Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional

Transition path sampling (TPS), which involves finding probable paths connecting two points on an energy landscape, remains a challenge due to the complexity of real-world atomistic systems. Current machine learning approaches rely on expensive training procedures and under-utilize growing quantities of atomistic data, limiting scalability and generalization. Generative models of atomistic conformational ensembles sample temporally independent states from energy landscapes, but their application to TPS remains mostly unexplored. In this work, we address TPS by interpreting candidate paths as trajectories sampled from stochastic dynamics induced by the learned score function of generative models, namely denoising diffusion and flow matching. Under these dynamics, finding high-likelihood transition paths becomes equivalent to minimizing the Onsager-Machlup (OM) action functional, enabling us to repurpose pre-trained generative models for TPS in a zero-shot fashion. We demonstrate our approach on a Müller-Brown potential and several fast-folding proteins, where we obtain diverse, physically realistic transition pathways, as well as tetrapeptides, where we demonstrate successful TPS on systems not seen by the generative model during training. Our method can be easily incorporated into new generative models, making it practically relevant as models continue to scale and improve.

Understanding how molecules transition from one stable configuration to another—like how a protein folds or a chemical reaction occurs—is a fundamental challenge in biology and chemistry with tangible applications that have the potential to improve technology and human health. These molecular transitions are rare and hard to observe directly, making them difficult to simulate even with powerful computers. Our research tackles this by reimagining how to find these molecular transitions using machine learning. Specifically, we show that existing generative AI models, originally trained to create static molecular structures, can be repurposed—without retraining—to predict the entire pathway a molecule might follow during such a transition. We do this by combining these models with principles from physics, using a tool called the Onsager-Machlup action to identify the most likely transition paths. This lets us generate physically accurate and diverse transition pathways far more efficiently than traditional simulation methods. Our method works with any generative AI model that produces molecular structures, and we’re sharing our code to encourage others to apply it as these models continue to improve. By enabling faster, more scalable simulations of rare molecular events, our work opens new avenues for drug discovery, protein design, and understanding the physical mechanisms of life.