I expect everything is a bit touch and go at the moment with the site migration, and that things will be rolled out slowly in batches, but when/if you do put together an official Best Practices for Mitochondrial Short Variants document, would it be possible for more comprehensive explanations / usage cases / maybe an algorithm/process flowchart? Something along these lines maybe: https://gatk.broadinstitute.org/hc/en-us/articles/360035894731-Somatic-short-variant-discovery-SNVs-Indels- ??
When I reverse engineered a command line pipe from the Terra vcfs, I had to pull from and piece together several different sources of info including but not limited to:
The actual GATK workflow and wdl scripts: https://github.com/gatk-workflows/gatk4-mitochondria-pipeline
The old announcement Mitochondrial Analysis discussion thread: https://gatkforums.broadinstitute.org/gatk/discussion/23598/new-mitochondrial-analysis-with-mutect2#latest
The usage docs for Mutect2, FilterMutectCalls, and VariantFiltration.
Mutect2 error reports: https://gatkforums.broadinstitute.org/gatk/discussion/23863/mutect2-mitochondria-flag-doesnt-work -- (P.S. This error is still being perpetuated in the newer Mutect2 tool index usage cases. Isn't the argument --mitochondria-mode not --mitochondria?).
And also the Terra input/output guides, which can be even more obtuse.
Sidenote: I've been working with this pipeline for several months now and I still don't quite understand how some of the bits and pieces work. Can someone explain in detail the algorithm/logic behind NuMT filtration using median autosomal coverage as a filtering statistic? Is it basically the same as https://gatk.broadinstitute.org/hc/en-us/articles/360036732951-PolymorphicNuMT in 188.8.131.52??
Please sign in to leave a comment.