Heads up:
We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
If you happen to see a question you know the answer to, please do chime in and help your fellow community members. We encourage our fourm members to be more involved, jump in and help out your fellow researchers with their questions. GATK forum is a community forum and helping each other with using GATK tools and research is the cornerstone of our success as a genomics research community.We appreciate your help!

Test-drive the GATK tools and Best Practices pipelines on Terra

Check out this blog post to learn how you can get started with GATK and try out the pipelines in preconfigured workspaces (with a user-friendly interface!) without having to install anything.

Discrepencies in quality scores and read depths

I was given reduced.bam files, and asked to rerun GATK on our targeted sequencing project (both for learning purposes, and to run some additional downstream analyses), and I'm having difficulty replicating what the previous person did exactly. I traced the root of the problem to the UnifiedGenotyper step --when comparing the vcf files output, I see some (not too many, but a few) discordances, namely, values for QUAL are different (although mostly correlated), and values for AD, DP and PL under INFO are different as well. Genotype calls and genotype qualities, however, are mostly concordant, except for a very few indel/MNP sites. The new run also lists more number of variants that passes the quality filtering.

Comparing the old run (from log files) to my run, I can spot only a few differences, so I was wondering if you have any insights whether any of these, or which, could cause the observed discrepancies: First of all, the old analyses were done in December 2012, and I think GATK2.0 was used, whereas I use GATK2.7. Is it possible that new versions of GATK calculates the quality scores differently, particularly, for indels..? Second difference is, the person who ran it before had scattered the intervals into 40 counts (and gathered it after UnifiedGenotyper), whereas I ran it as one piece. Also he had -dcov 75 in his command line, whereas I omitted that.

Any insight is much appreciated!

Thanks a lot,


Best Answer


Sign In or Register to comment.