We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
This section of the forum is now closed; we are working on a new support model for WDL that we will share here shortly. For Cromwell-specific issues, see the Cromwell docs and post questions on Github.
Bias Filtering in Muetect2.wdl provided by GATK github repo, didn´t work
I locally tested mutect2.wdl (from: GATK Github Repo) with match json file in this repo. (mutect2.exome.inputs.json).
For small testing data (20G, HCC1143.bam and HCC1143_normal.bam, downloaded from gs://gatk-best-practices/somatic-b37/), it finished in 15 mins. According Terra estimated running time is around 2 hours. I guess the main reason is that I didnt enable orientation_bias_filter and oncotator. (in JSON file, "Mutect2.run_orientation_bias_filter": "False", "Mutect2.run_oncotator": "False").
However, after I changed "Mutect2.run_orientation_bias_filter": "True", this workflow still finished in 15 mins and checked the output file, it still showed: "Mutect2.read_orientation_model_params": null
But I did find: "Mutect2.filtered_vcf": "/media/volume2/gatk4-somatic-snvs-indels/m2_orientation_filter/cromwell-executions/Mutect2/08223435-4a59-4efa-8b16-f2a8e90b6df3/call-Filter/execution/HCC1143-filtered.vcf"
Does anyone knows why this happened? What else should I add to enable the bias_filter? So, orientation_bias_filter almost didnt cost time?