VQSR/tasks/VQSR.wdl

task VQSR {
	String SENTIEON_INSTALL_DIR
	String fasta
	File vcf
	File vcf_index
	File ref_dir
	String dbsnp
	String dbsnp_Mill
	String dbsnp_1000G_omni
	String dbsnp_hapmap
	String dbsnp_1000G_phase1
	String sample
	String docker
	String cluster_config
	String disk_size

command <<<
		set -o pipefail
		set -e
		export SENTIEON_LICENSE=192.168.0.55:8990

		resource_text="--resource ${dbsnp_1000G_phase1} \
              --resource_param 1000G,known=false,training=true,truth=false,prior=10.0 "
		resource_text="$resource_text --resource ${dbsnp_1000G_omni} \
              --resource_param omni,known=false,training=true,truth=true,prior=12.0 "
		resource_text="$resource_text --resource ${dbsnp} \
              --resource_param dbsnp,known=true,training=false,truth=false,prior=2.0 "
		resource_text="$resource_text --resource ${dbsnp_hapmap} \
              --resource_param hapmap,known=false,training=true,truth=true,prior=15.0"

        annotation_array="DP QD FS SOR MQ MQRankSum ReadPosRankSum"
        for annotation in $annotation_array; do
		  annotate_text="$annotate_text --annotation $annotation"
		done

		${SENTIEON_INSTALL_DIR}/bin/sentieon driver -r ${ref_dir}/${fasta} --algo VarCal -v ${vcf} $resource_text $annotate_text --var_type SNP --plot_file ${sample}.vqsrSNP.hc.plotfile --tranches_file ${sample}.vqsrSNP.hc.tranches ${sample}.vqsrSNP.hc.recal
  		${SENTIEON_INSTALL_DIR}/bin/sentieon driver -r ${ref_dir}/${fasta} --algo ApplyVarCal -v ${vcf} --var_type SNP --tranches_file ${sample}.vqsrSNP.hc.tranches --sensitivity 99.0 --recal ${sample}.vqsrSNP.hc.recal ${sample}.vqsrSNP.hc.recaled.vcf.gz
  		${SENTIEON_INSTALL_DIR}/bin/sentieon plot vqsr -o ${sample}.vqsrSNP.pdf ${sample}.vqsrSNP.hc.plotfile

  		resource_text="$resource_text --resource ${dbsnp_Mill} \
              --resource_param Mills,known=false,training=true,truth=true,prior=12.0 "
		resource_text="$resource_text --resource ${dbsnp} \
              --resource_param dbsnp,known=true,training=false,truth=false,prior=2.0 "

        annotation_array="QD DP FS SOR MQ ReadPosRankSum"
		annotate_text=""
		for annotation in $annotation_array; do
  		  annotate_text="$annotate_text --annotation $annotation"
		done

		${SENTIEON_INSTALL_DIR}/bin/sentieon driver -r ${ref_dir}/${fasta} --algo VarCal -v ${vcf} $resource_text $annotate_text --var_type INDEL --plot_file ${sample}.vqsrINDEL.hc.plotfile --max_gaussians 4 --tranches_file ${sample}.vqsrINDEL.hc.tranches ${sample}.vqsrINDEL.hc.recal
		${SENTIEON_INSTALL_DIR}/bin/sentieon driver -r ${ref_dir}/${fasta} --algo ApplyVarCal -v ${sample}.vqsrSNP.hc.recaled.vcf.gz --var_type INDEL --recal ${sample}.vqsrINDEL.hc.recal --tranches_file ${sample}.vqsrINDEL.hc.tranches --sensitivity 99.0 ${sample}.vqsrSNPINDEL.hc.recaled.vcf.gz
		${SENTIEON_INSTALL_DIR}/bin/sentieon plot vqsr -o ${sample}.vqsrINDEL.VQSR.pdf ${sample}.vqsrINDEL.hc.plotfile

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

	output {
	File filtered_vcf = "${sample}.vqsrSNPINDEL.hc.recaled.vcf.gz"
	}
}