irene
New Member
Posts: 5
|
Post by irene on Aug 18, 2014 13:33:22 GMT
Dear all,
I have a question regarding ethnic groups and GCTA estimations.
More precisely, I have a sample of N= 5,320 individuals but half of them are Caucasian, the rest are different ethnic groups (some African, some Asian, some mixed).
So far, I have been running GCTA only on the Caucasian sub-sample. But I get large SE and I would like to see if/ how I can use the rest of my sample.
Is there any way to overcome this problem?
I think that constructing the GRM with all individuals and using ethnicity (or PCAs) as covariate will not work.
Would it make sense to select only the SNPs that have the same/ similar frequency in all populations?
Is there maybe another solution?
Thank you in advance!
|
|