Protein design improves stability across shape changes
Robust Biomolecular Complex Design Across Protein Conformational Landscapes
Artificial Intelligence
Summary
Proteins can change their shapes, but designing molecules to fit them usually focuses on just one shape. This causes problems when the protein shifts and the designed molecule no longer fits well. The authors created FlexEvo, a way to tweak protein designs quickly so they work well across different shapes without needing to redesign from scratch. It balances keeping the original good features and avoiding clashes when the protein moves. Testing shows this method greatly reduces loss of performance while adding only a little bit of extra design time.
What this means in practice
- •For protein engineers: Improve designed protein binders to remain effective despite changes in target protein shapes without retraining models.
- •For pharmaceutical development teams: Enhance robustness of therapeutic molecule design for flexible protein targets, reducing failure due to protein structural changes.
Authors
Qingyuan Zeng, Zongqi Xu, Anglin Liu, Ziqi Gong, Pengxiang Cai, Zixin Guan, Yunan Chen, Sen Gao, Min Zhou, Jintai Chen
Abstract
Proteins populate conformational ensembles, yet structure-based biomolecular design typically optimizes candidates against a single target conformation. Consequently, a candidate that fits one state can lose favorable interactions or develop steric clashes when the target adopts another. We introduce FlexEvo, a model-agnostic evolutionary framework that adapts candidates once at inference time from a single target conformation to improve compatibility with alternative natural conformations unseen during adaptation, without retraining the source model or requiring a conformational ensemble. FlexEvo casts cross-state adaptation as geometry-constrained bi-objective optimization, balancing preservation of input-state interactions against robustness to plausible conformational perturbations. To limit the search space and reduce invalid structural edits, geometry-derived FlexBoxes define protected anchor regions, adaptable regions for local exploration, and forbidden regions for clash avoidance. A unified all-atom representation supports topology-preserving adaptation across diverse binder categories, while Pareto selection preserves nondominated candidates across the two objectives. We evaluate FlexEvo across multiple generation baselines and nine representative binder categories spanning diverse molecular sizes and structural topologies. FlexEvo reduces the category-balanced mean relative performance degradation from 47.8% to 4.4%, while adding only 1.4--3.1 minutes of adaptation per sample. These results establish single-state inference-time adaptation as a practical route toward robust biomolecular complex design across protein conformational landscapes.