Abstract
Predicting the ligand’s binding conformation within a target protein is a pivotal step in drug discovery. Despite its speed, molecular docking is ill-suited for blind docking where the pocket is unknown. Recently, deep generative models, especially diffusion models, have been proposed for accurate blind docking. However, it is found that while deep generative models excel in locating the pocket, they still lag behind traditional methods in terms of conformation generation. In this study, we introduce a blind docking approach named DiffSim to seamlessly integrate the diffusion model with molecular dynamics (MD) simulation. By aligning reverse diffusion sampling with MD simulation trajectories, DiffSim aim to generate ligand conformations informed by MD-modelled protein-ligand interactions. We demonstrate that the diffusion model can essentially be a coarse-grained simulator for MD simulation. Empirical results demonstrate the effectiveness of the approach and highlight the potential of combining physics-informed MD simulation with deep learning models in drug discovery.