Generating Structural Ensembles of Disordered Proteins with Diffusion Models
- 1University of Toronto
Abstract
Generative models of protein structures have accelerated scientific discovery. However, the leading structure-prediction models are trained primarily on crystal structures, limiting their ability to sample the conformations of intrinsically disordered proteins (IDPs). All-atom molecular dynamics (MD) simulations can probe the diversity of IDP conformational ensembles, but these simulations are slow and costly. Leveraging the speed of generative models and the accuracy of MD, we develop Zephyr, an SE(3) equivariant diffusion model trained on simulated IDP ensembles. The model generates the structure of a single protein sequence in two phases: a coarse-grained structure is first sampled and heavy atoms are subsequently denoised onto this structure. We evaluate our method on two fronts: ensemble accuracy and sampling efficiency. Across five test systems, the model reproduces key geometric and energetic metrics of the MD ensembles, while sampling at least 63 times faster.
Related
- Enhancing simulations of disordered protein ensembles with chemical shift restraints in coarse-grained models
- Rational design of disordered proteins for sequence–function investigation
- DL3D: visual representation of conformational ensembles of domain-linker-domain proteins
- Predictive all-atom simulations of disordered proteins and biomolecular condensates through osmometry-guided force-field optimization
- Solvent-Isotope Effects in Biomolecular Phase Separation and Fibrillation of Disordered Proteins and Peptides
BibTeX
@misc{trottier2026generating,
title = {{Generating Structural Ensembles of Disordered Proteins with Diffusion Models}},
author = {O. Trottier and J. H. Gwozdecky and S. Rauscher},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.02.756250},
url = {https://www.biorxiv.org/content/10.64898/2026.10.02.756250v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Generating Structural Ensembles of Disordered Proteins with Diffusion Models
AU - O. Trottier
AU - J. H. Gwozdecky
AU - S. Rauscher
PY - 2026
DA - 2026/10/02
PB - bioRxiv
DO - 10.64898/2026.10.02.756250
UR - https://www.biorxiv.org/content/10.64898/2026.10.02.756250v1
AB - Generative models of protein structures have accelerated scientific discovery. However, the leading structure-prediction models are trained primarily on crystal structures, limiting their ability to sample the conformations of intrinsically disordered proteins (IDPs). All-atom molecular dynamics (MD) simulations can probe the diversity of IDP conformational ensembles, but these simulations are slow and costly. Leveraging the speed of generative models and the accuracy of MD, we develop Zephyr, an SE(3) equivariant diffusion model trained on simulated IDP ensembles. The model generates the structure of a single protein sequence in two phases: a coarse-grained structure is first sampled and heavy atoms are subsequently denoised onto this structure. We evaluate our method on two fronts: ensemble accuracy and sampling efficiency. Across five test systems, the model reproduces key geometric and energetic metrics of the MD ensembles, while sampling at least 63 times faster.
ER -