Track genomics, neuroscience and molecular biology.
Search papers, labs, and topics across Lattice.
FlexEvo is a model-agnostic, inference-time evolutionary framework that adapts biomolecular complexes designed on a single target conformation to maintain binding affinity across unseen conformational states without retraining. Current generative biomolecular design methods optimize against rigid, static structures, leading to catastrophic steric clashes and affinity loss when proteins transition between natural physiological states. By framing adaptation as geometry-constrained bi-objective optimization, FlexEvo slashes category-balanced performance degradation across alternative conformations from 47.8% to 4.4% while adding only 1.4–3.1 minutes of compute per sample.
Static protein binder designs typically lose nearly half their binding performance against alternate target conformations, but lightweight inference-time evolutionary adaptation cuts that degradation down to 4.4% in under three minutes without needing ensemble data.
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.