I am performing eQTL analysis, and wish to be answered for a quick question. In conducting eQTL analysis, biologically, is it fine to include subjects from all ethnic groups (to increase statistical power; not just European), despite the difference in LD structure across the population just by having covariates of several principal components for ancestry information? My current sample size is 95, but 60 by excluding non-European subjects.
When I did perform polygenic risk scoring, I did include only European subjects since a major ethnic group of GWAS subjects in my filed is European. I would like to learn how eQTL analyses deal with this in general. By the way, I have posted it to Biostars as well because it is also highly relevant to Bioinformatics. Thanks!