De novo quality scores

KathKath Posts: 43Member

Hello,

I was just wondering if anyone uses GATK's SelectVariants walker to call de novo mutations (Mendelian violations) and, if so, what -mvq cut-off do they use? My data is exome sequencing with a large range of read depths - from mean target coverage of 14X to >50X.

Thanks,

Kath

Answers

  • Geraldine_VdAuweraGeraldine_VdAuwera Posts: 8,147Administrator, GATK Dev admin

    Hi Kath,

    Just to be clear about terminology, SelectVariants doesn't "call" anything, that's what UnifiedGenotyper and HaplotypeCaller do. SelectVariants allows you to select a subset of previously called variants that match certain criteria.

    We don't currently have any specific recommendations on parameters for selecting de novo mutations. My personal approach would be to use a rather low mvq value, then apply VariantFiltration to discriminate between artifacts and real mutations of interest.

    Geraldine Van der Auwera, PhD

  • KathKath Posts: 43Member

    Thanks for your reply. I don't have much of a feel for what parameters people typically use with VariantFiltration for narrowing down to real variants. In the example you have AB<0.2 and MQ0>50 - is this typical criteria? I am using a multi-sample vcf. If a variant exceeds the thresholds in sample A but not sample B, will the likely genotype in sample B still be presented in the output vcf?

    Kath

  • Geraldine_VdAuweraGeraldine_VdAuwera Posts: 8,147Administrator, GATK Dev admin

    We provide some standard threshold values in the documentation; there's a section on that at the very end of the Best Practices document.

    Geraldine Van der Auwera, PhD

  • KathKath Posts: 43Member

    Thank you! Sorry I didn't notice that before.

Sign In or Register to comment.