quartet-wgs-germline-qc/tasks/split_gvcf_files.wdl

task split_gvcf_files {
	File gvcf
	String project
	String docker
	String cluster_config
	String disk_size
	
	command <<<

		cat ${gvcf} | grep '#CHROM' | sed s'/\t/\n/g' > name

		ncol=`cat name | wc -l`

		sed -i '1,9d' name

		for i in $(seq 1 $ncol);  do  cat ${gvcf}| cut -f1-9,$i > $i.splited.vcf; done

		ls *splited.vcf | sort -n | paste - name > rename

		cat rename | while read a b
		do
			mv $a $b.vcf
			if [[ $b.vcf =~ "LCL5_1" ]];then
				cp $b.vcf ${project}.LCL5_1.vcf
		    elif [[ $b.vcf =~ "LCL5_2" ]]; then
		    	cp $b.vcf ${project}.LCL5_2.vcf
	        elif [[ $b.vcf =~ "LCL5_3" ]]; then
	        	cp $b.vcf ${project}.LCL5_3.vcf
		    elif [[ $b.vcf =~ "LCL6_1" ]]; then
				cp $b.vcf ${project}.LCL6_1.vcf
		    elif [[ $b.vcf =~ "LCL6_2" ]]; then
				cp $b.vcf ${project}.LCL6_2.vcf
		    elif [[ $b.vcf =~ "LCL6_3" ]]; then
				cp $b.vcf ${project}.LCL6_3.vcf
		    elif [[ $b.vcf =~ "LCL7_1" ]]; then
				cp $b.vcf ${project}.LCL7_1.vcf
		    elif [[ $b.vcf =~ "LCL7_2" ]]; then
				cp $b.vcf ${project}.LCL7_2.vcf
		    elif [[ $b.vcf =~ "LCL7_3" ]]; then
				cp $b.vcf ${project}.LCL7_3.vcf
		    elif [[ $b.vcf =~ "LCL8_1" ]]; then
				cp $b.vcf ${project}.LCL8_1.vcf
		    elif [[ $b.vcf =~ "LCL8_2" ]]; then
				cp $b.vcf ${project}.LCL8_2.vcf
		    elif [[ $b.vcf =~ "LCL8_3" ]]; then
				cp $b.vcf ${project}.LCL8_3.vcf
	        fi
		done

        cat ${project}.LCL5_1.vcf | grep -v '#' > LCL5_1.body
        cat ${project}.LCL5_2.vcf | grep -v '#' > LCL5_2.body
        cat ${project}.LCL5_3.vcf | grep -v '#' > LCL5_3.body

        cat ${project}.LCL6_1.vcf | grep -v '#' | cut -f 10 > LCL6_1.body
        cat ${project}.LCL6_2.vcf | grep -v '#' | cut -f 10 > LCL6_2.body
        cat ${project}.LCL6_3.vcf | grep -v '#' | cut -f 10 > LCL6_3.body

        cat ${project}.LCL7_1.vcf | grep -v '#' | cut -f 10 > LCL7_1.body
        cat ${project}.LCL7_2.vcf | grep -v '#' | cut -f 10 > LCL7_2.body
        cat ${project}.LCL7_3.vcf | grep -v '#' | cut -f 10 > LCL7_3.body

        cat ${project}.LCL8_1.vcf | grep -v '#' | cut -f 10 > LCL8_1.body
        cat ${project}.LCL8_2.vcf | grep -v '#' | cut -f 10 > LCL8_2.body
        cat ${project}.LCL8_3.vcf | grep -v '#' | cut -f 10 > LCL8_3.body

		echo -e "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tLCL5\tLCL6\tLCL7\tLCL8" > header_name

		cat ${project}.LCL5_1.vcf | grep '##' | cat - header_name > header

		paste LCL5_1.body LCL6_1.body LCL7_1.body LCL8_1.body > family_1.body
		paste LCL5_2.body LCL6_2.body LCL7_2.body LCL8_2.body > family_2.body
		paste LCL5_3.body LCL6_3.body LCL7_3.body LCL8_3.body > family_3.body

		cat header family_1.body > ${project}.1.family.vcf
		cat header family_2.body > ${project}.2.family.vcf
		cat header family_3.body > ${project}.3.family.vcf




	>>>

	runtime {
		docker:docker
		cluster: cluster_config
		systemDisk: "cloud_ssd 40"
		dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
	}
	output {
		Array[File] splited_vcf = glob("*.vcf")
		Array[File] family_vcf = glob("*.family.vcf")
		File LCL5_1 = "${project}.LCL5_1.vcf"
		File LCL5_2 = "${project}.LCL5_2.vcf"
		File LCL5_3 = "${project}.LCL5_3.vcf"
		File LCL6_1 = "${project}.LCL6_1.vcf"
		File LCL6_2 = "${project}.LCL6_2.vcf"
		File LCL6_3 = "${project}.LCL6_3.vcf"
		File LCL7_1 = "${project}.LCL7_1.vcf"
		File LCL7_2 = "${project}.LCL7_2.vcf"
		File LCL7_3 = "${project}.LCL7_3.vcf"
		File LCL8_1 = "${project}.LCL8_1.vcf"
		File LCL8_2 = "${project}.LCL8_2.vcf"
		File LCL8_3 = "${project}.LCL8_3.vcf"
	}
}