Generative model predicts peptide assembly structures and dynamics
SupraTITO: Transferable Generative Molecular Dynamics for Supramolecular Systems
Machine Learning
Summary
Peptides form complex structures by sticking together, but predicting how and when this happens is very hard because it involves many molecules interacting slowly. The authors developed SupraTITO, a smart computer model that learns from detailed simulations and can predict how peptides self-assemble across different sequences and concentrations. This model can run faster than traditional methods while keeping predictions accurate over long times. SupraTITO helps scientists explore peptide behaviors that were previously too slow or costly to simulate in detail.
What this means in practice
- •For pharmaceutical formulation teams: Predict peptide-based drug formulations by simulating their assembly without costly full molecular dynamics.
- •For materials engineers: Design peptide materials by modeling assembly processes efficiently across different sequences and concentrations.
Authors
Weilong Chen, Nuno Costa, Julija Zavadlav
Abstract
Peptide sequence governs both the structures formed through supramolecular assembly and the dynamics by which they emerge, but predicting either requires resolving slow collective processes among many interacting molecules. Molecular dynamics (MD) provides microscopic insight into these processes, yet the long timescales of assembly and the vast peptide sequence space make systematic exploration computationally demanding. We introduce SupraTITO, a transferable generative molecular dynamics (GenMD) framework for supramolecular systems, demonstrated through peptide self-assembly. SupraTITO learns transferable implicit transfer operators (TITO) conditioned on peptide sequence, molecular topology, and periodic geometry, allowing configurations to be propagated over physical intervals much longer than an MD integration step. On a comprehensive dipeptide benchmark, SupraTITO generalizes to held-out sequences and reproduces sequence-dependent structures and dynamics while maintaining molecular integrity over long rollouts. Compared with direct ensemble prediction trained on the same trajectory data, SupraTITO more accurately reproduces assembly structures while also resolving their temporal evolution. The learned dynamics generalize across peptide concentrations, including dilute conditions not represented during training. These results extend transferable GenMD to collective dynamics in periodic supramolecular systems and provide a foundation for modeling related processes beyond peptide assembly.