Calling Genotypes On Multiple Trios

Question

[a11msp]

Hi everyone,

I have a set of eight Illumina-sequenced trios for an organism, for which a reference genome is available. I'd now like to call genotypes on these data taking advantage of both a "classic" multisample and trio-aware approaches. I understand that bcftools (part of samtools suite) can do trio-aware calling, but I'm not sure it can work with multiple trios. At the same time, GATK (and samtools too?) can run a Bayesian model based on observed allele frequencies assuming a flat population structure.

So my question is - are there any tools that can combine both approaches?

Thanks a lot!

[Zev.Kronenberg]

You can always hard filter by taking the union of the parents intersected with the child. The child must contain only SNVs found in the parents.

[a11msp]

Thanks - I currently do something along these lines, but I was wondering if a more general way of taking this info into account when computing SNP quality scores was available...

[Zev.Kronenberg]

mpileup in samtools can leverage calling multiple individuals. I believe this is beneficial when there is evidence of a SNP, but low coverage. http://samtools.sourceforge.net/mpileup.shtml

Answer 1

[Bioinfosm]

VAAST is one tool, especially its upcoming p-VAAST version that can be of use for that analysis.

[Zev.Kronenberg]

Yes. p-VAAST is ideal for disease gene hunting in families. We are happy to hear that people are adopting VAAST.

[a11msp]

Thanks, will try out VAAST! Could you let me know a bit more about the upcoming p-VAAST and when it's expected to appear (which would be very good to know for my project's logistics)

[Bioinfosm]

from what I heard, its coming in May-Jun timeframe

[Zev.Kronenberg]

It's close. We are testing it on some real datasets now.

[a11msp]

Great, thanks! And what will be the key difference from the existing VAAST?