VariantRecalibrator and CalculateGenotypePosteriors requires the same 1000G vcf file?

bliu2bliu2 StanfordMember

Hello,

I am using the best practice pipeline to call genotypes from whole-genome sequencing data on 67 human samples. In the examples on the website, VariantRecalibrator uses 1000G_phase1.snps.high_confidence.hg19.sites.vcf as one of the training examples. However, CalculateGenotypePosteriors uses 1000G_phase3_v4_20130502.sites.vcf as the supporting file. Is this intentional? Does it matter which version of 1000G vcf is supplied to CalculateGenotypePosteriors?

Issue · Github
by Sheila

Issue Number
733
State
closed
Last Updated
Assignee
Array
Milestone
Array
Closed By
vdauwera

Answers

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie

    Hi there,

    The origin of the difference is purely historical. The CalculateGenotypePosteriors tool (and associated genotype refinement method) was written and put into play much later than the VQSR tools, so the developers used the latest dataset available at that time. Whereas the supporting datasets for VQSR haven't been updated in quite a while. We are currently looking into updating the recommendations to reflect more recent datasets; my intuition is that it will not make a big difference which file you use, but it's possible that marginal cases may see some minor effects.

Sign In or Register to comment.