search

stephenslab / susieR

R package for "sum of single effects" regression.
https://stephenslab.github.io/susieR
Other
176 stars 45 forks source link

How to do multi-enthnic fine-mapping? #59

Closed gaow closed 3 years ago

gaow commented 5 years ago

I think this has been done before in the context of single effect model: basically combining the BF obtained before calculating PIP. This would work if the cross population causal variable is the same -- which should mostly be the case.

Potentially we can do the same for each SER update of SuSiE. Is there a problem for it? The code and interface might be a bit messy, though.

stephens999 commented 5 years ago

i would suggest treating this a special case of the multivariate version (at least initially).