add samtools to build bam index

defaults

@@ -2,6 +2,8 @@
     "fasta": "oss://pgx-reference-data/GRCh38.d1.vd1/GRCh38.d1.vd1.fa",
     "faidx": "oss://pgx-reference-data/GRCh38.d1.vd1/GRCh38.d1.vd1.fa.fai",
     "docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/cnvkit:0.9.7",
+    "samtools_docker": "registry.cn-shanghai.aliyuncs.com/pgx-docker-registry/samtools:v1.3.1",
+    "samtools_cluster": "OnDemand bcs.a2.large img-ubuntu-vpc",
     "disk_size": "400",
     "cluster_config": "OnDemand bcs.a2.7xlarge img-ubuntu-vpc",
     "bed": "oss://pgx-reference-data/bed/hg38/Exome-Agilent_V6_chr.bed",

inputs

@@ -1,15 +1,15 @@
 {
   "{{ project_name }}.sample_id": "{{ sample_id }}",
   "{{ project_name }}.normal_bam": {{ normal_bam | tojson }},
-  "{{ project_name }}.normal_bai": {{ normal_bai | tojson }},
   "{{ project_name }}.tumor_bam": {{ tumor_bam | tojson }},
-  "{{ project_name }}.tumor_bai": {{ tumor_bai | tojson }},
   "{{ project_name }}.fasta": "{{ fasta }}",
   "{{ project_name }}.faidx": "{{ faidx }}",
   "{{ project_name }}.ref_flat": "{{ ref_flat }}",
   "{{ project_name }}.method": "{{ method }}",
   "{{ project_name }}.segment_method": "{{ segment_method }}",
   "{{ project_name }}.reference": "{{ reference }}",
+  "{{ project_name }}.samtools_docker": "{{ samtools_docker }}",
+  "{{ project_name }}.samtools_cluster": "{{ samtools_cluster }}",
   "{{ project_name }}.docker": "{{ docker }}",
   "{{ project_name }}.bed": "{{ bed }}",
   "{{ project_name }}.disk_size": "{{ disk_size }}",

tasks/batch.wdl

@@ -1,9 +1,8 @@
 task batch {
     String sample_id
     Array[File] tumor_bam
-    Array[File] tumor_bai
     Array[File] normal_bam
-    Array[File] normal_bai
+    Array[File] bam_index
     File bed
     File faidx
     File fasta
@@ -23,21 +22,24 @@ task batch {
     command <<<
         set -o pipefail
         set -e
+        nt=$(nproc)
+
         mkdir -p /cromwell_root/tmp/cnvkit
-        cp ${sep=' ' normal_bai} /cromwell_root/tmp/cnvkit
-        cp ${sep=' ' tumor_bai} /cromwell_root/tmp/cnvkit
+        cp -r ${sep=" " bam_index} /cromwell_root/tmp/cnvkit
+
         # must exist parameters
         cp ${fasta} /cromwell_root/tmp/cnvkit/hg38.fa
         cp ${faidx} /cromwell_root/tmp/cnvkit/hg38.fai
         cp ${bed} /cromwell_root/tmp/cnvkit/target.bed
         cp ${access_bed} /cromwell_root/tmp/cnvkit/access-mappable.bed
+
         # optional parameters
         if [ ${ref_flat} != "" ]; then cp ${ref_flat} /cromwell_root/tmp/cnvkit/ref_flat.txt; fi
         if [ ${reference} != "" ]; then cp ${reference} /cromwell_root/tmp/cnvkit/my_reference.cnn; fi
         
         cd /cromwell_root/tmp/cnvkit
         mkdir results
-        cnvkit.py batch ${sep=' ' tumor_bam} --normal ${sep=' ' normal_bam} \
+        cnvkit.py batch -p $nt ${sep=' ' tumor_bam} --normal ${sep=' ' normal_bam} \
         --method ${method} --segment-method ${segment_method} \
         --targets target.bed ${access_opt} ${annotate_opt} \
         --fasta hg38.fa ${reference_opt} \

tasks/samtools.wdl

@@ -0,0 +1,28 @@
+task samtools {
+  Array[File] tumor_bam
+  Array[File] normal_bam
+  String docker
+  String cluster
+  String disk_size
+
+  command <<<
+    set -o pipefail
+    set -e
+    nt=$(nproc)
+    mkdir -p /cromwell_root/tmp/samtools
+    cd /cromwell_root/tmp/samtools
+    /opt/conda/bin/samtools -@ $nt index ${sep=' ' tumor_bam}
+    /opt/conda/bin/samtools -@ $nt index ${sep=' ' normal_bam}
+  >>>
+
+  runtime {
+    docker: docker
+    cluster: cluster
+    systemDisk: "cloud_ssd 40"
+    dataDisk: "cloud_ssd " + disk_size + " /cromwell_root/"
+  }
+
+  output {
+    Array[File] bam_index = glob("/cromwell_root/tmp/samtools/*bai")
+  }
+}

workflow.wdl

@@ -1,13 +1,12 @@
 import "./tasks/access.wdl" as access
+import "./tasks/samtools.wdl" as samtools
 import "./tasks/batch.wdl" as batch
 import "./tasks/export.wdl" as export
 
 workflow {{ project_name }} {	
     String sample_id
     Array[File] tumor_bam
-    Array[File] tumor_bai
     Array[File] normal_bam
-    Array[File] normal_bai
     File bed
 	File faidx
 	File fasta
@@ -32,6 +31,15 @@ workflow {{ project_name }} {
         disk_size = disk_size
     }
 
+    call samtools.samtools as samtools {
+        input:
+        tumor_bam = tumor_bam,
+        normal_bam = normal_bam,
+        docker=samtools_docker,
+        cluster=samtools_cluster,
+        disk_size=disk_size
+    }
+
     call batch.batch as batch {
         input: 
         sample_id = sample_id,
@@ -42,9 +50,8 @@ workflow {{ project_name }} {
         segment_method = segment_method,
         reference = reference,
         tumor_bam = tumor_bam,
-        tumor_bai = tumor_bai,
         normal_bam = normal_bam,
-        normal_bai = normal_bai,
+        bam_index = samtools.bam_index,
         bed = bed,
         access_bed = access.access_bed,
         docker = docker,