1000 Genomes exomes as controls

Question

[knishimura]

I'm analyzing a small case-only exome sequencing study and am hoping to use publicly-available controls for statistical comparison with SKAT or the C-alpha test. People have suggested that I use the 1000 Genomes exomes. Does anyone know of any studies that have been published that use 1000G exomes as controls? I would like to review the methods that people have used.

Thanks!

[Pierre Lindenbaum]

tell us you searched something, what did you find ?

[mkaushal]

I am also working on an exome sequencing study with 122 affected samples and no controls. Could you please let me know how you eventually performed the rare variant association analysis?

[Dawe]

I have now the very same problem. Did you find any valuable resource? I'm reading the original paper by Neale trying to implement c-alpha test based on allele counts found in ExAC data. In principle you won't need individual genotypes, only allele counts. Unfortunately the methodology described has some quirks (apparently data in figure 2A don't match table 1)

[Filago]

Yes, in principle you do not need individual genotypes. But then you make the assumption of no LD between your variants, leading (in the worst case) to an inflation of false positives.