first commit

codescripts/callable_bed_voting.py

@@ -0,0 +1,55 @@
+# import modules
+import sys, argparse, os
+import fileinput
+from operator import itemgetter   
+
+parser = argparse.ArgumentParser(description="this script is to vote callable bed region")
+
+parser.add_argument('-bed', '--multiSampleBED', type=str, help='The bed file to get high confidence region',  required=True)
+parser.add_argument('-prefix', '--prefix', type=str, help='The output file you want to name',  required=True)
+
+args = parser.parse_args()
+
+# Rename input:
+input_file = args.multiSampleBED
+prefix = args.prefix
+
+consensus_filename = prefix + '.27consensus.bed'
+outCONSENSUS = open(consensus_filename,'w')
+filter_filename = prefix + '.filtered.bed'
+outFiltered = open(filter_filename,'w')
+#initial
+#sequence_tech = ['SEQ2000','SEQ2000','SEQ2000','SEQ2000','SEQ2000','SEQ500','SEQ500','SEQ500','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','Nova','XTen','XTen','XTen','XTen','XTen','XTen','XTen','XTen','XTen','XTen','XTen','XTen']
+#sequence_site = ['BGI','BGI','BGI','WGE','WGE','BGI','BGI','BGI','ARD','ARD','ARD','ARD','ARD','ARD','BRG','BRG','BRG','BRG','GAC','NVG','WUX','ARD','ARD','ARD','NVG','NVG','NVG','WUX','WUX','WUX','WUX','WUX','WUX']
+
+def consensus_bed(oneLine):
+	line = oneLine.strip()
+	strings = line.split('\t')
+	# replicate
+	SEQ2000_BGI = 1 if strings[5:8].count('1') > 1 else 0
+	T7_WGE = 1 if strings[8:11].count('1') > 1 else 0 
+	Nova_ARD_1 = 1 if strings[11:14].count('1') > 1 else 0
+	Nova_ARD_2 = 1 if strings[14:17].count('1') > 1 else 0		
+	Nova_BRG = 1 if strings[17:20].count('1') > 1 else 0
+	Nova_WUX = 1 if strings[20:23].count('1') > 1 else 0
+	XTen_ARD = 1 if strings[23:26].count('1') >1 else 0
+	XTen_NVG = 1 if strings[26:29].count('1') > 1 else 0
+	XTen_WUX = 1 if strings[29:32].count('1') > 1 else 0
+	# library
+	pcr = 1 if [SEQ2000_BGI,XTen_ARD,XTen_WUX,XTen_NVG].count(1) > 2 else 0
+	pcr_free = 1 if [T7_WGE,Nova_ARD_1,Nova_ARD_2,Nova_BRG,Nova_WUX].count(1) > 3 else 0
+	voted = 1 if [pcr,pcr_free].count(1) > 1 else 0 
+	# get consensus bed and tech specific bed
+	if voted == 1:
+		outCONSENSUS.write(oneLine)
+	else:
+		outFiltered.write(oneLine)
+
+for oneLine in fileinput.input(input_file):
+	consensus_bed(oneLine)
+
+
+outCONSENSUS.close()
+outFiltered.close()
+
+

codescripts/count_bed.py

@@ -0,0 +1,18 @@
+import sys,getopt
+import fileinput
+
+def process(line):
+	strings = line.strip().split('\t')
+	pos2 = int(strings[2])
+	pos1 = int(strings[1])
+	c = pos2 - pos1
+	return c
+	
+
+result = 0
+
+for line in fileinput.input(sys.argv[1]):
+	C = process(line)
+	result = result + C
+
+print(result)

inputs

@@ -0,0 +1,10 @@
+{
+  "{{ project_name }}.disk_size": "150",
+  "{{ project_name }}.inputSamplesFile": "{{ inputSamplesFile }}",
+  "{{ project_name }}.bedVote.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/high_confidence_call_manuscript:v1.1",
+  "{{ project_name }}.cluster_config": "OnDemand bcs.a2.large img-ubuntu-vpc",
+  "{{ project_name }}.mergeBed.docker": "registry-internal.cn-shanghai.aliyuncs.com/pgx-docker-registry/bedtools:v2.27.1",
+  "{{ project_name }}.CallableLoci.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/gatk:3.8-1",
+  "{{ project_name }}.quartet_sample": "{{ quartet_sample }}"
+}
+

tasks/CallableLoci.wdl

@@ -0,0 +1,29 @@
+task CallableLoci {
+	
+	File bed
+	String sample
+	String docker
+	String disk_size
+	String cluster_config
+	
+
+	command <<<
+
+		cat ${bed} | grep CALLABLE > ${sample}.CALLABLE.bed
+		>>>
+
+		runtime {
+		docker:docker
+    	cluster:cluster_config
+    	systemDisk: "cloud_ssd 40"
+    	dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+	}
+
+	output {
+		File callable_bed = "${sample}.CALLABLE.bed"
+	}
+}
+
+
+
+

tasks/bedVote.wdl

@@ -0,0 +1,31 @@
+task bedVote {
+	
+	File merged_bed
+	String sample
+	String docker
+	String disk_size
+	String cluster_config
+	
+
+	command <<<
+		
+		python /opt/callable_bed_voting.py -bed ${merged_bed} -prefix ${sample}
+
+		>>>
+
+		runtime {
+		docker:docker
+    	cluster:cluster_config
+    	systemDisk: "cloud_ssd 40"
+    	dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+	}
+
+	output {
+		File consensus_bed = "${sample}.27consensus.bed"
+		File filtered_bed = "${sample}.filtered.bed"
+	}
+}
+
+
+
+

tasks/gvcf_homo_ref.wdl

@@ -0,0 +1,33 @@
+task gvcf_homo_ref {
+	
+	File gvcf
+	String sample
+	String docker
+	String disk_size
+	String cluster_config
+	
+
+	command <<<
+
+		awk '{ if ($5 == "<NON_REF>") { print } }'  ${gvcf} | sed s'/:/\t/'g | awk '{ if($16 >= 30) { print }}' | cut -f1,2,8 | sed s'/END=//g' > ${sample}.filtered.bed
+
+		bedtools merged -i ${sample}.filtered.bed > ${sample}.homo_ref.bed 
+
+		>>>
+
+		runtime {
+		docker:docker
+    	cluster:cluster_config
+    	systemDisk: "cloud_ssd 40"
+    	dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+	}
+
+	output {
+		File filtered_bed = "${sample}.filtered.bed"
+		File homo_ref_bed = "${sample}.homo_ref.bed"
+	}
+}
+
+
+
+

tasks/mergeBed.wdl

@@ -0,0 +1,30 @@
+task mergeBed {
+	
+	Array[File] callable_bed
+	String sample
+	String docker
+	String disk_size
+	String cluster_config
+	
+
+	command <<<
+		
+		/opt/ccdg/bedtools-2.27.1/bin/bedtools multiinter -i ${sep=" " callable_bed} > ${sample}.CALLABLE.merged.bed
+
+		>>>
+
+		runtime {
+		docker:docker
+    	cluster:cluster_config
+    	systemDisk: "cloud_ssd 40"
+    	dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+	}
+
+	output {
+		File merged_bed = "${sample}.CALLABLE.merged.bed"
+	}
+}
+
+
+
+

workflow.wdl

@@ -0,0 +1,38 @@
+import "./tasks/gvcf_homo_ref.wdl" as gvcf_homo_ref
+import "./tasks/mergeBed.wdl" as mergeBed
+import "./tasks/bedVote.wdl" as bedVote
+
+
+workflow project_name {
+
+	File inputSamplesFile
+	Array[Array[File]] inputSamples = read_tsv(inputSamplesFile)
+	String quartet_sample
+	String disk_size
+	String cluster_config
+	
+	scatter (quartet in inputSamples){
+		call gvcf_homo_ref.gvcf_homo_ref as gvcf_homo_ref {
+			input: 
+			gvcf=quartet[0],
+			sample=quartet[1],
+			disk_size=disk_size,
+			cluster_config=cluster_config
+		}
+	}
+	call mergeBed.mergeBed as mergeBed {
+		input:
+		callable_bed=gvcf_homo_ref.homo_ref_bed,
+		sample=quartet_sample,
+		disk_size=disk_size,
+		cluster_config=cluster_config
+	}
+	call bedVote.bedVote as bedVote {
+		input:
+		merged_bed=mergeBed.merged_bed,
+		sample=quartet_sample,
+		disk_size=disk_size,
+		cluster_config=cluster_config
+	}
+}
+