There are four normalized gene expression data in syn5609488, and I don't know whether I should choose to include ancestry or not. My question is which expression data you are using in sherlock analysis(syn5585485) or eQTL analysis(syn4622659)?include ancestry or not? Thank you!

Created by xiaoyan li adelalady
Thank you very much for your reply.
"REQUIRE ANCESTRY" means that the expression is adjusted for SNP-inferred ancestry components, "EXCLUDE ANCESTRY" means that it was not. See [Fromer et al](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5083142/) for details. For our analysis, we did not pre-adjust for ancestry components (i.e. we used "EXCLUDE ANCESTRY"), but we included them in the eQTL regression model. Again, you can see the manuscript for details.
Sorry, Solly, I didn't clarify my question. I noticed that the sample size was the same in the expression data of 'EXCLUDE ANCESTRY + NoSVA' file and 'REQUIRE ANCESTRY + NoSVA' file, combined with what you mean, **can I understand** that the expression file for adding the ancestor's name refers to adjust for ancestry? And I want to make box-plots for SNPs based on the expression data(syn5609488) and imputed genotype(syn3275221), should I choose to include the expression data set of ancestors('REQUIRE ANCESTRY + NoSVA')? I'm really sorry to bother you again.
If you're doing eQTL analysis, I suspect you'll want to adjust for ancestry in some way, whether you use expression that has been pre-adjusted for ancestry, or include ancestry components directly in your eQTL model. Solly

CMC expression data -- how to choose page is loading…