修改input文件中的docker和workDir

Docker/Dockerfile

@@ -0,0 +1,9 @@
+FROM choppy/bioconductor-cn.mops:1.16.2
+
+COPY . /opt/mops
+RUN chmod a+x /opt/mops/cn_mops
+RUN ln -s /opt/mops/cn_mops /bin
+
+WORKDIR /data/
+
+CMD ["ls", "-al", "/opt"]

Docker/cn.mops.R

@@ -0,0 +1,140 @@
+###################
+##   cn.mops     ## 
+##  Jiyang Zhang ##
+##  2019.01.21   ##
+###################
+#
+#1 Introduction
+##The cn.mops package is part of the Bioconductor (http://www.bioconductor.org) project.
+##The package allows to detect copy number variations (CNVs) from next generation sequencing (NGS) data sets based on a generative model. 
+##Please visit http://www.bioinf.jku.at/software/cnmops/cnmops.htmlfor additional information.
+#
+#2 Set up
+#
+##2.1 load package
+library(cn.mops) #cn.mops_1.16.2
+#
+##2.2 load options
+#
+###2.2.1 Collect arguments
+args <- commandArgs(TRUE)
+#
+###2.2.2 Help section
+#### Default setting when no arguments passed
+if(length(args) < 1) {
+  args <- c("--help")
+}
+#### Help section
+if("--help" %in% args) {
+  cat("
+      The R Script
+      
+      Arguments:
+      --Tumor=the name of the TUMOR sample BAMfile - string
+      --Normal=the name of the NORMAL sample BAMfile - string
+	  --TumorSampleID=the name of the TUMOR sample  - string
+      --bed_file=BED (Browser Extensible Data) lines have four required fields included chr, start, end and gene annotation.  - string
+      --workDir=working directory   - character
+      --help              - print this text
+      
+      
+      Example:
+      Rscript cn.mops.R --Tumor=TumorSample.bam --Normal=NormalSample.bam --TumorSampleID=TumorSampleID --bed_file=bedFile --workDir=./ \n\n") #--outputDir=./ 
+  
+  q(save="no")
+}
+#
+####2.2.3 Parse arguments (we expect the form --arg=value)
+parseArgs <- function(x) strsplit(sub("^--", "", x), "=")
+argsDF <- as.data.frame(do.call("rbind", parseArgs(args)))
+argsL <- as.list(as.character(argsDF$V2))
+names(argsL) <- argsDF$V1
+args<-argsL
+#
+## Arg1 default
+if(is.null(args$Tumor)) {
+  stop("Tumor sample bamFile is missing")
+}else{
+  Tumor=args$Tumor
+}
+#
+## Arg2 default
+if(is.null(args$Normal)) {
+  stop("Normal sample bamFileis missing")
+}else{
+  Normal=args$Normal
+}
+#
+## Arg3 default
+if(is.null(args$TumorSampleID)) {
+  stop("TumorSampleID missing")
+}else{
+ TumorSampleID=args$TumorSampleID
+}
+#
+## Arg4 default
+if(is.null(args$bed_file)) {
+  stop("bed_file is missing")
+}else{
+  bed_file=args$bed_file
+}
+#
+## Arg5 default
+if(is.null(args$workDir)) {
+  stop("workDir is missing")
+}else{
+  workDir=args$workDir
+}
+#
+#3 Run cn.mops
+# Tumor <- c("Illumina_pt2_B1700.chr20_X.sorted.deduped.bam")
+# Normal <- c("Illumina_pt2_B17NC.chr20_X.sorted.deduped.bam")
+# workDir <- "/home/zhangjiyang/shiJzhiP/shijianzhiP2/scripts"
+# outputDir <- "/home/zhangjiyang/shiJzhiP/shijianzhiP2/scripts"
+# bed_file <- "Illumina_pt2_exonanno_exonfrom1_chr20_X.bed"
+#
+##3.1 BAM files and bed file as input.
+BAMFiles <- c(Tumor,Normal)
+segments <- read.table(paste0(workDir,"/", bed_file),sep="\t",as.is=TRUE)
+#
+##3.2 Get read counts from BAM.
+gr <- GRanges(segments[,1],IRanges(segments[,2],segments[,3]))
+#
+##3.3 The result object
+sample <- getSegmentReadCountsFromBAM(BAMFiles,GR=gr,mode="paired")
+resRef <- referencecn.mops(cases=sample[,1],controls=sample[,2],
+                           classes=c("CN0", "CN1", "CN2", "CN3", "CN4", "CN5", "CN6",
+                                     "CN7","CN8","CN16","CN32","CN64","CN128"),
+                           I = c(0.025, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 8, 16, 32, 64),
+                           segAlgorithm="DNAcopy")
+result<- calcIntegerCopyNumbers(resRef)
+CNVs <- as.data.frame(cnvs(result))
+colnames(segments) <- c("chr","start","end","anno")
+colnames(CNVs)[colnames(CNVs)=="seqnames"]  <- c("chr")
+bed.anno <- segments
+bed.anno$start_idx <- paste(bed.anno$chr,bed.anno$start,sep="_")
+bed.anno$end_idx <- paste(bed.anno$chr,bed.anno$end,sep="_")
+CNV_number <- nrow(CNVs)
+Res <- data.frame()
+for(i in 1:CNV_number) {
+  CNVs.start.idx <- paste(CNVs[i,]$chr,CNVs[i,]$start,sep="_")
+  CNVs.end.idx <- paste(CNVs[i,]$chr,CNVs[i,]$end,sep="_")
+  exon_start <- rownames(bed.anno[bed.anno$start_idx%in%CNVs.start.idx,])
+  exon_end <- rownames(bed.anno[bed.anno$end_idx%in%CNVs.end.idx,])
+  res <- CNVs[i,]
+  res$gene <- unlist(strsplit(bed.anno[exon_start,]$anno,split = "_"))[1]
+  res$exon_start <- unlist(strsplit(bed.anno[exon_start,]$anno,split = "_"))[2]
+  res$exon_end <-  unlist(strsplit(bed.anno[exon_end,]$anno,split = "_"))[2]
+  Res <- rbind(Res,res)
+}
+Res <- data.frame(sampleName=Res$sampleName,chr=Res$chr,
+                  start=Res$start,end=Res$end,gene=Res$gene,
+                  exon_start=Res$exon_start,exon_end=Res$exon_end,
+                  CN=Res$CN)
+#
+##3.4 Output
+write.table(Res,paste0(TumorSampleID,"cn.mops.res.txt"),sep="\t",
+           col.names=T,row.names=F,quote=F)
+
+
+

Docker/cn_mops

@@ -0,0 +1,28 @@
+#!/bin/bash
+
+# Wrapper to easily run an R script at the command line, with arguments,
+# like you're used to with Python/Perl/Ruby/Bash/anything else remotely sane.
+#
+# Usage:
+#   $(echo $0) -p <program_name> <args>
+
+set -eu -o pipefail
+
+export LC_ALL=en_US.UTF-8
+
+# Find original directory of bash script, resolving symlinks
+# http://stackoverflow.com/questions/59895/can-a-bash-script-tell-what-directory-its-stored-in/246128#246128
+SOURCE="${BASH_SOURCE[0]}"
+while [ -h "$SOURCE" ]; do # resolve $SOURCE until the file is no longer a symlink
+    DIR="$( cd -P "$( dirname "$SOURCE" )" && pwd )"
+    SOURCE="$(readlink "$SOURCE")"
+    [[ $SOURCE != /* ]] && SOURCE="$DIR/$SOURCE" # if $SOURCE was a relative symlink, we need to resolve it relative to the path where the symlink file was located
+done
+BINDIR="$( cd -P "$( dirname "$SOURCE" )" && pwd )"
+echo "Binary directory: ${BINDIR}"
+
+if [ -f "$BINDIR/cn.mops.R" ];then
+    Rscript "$BINDIR/cn.mops.R" "$@"
+else
+    echo "No such r script: $BINDIR/cn.mops.R"
+fi

inputs

@@ -1,9 +1,8 @@
 {
   "{{ project_name }}.TumorBam": "{{ TumorBam }}",
   "{{ project_name }}.TumorSampleID": "{{ TumorSampleID }}",
-  "{{ project_name }}.docker": "{{ docker }}",
-  "{{ project_name }}.cluster_config": "{{ cluster_config }}",
-  "{{ project_name }}.workDir": "{{ workDir }}",
+  "{{ project_name }}.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/cn_mops:v0.1.0",
+  "{{ project_name }}.cluster_config": "OnDemand bcs.a2.3xlarge img-ubuntu-vpc",
   "{{ project_name }}.NormalBam": "{{ NormalBam }}",
   "{{ project_name }}.disk_size": "{{ disk_size }}",
   "{{ project_name }}.bed_file": "{{ bed_file }}"

tasks/cn_mops.wdl

@@ -3,13 +3,12 @@ task CNV {
 	String NormalBam
 	String TumorSampleID
 	String bed_file
-	String workDir
 	String docker
 	String disk_size
 	String cluster_config
 	
 	command <<<
-		Rscript cn.mops.R --Tumor=${TumorBam} --Normal=${NormalBam} --TumorID==${TumorSampleID} --bed_file=${bed_file} --workDir=${workDir}
+		cn_mops --Tumor=${TumorBam} --Normal=${NormalBam} --TumorID==${TumorSampleID} --bed_file=${bed_file} --workDir=.
 	>>>
 
 	runtime {
@@ -19,6 +18,6 @@ task CNV {
     	dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
 	}
 	output {
-		File sorted_bam = "${TumorSampleID}.cn.mops.res.txt"
+		File cn_mops = "${TumorSampleID}.cn.mops.res.txt"
 	}
 }