PocketAnchor: Learning structure-based pocket representations for protein-ligand interaction prediction

Protein-ligand interactions are essential for cellular activities and drug discovery processes. Appropriately and effectively representing protein features is of vital importance for developing computational approaches, especially data-driven methods, for predicting protein-ligand interactions. However, existing approaches may not fully investigate the features of the ligand-occupying regions in the protein pockets. Here, we design a structure-based protein representation method, named PocketAnchor, for capturing the local environmental and spatial features of protein pockets to facilitate protein-ligand interaction-related learning tasks. We define “anchors” as probe points reaching into the cavities and those located near the surface of proteins, and we design a specific message passing strategy for gathering local information from the atoms and surface neighboring these anchors. Comprehensive evaluation of our method demonstrated its successful applications in pocket detection and binding affinity prediction, which indicated that our anchor-based approach can provide effective protein feature representations for improving the prediction of protein-ligand interactions. [Display omitted] •PocketAnchor describes protein pockets using probe points named “anchors”•PocketAnchor learns reasonable anchor features via message passing neural networks•PocketAnchor improves the success rates of pocket detection•PocketAnchor learns effective feature representation for improving affinity prediction Molecular representation is important for developing deep learning-based methods in modeling biomolecules. To improve current protein-ligand interaction prediction methods, Li et al. designed PocketAnchor to represent the subpocket-level structural features of protein binding sites, which were proven to be effective for downstream pocket detection and affinity prediction tasks.