Mitochodrial reads filtered by cellranger atac?

[tkiehl]

I'm trying to determine if the mitochondrial reads are actually being filtered out during this process. In using the instructions for building a custom reference and following guidance from 10x it would appear the the mitochondrial reads are being filtered out before mgatk can see them.

In the config file example on these mgatk instructions ( https://github.com/caleblareau/mgatk/wiki/Increasing-coverage-from-10x-processing ) there is a line: non_nuclear_contigs: ["chrM"]

Per this faq from 10x ( https://kb.10xgenomics.com/hc/en-us/articles/7461773747853-Does-Cell-Ranger-ATAC-filter-out-mitochondrial-reads ) it would appear that this configuration would result in the filtering of mitochondrial reads.

This seems to bear out in my hands as my mtdna depth plots indicate very low mitochondrial read counts. At the same time, I may be misunderstanding. In one example, if I sort the bam file I get 2.4 million reads on chrM but if I look at the peak_annotation.tsv, there are no peaks annotated for chrM

Hopefully I've provided enough information here for someone to clarify this for me. Thanks!