This folder contains data for the paper "Computing disease-specific gene embeddings via constrained optimization", submitted to The 2022 ICML Workshop on Computational Biology.
Notebook with the interesting code part is available at this link
The generated embedding can be viewed on Tensorboard
PCG_AML.csv : information on genes that relate with Acute Myeloid Leukemia cells funcional states. Data re-arranged from CancerSEA.
E_sim.npy : edges related to similar genes; information derived from metabolic pathway. Information derived from KEGG
genes_pathways.txt : genes for which we know at least one matabolic pathway.
count_only_mal_D0_small.npy.zip : count matrix related only to genes for which we know any relation with an AML cells functional state. Data re-arranged from GSE116256.
genes_only_mal_D0_small.txt : genes associated with the count matrix related only to genes for which we know any relation with an AML functional state.
count_only_mal_D0.npy.zip : count matrix related to all the genes availble in the consider GEO series. Data re-arranged from GSE116256.
genes_only_mal_D0.txt: genes associated with the count matrix related to all the genes availble in the considered GEO series.