ExAC includes WES of phenotyped population

Question

[CY]

I have a question regarding ExAC.

I am trying to retrieve SNP with corresponding allele frequency for my own research. However, I noticed that ExAC dataset includes a fair amount of WES of phenotyped population including diabetes, Schizophrenia & Bipolar and Myocardial condition.

Wouldn't the SNP allele frequency be biased by these phenotyped population?

Also, ExAC states that they specifically excluded severe pediatric diseases. Why is this condition so special and has to be excluded?

Thanks

[Amitm]

I don't know regarding the check for paediatric disease samples, but ExAC (rel. 0.3 onwards) does maintain 2 subsets - 1) excluding TCGA cohorts 2) excluding Pyschiatric cohorts

ftp://ftp.broadinstitute.org/pub/ExAC_release/release1/subsets/

Answer 1

[sure]

From my limited understanding, ExAC was created to have a reasonable comparison data set (Reference) for childhood-onset Mendelian diseases. It will make sense not to include individuals with severe pediatric diseases. SNP allele frequency will be biased to those SNPs which are associated with above phenotypes for others it should be fine.

Answer 2

[Chris Miller]

Sure - ExAC is useful for lots of things, but it is not designed to be a control for common disease studies. That requires sequencing a well-phenotyped cohort of individuals known to be negative for your trait of interest.

[CY]

Thanks for sharing. However, I think, as singh opinion, it could be a control for Mendelian diseases, especially with high penetrance, right?