Flexibility-Conditioned Protein Structure Design with Flow Matching
This work addresses the problem of designing proteins with specific dynamic properties for applications in catalysis and molecular recognition, representing an incremental step beyond static property generation.
The authors tackled the limitation of generating proteins with only static properties by proposing a framework to condition structure generation on flexibility, crucial for functionalities like catalysis, and demonstrated that their model FliPS generates novel and diverse protein backbones with desired flexibility, verified by Molecular Dynamics simulations.
Recent advances in geometric deep learning and generative modeling have enabled the design of novel proteins with a wide range of desired properties. However, current state-of-the-art approaches are typically restricted to generating proteins with only static target properties, such as motifs and symmetries. In this work, we take a step towards overcoming this limitation by proposing a framework to condition structure generation on flexibility, which is crucial for key functionalities such as catalysis or molecular recognition. We first introduce BackFlip, an equivariant neural network for predicting per-residue flexibility from an input backbone structure. Relying on BackFlip, we propose FliPS, an SE(3)-equivariant conditional flow matching model that solves the inverse problem, that is, generating backbones that display a target flexibility profile. In our experiments, we show that FliPS is able to generate novel and diverse protein backbones with the desired flexibility, verified by Molecular Dynamics (MD) simulations. FliPS and BackFlip are available at https://github.com/graeter-group/flips .