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!

can depth of bam file at some position is lower than depth of vcf?

mankyuamankyua Member
edited August 2017 in Ask the GATK team

I aligned reads to reference with BWA-mem and called variants using HC.

In my understanding, depth of bam can be reduced in vcf file because of some filtering options. However, can depth of bam file at some position is lower than depth of vcf? for example, in a bam file, there is only one read is aligned to reference. reference is G and alternative is A. So, if I understand about variant calling, AD should be 0:1, isn't it? But AD is 0:2. How can I interpreter this situation?

I want to upload my files but I cannot open the attach image option above.

Many thanks



  • EADGEADG KielMember ✭✭✭

    Hi @mankyua,

    the HaplotypeCaller does some realignment, further information about this you can get in the docs. You can use -bamout [Filename] to see the result of the realignment. If you want to limit the output to your region of interest, because the HC slows down with the bamout option activated, you can use the -L chr:Start:End option.

    Do a comparison of the origninal bam and the one from bamout in IGV.

    I think you will see that the HC realigned your read at the specific position and therefor you get an other depth then in the original bam.

    Hope this helps,

    Greetings EADG

  • Many thanks! Kiel!

    It's good information for me!

Sign In or Register to comment.