I've imputed genome-wide SNPs using the Phase 3 1000 Genomes Project data as a reference set, and I want to use this data to conduct an eQTL analysis. I want to do it in R, owing to the flexibility it affords me, and also the R package Matrix eQTL is supposedly exponentially faster than any other software with which to conduct this type of analysis.

The trouble is, I can't find any kind of post-imputation pipeline for the best method to do this. GTOOL, PLINK, SNPTEST, etc, there seem to be a variety of ways of handling the output of IMPUTE2, but I don't know what the best way for loading it into R would be.

Also, is it generally recommended to keep the files split up by chromosome or merge them into a single one?