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

Abstract Modeling and predicting protein-ligand interactions have a wide range of applications in drug discovery and biological research. Appropriate and effective protein feature representations are of vital importance for developing computational approaches, especially data-driven methods, for predicting protein-ligand interactions. However, existing sequence-based protein representation methods often fail to explicitly learn the spatial features of proteins, while current structure-based methods do not fully investigate the ligand-occupying regions in protein pockets. In this work, we propose a novel 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 anchor points. Comprehensive evaluation of our method demonstrated that it can be successfully applied to detect the ligand binding sites on a protein surface and greatly outperform existing baseline methods. Our anchor-based model also achieved state-of-the-art performance in the protein-ligand binding affinity prediction task and exhibited great generalization ability for novel proteins. Further analyses illustrated that the anchor features learned by PocketAnchor can successfully capture the geometric and chemical properties of subpockets. In summary, our anchor-based approach can provide effective protein feature representations for developing computational methods to improve the prediction of protein-ligand interactions.