benchmarking/tasks/benchmark.wdl

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


	command <<<
		set -o pipefail
		set -e
		nt=$(nproc)
		mkdir -p /cromwell_root/tmp
		cp -r ${ref_dir} /cromwell_root/tmp/

		export HGREF=/cromwell_root/tmp/reference_data/GRCh38.d1.vd1.fa

		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 > ${sample}.filtered.vcf

		/opt/rtg-tools/dist/rtg-tools-3.10.1-4d58ead/rtg bgzip ${sample}.filtered.vcf -c > ${sample}.filtered.rtg.vcf.gz
		/opt/rtg-tools/dist/rtg-tools-3.10.1-4d58ead/rtg index -f vcf ${sample}.filtered.rtg.vcf.gz

		if [ ${sample_mark} == "LCL5" ];then
			/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL5.afterfilterdiffbed.vcf.gz ${sample}.filtered.rtg.vcf.gz -f ${benchmarking_dir}/LCL5.high.confidence.bed.gz --threads $nt -o ${sample} -r ${ref_dir}/${fasta}
	    elif [ ${sample_mark} == "LCL6" ]; then
	    	/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL6.afterfilterdiffbed.vcf.gz ${sample}.filtered.rtg.vcf.gz -f ${benchmarking_dir}/LCL6.high.confidence.bed.gz --threads $nt -o ${sample} -r ${ref_dir}/${fasta}
        elif [ ${sample_mark} == "LCL7" ]; then
        	/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL7.afterfilterdiffbed.vcf.gz ${sample}.filtered.rtg.vcf.gz -f ${benchmarking_dir}/LCL7.high.confidence.bed.gz --threads $nt -o ${sample} -r ${ref_dir}/${fasta}
	    elif [ ${sample_mark} == "LCL8" ]; then
			/opt/hap.py/bin/hap.py ${benchmarking_dir}/LCL8.afterfilterdiffbed.vcf.gz ${sample}.filtered.rtg.vcf.gz -f ${benchmarking_dir}/LCL8.high.confidence.bed.gz --threads $nt -o ${sample} -r ${ref_dir}/${fasta}
        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 rtg_vcf = "${sample}.filtered.rtg.vcf.gz"
		File rtg_vcf_index = "${sample}.filtered.rtg.vcf.gz.tbi"
		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"
	}
}