REQUIRED for all errors and issues:
a) GATK version used: 4.5.0.0 (but also applies to 4.6.0.0 and current git master branch)
b) Exact command used: gatk Mutect2 --mitochondria-mode
c) Entire program log: N/A

Hi,

I have been trying to use tumour-normal matched samples with Mutect2 in mitochondria mode by manually setting the flags that --mitochondria-mode automatically sets for you (since the documentation says that --mitochondria-mode only allows a single sample).

When testing on tumour-only samples, I noticed that the variants called did not match when manually setting the flags vs. using --mitochondria-mode. The guidance given in the Mitochondria Best Practices Workflow (https://gatk.broadinstitute.org/hc/en-us/articles/4403870837275-Mitochondrial-short-variant-discovery-SNVs-Indels) and the Mutect2 documentation (https://gatk.broadinstitute.org/hc/en-us/articles/27007991962907-Mutect2) states:

Specifically, the mode sets --initial-tumor-lod to 0, --tumor-lod-to-emit to 0, --af-of-alleles-not-in-resource to 4e-3, and the advanced parameter --pruning-lod-threshold to -4.

From looking at the source code of Mutect2, --pruning-lod-threshold is actually set to -4 * ln(10) and two further flags are also set: --recover-all-dangling-branches true and --annotation “OriginalAlignment”. Furthermore, the documentation also states

"Mutect2 automatically sets parameters appropriately for calling on mitochondria with the --mitochondria flag",

but the flag is actually --mitochondria-mode.

Is the 'correct' value of --pruning-lod-threshold in mitochondria mode -4 or -4 * ln(10)?

Many thanks,

Tom