Identify Germline Mutations

Question

[yliuboston]

Hi everyone,

I am now trying to identify germline mutations from tumor and matched normal samples in TCGA and also I need to know this germline mutation occur in exactly which samples. I know some software like VarScan can do this, but they are all working on .bam file, I only have some .vcf files, any software or protocol can do the following? 1. Identify germline mutation in a group of tumor and matched normal. 2. For each of germline mutations identified, tell which sample they call.

Thanks in advance!

Answer 1

[Charles Warden]

1) VarScan actually uses .pileup (it doesn't directly read .bam file)

2) If you already have a list of variants in a .vcf file, I don't really think you can call somatic mutations by doing anything beyond comparing overlap between two files (which is something I would do using a custom Perl script). If at all possible, I would try to find a more raw form of data and use standard tools like VarScan, MuTect, SomaticSniper, etc.

[karl.stamm]

It smells like a homework problem. He has two vcf and just needs to subtract.

[yliuboston]

Thanks a lot for your explanation. I think I have a .vcf file that have compared one tumor and its matched normal, so I guess the "subtraction" job has been done. But I am not quite sure among all those variants called, which germline mutations that I should be confirmed to use in the following analysis, like filter=pass, or based on DP bigger than a threshold? If I donot have access to a more raw form of data, is there a standard way to do this? Thanks in advance!