Multi-state Protein Design with DynamicMPNN

TL;DR

DynamicMPNN is a novel inverse folding model designed to generate protein sequences compatible with multiple conformations, significantly improving multi-state protein design accuracy over existing methods.

Contribution

It introduces a joint learning approach for multi-state protein design, outperforming previous models in accuracy and sequence recovery.

Findings

Outperforms ProteinMPNN by up to 25% on decoy-normalized RMSD.

Achieves 12% higher sequence recovery on multi-state benchmark.

Trained on 46,033 conformational pairs covering 75% of CATH superfamilies.

Abstract

Structural biology has long been dominated by the one sequence, one structure, one function paradigm, yet many critical biological processes - from enzyme catalysis to membrane transport - depend on proteins that adopt multiple conformational states. Existing multi-state design approaches rely on post-hoc aggregation of single-state predictions, achieving poor experimental success rates compared to single-state design. We introduce DynamicMPNN, an inverse folding model explicitly trained to generate sequences compatible with multiple conformations through joint learning across conformational ensembles. Trained on 46,033 conformational pairs covering 75% of CATH superfamilies and evaluated using Alphafold 3, DynamicMPNN outperforms ProteinMPNN by up to 25% on decoy-normalized RMSD and by 12% on sequence recovery across our challenging multi-state protein benchmark.