Representing local protein environments with atomistic foundation models

2025-05-29

AI Summary

This study addresses the challenge of efficiently representing highly variable local protein environments—structurally and chemically heterogeneous regions where existing methods struggle to jointly encode geometric, sequential, and protonation-state information. To this end, we propose the first local protein environment encoding framework leveraging emergent representations from intermediate layers of an atomic foundation model (AFM), integrating secondary structure, residue identity, and protonation state within a physically constrained architecture for nuclear magnetic resonance (NMR) chemical shift prediction. Our method synergistically combines geometric deep learning with structural embedding space modeling, enabling the first data-driven learning of biophysical environmental priors. On the NMR chemical shift prediction task, it achieves state-of-the-art accuracy. The learned representation space exhibits strong structural semantic coherence, significantly outperforming conventional molecular simulation–based and sequence-based approaches.

Technology Category

• Machine Learning: Representation Learning
• Natural Language Processing: Learning & Optimization for NLP
• Computer Vision: Representation Learning for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

The local structure of a protein strongly impacts its function and interactions with other molecules. Therefore, a concise, informative representation of a local protein environment is essential for modeling and designing proteins and biomolecular interactions. However, these environments' extensive structural and chemical variability makes them challenging to model, and such representations remain under-explored. In this work, we propose a novel representation for a local protein environment derived from the intermediate features of atomistic foundation models (AFMs). We demonstrate that this embedding effectively captures both local structure (e.g., secondary motifs), and chemical features (e.g., amino-acid identity and protonation state). We further show that the AFM-derived representation space exhibits meaningful structure, enabling the construction of data-driven priors over the distribution of biomolecular environments. Finally, in the context of biomolecular NMR spectroscopy, we demonstrate that the proposed representations enable a first-of-its-kind physics-informed chemical shift predictor that achieves state-of-the-art accuracy. Our results demonstrate the surprising effectiveness of atomistic foundation models and their emergent representations for protein modeling beyond traditional molecular simulations. We believe this will open new lines of work in constructing effective functional representations for protein environments.

Problem

• Modeling diverse local protein structures and chemical features
• Developing concise representations for protein function and interactions
• Creating accurate physics-informed chemical shift predictors

Innovation

• Uses atomistic foundation models for protein representation
• Captures local structure and chemical features effectively
• Enables physics-informed chemical shift predictor