I'm running GWAS analyses using GAPIT, and I'd like to fit for example location or other grouping variables as a covariate. Any idea how to do this? I tried to fit an incidence matrix but I keep getting the following error.

Error in t(KWO) %% ginv(KW) %% UW : requires numeric/complex matrix/vector arguments

fitting 0 & 1 for M & F seems to work but fitting (1,0) (0,1) (0,0) for groups A, B, C doesn't, and Gapit forums weren't much help also!

Many thanks,

Arjan