Flexibility-conditioned protein structure design with flow matching

Vsevolod Viliuga, Leif Seute, Nicolas Wolf, Simon Wagner, Arne Elofsson, Jan Stühmer, Frauke Gräter
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:61513-61533, 2025.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-viliuga25a, title = {Flexibility-conditioned protein structure design with flow matching}, author = {Viliuga, Vsevolod and Seute, Leif and Wolf, Nicolas and Wagner, Simon and Elofsson, Arne and St\"{u}hmer, Jan and Gr\"{a}ter, Frauke}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {61513--61533}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/viliuga25a/viliuga25a.pdf}, url = {https://proceedings.mlr.press/v267/viliuga25a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T Flexibility-conditioned protein structure design with flow matching %A Vsevolod Viliuga %A Leif Seute %A Nicolas Wolf %A Simon Wagner %A Arne Elofsson %A Jan Stühmer %A Frauke Gräter %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-viliuga25a %I PMLR %P 61513--61533 %U https://proceedings.mlr.press/v267/viliuga25a.html %V 267 %X 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.
APA
Viliuga, V., Seute, L., Wolf, N., Wagner, S., Elofsson, A., Stühmer, J. & Gräter, F.. (2025). Flexibility-conditioned protein structure design with flow matching. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:61513-61533 Available from https://proceedings.mlr.press/v267/viliuga25a.html.

Related Material