bioRxivJClub
Protein Pocket Coengagement Prioritizes Drug-Drug Interactions
bioRxiv · · Preprint · Open access
Skolnick, Srinivasan
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
Pocket coengagement predicts documented drug-drug interactions: confidence-summed COLIG ranking of 21,945 drug pairs achieves 49.3% top-5% precision versus DDInter.
- Why it matters: DDIs are mostly catalogued empirically, leaving limited structural rationale for why two drugs interact, and whether pocket-level ligand engagement alone can recover known interactions.
- What they did: They used composite protein multiple-ligand (COLIG) assignments to score 21,945 pairs from 210 drugs by predicted engagement of each protein’s top-ranked pocket, without training on clinical DDI labels.
- The result: Top-pocket engagement reached 50.3% precision against DDInter, and class-dependent hypotheses emerge for direct versus noncontact ligand engagement, though causal mechanisms remain unestablished.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access