merge

codescripts/merge_two_family_with_genotype.py

+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('-genotype', '--genotype', type=str, help='Genotype information of a set of four family members',  required=True)
+parser.add_argument('-family', '--family', type=str, help='family name',  required=True)
+
+
+args = parser.parse_args()
+lcl5 = args.LCL5
+lcl6 = args.LCL6
+genotype = args.genotype
+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)
+genotype_dat = pd.read_table(genotype)
+merged_df = pd.merge(lcl5_dat, lcl6_dat,  how='outer', left_on=['#CHROM','POS'], right_on = ['#CHROM','POS'])
+merged_genotype_df = pd.merge(merged_df, genotype_dat,  how='outer', left_on=['#CHROM','POS'], right_on = ['#CHROM','POS'])
+
+merged_genotype_df_sub = merged_genotype_df.iloc[:,[0,1,22,23,24,25,26,27,7,17]]
+merged_genotype_df_sub.columns = ['CHROM', 'POS', 'REF', 'ALT','LCL5','LCL6','LCL7','LCL8', 'TRIO5', 'TRIO6']
+
+for row in merged_genotype_df_sub.itertuples():
+	# sister
+	if row.LCL5 == row.LCL6:
+		if row.LCL5 == './.':
+			mendelian = 'noInfo'
+			sister_count = "no"
+		elif row.LCL5 == '0/0':
+			mendelian = 'Ref'
+			sister_count = "no"
+		else:
+			mendelian = '1'
+			sister_count = "yes_same"
+
+	else:
+		mendelian = '0'
+		if (row.LCL5 == './.' or row.LCL5 == '0/0') and (row.LCL6 == './.' or row.LCL6 == '0/0'):
+			sister_count = "no"
+		else:
+			sister_count = "yes_diff"
+	# family trio5
+	if row.LCL5 == row. LCL7 == row.LCL8 == './.':
+		mendelian = mendelian + ':noInfo'
+	elif row.LCL5 == row. LCL7 == row.LCL8 == '0/0':
+		mendelian = mendelian + ':Ref'
+	elif pd.isnull(row.TRIO5) == True:
+		mendelian = mendelian + ':unVBT'
+	else:
+		mendelian = mendelian + ':' + row.TRIO5.split('=')[1]
+	# family trio6
+	if row.LCL6 == row. LCL7 == row.LCL8 == './.':
+		mendelian = mendelian + ':noInfo'
+	elif row.LCL6 == row. LCL7 == row.LCL8 == '0/0':
+		mendelian = mendelian + ':Ref'
+	elif pd.isnull(row.TRIO6) == True:
+		mendelian = mendelian + ':unVBT'
+	else:
+		mendelian =  mendelian + ':' + row.TRIO6.split('=')[1]
+	# not count into family
+	if (row.LCL5 == './.' or row.LCL5 == '0/0') and (row.LCL6 == './.' or row.LCL6 == '0/0') and (row.LCL7 == './.' or row.LCL7 == '0/0') and (row.LCL8 == './.' or row.LCL8 == '0/0'):
+		mendelian_count = "no"
+	else:
+		mendelian_count = "yes"
+	outline = row.CHROM + '\t' + str(row.POS) + '\t' + row.REF + '\t' + row.ALT + '\t' + row.LCL5 + '\t' + row.LCL6 + '\t' + row.LCL7 + '\t' + row.LCL8 + '\t' + str(row.TRIO5) + '\t' + str(row.TRIO6) + '\t' + str(mendelian) + '\t' + str(mendelian_count) + '\t' + str(sister_count) + '\n'
+	family_file.write(outline)

inputs

 {
   "{{ project_name }}.fasta": "GRCh38.d1.vd1.fa",
+  "{{ project_name }}.two_family_merge.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/high_confidence_call_manuscript:v1.4",
+  "{{ project_name }}.family_name": "{{ family_name }}",
+  "{{ project_name }}.sister.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/vbt:v1.1",
   "{{ project_name }}.disk_size": "150",
+  "{{ project_name }}.merge.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/high_confidence_call_manuscript:v1.4",
   "{{ project_name }}.inputSamplesFile": "{{ inputSamplesFile }}",
-  "{{ project_name }}.docker": "registry-vpc.cn-shanghai.aliyuncs.com/pgx-docker-registry/vbt:v1.1",
   "{{ project_name }}.cluster_config": "OnDemand bcs.b4.xlarge img-ubuntu-vpc",
   "{{ project_name }}.ref_dir": "oss://chinese-quartet/quartet-storage-data/reference_data/"
-}
+}

tasks/VCFinfo.wdl

-task VCFinfo {
-	File repeat_annotated_vcf
-	String sample
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		python /opt/variants_quality_location_intergration.py -vcf ${repeat_annotated_vcf} -prefix ${sample}
-
-		cat ${sample}_variant_quality_location.txt | grep '#CHROM' > header
-
-		for i in chr1 chr2 chr3 chr4 chr5 chr6 chr7 chr8 chr9 chr10 chr11 chr12 chr13 chr14 chr15 chr16 chr17 chr18 chr19 chr20 chr21 chr22 chrX
-		do
-			cat ${sample}_variant_quality_location.txt | grep -w $i | cat header - > ${sample}.$i.vcfInfo.txt
-		done
-
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File extracted_info = "${sample}_variant_quality_location.txt"
-		Array[File] chromo_vcfInfo = glob("*.vcfInfo.txt")
-	}
-}

tasks/VCFrename.wdl

-task VCFrename {
-	File trio_vcf_gz
-	File trio_vcf_idx
-	String mother_name
-	String father_name
-	String child_name
-	String family_name
-	String child
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-		echo "MOTHER ${mother_name}.${child}
-		FATHER ${father_name}.${child}
-		CHILD ${child_name}" > rename.txt
-
-		rtg vcfannotate -i ${trio_vcf_gz} -o ${family_name}.${child}.rename.vcf.gz --relabel=rename.txt
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File rename_trio_vcf_gz = "${family_name}.${child}.rename.vcf.gz"
-		File rename_trio_vcf_idx = "${family_name}.${child}.rename.vcf.gz.tbi"
-	}
-}

tasks/bed_annotation.wdl

-task bed_annotation {
-	File merged_vcf_gz
-	File merged_vcf_idx
-	File repeat_bed
-	String sample
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		rtg vcfannotate --bed-info=${repeat_bed} -i ${merged_vcf_gz} -o ${sample}.mendelian.merged.repeatAnno.vcf.gz
-
-		gunzip ${sample}.mendelian.merged.repeatAnno.vcf.gz
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File repeat_annotated_vcf = "${sample}.mendelian.merged.repeatAnno.vcf"
-	}
-}

tasks/final_result.wdl

-task FinalResult {
-	File extracted_info
-	File annotated_txt
-	String prefix = basename(annotated_txt,".mendelian.txt")
-	String sample
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		python /opt/FinalResult2VCF.py -vcfInfo ${extracted_info} -mendelianInfo ${annotated_txt} -prefix ${prefix} -sample ${sample}
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File benchmarking_calls = "${prefix}_benchmarking_calls.vcf"
-		File all_info = "${prefix}_all_sample_information.vcf"
-	}
-}

tasks/mendelian.wdl

-task mendelian {
-	File child_vcf
-	File LCL7_vcf
-	File LCL8_vcf
-	String LCL7_name
-	String LCL8_name
-	String child_name
-	File ref_dir
-	String fasta
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-		export LD_LIBRARY_PATH=/opt/htslib-1.9
-		nt=$(nproc)
-		mkdir VBT
-
-		/opt/VBT-TrioAnalysis/vbt mendelian -ref ${ref_dir}/${fasta} -mother ${LCL8_vcf} -father ${LCL7_vcf} -child ${child_vcf} -outDir VBT -out-prefix ${child_name}.family --output-violation-regions -thread-count $nt
-
-		cat VBT/${child_name}.family_trio.vcf > ${child_name}.family.vcf
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		Array[File] vbt_mendelian = glob("VBT/*")
-		File trio_vcf = "${child_name}.family.vcf"
-	}
-}

tasks/merge.wdl

 task merge {
-	Array[File] family_vcf_gz
-	Array[File] family_vcf_idx
-	String sample
+	Array[File] family_mendelian_info
+	String family_name
 	String docker
 	String cluster_config
 	String disk_size
 	
 	command <<<
 
-		rtg vcfmerge --force-merge-all -o ${sample}.merged.vcf.gz ${sep=" " family_vcf_gz}
+		cat ${sep=" " family_mendelian_info} | sort -k1,1 -k2,2n > ${family_name}.mendelian.txt
 
-		zcat ${sample}.merged.vcf.gz | grep -v '#' | cut -f1-2 | sed s'/\t/_/g' | sort | uniq -c | sed 's/\s\+/\t/g' | awk '{ if ($1 != 1) { print } }' | cut -f3 > ${sample}.vcf_dup.txt
+		cat ${family_name}.mendelian.txt | cut -f13 | sort | uniq -c > ${family_name}.sister.reproducibility.txt
+
+		cat ${family_name}.mendelian.txt | cut -f11 | sort | uniq -c | grep '1:1:1' > ${family_name}.mendelian.txt
+		cat ${family_name}.mendelian.txt | cut -f11 | sort | uniq -c | grep 'Ref:1:1' >> ${family_name}.mendelian.txt
+		cat ${family_name}.mendelian.txt | cut -f12 | sort | uniq -c | grep 'yes' >> ${family_name}.mendelian.txt
 
 	>>>
 
 		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
 	}
 	output {
-		File merged_vcf_gz = "${sample}.merged.vcf.gz"
-		File merged_vcf_idx = "${sample}.merged.vcf.gz.tbi"
-		File vcf_dup = "${sample}.vcf_dup.txt"
+		File family_all_info = "${family_name}.mendelian.txt"
+		File sister_consistency = "${family_name}.sister.reproducibility.txt"
+		File family_mendelian = "${family_name}.mendelian.txt"
 	}
 }

tasks/mergeSister.wdl

-task mergeSister {
-	File LCL5_trio_vcf_gz
-	File LCL5_trio_vcf_idx
-	File LCL6_trio_vcf_gz
-	File LCL6_trio_vcf_idx
-	String family_name
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-		rtg vcfmerge -o LCL5.LCL6.merged.vcf.gz ${LCL5_trio_vcf_gz} ${LCL6_trio_vcf_gz}
-
-		rtg vcfmerge -o LCL6.LCL5.merged.vcf.gz ${LCL6_trio_vcf_gz} ${LCL5_trio_vcf_gz}
-
-		zcat LCL5.LCL6.merged.vcf.gz | grep '##' > header
-		zcat LCL5.LCL6.merged.vcf.gz | grep -v '##' | cut -f8 > LCL5.mendelian
-		zcat LCL6.LCL5.merged.vcf.gz | grep -v '##' | paste - LCL5.mendelian > body
-
-		cat header body > ${family_name}.trio.info.vcf
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-
-	output {
-		File family_mendelian_info = "${family_name}.trio.info.vcf"
-	}
-
-}

tasks/mergeVCFInfo.wdl

-task mergeVCFInfo {
-	Array[File] vcf_gz
-	Array[File] vcf_idx
-	String sample
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		rtg vcfmerge --force-merge-all -o ${sample}.merged.info.vcf.gz ${sep=" " vcf_gz}
- 
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File merged_vcf = "${sample}.merged.info.vcf.gz"
-		File merged_vcf_idx = "${sample}.merged.info.vcf.gz.tbi"
-	}
-}

tasks/merge_info.wdl

-task merge_info {
-	File vcfInfo
-	File mendelianInfo
-	String sample
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		python /opt/merge_mendelian_vcfinfo.py -vcfInfo ${vcfInfo} -mendelianInfo ${mendelianInfo} -sample ${sample}
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File all_info = "${sample}_mendelian_vcfInfo.vcf"
-	}
-}

tasks/oneClass.wdl

-task oneClass {
-	File snv_train_vcf
-	File snv_test_vcf
-	File indel_train_vcf
-	File indel_test_vcf
-	String sampleName = basename(snv_train_vcf,".normed.snv.train.txt")
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-	python /opt/oneClass.py -train ${snv_train_vcf} -test ${snv_test_vcf} -name ${sampleName}_snv
-
-	python /opt/oneClass.py -train ${indel_train_vcf} -test ${indel_test_vcf} -name ${sampleName}_indel	
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-
-	output {
-		File snv_true_txt = "${sampleName}_snv_predicted_true.txt"
-		File snv_false_txt = "${sampleName}_snv_predicted_false.txt"
-		File snv_true_bed = "${sampleName}_snv_predicted_true.bed"
-		File snv_false_bed = "${sampleName}_snv_predicted_false.bed"
-		File snv_padding = "${sampleName}_snv_padding.bed"
-		File indel_true_txt = "${sampleName}_indel_predicted_true.txt"
-		File indel_false_txt = "${sampleName}_indel_predicted_false.txt"
-		File indel_true_bed = "${sampleName}_indel_predicted_true.bed"
-		File indel_false_bed = "${sampleName}_indel_predicted_false.bed"
-		File indel_padding = "${sampleName}_indel_padding.bed"
-	}
-}
-

tasks/reformVCF.wdl

-task reformVCF {
-	File family_mendelian_info
-	File family_name
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-	python /opt/reformVCF.py -vcf ${family_mendelian_info} -name ${family_name}
-
-	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 '\./\.' > ${family_name}.LCL8.txt
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-
-	output {
-		File LCL5_family_info = "${family_name}.LCL5.vcf"
-		File LCL6_family_info = "${family_name}.LCL6.vcf"
-		File LCL7_family_info = "${family_name}.LCL7.vcf"
-		File LCL8_family_info = "${family_name}.LCL8.vcf"
-		File family_info = "${family_name}.vcf"
-		File LCL5_family_info_txt = "${family_name}.LCL5.txt"
-		File LCL6_family_info_txt = "${family_name}.LCL6.txt"
-		File LCL7_family_info_txt = "${family_name}.LCL7.txt"
-		File LCL8_family_info_txt = "${family_name}.LCL8.txt"
-	}
-}
-

tasks/sister.wdl

 	String LCL7_name
 	String LCL8_name
 	String fasta
-	String family_name
+	String family_chromo_name
 	String docker
 	String cluster_config
 	String disk_size
 		cat ${LCL7_vcf} | grep -v '##' | cut -f10 > F7
 		cat ${LCL8_vcf} | grep -v '##' | cut -f10 > M8
 		cat ${LCL5_vcf} | grep -v '##' | paste - D6 F7 M8 > body
-		cat ${LCL5_vcf} | grep '##' | cat - body > ${family_name}.vcf
+		cat ${LCL5_vcf} | grep '##' | cat - body > ${family_chromo_name}.vcf
 	# prepare ped file, D5
-		echo "${family_name}	${LCL8_name}	0	0	2	-9
-		${family_name}	${LCL7_name}	0	0	1	-9
-		${family_name}	${LCL5_name}	${LCL7_name}	${LCL8_name}	2	-9" > ${family_name}.D5.ped
+		echo "${family_chromo_name}	${LCL8_name}	0	0	2	-9
+		${family_chromo_name}	${LCL7_name}	0	0	1	-9
+		${family_chromo_name}	${LCL5_name}	${LCL7_name}	${LCL8_name}	2	-9" > ${family_chromo_name}.D5.ped
 
 		mkdir VBT_D5
-		/opt/VBT-TrioAnalysis/vbt mendelian -ref ${ref_dir}/${fasta} -mother ${family_name}.vcf -father ${family_name}.vcf -child ${family_name}.vcf -pedigree ${family_name}.D5.ped -outDir VBT_D5 -out-prefix ${family_name}.D5 --output-violation-regions -thread-count $nt
+		/opt/VBT-TrioAnalysis/vbt mendelian -ref ${ref_dir}/${fasta} -mother ${family_chromo_name}.vcf -father ${family_chromo_name}.vcf -child ${family_chromo_name}.vcf -pedigree ${family_chromo_name}.D5.ped -outDir VBT_D5 -out-prefix ${family_chromo_name}.D5 --output-violation-regions -thread-count $nt
+
+		cat VBT_D5/${family_chromo_name}.D5_trio.vcf > ${family_chromo_name}.D5.vcf
 	# prepare ped file, D6
-		echo "${family_name}	${LCL8_name}	0	0	2	-9
-		${family_name}	${LCL7_name}	0	0	1	-9
-		${family_name}	${LCL6_name}	${LCL7_name}	${LCL8_name}	2	-9" > ${family_name}.D6.ped
+		echo "${family_chromo_name}	${LCL8_name}	0	0	2	-9
+		${family_chromo_name}	${LCL7_name}	0	0	1	-9
+		${family_chromo_name}	${LCL6_name}	${LCL7_name}	${LCL8_name}	2	-9" > ${family_chromo_name}.D6.ped
 
 		mkdir VBT_D6
-		/opt/VBT-TrioAnalysis/vbt mendelian -ref ${ref_dir}/${fasta} -mother ${family_name}.vcf -father ${family_name}.vcf -child ${family_name}.vcf -pedigree ${family_name}.D6.ped -outDir VBT_D6 -out-prefix ${family_name}.D6 --output-violation-regions -thread-count $nt
+		/opt/VBT-TrioAnalysis/vbt mendelian -ref ${ref_dir}/${fasta} -mother ${family_chromo_name}.vcf -father ${family_chromo_name}.vcf -child ${family_chromo_name}.vcf -pedigree ${family_chromo_name}.D6.ped -outDir VBT_D6 -out-prefix ${family_chromo_name}.D6 --output-violation-regions -thread-count $nt
+
+		cat VBT_D6/${family_chromo_name}.D6_trio.vcf > ${family_chromo_name}.D6.vcf
 	>>>
 
 	runtime {
 	output {
 		Array[File] D5_mendelian = glob("VBT_D5/*")
 		Array[File] D6_mendelian = glob("VBT_D6/*")
-		File family_vcf = "${family_name}.vcf"
+		File D5_trio_vcf = "${family_chromo_name}.D5.vcf"
+		File D6_trio_vcf = "${family_chromo_name}.D6.vcf"
+		File family_vcf = "${family_chromo_name}.vcf"
 	}
 }

tasks/two_family_merge.wdl

 task two_family_merge {
 	File LCL5_trio_vcf
 	File LCL6_trio_vcf
-	String family_name
+	File genotype_file
+	String family_chromo_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}
+		cat ${LCL5_trio_vcf} | grep -v '##' > ${family_chromo_name}.LCL5.txt
+		cat ${LCL6_trio_vcf} | grep -v '##' > ${family_chromo_name}.LCL6.txt
+		cat ${genotype_file} | grep -v '##' | awk '
+			BEGIN { OFS = "\t" }
+			NF > 2 && FNR > 1 { 
+				for ( i=9; i<=NF; i++ ) { 
+					split($i,a,":") ;$i = a[1];
+				} 
+			} 
+			{ print }
+		' | cut -f1,2,4,5,10- > ${family_chromo_name}.genotype.txt
+		python /opt/merge_two_family_with_genotype.py -LCL5 ${family_chromo_name}.LCL5.txt -LCL6 ${family_chromo_name}.LCL6.txt -genotype ${family_chromo_name}.genotype.txt -family ${family_chromo_name}
 	>>>
 
 	runtime {
 	}
 
 	output {
-		File family_mendelian_info = "${family_name}.txt"
+		File family_mendelian_info = "${family_chromo_name}.txt"
 	}
 
 }

tasks/variantsNorm.wdl

-task variantsNorm {
-	File vcf
-	File ref_dir
-	String fasta
-	String sampleName
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-
-		cat ${vcf} | grep '#' > header
-		cat ${vcf} | grep -v '#' > body
-		cat body | grep -w '^chr1\|^chr2\|^chr3\|^chr4\|^chr5\|^chr6\|^chr7\|^chr8\|^chr9\|^chr10\|^chr11\|^chr12\|^chr13\|^chr14\|^chr15\|^chr16\|^chr17\|^chr18\|^chr19\|^chr20\|^chr21\|^chr22\|^chrX' > body.filtered
-		cat header body.filtered > ${sampleName}.filtered.vcf
-
-		/opt/hall-lab/bcftools-1.9/bin/bcftools norm -f ${ref_dir}/${fasta} ${sampleName}.filtered.vcf > ${sampleName}.normed.vcf
-
-		cat ${sampleName}.normed.vcf | grep -v '##' > ${sampleName}.normed.txt
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File normed_vcf = "${sampleName}.normed.vcf"
-		File normed_txt = "${sampleName}.normed.txt"
-	}
-}

tasks/votes.wdl

-task votes {
-	Array[File] family_mendelian_info
-	File vcf 
-	String chromo
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-		mkdir temp
-		for i in ${sep=" " family_mendelian_info}
-		do
-			cp $i temp
-		done
-
-		cat ${vcf} | grep -v '##' > vcf_info.txt
-
-		python /opt/voted_by_vcfinfo_mendelianinfo.py -folder ./temp -vcf vcf_info.txt
-
-		cp LCL5_voted.vcf LCL5.${chromo}.voted.vcf
-		cp LCL6_voted.vcf LCL6.${chromo}.voted.vcf
-		cp LCL7_voted.vcf LCL7.${chromo}.voted.vcf
-		cp LCL8_voted.vcf LCL8.${chromo}.voted.vcf
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File LCL5_voted_vcf = "LCL5.${chromo}.voted.vcf"
-		File LCL6_voted_vcf = "LCL6.${chromo}.voted.vcf"
-		File LCL7_voted_vcf = "LCL7.${chromo}.voted.vcf"
-		File LCL8_voted_vcf = "LCL8.${chromo}.voted.vcf"
-		File all_sample_info = "all_sample_information.txt"
-	}
-}
-

tasks/zipIndex.wdl

-task zipIndex {
-	File vcf
-	String vcf_name = basename(vcf,".vcf")
-	String docker
-	String cluster_config
-	String disk_size
-	
-	command <<<
-		rtg bgzip ${vcf} -c > ${vcf_name}.vcf.gz
-		rtg index -f vcf ${vcf_name}.vcf.gz
-
-	>>>
-
-	runtime {
-		docker:docker
-		cluster: cluster_config
-		systemDisk: "cloud_ssd 40"
-		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
-	}
-	output {
-		File vcf_gz = "${vcf_name}.vcf.gz"
-		File vcf_idx = "${vcf_name}.vcf.gz.tbi"
-	}
-}

workflow.wdl

 import "./tasks/sister.wdl" as sister
+import "./tasks/two_family_merge.wdl" as two_family_merge
+import "./tasks/merge.wdl" as merge
 
 workflow {{ project_name }} {
 	File inputSamplesFile
 	Array[Array[File]] inputSamples = read_tsv(inputSamplesFile)
 	File ref_dir
-	String docker
+	String family_name
 	String fasta
 	String cluster_config
 	String disk_size
 			LCL7_name=quartet[6],
 			LCL8_name=quartet[7],
 			fasta=fasta,
-			family_name=quartet[8],
-			docker=docker,
+			family_chromo_name=quartet[8],
 			cluster_config=cluster_config,
 			disk_size=disk_size
 		}
+		call two_family_merge.two_family_merge as two_family_merge {
+			input:
+			LCL5_trio_vcf=sister.D5_trio_vcf,
+			LCL6_trio_vcf=sister.D6_trio_vcf,
+			genotype_file=sister.family_vcf,
+			family_chromo_name=quartet[8],
+			cluster_config=cluster_config,
+			disk_size=disk_size
+		}
+	}
+	call merge.merge as merge {
+		input:
+		family_mendelian_info=two_family_merge.family_mendelian_info,
+		family_name=family_name,
+		cluster_config=cluster_config,
+		disk_size=disk_size
 	}
 }