To optimize computation time, the Script 2 is made to launch a job array on the cluster. To do that, we cut the genome in blocks with a fixed SNPs number, and we launch CMS on each blocks in parallel.
The number of blocks should depend on the number of phenotypes and the number of individuals.
