(How to) Call common and rare germline copy number variants - which bamfile

February 18, 2020 12:01

This is a general question. When using the above mentioned pipeline, which bam-files are you supposed to use. Do you go through the whole, realign, markduplicate, deduplicate or do you use the bamfile you get from your alignment step (in my case bwa mem)?

This is for whole genome germline data.

Thanks!

6 comments

Bhanu Gandham

February 19, 2020 15:03
Hi Sophie Agger

Take a look at this doc: https://gatk.broadinstitute.org/hc/en-us/articles/360035531152--How-to-Call-common-and-rare-germline-copy-number-variants
-1

Comment actions Permalink
Sophie Agger

February 20, 2020 07:43

Edited
Hi Bhanu,

That's where I started, I couldn't find the answer there which is why I tried the forum. Since the sample files are just called .bam I assumed that I was supposed to use the raw bam file, but my supervisor thought that I should use the cleaned up bam file, so now I'm unsure.
0

Comment actions Permalink
manolis

February 21, 2020 16:49

Edited
Hi,

personally I tested the gCNV pipeline using the bam file created from the BQSR steps. I think that is good practices to run first all pre-processing steps and then use your bam for further downstream analysis.

Best
0

Comment actions Permalink
Bhanu Gandham

February 22, 2020 14:56
Thank you for your invaluable input manolis.
-1

Comment actions Permalink
Andrey Smirnov

February 25, 2020 17:08
Sophie Agger I'm one of the developers of GATK gCNV. We have always used BAMs that are preprocessed using GATK best practices for running GATK gCNV, so I would recommend doing that.
1

Comment actions Permalink
Sophie Agger

March 02, 2020 11:49
Thank you Andrey :)
0

Comment actions Permalink