Structure aware learning improves protein mutation design
Structure-aware Reinforcement Learning for Protein Directed Evolution
Machine LearningArtificial Intelligence
Summary
Designing better proteins by changing their building blocks is important but tough, especially because proteins fold into complex shapes. Existing methods mostly focus on the protein's sequence but ignore its 3D shape and interactions. The authors developed a new method called StructEvo that uses a special way to guess how protein shapes change when mutated, helping it better navigate possible changes. This method outperforms older techniques and helps discover important mutation patterns in proteins.
What this means in practice
- •For biotechnology developers: Produce optimized proteins by efficiently exploring mutations with structural awareness for improved function.
- •For drug discovery teams: Advance engineering of therapeutic proteins by leveraging structural mutation predictions to identify effective variants.
Authors
Zikun Nie, Suyuan Zhao, Yizhen Luo, Siqi Fan, Zaiqing Nie
Abstract
Protein optimization remains a longstanding goal in life sciences. Existing machine learning-assisted directed evolution (MLDE) methods primarily rely on sequence-only features, overlooking the critical spatial constraints and co-evolutionary interactions encoded in protein structures. However, directly integrating structural information remains challenging due to the scarcity of reliable mutant structures. To address these issues, we propose StructEvo, a novel structure-aware reinforcement learning framework for protein directed evolution. StructEvo employs a delta-structure fusion encoder to approximate mutant structure features via feature differences, enabling dynamic incorporation of spatial knowledge. The vast mutation space is then decomposed into manageable subspaces through a structure-aligned hierarchical action network, while a geometric constraint further stabilizes delta feature learning. Our approach outperforms prior state-of-the-art methods by 9.2% and 16.3% on two challenging optimization benchmarks, and further identifies an experimentally validated epistasis pattern in GFP, highlighting the importance of structural guidance for effective protein directed evolution.