variant calling

Question

[Elisa]

Hi everyone,

I must perform somatic variant calling on ovarian samples (targeted sequencing performed with Illumina). I do not have the control match; it is still correct to use Mutect2 or do you have any other tool to suggest me ? Moreover, I'd like to know if I should perform local realignment around indels.

Thanks in advance

Answer 1

[g.papp-co]

Hi!

You can use Mutect2 with tumor only mode using a PoN and germline resource.

https://gatk.broadinstitute.org/hc/en-us/articles/360037593851-Mutect2

Consider filtering as well, but from what i have read from others, the whole tumor-only procedure is not advisable.

As for the realignment, i think's it's not neccesary anymore but please read: https://gatk.broadinstitute.org/hc/en-us pipeline part.

[Elisa]

It is not really clear to me how to build or where to find PoN and a germline resource.

[g.papp-co]

Germline resources can be downloaded from https://gnomad.broadinstitute.org/ website as a google cloud public datasets. I have never used PoN but there is a little article about it: https://gatk.broadinstitute.org/hc/en-us/articles/360035890631-Panel-of-Normals-PON- and also some download links.

Answer 2

[Kevin Blighe]

freebayes can do this, too, and it's easy to specify genomic intervals. For example:

freebayes \
  --region chr3:178148114-180240093 \
  --fasta-reference ref_genome.fa \
  --bam reheader.bam \
  --vcf freebayes_PIK3CA.vcf ;

Obviously there are limitations to the results produced.

Kevin