Hi All,
Got a question here when using PLINK for GWAS analysis. Appreciated for your opinions.
I recently run a logistic regression analysis for a set of GWAS SNPs, and tried to include first ten PCs as covariates. However, this returned "NA" as results in the fileds of e.g. OR, SE, P etc. When including the first three PCs, it was the same, i.e. NA returned. When only included a single PC, say PC5, it worked.
This is the first time I had this problem, it seems to me somehow I over correct it by including several PCs at the same time. Not sure if any people had experienced similar issue, would like to know what is going on exactly. Thanks for any opinions.
Guan
Maybe multi-collinearity? Why does my linear/logistic regression output have all NA's? http://pngu.mgh.harvard.edu/~purcell/plink/faq.shtml#faq6