Speeding up Exon skipping and intron retention on HPC

[rugilemat]

Hi,

I've been trying run swan as a script on our HPC and it seems that es_df = sg.find_es_genes(verbose=True) and ir_df = sg.find_ir_genes(verbose=True) take a very long time to run (48+ hours) as it has 40+ k edges to test. I've tried going down multiprocessing route but not sure that's the best way.

I'm wondering if you had any advice on trying to speed this up, especially in HPC environment?