To greatly shorten enrichment analysis time, process databases with the optimizer module. It will pre-calculate overlap statistics for each set of background SNPs in the [dir]/custom_data/backgrounds/[org] folder vs. each genome annotation track in the [dir]/grsnp_db*/[org] folder. These statistics take considerable amount of time and pre-calculating them in advance greatly speeds up GenomeRunner.
These pre-calculated statistics are stored in the [dir]/grsnp_db/[organism]/bkg_overlaps.gr files. Before running the optimizer, be sure to place some background SNP sets in the [dir]/custom_data/backgrounds/[organism]/ folder.
We run the optimizer module as:
python -m grsnp.optimizer -g hg19 -d [dir] -w 1
The -w argument specifies number of workers to perform statistical calculations. See the Parallelization section for more details.