Wdlupdate (#6)

* updated wdl/json to gatk4.0.1.1 release * add info for Readme * Update wdl to gatk4.0.1.2 Release * Update docker to gatk4.0.1.2 * added mutect2_pon.wdl, added Parameter descriptions to readme * Update mutect2_pon.inputs.json * Update mutect2.exome.inputs.json * updated mutect2.wdl to gatk4.0.2.0 * Update to gatk4.0.4.0 , Addtion of NIO supported mutect2 version
gatk-workflows · May 1, 2018 · 1c458a9 · 1c458a9
1 parent d9e4008
commit 1c458a9
Show file tree

Hide file tree

Showing 5 changed files with 996 additions and 12 deletions.
diff --git a/README.md b/README.md
@@ -5,7 +5,7 @@ Workflows for somatic short variant analysis with GATK4.
 
 ### mutect2 :
 Implements Somatic short variant discovery using [GATK Best Practices](https://software.broadinstitute.org/gatk/best-practices/workflow).
-
+Note: Also provided in this repo is mutect2_nio which is a NIO supported version of the wdl.  
 
 #### Requirements/expectations
 - Tumor bam and index
@@ -40,7 +40,7 @@ Used to validate mutect2 workflow.
 - GATK4 or later 
 
 Cromwell version support 
-- Successfully tested on v30
+- Successfully tested on v31
 
 
 ### Parameter descriptions :

diff --git a/mutect2-normal-normal.inputs.json b/mutect2-normal-normal.inputs.json
@@ -34,6 +34,6 @@
   "Mutect2NormalNormal.scatter_count": "10",
 
   "##_COMMENT4": "Docker",
-  "Mutect2NormalNormal.gatk_docker": "broadinstitute/gatk:4.0.3.0"
+  "Mutect2NormalNormal.gatk_docker": "broadinstitute/gatk:4.0.4.0"
 }
 
diff --git a/mutect2.exome.inputs.json b/mutect2.exome.inputs.json
@@ -1,7 +1,7 @@
 {
   "##_COMMENT1": "Runtime",
   "Mutect2.oncotator_docker": "broadinstitute/oncotator:1.9.6.1",
-  "Mutect2.gatk_docker": "broadinstitute/gatk:4.0.2.0",
+  "Mutect2.gatk_docker": "broadinstitute/gatk:4.0.4.0",
 
   "##_COMMENT2": "Workflow options",
   "Mutect2.intervals": "gs://gatk-best-practices/somatic-b37/whole_exome_agilent_1.1_refseq_plus_3_boosters.Homo_sapiens_assembly19.baits.interval_list",

diff --git a/mutect2.wdl b/mutect2.wdl
@@ -40,6 +40,7 @@
 ## ** Secondary resources ** (for optional tasks)
 ## onco_ds_tar_gz, default_config_file: Oncotator datasources and config file
 ## sequencing_center, sequence_source: metadata for Oncotator
+## filter_oncotator_maf: Whether the MAF generated by oncotator should have the filtered variants removed. Default: true
 ##
 ## Outputs :
 ## - One VCF file and its index with primary filtering applied; secondary filtering and functional annotation if requested; a bamout.bam
@@ -83,6 +84,8 @@ workflow Mutect2 {
     Boolean make_bamout_or_default = select_first([make_bamout, false])
     Boolean? compress_vcfs
     Boolean compress = select_first([compress_vcfs, false])
+    File? gga_vcf
+    File? gga_vcf_idx
 
     # oncotator inputs
     Boolean? run_oncotator
@@ -109,7 +112,11 @@ workflow Mutect2 {
     String gatk_docker
     String basic_bash_docker = "ubuntu:16.04"
     String? oncotator_docker
-    String oncotator_docker_or_default = select_first([oncotator_docker, "broadinstitute/oncotator:1.9.6.1"])
+    String oncotator_docker_or_default = select_first([oncotator_docker, "broadinstitute/oncotator:1.9.8.0"])
+    Boolean? filter_oncotator_maf
+    Boolean filter_oncotator_maf_or_default = select_first([filter_oncotator_maf, true])
+    String? oncotator_extra_args
+
     Int? preemptible_attempts
 
     # Use as a last resort to increase the disk given to every task in case of ill behaving data
@@ -177,9 +184,10 @@ workflow Mutect2 {
                 m2_extra_args = m2_extra_args,
                 make_bamout = make_bamout_or_default,
                 compress = compress,
+                gga_vcf = gga_vcf,
+                gga_vcf_idx = gga_vcf_idx,
                 gatk_override = gatk_override,
                 gatk_docker = gatk_docker,
-                preemptible_attempts = preemptible_attempts,
                 disk_space = tumor_bam_size + normal_bam_size + ref_size + gnomad_vcf_size + m2_output_size + disk_pad
         }
 
@@ -270,6 +278,7 @@ workflow Mutect2 {
             compress = compress,
             preemptible_attempts = preemptible_attempts,
             contamination_table = CalculateContamination.contamination_table,
+            maf_segments = CalculateContamination.maf_segments,
             m2_extra_filtering_args = m2_extra_filtering_args,
             disk_space = ceil(size(MergeVCFs.merged_vcf, "GB") * small_input_to_output_multiplier) + disk_pad
     }
@@ -307,7 +316,9 @@ workflow Mutect2 {
                 control_id = M2.normal_sample[0],
                 oncotator_docker = oncotator_docker_or_default,
                 preemptible_attempts = preemptible_attempts,
-                disk_space = ceil(size(oncotate_vcf_input, "GB") * large_input_to_output_multiplier) + onco_tar_size + disk_pad
+                disk_space = ceil(size(oncotate_vcf_input, "GB") * large_input_to_output_multiplier) + onco_tar_size + disk_pad,
+                filter_maf = filter_oncotator_maf_or_default,
+                oncotator_extra_args = oncotator_extra_args
         }
     }
 
@@ -417,6 +428,8 @@ task M2 {
     String? m2_extra_args
     Boolean? make_bamout
     Boolean compress
+    File? gga_vcf
+    File? gga_vcf_idx
 
     String output_vcf = "output" + if compress then ".vcf.gz" else ".vcf"
     String output_vcf_index = output_vcf + if compress then ".tbi" else ".idx"
@@ -460,6 +473,7 @@ task M2 {
             ${"--germline-resource " + gnomad} \
             ${"-pon " + pon} \
             ${"-L " + intervals} \
+            ${"--genotyping-mode GENOTYPE_GIVEN_ALLELES --alleles " + gga_vcf} \
             -O "${output_vcf}" \
             ${true='--bam-output bamout.bam' false='' make_bamout} \
             ${m2_extra_args}
@@ -555,8 +569,15 @@ task MergeBamOuts {
         set -e
         export GATK_LOCAL_JAR=${default="/root/gatk.jar" gatk_override}
         gatk --java-options "-Xmx${command_mem}m" GatherBamFiles \
-            -I ${sep=" -I " bam_outs} -O ${output_vcf_name}.out.bam -R ${ref_fasta}
-        samtools index ${output_vcf_name}.out.bam ${output_vcf_name}.out.bam.bai
+            -I ${sep=" -I " bam_outs} -O unsorted.out.bam -R ${ref_fasta}
+
+        # We must sort because adjacent scatters may have overlapping (padded) assembly regions, hence
+        # overlapping bamouts
+
+        gatk --java-options "-Xmx${command_mem}m" SortSam -I unsorted.out.bam \
+            -O ${output_vcf_name}.out.bam \
+            --SORT_ORDER coordinate
+        gatk --java-options "-Xmx${command_mem}m" BuildBamIndex -I ${output_vcf_name}.out.bam
     >>>
 
     runtime {
@@ -569,7 +590,7 @@ task MergeBamOuts {
 
     output {
         File merged_bam_out = "${output_vcf_name}.out.bam"
-        File merged_bam_out_index = "${output_vcf_name}.out.bam.bai"
+        File merged_bam_out_index = "${output_vcf_name}.out.bai"
     }
 }
 
@@ -650,7 +671,7 @@ task CalculateContamination {
         fi
 
         gatk --java-options "-Xmx${command_mem}m" GetPileupSummaries -R ${ref_fasta} -I ${tumor_bam} ${"-L " + intervals} -V ${variants_for_contamination} -O pileups.table
-        gatk --java-options "-Xmx${command_mem}m" CalculateContamination -I pileups.table -O contamination.table $NORMAL_CMD
+        gatk --java-options "-Xmx${command_mem}m" CalculateContamination -I pileups.table -O contamination.table --tumor-segmentation segments.table $NORMAL_CMD
     }
 
     runtime {
@@ -663,6 +684,7 @@ task CalculateContamination {
     output {
         File pileups = "pileups.table"
         File contamination_table = "contamination.table"
+        File maf_segments = "segments.table"
     }
 }
 
@@ -676,6 +698,7 @@ task Filter {
     String output_vcf = output_name + if compress then ".vcf.gz" else ".vcf"
     String output_vcf_index = output_vcf + if compress then ".tbi" else ".idx"
     File? contamination_table
+    File? maf_segments
     String? m2_extra_filtering_args
 
     File? gatk_override
@@ -700,6 +723,7 @@ task Filter {
         gatk --java-options "-Xmx${command_mem}m" FilterMutectCalls -V ${unfiltered_vcf} \
       	    -O ${output_vcf} \
       	    ${"--contamination-table " + contamination_table} \
+      	    ${"--tumor-segmentation " + maf_segments} \
       	    ${m2_extra_filtering_args}
     }
 
@@ -778,6 +802,7 @@ task oncotate_m2 {
     File? default_config_file
     String case_id
     String? control_id
+    String? oncotator_extra_args
 
     # runtime
     String oncotator_docker
@@ -787,6 +812,10 @@ task oncotate_m2 {
     Int? cpu
     Boolean use_ssd = false
 
+    Boolean? filter_maf
+    Boolean is_filter_maf = select_first([filter_maf, true])
+    String filter_maf_args = if (is_filter_maf) then " --collapse-filter-cols --prune-filter-cols " else ""
+
     # Mem is in units of GB but our command and memory runtime values are in MB
     Int machine_mem = if defined(mem) then mem * 1000 else 3500
     Int command_mem = machine_mem - 500
@@ -818,7 +847,9 @@ task oncotate_m2 {
             -a source:${default="Unknown" sequence_source} \
             -a normal_barcode:${control_id} \
             -a tumor_barcode:${case_id} \
-            ${"--default_config " + default_config_file}
+            ${"--default_config " + default_config_file} \
+            ${filter_maf_args} \
+            ${oncotator_extra_args}
     >>>
 
     runtime {
@@ -947,3 +978,4 @@ task Funcotate {
         File funcotated_vcf_index = "${output_vcf_index}"
     }
 }
+