Update: add vcf2vcf step into VEP

tasks/VEP.wdl

@@ -22,20 +22,22 @@ task VEP {
     set -e
     nt=$(nproc)
 
-    awk -F'\t' '{if(($1~"^#")||($1!~"^#" && $7=="PASS")){print $0}}' ${vcf} > ${basename}.PASS.vcf
+    awk -F'\t' '{if(($1~"^#")||($1!~"^#" && $7=="PASS")){print $0}}' ${vcf} > ${sample_id}.vcf
     
-    ncol=`awk -F'\t' '{if($1!~"^#"){print NF}}' ${basename}.PASS.vcf | uniq`
+    # Judge the SAMPLE info of vcf file
+    ncol=`awk -F'\t' '{if($1!~"^#"){print NF}}' ${sample_id}.vcf | uniq`
     if [ $ncol -lt 11 ]; then
-      vcf2maf_ID="--tumor-id ${tumor_id} --normal-id ${normal_id}"
+      SAMPLE="--tumor-id ${tumor_id} --normal-id ${normal_id}"
     else
-      vcf2maf_ID="--tumor-id ${tumor_id}"
+      SAMPLE="--tumor-id ${sample_id}"
     fi
 
-    nrow=`awk -F'\t' '{if($1~"^chr"){print $0}}' ${basename}.PASS.vcf | wc -l`
+    # Set the buffer_size based on the data size
+    nrow=`awk -F'\t' '{if($1~"^chr"){print $0}}' ${sample_id}.vcf | wc -l`
     if [ $nrow -lt 5000 ]; then
       buffer_size="--buffer_size 5000"
     else
-      buffer_size="--buffer_size 500"
+      buffer_size="--buffer_size 1000"
     fi
 
     # Extract the BND variants from VCF
@@ -55,7 +57,7 @@ task VEP {
 
     # vep
     # perl ${vep_path}/vep \
-    # --input_file ${basename}.PASS.vcf --output_file ${basename}.PASS.vep.vcf \
+    # --input_file ${sample_id}.vcf --output_file ${basename}.PASS.vep.vcf \
     # --fasta ${ref_dir}/${fasta} \
     # --dir ${cache} \
     # --assembly ${ncbi_build} \
@@ -68,6 +70,13 @@ task VEP {
     # --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 --flag_pick_allele --pick_order canonical,tsl,biotype,rank,ccds,length --force_overwrite --offline --pubmed --regulatory
     
+    # vcf2vcf: transfer into a standardized format
+    perl ${vcf2maf_path}/vcf2vcf.pl \
+    --input-vcf ${sample_id}.vcf --output-vcf ${basename}.norm.vcf \
+    $SAMPLE \
+    --ref-fasta ${reference}
+    
+    # VEP annotation
     perl ${vep_path}/vep --format vcf --vcf \
     --assembly ${ncbi_build} \
     --species ${species} \
@@ -75,14 +84,14 @@ task VEP {
     --offline \
     --cache --dir_cache ${cache} \
     --fasta ${ref_dir}/${fasta} \
-    --input_file ${basename}.PASS.vcf --output_file ${basename}.PASS.vep.vcf
-
+    $buffer_size \
+    --input_file ${basename}.norm.vcf --output_file ${basename}.vep.vcf
 
     # vcf2maf
     perl ${vcf2maf_path}/vcf2maf.pl \
     --inhibit-vep \
-    --input-vcf ${basename}.PASS.vep.vcf --output-maf ${basename}.PASS.maf \
-    $vcf2maf_ID \
+    --input-vcf ${basename}.vep.vcf --output-maf ${basename}.maf \
+    $SAMPLE \
     --ref-fasta ${ref_dir}/${fasta} \
     --ncbi-build ${ncbi_build} \
     --species ${species} \
@@ -97,8 +106,7 @@ task VEP {
   }
 
   output {
-    File input_vcf = "${basename}.PASS.vcf"
-    File vep_vcf = "${basename}.PASS.vep.vcf"
-    File maf = "${basename}.PASS.maf"
+    File vep_vcf = "${basename}.vep.vcf"
+    File maf = "${basename}.maf"
   }
 }