Background information and recommendations for phasing ASW whole genomes

Question

[vlaufer]

Hello Biostars,

I have a cohort of whole genomes that I am about to phase.

Before doing so, I am seeking recommendations on how to do this optimally.

The whole genomes I have are mostly ASW genomes, with a few Caucasians, and I was planning to use BEAGLE. Currently, I had planned to carry out the following procedure:

1. Download the CGI-sequenced CEU, YRI, and ASW genomes from 1kG
2. Phase these control genomes first (I do not believe the CGI genomes are phased, and these were the ones I was going to use).
3. Using these as a reference, phase the case ASW genomes

I had planned to use something like the following:

java -jar b4.r1274.jar ref=1kG.vcf gt=case.vcf out=out.gt

Is this a bad plan? Does anyone have recommendations about a better procedure? Better program? Good place to do background reading on the number and type of phased genomes it is helpful to have before phasing your own?

Thank you

Answer 1

[Zev.Kronenberg]

There is nothing wrong with what you are proposing to do.

Alternatively you could use the 1KG AWS as a reference panel to improve your phasing. You wouldn't need to re-phase the 1KG data, just tell BEAGLE to use it as a reference panel.

[vlaufer]

Thank you very much. I've seen your comments elsewhere as well, and I wanted to thank you for responding here as well.

Local ancestry estimates, here I come!