deeprank
deeprank copied to clipboard
shall we disable one grid for each chain
It does not make much sense to have one grid for each chain. We care about the interactions between atoms no matter to which chain an atom belongs to. This is also done in:
Ragoza, Matthew, et al. "Protein–ligand scoring with convolutional neural networks." Journal of chemical information and modeling 57.4 (2017): 942-957.
Using one grid for all chains will also allow DeepRank to deal with protein complexes with 1, 2, 3, and more chains, instead of limiting to complexes with two chains.