Filter coding germline variants from 100 human WGS gVCFs

Hi there,

My aim is to filter germline coding variants from 100 human WGS gVCFs. I was wondering what would be the best approach. I am using GATK Best Practices (version 4.1.0.0) on two local servers.
1/ use a list of exons as intervals for GenomicsDBImport (~240.000 intervals), and then use GenotypeGVCFs? If it’s the best approach, is the number of intervals appropriate? My experience is that GenomicsDBImport runs quite slow with "only" 650 intervals.
2/ Import “everything” in GenomicsDBimport, use GenotypeGVCFs, and finally restrict the VCF to coding regions with SelectVariants or another GATK function? If it’s the best approach, what would be the appropriate number of intervals for GenomicsDBImport to speed up the process?

Thanks!

Eric

Best Answer

Answers

Sign In or Register to comment.