Battenberg Sex Assignment For Metapipeline

[graceooh]

Describe the issue
Hi, thank you for your work on this pipeline.

It looks like chr X for males are being reported wrong for battenberg. Battenberg is often reporting major copy number 1 and minor copy number 1 for my samples, which is not correct since all my cohort is male. It looks like issues with allele counts on X for males have been documented here (https://uclahs-cds.atlassian.net/wiki/spaces/BOUTROS/pages/3639614/Battenberg+How-To ). But this document was created back in 2017. Do any of you know if this is still a battenberg issue?

Also, after checking 3 samples from my cohort, it looks like only a small part of chr X is being reported with copy numbers (~0.1 % of the whole results) so I don't think it's too big of a problem. We can just let users be aware that chr X will be wrong or switch off chr X copy number calling for males as it can have downstream implications in SRC.

Some of my outputs are here:

Example file 1: /hot/data/unregistered/GAP6-Oh-PRAD-OSCR/pipeline_outputs/metapipeline-DNA/metapipeline-DNA-6.0.0/PRAD0000059/main_workflow/output/call-sCNA-3.2.0/RME-01-MR04/Battenberg-2.2.9/output/Battenberg-2.2.9_PRAD0000059_RME-01-MR04_subclones.txt

Example file 2: /hot/data/unregistered/GAP6-Oh-PRAD-OSCR/pipeline_outputs/metapipeline-DNA/metapipeline-DNA-6.0.0/PRAD0000059/main_workflow/output/call-sCNA-3.2.0/JHU4-Bone/Battenberg-2.2.9/output/Battenberg-2.2.9_PRAD0000059_JHU4-Bone_subclones.txt

[Faizal-Eeman]

There are two possible cases I can think of where both major and minor copy numbers are =1

1. The coordinates for the CNA segments in question belong to Pseudoautosomal Regions (PAR) which are regions of similarity between chrX and chrY making the region seem diploid to most variant callers. There are two chrX CNAs in example 1 and example 2 you've provided, one CNA overlaps with PAR1 and the other with PAR2 which explains why we see a copy number for minor allele when only major allele is expected for a male (XY) case. See Pseudoautosomal Regions in this project PR.
2. The sample in question may have Klinefelter syndrome (XXY) which could lead Battenberg to report both major and minor copy number =1 in chrX as opposed to in a typical male XY sample where only major copy number =1. This can be verified by doing a coverage comparison of chrX vs chrY to observe 2:1 coverage ratio.

Suggestions:

• If all chrX CNA called by Battenberg in the cohort overlap with PAR, then they can be excluded.

• If there are non-PAR chrX CNAs called by Battenberg and both major/minor copy number =1, then the sample should be checked for a XXY case.

We can just let users be aware that chr X will be wrong or switch off chr X copy number calling for males as it can have downstream implications in SRC.

Yes, this is a good suggestion to improve documentation for pipeline-call-sCNA.