Deep Learning for Protein-Ligand Docking: Are We There Yet?

The effects of ligand binding on protein structures and their functions carry numerous implications for modern biomedical research and biotechnology development efforts such as drug discovery. Although several deep learning (DL) methods and benchmarks designed for protein-ligand docking have recently been introduced, to date no prior works have systematically studied the behavior of docking methods within the context of (1) using predicted (apo) protein structures for docking (e.g., for applicability to unknown structures); (2) docking multiple ligands concurrently to a given target protein (e.g., for enzyme design); and (3) having no prior knowledge of binding pockets (e.g., for unknown pocket generalization). To enable a deeper understanding of docking methods' real-world utility, we introduce PoseBench, the first comprehensive benchmark for protein-ligand docking. PoseBench enables researchers to rigorously and systematically evaluate DL docking methods for apo-to-holo protein-ligand docking and protein-ligand structure generation using single and multi-ligand benchmark datasets, the latter of which we introduce for the first time to the DL community. Empirically, using PoseBench, we find that (1) DL methods consistently outperform conventional docking algorithms; (2) most recent DL docking methods fail to generalize to multi-ligand protein targets; and (3) training DL methods with physics-informed loss functions on diverse clusters of protein-ligand complexes is a promising direction for future work. Code, data, tutorials, and benchmark results are available at https://github.com/BioinfoMachineLearning/PoseBench.