Adding oncotator filtering enabled in M2 WDL. (#4423)

* Adding oncotator filtering enabled in M2 WDL. * Fixing cromwell test oncotator version * Changing default oncotator docker to `1.9.8.0`
broadinstitute · Mar 21, 2018 · 2380b73 · 2380b73
1 parent 29b6438
commit 2380b73
Show file tree

Hide file tree

Showing 3 changed files with 19 additions and 4 deletions.
diff --git a/scripts/m2_cromwell_tests/test_m2_wdl_multi.json b/scripts/m2_cromwell_tests/test_m2_wdl_multi.json
@@ -1,6 +1,6 @@
 {
   "Mutect2_Multi.gatk_docker": "__GATK_DOCKER__",
-  "Mutect2_Multi.oncotator_docker": "broadinstitute/oncotator:1.9.6.1",
+  "Mutect2_Multi.oncotator_docker": "broadinstitute/oncotator:1.9.7.0",
   "Mutect2_Multi.intervals": "/home/travis/build/broadinstitute/gatk/scripts/m2_cromwell_tests/interval_list.interval_list",
   "Mutect2_Multi.ref_fasta": "/home/travis/build/broadinstitute/gatk/src/test/resources/large/human_g1k_v37.20.21.fasta",
   "Mutect2_Multi.ref_fai": "/home/travis/build/broadinstitute/gatk/src/test/resources/large/human_g1k_v37.20.21.fasta.fai",

diff --git a/scripts/mutect2_wdl/README.md b/scripts/mutect2_wdl/README.md
@@ -118,6 +118,7 @@ Recommended default values (where possible) are found in ``mutect2_template.json
 - ``Mutect2.sequencing_center`` -- Please see parameter description above in the mutect2_multi_sample.   
 - ``Mutect2.sequence_source`` -- Please see parameter description above in the mutect2_multi_sample.
 - ``Mutect2.default_config_file`` -- Please see parameter description above in the mutect2_multi_sample.
+- ``Mutect2.filter_oncotator_maf`` -- (optional, default true) Whether Oncotator should remove filtered variants when rendering the MAF.  Ignored if `run_oncotator` is false. 
 
 #### mutect2-replicate-validation
 

diff --git a/scripts/mutect2_wdl/mutect2.wdl b/scripts/mutect2_wdl/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
@@ -109,7 +110,11 @@ 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
@@ -308,7 +313,9 @@ 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
         }
     }
 
@@ -782,6 +789,7 @@ task oncotate_m2 {
     File? default_config_file
     String case_id
     String? control_id
+    String? oncotator_extra_args
 
     # runtime
     String oncotator_docker
@@ -791,6 +799,10 @@ 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
@@ -822,7 +834,9 @@ 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 {