search

mskcc / tempo

CCS research pipeline to process WES and WGS TN pairs
https://cmotempo.netlify.com/
12 stars 5 forks source link

Postprocess of germline mutation calls #269

Closed kpjonsson closed 5 years ago

kpjonsson commented 5 years ago

Re: #247 and other things not yet discussed at length.

Finalize post-processing and annotation of germline calls. Something like this: image

This is on me to sort out.

kpjonsson commented 5 years ago

As per discussion in planning meeting, we'll limit post-processing of germline calls to filtering of common variation and sequencing artifacts. This also includes genotyping of the tumor.

evanbiederstedt commented 5 years ago

@kpjonsson I forget what we agreed to use for filtering....dbSNP.vcf?

kpjonsson commented 5 years ago

I'm not sure either. What exactly are you referring to?

evanbiederstedt commented 5 years ago

@kpjonsson I'm thinking about the GATK4 standards for doing germline QC. I think they are still using dbSNP.vcf, which are AFs from the 1K genome project.

https://software.broadinstitute.org/gatk/documentation/tooldocs/3.8-0/org_broadinstitute_gatk_tools_walkers_haplotypecaller_HaplotypeCaller.php

What exactly are you referring to?

This is one way to annotate/filter out common variants. I'm guessing you would try something more sophisticated.

Related discussion: https://github.com/mskcc/vaporware/issues/238

kpjonsson commented 5 years ago

I think that's only used for the sake of annotation. I don't see why we wouldn't use gnomAD here as for the somatic variants (which is what I planned).

evanbiederstedt commented 5 years ago

I don't see why we wouldn't use gnomAD here as for the somatic variants (which is what I planned).

Yes, I had suspected this. I agree that this makes sense.

kpjonsson commented 5 years ago

Closed due to #342.