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!

Contaminaton very high in Mutect mitochondrial pipeline

Hi, I ran the mitochondrial pipeline on a standard Truseq WGS genome sample, but I got contamination 0.98. See here for contamination metrics:

SampleID Contamination MinorHG MinorLevel MinorSNPs MinorHGvariants MajorHG MajorLevel MajorSNPs MajorHGvariants VerifyScore meanCovVar
UDP-1003_genome.realigned.bam None U8b1b2 0.988 24/30 3 U8b1b2 0.000 24/30 1 0.0 3260.0

Not sure how this is possible or what this means. i.e. how is the minor HG 0.98 but the major one 0 ?

Answers

  • GERGER Member

    Quick update -- I ran haplochecker on its own on the aligned BAM file of the Mutect Mitochondrial pipeline and it gave the same illogical results above. So the problem is with the haplochecker tool.

    Notice also that it is outputting the MinorHG and MajorHG as the same.

    So I think the new Mutect Mitochondrial pipeline cannot rely on haplochecker due to this issue. Maybe you can use haplocheck or some other tool. Until this is fixed, the Mitochondrial pipeline won't work.

  • mshandmshand Member, Broadie, Dev ✭✭✭

    Hi @GER,

    Thank you for reporting this! Yes, I believe there is a bug in the version of haplochecker that we are currently using. We're testing the latest version of haplochecker which shouldn't have this issue, and once we've sorted it out we can update the pipeline to use that version. We can also provide a way to input a separate contamination estimate (such as from the autosome or another mitochondrial contamination tool) to add some more flexibility.

  • GERGER Member

    Yes thanks. I actually posted the issue to the haplochecker github and the author responded to upgrade to the newest version of haplocheck. I made a docker for it and it runs ok already as part of your wdl pipeline. Just requires some minor fixes to processing the output.

    But... now I found a new bug in the new haplocheck where it reports Major HG level = 0 when there are no heteroplasmies found: https://github.com/genepi/haplocheck/issues/7. It shoudl be 1.0

    I also opened another issue on haplocheck's github to clarify how the old haplochecker contamination output corresponds to the new one, because they are different. In case you have any insight into this: https://github.com/genepi/haplocheck/issues/6

Sign In or Register to comment.