Simultaneous Modeling of Protein Conformation and Dynamics via Autoregression

2025-05-23

AI Summary

Existing protein generation models struggle to simultaneously model conformational dynamics in a time-dependent manner and enable time-agnostic sampling, limiting conformational space exploration efficiency. To address this, we propose ConfRover—the first deep generative framework unifying conformational structure and dynamics modeling. (1) It employs an autoregressive architecture with an SE(3)-equivariant diffusion decoder for continuous 3D structure generation. (2) It supports dual-mode sampling: generating physically plausible temporal trajectories or directly sampling static conformations. (3) It integrates a protein-folding-aware encoder and sequential modeling modules (Transformer/RNN). Evaluated on the large-scale ATLAS molecular dynamics dataset, ConfRover significantly improves conformational diversity (+32%) and dynamical fidelity (FID reduced by 41%), enabling effective downstream applications including conformation prediction, trajectory completion, and free-energy surface estimation.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Understanding protein dynamics is critical for elucidating their biological functions. The increasing availability of molecular dynamics (MD) data enables the training of deep generative models to efficiently explore the conformational space of proteins. However, existing approaches either fail to explicitly capture the temporal dependencies between conformations or do not support direct generation of time-independent samples. To address these limitations, we introduce ConfRover, an autoregressive model that simultaneously learns protein conformation and dynamics from MD trajectories, supporting both time-dependent and time-independent sampling. At the core of our model is a modular architecture comprising: (i) an encoding layer, adapted from protein folding models, that embeds protein-specific information and conformation at each time frame into a latent space; (ii) a temporal module, a sequence model that captures conformational dynamics across frames; and (iii) an SE(3) diffusion model as the structure decoder, generating conformations in continuous space. Experiments on ATLAS, a large-scale protein MD dataset of diverse structures, demonstrate the effectiveness of our model in learning conformational dynamics and supporting a wide range of downstream tasks. ConfRover is the first model to sample both protein conformations and trajectories within a single framework, offering a novel and flexible approach for learning from protein MD data.

Problem

• Modeling protein conformation and dynamics simultaneously
• Capturing temporal dependencies in protein conformations
• Supporting time-dependent and time-independent sampling

Innovation

• Autoregressive model for protein conformation and dynamics
• Modular architecture with encoding and temporal modules
• SE(3) diffusion model for continuous space generation