Somaticindeldetector Vs Unifiedgenotyper

Question

[Leandro Batista]

Hey Guys,

I've been establishing my own first pipeline for calling variants: Indels and SNPs. For that, I used mainly GATK tools.

I'm doing this for whole genome and whole exome sequences of mouse inbred strains.

So I'd like to hear your opinion about calling Indels with SomaticIndelDetector (--unpaired flag for simple sample) or UnifiedGenotyper (INDEL mode).

Which one gives better results and is more appropriate for my study?

[Russh]

Could you expand on why you're considering SomaticIndelDetector please? I would have thought, given your application, you'd be following up germline variation rather than somatic mutations wouldn't you?

[Leandro Batista]

My understanding is that SomaticIndelDetector only makes that difference (germline/somatic indel) when analyzing two samples (tumor and normal). In the unpaired mode, I guess it just calls Indels...