Quality_control/tasks/benchmark.wdl

task benchmark {
	File gzvcf
	File gzvcf_index
	File benchmarking_dir
	File ref_dir
	String sample = basename(gzvcf,".vcf.gz")
	String sample_mark
	String fasta
	String docker
	String cluster_config
	String disk_size


	command <<<
		set -o pipefail
		set -e
		nt=$(nproc)
		export HGREF=/cromwell_inputs/*/reference_data/GRCh38.d1.vd1.fa
		if [ ${sample_mark} == "LCL5" ];then
			/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL5.vcf.gz ${gzvcf} -f ${benchmarking_dir}/LCL5.bed.gz --threads $nt -o ${sample}
	    elif [ ${sample_mark} == "LCL6" ]; then
	    	/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL6.vcf.gz ${gzvcf} -f ${benchmarking_dir}/LCL6.bed.gz --threads $nt -o ${sample}
        elif [ ${sample_mark} == "LCL7" ]; then
        	/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL7.vcf.gz ${gzvcf} -f ${benchmarking_dir}/LCL6.bed.gz --threads $nt -o ${sample}
	    elif [ ${sample_mark} == "LCL8" ]; then
			/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL8.vcf.gz ${gzvcf} -f ${benchmarking_dir}/LCL6.bed.gz --threads $nt -o ${sample}
        else
        	echo "only for quartet samples"
        fi		
	>>>

	runtime {
		docker:docker
		cluster:cluster_config
		systemDisk:"cloud_ssd 40"
		dataDisk:"cloud_ssd " + disk_size + " /cromwell_root/"
	}

	output {
		File gzip_vcf = "${sample}.vcf.gz"
		File gzip_vcf_index = "${sample}.vcf.gz.tbi"
		File roc_all_csv = "${sample}.roc.all.csv.gz"
		File roc_indel = "${sample}.roc.Locations.INDEL.csv.gz"
		File roc_indel_pass = "${sample}.roc.Locations.INDEL.PASS.csv.gz"
		File roc_snp = "${sample}.roc.Locations.SNP.csv.gz"
		File roc_snp_pass = "${sample}.roc.Locations.SNP.PASS.csv.gz"
		File summary = "${sample}.summary.csv"
		File extended = "${sample}.extended.csv"
		File metrics = "${sample}.metrics.json.gz"
	}
}