Substrate/Product data for Enzymes

[jozhang97]

Hi,

Thanks for the useful repo and nice presentation at NeurIPS!

I'm looking to get more information about the ligand (e.g. pdb id, pubchem id, smiles) for the enzyme commission task. I downloaded but could not find this information in the annotation file here https://github.com/BorgwardtLab/proteinshake/blob/6542d3c51f4b20b6b071275a7b8fe9055709e8b3/proteinshake/datasets/enzyme_commission.py#L52

Perhaps this is because the enzyme was not crystalized with the substrate. But the enzymes are quite well documented (there are Gene Oncotology and EC numbers), so I suspect that the ligand information is somewhere. Do you have any recommendations on how to find it?

[timkucera]

Thanks for the issue! We're working with high pressure on the next version release, where the different databases will be more accessibly integrated (including ligand information). I would suggest to wait for that release, if you need the data urgently you can retrieve them from BRENDA (https://www.brenda-enzymes.org/ecexplorer.php?browser=1) or Rhea (https://www.rhea-db.org/)

[jozhang97]

Thank you for your contribution!