Joint genotyping for somatic variants by using unrelated samples
Entering edit mode
5 months ago
nkabo ▴ 60

Dear all,

I am new in the genomics data analysis field, I have a question regarding the tools for joint genotyping. I have WES data for 60 case samples (human, none of them are related). I am trying to find possible rare variants and I would like to do joint genotyping. Samples are unrelated and variants are somatic.

Could you offer a program or tool for this task? I usually follow GATK Best Practices Guidelines, but I could not see a tool for that.

Many thanks

variantanalysis NGS WES WGS • 354 views
Entering edit mode

It seems no such program exists. (I'd loved to be proved wrong, though!) GATK's Mutect2 does have some joint genotyping capabilities, but only within the same individual.

Entering edit mode

hi nkabo,

sorry, im not totally clear on what you mean. do you mean you currently have 60 individual BAM fils and you would like to do joint variant calling?

In other words, it seems like you are trying to get to a multisample VCF, but where are you at present? There are steps you can take during alignment as well as during variant calling.

also, i assume that, since you are mentioning somatic variation, you looking at some kind of cancer. is that right? or, alternatively, are. you looking at de novo variants that occur in development and that are therefore not inherited? the reason i ask is, if you're looking at cancer and you know the subtype, there is a hell of a lot more sophisticated analysis possible based on our understanding of tumor mutational signatures. in other words, you can look for the types of variants that tend to arise in that cancer.

its also necessary to know because, if it is somatic/cancer/something like that, you likely wont have 100% pure sample, youll have a mixture - possibly of tumor and normal, possibly of different tumor clones etc.

the best pipeline to use is something i can tell you (and provide citations), but not without knowing the answer to these questions.


Entering edit mode


Many thanks for your reply. I have 60 BAM files at the moment, all are from patients.

I am working on a rare neurological disorder and looking for de novo variants (not inherited) and somatic variants, samples are not extracted from tumors and purity is in a good level.

Is there any pipeline possible that you could suggest in this case?

Thanks a lot!


Login before adding your answer.

Traffic: 1409 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6