ReactomeGS generation

[IwanParf]

Hello,
Thank you for setting up this data source! I noticed that the reactomeGS geneset matrix contains one pathway with a missing name (NA, row 29), which leads to an error when trying to plot the enrichment, just FYI.

My actual question is how the matrix was set up. the manual says it is based on Reactome v59, which had info on 2,080 pathways involving 10,624 proteins encoded by 10,381 different human genes. The matrix itself consists of 1308 rows (pathways) and 18818 ensemblIDs (genes). I was wondering how you extended the pathway info by app. 8500 ensembl IDs, which of course are to some extent redundant IDs that link to genes that are covered in reactome. I would be interested in how this matching was done to basically have a more up-to date pathway database, because in the latest v82 they have 2600+ pathways.

Thanks a lot!

Iwan