=$classJson','
+		classJson=$classJson' "position":"'$(cut -d'_' -f1 <<< $classCat)'","color_map": {"type":"'$classColorType'"}}'
+		if [ $classIter -lt $classSize ]		
+		then
+			classJson=$classJson','
+		fi
+  	fi
+done
+classJson=$classJson']'
+#END: Construct JSON for classification files
+
+parmJson=$parmJson$matrixJson$rowConfigJson$colConfigJson$classJson
+parmJson=$parmJson'}'
+#echo "HEATMAP PARAMETERS JSON: "$parmJson	
 
 #run R to cluster matrix
-output="$(R --slave --vanilla --file=$tooldir/CHM.R --args $4 $5 $6 $7 $8 $9 ${10} $tdir/ROfile.txt $tdir/COfile.txt $tdir/RDfile.txt $tdir/CDfile.txt ${13} ${14} ${15} ${16} 2>&1)"
+output="$(R --slave --vanilla --file=$tooldir/CHM.R --args $inputMatrix $rowOrder $rowDistance $rowAgglomeration $colOrder $colDistance $colAgglomeration $rowOrderFile $colOrderFile $rowDendroFile $colDendroFile $rowCuts $colCuts $rowLabels $colLabels 2>&1)"
 rc=$?;
 if [ $rc != 0 ]
 then
@@ -24,33 +141,7 @@
   exit $rc;
 fi
 
-#there are a variable number of triplicate parameters for classification bars
-count=0
-classifications=''
-
-#if row cut was done, add that autogenerated classification
-if [ ${13} -gt 1 ]
-then
-  classifications="Class $tdir/ROfile.txt.cut row_categorical"
-fi
-
-#if col cut was done, add that autogenerated classification
-if [ ${14} -gt 1 ]
-then
-  classifications="$classifications Class $tdir/COfile.txt.cut col_categorical"
-fi
-
-#now add the user provided classification files 
-for i in "$@"; do
-  if [ $count -gt 16 ]
-  then
-    classifications=$classifications' '$i
-  fi
-  count=$((count+1))
-done
-echo "classifications " $classifications
-echo "${11} " ${11}
 #call java program to generate NGCHM viewer files.
-java -jar $tooldir/GalaxyMapGen.jar "${1}" "${2}" "${3}" DataLayer1 $4 linear ${15} ${16} $5 $6 $7 $tdir/ROfile.txt $tdir/RDfile.txt $8 $9 ${10} $tdir/COfile.txt $tdir/CDfile.txt ${11} $classifications
+java -jar $tooldir/GalaxyMapGen.jar "$parmJson"
 #clean up tempdir
 rm -rf $tdir
--- a/mda_heatmap_gen.xml	Thu Nov 08 14:59:04 2018 -0500
+++ b/mda_heatmap_gen.xml	Thu Jun 20 11:30:12 2019 -0400
@@ -2,16 +2,17 @@
 <tool id="mda_heatmap_gen" name="NG-CHM Generator" version="2.3">
    <requirements>
        <requirement type="package" version="3.4.1">r-base</requirement> 
-	   <requirement type="package" version="8">openjdk</requirement>
+		<requirement type="package" version="8.0.144">openjdk</requirement>
    </requirements>
    <description>Create Clustered Heat Maps</description>
-   <command interpreter="bash" detect_errors="aggressive">$__tool_directory__/heatmap.sh  "standard" "Heat_Map_$hmname" "$hmdesc" '$inputmatrix' ${d_rows.rowOrderMethod} ${d_rows.rowDistanceMeasure} ${d_rows.rowAgglomerationMethod} ${d_cols.columnOrderMethod} ${d_cols.columnDistanceMeasure} ${d_cols.columnAgglomerationMethod} $summarymethod '$__tool_directory__;$__tool_data_path__/' 0 0 labels labels 'None'
+   <command interpreter="bash" detect_errors="aggressive">$__tool_directory__/heatmap.sh  "$__tool_directory__" "$__tool_data_path__/" "chm_name|Heat_Map_$hmname" "chm_description|$hmdesc" 
+	"matrix_files|path|$inputmatrix|name|datalayer|summary_method|$summarymethod"
+	"row_configuration|order_method|${d_rows.rowOrderMethod}|distance_metric|${d_rows.rowDistanceMeasure}|agglomeration_method|${d_rows.rowAgglomerationMethod}|tree_covar_cuts|0|data_type|labels"    
+	"col_configuration|order_method|${d_cols.columnOrderMethod}|distance_metric|${d_cols.columnDistanceMeasure}|agglomeration_method|${d_cols.columnAgglomerationMethod}|tree_covar_cuts|0|data_type|labels"    
     #for $op in $operations
-       '${op.class_name}'
-       '${op.repeatinput.file_name}'
-       '${op.cat}'
-      #end for
- 	'$output' 
+       'classification|name|${op.class_name}|path|${op.repeatinput.file_name}|category|${op.cat}'
+    #end for
+ 	'output_location|$output' 
  </command>
 	<stdio>
       <exit_code range="1:" level="fatal" />
@@ -22,11 +23,6 @@
            <sanitizer>
               <valid>
                 <add preset="string.printable"/>
-<!--            <add value="string.letters"/>
-                <add value="string.digits"/>
-                <add value="-"/>
-                <add value="_"/>
--->
             	<remove value="&quot;"/>
             	<remove value="&apos;"/>
                 <remove value=" "/> 
@@ -59,12 +55,12 @@
         <when value="Hierarchical">
 			<param name="rowDistanceMeasure" type="select"  label="Row Distance Metric" help="For clustering, select the method of determining distance between rows">
 				<option value="euclidean">Euclidean</option>
-				<option value="binary">Binary</option>
+				<!-- <option value="binary">Binary</option> ** breaks dendrogram  -->
 				<option value="manhattan">Manhattan</option>
 				<option value="maximum">Maximum</option>
-				<option value="canberra">Canberra</option>	    
+				<!-- <option value="canberra">Canberra</option> ** breaks dendrogram  -->
 				<option value="minkowski">Minkowski</option>	    
-				<option value="correlation">Correlation</option>	    
+				<!-- <option value="correlation">Correlation</option>	** breaks dendrogram -->    
 			</param>
 			<param name="rowAgglomerationMethod" type="select"  label="Row Clustering Method" help="For clustering, select algorithm for building clusters.">
 				<option value="average">Average Linkage</option>
@@ -72,8 +68,8 @@
 				<option value="single">Single Linkage</option>
 				<option value="ward" selected="true">Ward</option>
 				<option value="mcquitty">Mcquitty</option>	    
-				<option value="median">Median</option>	    
-				<option value="centroid">Centroid</option>	    
+				<!-- <option value="median">Median</option>	   ** breaks dendrogram 
+				<option value="centroid">Centroid</option>	** breaks dendrogram     -->
 			</param>
         </when>
         <when value="Original">
@@ -94,12 +90,12 @@
         <when value="Hierarchical">
 			<param name="columnDistanceMeasure" type="select"  label="Column Distance Metric" help="For clustering, select the method of determining distance between columns">
 				<option value="euclidean">Euclidean</option>
-				<option value="binary">Binary</option>
+				<!-- <option value="binary">Binary</option> ** breaks dendrogram  -->
 				<option value="manhattan">Manhattan</option>
 				<option value="maximum">Maximum</option>
-				<option value="canberra">Canberra</option>	    
+				<!-- <option value="canberra">Canberra</option> ** breaks dendrogram  -->	    
 				<option value="minkowski">Minkowski</option>	    
-				<option value="correlation">Correlation</option>	    
+				<!-- <option value="correlation">Correlation</option>	** breaks dendrogram     -->
 			</param>
 			<param name="columnAgglomerationMethod" type="select"  label="Column Clustering Method" help="For clustering, select algorithm for building clusters.">
 				<option value="average">Average Linkage</option>
@@ -107,8 +103,8 @@
 				<option value="single">Single Linkage</option>
 				<option value="ward" selected="true">Ward</option>
 				<option value="mcquitty">Mcquitty</option>	    
-				<option value="median">Median</option>	    
-				<option value="centroid">Centroid</option>	    
+				<!-- <option value="median">Median</option>	  ** breaks dendrogram  
+				<option value="centroid">Centroid</option>	  ** breaks dendrogram   -->
 			</param>
         </when>
         <when value="Original">
@@ -133,9 +129,9 @@
         </param>
         <param name="repeatinput" type="data" format="Tabular" label="Covariate File" help="Tab delimited text file with row or column label and covariate value on each line."/>
 	<param name="cat" type="select" label="Axis Covariate Type" help="Identify the covariate as belonging to rows or columns and containing categorical or continuous values.">
-	  <option value="row_categorical" >Row Categorical</option>
+	  <option value="row_discrete" >Row Categorical</option>
 	  <option value="row_continuous" >Row Continuous</option>
-	  <option value="column_categorical" >Column Categorical</option>
+	  <option value="column_discrete" >Column Categorical</option>
 	  <option value="column_continuous" >Column Continuous</option>
 	</param>
     </repeat>       
Binary file mda_heatmap_viz.zip has changed