Aligning Protein Conformation Ensemble Generation with Physical Feedback

TL;DR

This paper introduces Energy-based Alignment (EBA), a novel method that integrates physical feedback into generative models for protein conformation sampling, significantly improving the physical plausibility and quality of generated protein ensembles.

Contribution

The paper presents EBA, a new approach that effectively incorporates physical model feedback into generative protein structure models, addressing the challenge of intractable energy-based optimization.

Findings

EBA achieves state-of-the-art results on the MD ensemble benchmark.

EBA improves the physical plausibility of generated protein structures.

The method enhances model predictions for protein conformational sampling.

Abstract

Protein dynamics play a crucial role in protein biological functions and properties, and their traditional study typically relies on time-consuming molecular dynamics (MD) simulations conducted in silico. Recent advances in generative modeling, particularly denoising diffusion models, have enabled efficient accurate protein structure prediction and conformation sampling by learning distributions over crystallographic structures. However, effectively integrating physical supervision into these data-driven approaches remains challenging, as standard energy-based objectives often lead to intractable optimization. In this paper, we introduce Energy-based Alignment (EBA), a method that aligns generative models with feedback from physical models, efficiently calibrating them to appropriately balance conformational states based on their energy differences. Experimental results on the MD ensemble benchmark demonstrate that EBA achieves state-of-the-art performance in generating high-quality protein ensembles. By improving the physical plausibility of generated structures, our approach enhances model predictions and holds promise for applications in structural biology and drug discovery.