We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

Maximizing sensitivity of HaplotypeCaller for pooled sample

HESmithHESmith National Institutes of HealthMember

I'm attempting to call variants (primarily SNPs) with HaplotypeCaller from a pooled sample containing 95% wild-type and 5% polymorphic strain data (C. elegans, SE-50bp, 20-fold genomes). Per guidelines, the '''-ploidy 40''' flag was used, but it's only detecting ~1/4 the number of SNPs as other variant callers (e.g., FreeBayes and VarScan2). I can validate by prior annotation (the polymorphic strain has been sequenced) so I'm not concerned about false-positives. What additional parameters should be used to increase sensitivity?


  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    Some ways to increase sensitivity are to reduce minPruning, allow more alleles, things like that. But keep in mind HC isn't really designed for that type of design so it's not necessarily guaranteed you'll get to where you want to go. Have you already done some validation to gauge current performance? Meaning, of the N variants you think you're missing, ho many are actually real? Would help to know what is the target improvement.

Sign In or Register to comment.