Description of the bug

When there is no germline_resource specified, FILTERMUTECTCALLS is not run in the tumor-only somatic variant calling mode of the MUTECT2 workflow. This appears to happen because in the absence of the germline resource, GETPILEUPSUMMARIES, GATHERPILEUPSUMMARIES, and CALCULATECONTAMINATION are not run, each of which generates an output that is passed to FILTERMUTECTCALLS. However, FILTERMUTECTCALLS does not actually require these inputs to run. Nevertheless, FILTERMUTECTCALLS is never triggered/started in this circumstance.

In my case I am using the GRCm38 genome, for which no standard germline resource exists. Let me know if you need any other information from me, and thank you for your help!

Best,
Branden

Edit:
May I add that everything up to and including LEARNREADORIENTATIONMODEL completes without issue. The problem is with getting FILTERMUTECTCALLS to even attempt to run. I have tried providing various dummy values like empty paths, to no avail -- although perhaps I am not approaching that in the right way.

Command used and terminal output

nextflow run ../../pipelines/sarek -profile singularity -work-dir path/to/work/dir --input /path/to/input.csv --genome GRCm38 --tools "mutect2,merge" --outdir /path/to/output

Relevant files

No response

System information

Nextflow version 23.10.1.
Running on an HPC.
Slurm executor.
Singularity for containerization.
Sarek version 3.4.2.