AlphaFolding: 4D Diffusion for Dynamic Protein Structure Prediction with Reference and Motion Guidance

TL;DR

AlphaFolding introduces a novel 4D diffusion model that predicts dynamic protein structures over time, integrating molecular dynamics data, reference structures, and motion guidance to capture conformational changes accurately.

Contribution

This work is the first to develop a diffusion-based model for predicting multi-step protein trajectories, incorporating reference and motion guidance for dynamic structure generation.

Findings

High accuracy in predicting dynamic 3D structures of proteins with up to 256 amino acids.

Effective modeling of local flexibility and large conformational changes.

Demonstrated ability to generate coherent protein trajectories over multiple time steps.

Abstract

Protein structure prediction is pivotal for understanding the structure-function relationship of proteins, advancing biological research, and facilitating pharmaceutical development and experimental design. While deep learning methods and the expanded availability of experimental 3D protein structures have accelerated structure prediction, the dynamic nature of protein structures has received limited attention. This study introduces an innovative 4D diffusion model incorporating molecular dynamics (MD) simulation data to learn dynamic protein structures. Our approach is distinguished by the following components: (1) a unified diffusion model capable of generating dynamic protein structures, including both the backbone and side chains, utilizing atomic grouping and side-chain dihedral angle predictions; (2) a reference network that enhances structural consistency by integrating the latent embeddings of the initial 3D protein structures; and (3) a motion alignment module aimed at improving temporal structural coherence across multiple time steps. To our knowledge, this is the first diffusion-based model aimed at predicting protein trajectories across multiple time steps simultaneously. Validation on benchmark datasets demonstrates that our model exhibits high accuracy in predicting dynamic 3D structures of proteins containing up to 256 amino acids over 32 time steps, effectively capturing both local flexibility in stable states and significant conformational changes. URL: https://fudan-generative-vision.github.io/AlphaFolding/#/