Update: add VEP

defaults

@@ -18,6 +18,11 @@
   "sentieon_docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/sentieon-genomics:v202010.02",
   "varscan_docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/varscan2:v2.4.3",
   "annovar_docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/annovar:v2019.10",
+  "vep_docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/vep:v104.0",
+  "cache": "oss://pgx-reference-data/ensembl_vep/",
+  "vep_path": "/opt/vep/ensembl-vep",
+  "species": "homo_sapiens_merged",
+  "vcf2maf_path": "/opt/mskcc-vcf2maf",
   "tnseq_pon": "",
   "tnscope_pon": "",
   "disk_size": "200",
@@ -27,5 +32,6 @@
   "tnscope": false,
   "varscan": false,
   "annovar": false,
+  "vep": false,
   "only_pass": true
 }

inputs

@@ -23,6 +23,11 @@
   "{{ project_name }}.annovar_database": "{{ annovar_database }}",
   "{{ project_name }}.tnseq_pon": "{{ tnseq_pon }}",
   "{{ project_name }}.tnscope_pon": "{{ tnscope_pon }}",
+  "{{ project_name }}.vep_path": "{{ vep_path }}",
+  "{{ project_name }}.cache": "{{ cache }}",
+  "{{ project_name }}.species": "{{ species }}",
+  "{{ project_name }}.vcf2maf_path": "{{ vcf2maf_path }}",
+  "{{ project_name }}.vep_docker": "{{ vep_docker }}",
   "{{ project_name }}.disk_size": "{{ disk_size }}",
   "{{ project_name }}.cluster_config": "{{ cluster_config }}",
   "{{ project_name }}.haplotyper": {{ haplotyper | tojson }},
@@ -30,5 +35,6 @@
   "{{ project_name }}.tnscope": {{ tnscope | tojson }},
   "{{ project_name }}.varscan": {{ varscan | tojson }},
   "{{ project_name }}.annovar": {{ annovar | tojson }},
-  "{{ project_name }}.only_pass": {{ only_pass | tojson }}
+  "{{ project_name }}.only_pass": {{ only_pass | tojson }},
+  "{{ project_name }}.vep": {{ vep | tojson }}
 }

tasks/VEP.wdl

@@ -0,0 +1,77 @@
+task vcf2maf {
+
+  File vcf
+  String sample_id
+  String basename = basename(vcf,".vcf")
+  String tumor_id
+  String normal_id
+  File ref_dir
+  String fasta
+  String vep_path
+  File cache
+  String hg
+  String species
+  Boolean only_pass
+  String vcf2maf_path
+  String docker
+  String cluster_config
+  String disk_size
+
+
+  command <<<
+    set -o pipefail
+    set -e
+    nt=$(nproc)
+
+    if [ only_pass ]; then
+      awk -F'\t' '{if(($1~"^#")||($1!~"^#" && $7=="PASS")){print $0}}' ${vcf} > ${sample_id}.INPUT.vcf
+    else
+      cp ${vcf} ${sample_id}.INPUT.vcf
+    fi
+
+    # Define ncbi_build
+    if [ hg == "hg19" ]; then
+      ncbi_build="GRCh37"
+    elif [ hg == "hg38" ]; then
+      ncbi_build="GRCh38"
+    fi
+
+    # Extract the BND variants from VCF
+    awk -F'\t' '{if(($1~"^#")||($8!~".*SVTYPE=BND.*")){print $0}}' ${sample_id}.INPUT.vcf > ${sample_id}.INPUT.vcf2maf.vcf
+    awk -F'\t' '{if(($1~"^#")||($8~".*SVTYPE=BND.*")){print $0}}' ${sample_id}.INPUT.vcf > ${sample_id}.INPUT.VEP.vcf
+
+    # vcf2maf
+    perl ${vcf2maf_path}/vcf2maf.pl \
+    --input-vcf ${sample_id}.INPUT.vcf2maf.vcf --output-maf ${basename}.maf \
+    --tumor-id ${tumor_id} --normal-id ${normal_id} \
+    --ref-fasta ${ref_dir}/${fasta} \
+    --vep-path ${vep_path} \
+    --vep-data ${cache} \
+    --ncbi-build $ncbi_build \
+    --species ${species} \
+    --vep-fork $nt
+
+    # vep: only annotate the BND
+    perl ${vep_path}/vep \
+    --input_file ${sample_id}.INPUT.VEP.vcf --output_file ${basename}.BND.VEP.txt \
+    --fasta ${ref_dir}/${fasta} \
+    --dir ${cache} \
+    --assembly $ncbi_build \
+    --species ${species} \
+    --fork $nt \
+    --no_progress --no_stats --buffer_size 5000 --sift b --ccds --uniprot --hgvs --symbol --numbers --domains --gene_phenotype --canonical --protein --biotype --uniprot --tsl --variant_class --shift_hgvs 1 --check_existing --total_length --allele_number --no_escape --xref_refseq --failed 1 --tab --flag_pick_allele --pick_order canonical,tsl,biotype,rank,ccds,length --format vcf --force_overwrite --offline --pubmed --regulatory
+  >>>
+  
+  runtime {
+    docker: docker
+    cluster: cluster_config
+    systemDisk: "cloud_ssd 40"
+    dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+  }
+
+  output {
+    File input_vcf = "${sample_id}.INPUT.vcf"
+    File maf = "${basename}.${hg}.maf"
+    File bnd_vep = "${basename}.BND.VEP.txt"
+  }
+}

workflow.wdl

@@ -10,6 +10,7 @@ import "./tasks/somatic.wdl" as somatic
 import "./tasks/processSomatic.wdl" as processSomatic
 import "./tasks/somaticFilter.wdl" as somaticFilter
 import "./tasks/ANNOVAR.wdl" as ANNOVAR
+import "./tasks/VEP.wdl" as VEP
 
 
 workflow {{ project_name }} {
@@ -25,6 +26,7 @@ workflow {{ project_name }} {
   String sentieon_docker
   String varscan_docker
   String annovar_docker
+  String vep_docker
 
   File ref_dir
   String fasta
@@ -42,6 +44,11 @@ workflow {{ project_name }} {
   File? tnseq_pon
   File? tnscope_pon
 
+  String vep_path
+  File cache
+  String species
+  String vcf2maf_path
+
   String disk_size
   String cluster_config
 
@@ -51,6 +58,7 @@ workflow {{ project_name }} {
   Boolean varscan
 
   Boolean annovar
+  Boolean vep
   Boolean only_pass
 
   if (tumor_fastq_1!= "") {
@@ -263,6 +271,27 @@ workflow {{ project_name }} {
         disk_size=disk_size
       }
     }
+
+    if (vep) {
+      call VEP.VEP as TNseq_VEP {
+        input: 
+        vcf=TNseq.TNseq_vcf,
+        hg=hg,
+        only_pass=only_pass,
+        sample_id=sample_id,
+        tumor_id=sample_id + "_tumor",
+        normal_id=sample_id + "_normal",
+        ref_dir=ref_dir,
+        fasta=fasta,
+        vep_path=vep_path,
+        cache=cache,
+        species=species,
+        vcf2maf_path=vcf2maf_path,
+        docker=vep_docker,
+        cluster_config=cluster_config,
+        disk_size=disk_size
+      }
+    }
   }
 
   if (tnscope) {
@@ -303,6 +332,27 @@ workflow {{ project_name }} {
         disk_size=disk_size
       }
     }
+
+    if (vep) {
+      call VEP.VEP as TNscope_VEP {
+        input: 
+        vcf=TNscope.TNscope_vcf,
+        hg=hg,
+        only_pass=only_pass,
+        sample_id=sample_id,
+        tumor_id=sample_id + "_tumor",
+        normal_id=sample_id + "_normal",
+        ref_dir=ref_dir,
+        fasta=fasta,
+        vep_path=vep_path,
+        cache=cache,
+        species=species,
+        vcf2maf_path=vcf2maf_path,
+        docker=vep_docker,
+        cluster_config=cluster_config,
+        disk_size=disk_size
+      }
+    }
   }
 
   if (varscan) {
@@ -325,7 +375,7 @@ workflow {{ project_name }} {
       sample=sample_id,
       varscan_snp=somatic.varscan_snp,
       varscan_indel=somatic.varscan_indel,   
-      docker=varscan_docker,	
+      docker=varscan_docker,
       disk_size=disk_size,
       cluster_config=cluster_config
     }
@@ -340,7 +390,7 @@ workflow {{ project_name }} {
       varscan_indel_germline_hc=processSomatic.varscan_indel_germline_hc,
       varscan_indel_loh_hc=processSomatic.varscan_indel_loh_hc,
       varscan_indel=somatic.varscan_indel,   
-      docker=varscan_docker,	
+      docker=varscan_docker,
       disk_size=disk_size,
       cluster_config=cluster_config
     }
@@ -357,5 +407,26 @@ workflow {{ project_name }} {
         disk_size=disk_size
       }
     }
+
+    if (vep) {
+      call VEP.VEP as VarScan_VEP {
+        input: 
+        vcf=VarScan.VarScan_vcf,
+        hg=hg,
+        only_pass=only_pass,
+        sample_id=sample_id,
+        tumor_id=sample_id + "_tumor",
+        normal_id=sample_id + "_normal",
+        ref_dir=ref_dir,
+        fasta=fasta,
+        vep_path=vep_path,
+        cache=cache,
+        species=species,
+        vcf2maf_path=vcf2maf_path,
+        docker=vep_docker,
+        cluster_config=cluster_config,
+        disk_size=disk_size
+      }
+    }
   }
 }