Physically Grounded Generative Modeling of All-Atom Biomolecular Dynamics

Predicting the kinetic pathways of biomolecular systems at all-atom resolution is crucial for understanding protein function and drug efficacy, yet this task is hindered by the immense computational cost of conventional molecular dynamics (MD) simulations. While deep learning has revolutionized static structure prediction and equilibrium ensemble sampling, simulating the kinetics of conformational transitions remains a critical challenge. We introduce BioKinema, a physically grounded generative model that predicts continuous-time, all-atom biomolecular trajectories at a fraction of the cost of traditional simulations. In particular, BioKinema utilizes a spatial-temporal diffusion architecture motivated by the exponential decay of correlations characteristic of Langevin dynamics, and is explicitly trained to generate trajectories with the correct kinetics and thermodynamics. It employs a hierarchical forecasting-and-interpolation strategy to overcome the error accumulation that often plagues long-horizon generation. Through extensive validation, we demonstrate that BioKinema generates physically stable and dynamically accurate trajectories suitable for rigorous downstream analysis. For protein systems, it reproduces the equilibrium thermodynamics of the conformational ensemble and the underlying kinetics. For protein-ligand complexes, it successfully elucidates mechanisms such as ligand-driven conformational changes and allosteric interactions. Furthermore, BioKinema leverages enhanced sampling data to predict rare kinetic events, emerging as a powerful tool for estimating ligand unbinding pathways. Collectively, these results establish BioKinema as a computationally efficient complement to MD simulations that bridges the gap between static structure and dynamic function, enabling high-throughput exploration of the kinetic landscape for structural biology and drug discovery.