Controllable protein design through Feynman-Kac steering
By: Erik Hartman , Jonas Wallin , Johan Malmström and more
Potential Business Impact:
Designs proteins that stick to targets better.
Diffusion-based models have recently enabled the generation of realistic and diverse protein structures, yet they remain limited in their ability to steer outcomes toward specific functional or biochemical objectives, such as binding affinity or sequence composition. Here we extend the Feynman-Kac (FK) steering framework, an inference-time control approach, to diffusion-based protein design. By coupling FK steering with structure generation, the method guides sampling toward desirable structural or energetic features while maintaining the diversity of the underlying diffusion process. To enable simultaneous generation of both sequence and structure properties, rewards are computed on models refined through ProteinMPNN and all-atom relaxation. Applied to binder design, FK steering consistently improves predicted interface energetics across diverse targets with minimal computational overhead. More broadly, this work demonstrates that inference-time FK control generalizes diffusion-based protein design to arbitrary, non-differentiable, and reward-agnostic objectives, providing a unified and model-independent framework for guided molecular generation.
Similar Papers
Seek and You Shall Fold
Machine Learning (CS)
Creates protein shapes from experimental clues.
Flexibility-Conditioned Protein Structure Design with Flow Matching
Biomolecules
Designs proteins that can bend and move.
Feynman-Kac Correctors in Diffusion: Annealing, Guidance, and Product of Experts
Machine Learning (CS)
Lets AI create better pictures and molecules.