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!

How do MuTect1.1.4 and MuTect2 differ in Tumor only mode (no matched control) variant calling

What is the difference between MuTect1.1.4 and MuTect2 versions in Tumor only mode (no matched control) variant calling method? Is there a preference of using one over the other for this specific application? I would prefer not to use a PON vcf and run tumor only mode to identify variants and use COSMIC/dbSNP to filter out common SNPs. How do I interpret the KEEP vs REJECT calls in this case where there is no matched control to determine somatic variants?


  • davidbendavidben BostonMember, Broadie, Dev ✭✭✭

    @vsundare MuTect1 is obsolete. As of GATK 4, Mutect2 is a complete rewrite. We recommend that you use the best practices workflow using our AF-only gnomAD for the --germline-resource argument. dbSNP will give worse results. The PON is for removing technical artifacts, not germline variants, and therefore should be used when possible. When there is no matched normal a PASS call is Mutect2's best guess and inevitably has some chance of actually being a germline variant.

Sign In or Register to comment.