Graph Attention Site Prediction (GrASP) sc-PDB Dataset

The modified version of the sc-PDB dataset that was used for GrASP training. See the original sc-PDB database for raw input structures. ready_to_parse_mol2.zip: Protein and ligand structures after our additional processing was applied. processed.zip: Processed protein graphs used as graph neural network inputs. no_ligands.txt: List of systems that did not have any bound ligands according to the geometric criteria from P2Rank. scPDB_uniprot.pkl: Pickle containing UniProt ID(s) for each receptor, used to define train/test splits.