We've just added support multiple phenotypes. Suppose we wish to analyze 10 phenotypes. We could do 10 separate runs, but that would cost 10 times more in computer resources. (Likewise, 1000 phenotypes would take 1000 times more.) Happily, single_snp_scale can process multiple phenotypes almost as fast as 1. The only extra requirement that is the phenotypes must be prepared so that they have no missing values. See an example in the notebook: https://nbviewer.jupyter.org/github/fastlmm/FaST-LMM/blob/master/doc/ipynb/S... * Carl Carl Kadie, Ph.D. FaST-LMM Team