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!

Most of mapped reads are disappeared after Mutect2

HJKwonHJKwon Member
edited November 2017 in Ask the GATK team

Hi, I'm HJ.Kwon.

I found some strange result from Mutect2 run.
I'm analyzing Breast Cancer FFPE sample for research.

In final vcf file, I couldn't find ERBB4's mutation which is already confirmed by Tumor+Normal run.
So, I made bamout bam and compared it with original bam files through IGV.

In the IGV, I found all the reads mapped to ERBB4 are disappeared.
And, the same thing happened to PIK3CA except it has one region has remained.

I'm attaching two IGV screenshots.
All MAPQ values is higher than 40 in the original bam file.
Is a bamout bam file is the final bam file to call the variants, isn't it?
Then, I want to know how's bamout bam is generated, and I wonder if I can adjust some parameter not to lose many reads?

  1. PIK3CA (upper part is bamout.bam)

  1. ERBB4

Best Answer


  • SkyWarriorSkyWarrior TurkeyMember ✭✭✭

    bamout is not the file to call the variants. Bamout is the result of the called variants containing haplotypes and reads supporting the variant.

  • @SkyWarrior
    I happened to think it is used as the final variant calling because the detected variants in ERBB4 gene are appeared in bamout bam.
    I'm attaching another figure.
    In this figure, I could find mapped reads in bamout bam for ERBB4.

    In my original post, I cannot find any variants in the region where the bamout bam has no reads.

    Then, May I ask a question one more?
    How come the entire mapped reads could be disappeared in spite of high MAPQ?
    It must be very difficult to understand how Mutect2 works but, could you give me a simple explanation?

    Thank you in advance.

Sign In or Register to comment.