docker

codescripts/merge_two_family.py

@@ -0,0 +1,49 @@
+from __future__ import division
+import pandas as pd
+import sys, argparse, os
+import fileinput
+import re
+
+# input arguments
+parser = argparse.ArgumentParser(description="this script is to extract mendelian concordance information")
+
+parser.add_argument('-LCL5', '--LCL5', type=str, help='LCL5 family info',  required=True)
+parser.add_argument('-LCL6', '--LCL6', type=str, help='LCL6 family info',  required=True)
+parser.add_argument('-family', '--family', type=str, help='family name',  required=True)
+
+
+args = parser.parse_args()
+lcl5 = args.LCL5
+lcl6 = args.LCL6
+family = args.family
+
+
+# output file
+family_name = family + '.txt'
+
+family_file = open(family_name,'w')
+
+# input files
+lcl5_dat = pd.read_table(lcl5)
+lcl6_dat = pd.read_table(lcl6)
+
+merged_df = pd.merge(lcl5_dat, lcl6_dat,  how='outer', left_on=['#CHROM','POS'], right_on = ['#CHROM','POS'])
+
+
+for row in merged_df.itertuples():
+	if row.CHILD_x == row.CHILD_y:
+		mendelian = '1'
+	else:
+		mendelian = '0'
+	if pd.isnull(row.INFO_x) == True:
+		mendelian = mendelian + ':.'
+	else:
+		mendelian = mendelian + ':' + row.INFO_x.split('=')[1]
+	if pd.isnull(row.INFO_y) == True:
+		mendelian = mendelian + ':.'
+	else:
+		mendelian = mendelian + ':' + row.INFO_y.split('=')[1]
+
+
+	outline = row._1 + '\t' + str(row.POS) + '\t' + mendelian + '\n'
+	family_file.write(outline)

codescripts/reformVCF.py

@@ -83,14 +83,14 @@ def process(oneLine):
 	line = oneLine.rstrip()
 	strings = line.strip().split('\t')
 	# replace .
-	# LCL5 uniq
+	# LCL6 uniq
 	if strings[11] == '.':
 		strings[11] = '0/0'
 		strings[9] = strings[12]
 		strings[10] = strings[13]
 	else:
 		pass
-	# LCL6 uniq
+	# LCL5 uniq
 	if strings[14] == '.':
 		strings[14] = '0/0'
 		strings[12] = strings[9]

inputs

@@ -1,15 +1,10 @@
 {
   "{{ project_name }}.fasta": "GRCh38.d1.vd1.fa",
-  "{{ project_name }}.LCL5VCFrename.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/rtg-tools:latest",
+  "{{ project_name }}.two_family_merge.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/high_confidence_call_manuscript:v1.2",
   "{{ project_name }}.LCL6mendelian.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/vbt:v1.1",
-  "{{ project_name }}.mergeSister.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/rtg-tools:latest",
   "{{ project_name }}.LCL5mendelian.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/vbt:v1.1",
   "{{ project_name }}.disk_size": "150",
   "{{ project_name }}.inputSamplesFile": "{{ inputSamplesFile }}",
-  "{{ project_name }}.LCL6zipIndex.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/rtg-tools:latest",
-  "{{ project_name }}.LCL6VCFrename.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/rtg-tools:latest",
-  "{{ project_name }}.reformVCF.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/high_confidence_call_manuscript:v1.1",
-  "{{ project_name }}.LCL5zipIndex.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/rtg-tools:latest",
   "{{ project_name }}.cluster_config": "OnDemand bcs.b4.xlarge img-ubuntu-vpc",
   "{{ project_name }}.ref_dir": "oss://chinese-quartet/quartet-storage-data/reference_data/"
 }

tasks/reformVCF.wdl

@@ -9,10 +9,10 @@ task reformVCF {
 
 	python /opt/reformVCF.py -vcf ${family_mendelian_info} -name ${family_name}
 
-	cat ${family_name}.LCL5.vcf | grep -v '##' | grep -v '0/0' > ${family_name}.LCL5.txt
-	cat ${family_name}.LCL6.vcf | grep -v '##' | grep -v '0/0' > ${family_name}.LCL6.txt
-	cat ${family_name}.LCL7.vcf | grep -v '##' | grep -v '0/0' > ${family_name}.LCL7.txt
-	cat ${family_name}.LCL8.vcf | grep -v '##' | grep -v '0/0' > ${family_name}.LCL8.txt
+	cat ${family_name}.LCL5.vcf | grep -v '##' | grep -v '0/0' | grep -v '\./\.' > ${family_name}.LCL5.txt
+	cat ${family_name}.LCL6.vcf | grep -v '##' | grep -v '0/0' | grep -v '\./\.' > ${family_name}.LCL6.txt
+	cat ${family_name}.LCL7.vcf | grep -v '##' | grep -v '0/0' | grep -v '\./\.' > ${family_name}.LCL7.txt
+	cat ${family_name}.LCL8.vcf | grep -v '##' | grep -v '0/0' | grep -v '\./\.' > ${family_name}.LCL8.txt
 
 	>>>
 

tasks/two_family_merge.wdl

@@ -0,0 +1,26 @@
+task two_family_merge {
+	File LCL5_trio_vcf
+	File LCL6_trio_vcf
+	String family_name
+	String docker
+	String cluster_config
+	String disk_size
+	
+	command <<<
+		cat ${LCL5_trio_vcf} | grep -v '##' > ${family_name}.LCL5.txt
+		cat ${LCL6_trio_vcf} | grep -v '##' > ${family_name}.LCL6.txt
+		python opt/merge_two_family.py -LCL5 ${family_name}.LCL5.txt -LCL6 ${family_name}.LCL6.txt -family ${family_name}
+	>>>
+
+	runtime {
+		docker:docker
+		cluster: cluster_config
+		systemDisk: "cloud_ssd 40"
+		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+	}
+
+	output {
+		File family_mendelian_info = "${family_name}.txt"
+	}
+
+}

workflow.wdl

@@ -1,8 +1,5 @@
 import "./tasks/mendelian.wdl" as mendelian
-import "./tasks/zipIndex.wdl" as zipIndex
-import "./tasks/VCFrename.wdl" as VCFrename
-import "./tasks/mergeSister.wdl" as mergeSister
-import "./tasks/reformVCF.wdl" as reformVCF
+import "./tasks/two_family_merge.wdl" as two_family_merge
 
 
 workflow {{ project_name }} {
@@ -40,58 +37,13 @@ workflow {{ project_name }} {
 			cluster_config=cluster_config,
 			disk_size=disk_size
 		}
-		call zipIndex.zipIndex as LCL5zipIndex {
+		call two_family_merge.two_family_merge as two_family_merge {
 			input:
-			vcf=LCL5mendelian.trio_vcf,
-			cluster_config=cluster_config,
-			disk_size=disk_size
-		}
-		call zipIndex.zipIndex as LCL6zipIndex {
-			input:
-			vcf=LCL6mendelian.trio_vcf,
-			cluster_config=cluster_config,
-			disk_size=disk_size
-		}		
-		call VCFrename.VCFrename as LCL5VCFrename {
-			input:
-			trio_vcf_gz=LCL5zipIndex.vcf_gz,
-			trio_vcf_idx=LCL5zipIndex.vcf_idx,
-			mother_name=quartet[7],
-			father_name=quartet[6],
-			child_name=quartet[4],
+			LCL5_trio_vcf=LCL5mendelian.trio_vcf,
+			LCL6_trio_vcf=LCL6mendelian.trio_vcf,
 			family_name=quartet[8],
-			child="LCL5",
 			cluster_config=cluster_config,
 			disk_size=disk_size
 		}
-		call VCFrename.VCFrename as LCL6VCFrename {
-			input:
-			trio_vcf_gz=LCL6zipIndex.vcf_gz,
-			trio_vcf_idx=LCL6zipIndex.vcf_idx,
-			mother_name=quartet[7],
-			father_name=quartet[6],
-			child_name=quartet[5],
-			family_name=quartet[8],
-			child="LCL6",
-			cluster_config=cluster_config,
-			disk_size=disk_size
-		}
-		call mergeSister.mergeSister as mergeSister {
-			input:
-			LCL5_trio_vcf_gz=LCL5VCFrename.rename_trio_vcf_gz,
-			LCL5_trio_vcf_idx=LCL5VCFrename.rename_trio_vcf_idx,
-			LCL6_trio_vcf_gz=LCL6VCFrename.rename_trio_vcf_gz,
-			LCL6_trio_vcf_idx=LCL6VCFrename.rename_trio_vcf_idx,
-			family_name=quartet[8],
-			cluster_config=cluster_config,
-			disk_size=disk_size
-		}
-		call reformVCF.reformVCF as reformVCF {
-			input:
-			family_mendelian_info=mergeSister.family_mendelian_info,
-			family_name=quartet[8],
-			cluster_config=cluster_config,
-			disk_size=disk_size			
-		}
 	}
 }