Preprocessed CMAP dataset for UNAGI in-silico perturbation

These are preprocessed drug/compound info files from CMAP LINCS 2020 database, https://clue.io/data/CMap2020#LINCS2020. 'cmap_drug_target.npy' uses the direct drug target genes provided in CMAP LINCS 2020. 'cmap_drug_treated_res_cutoff.npy' uses genes which is up/down-regulated significantly after individual drug treatments in CMAP LINCS 2020. The level 5 MODZ score was used to determine the extent of gene expression changes after treatments. Top 5% significantly changed drug-gene pairs are kept. NOTE: Running perturbation with this file could take a long time. To reduce the run time, you can use only a subset of this data. e.g. running 'use_only_drugs.py' to run perturbation with only established drugs. 'cmap_direction_df.npy' indicates the direction of gene regulated by drugs after treatments. The level 5 MODZ score was used to indicate the direction of changes. 'use_only_drugs.py' filters out compounds without a formal drug name, e.g. 'BRD-xxxxx' will be removed. This script can extract established drugs for the repurposing purpose. NOTE: if you use 'cmap_drug_target.npy', you don't need to run this script.