SupraTITO: Transferable Generative Molecular Dynamics for Supramolecular Systems
SupraTITO learns transferable generative molecular dynamics for peptide self-assembly over long physical timescales.
The paper introduces SupraTITO, a transferable generative molecular-dynamics framework for supramolecular systems, demonstrated on peptide self-assembly. It learns implicit transfer operators conditioned on peptide sequence, molecular topology, and periodic geometry, propagating configurations over intervals much longer than a molecular-dynamics step. On a dipeptide benchmark it generalizes to held-out sequences, reproduces sequence-dependent structures and dynamics, and maintains molecular integrity over long rollouts. The learned dynamics also transfer across concentrations, including dilute conditions absent from training.
- Generative molecular-dynamics framework for peptide self-assembly.
- Operators are conditioned on sequence, topology, and periodic geometry.
- Generalizes to held-out dipeptide sequences and unseen concentrations.
- Reproduces assembly structure and temporal evolution better than direct ensemble prediction.
From the original article
1 min read at arxiv.orgPeptide 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.
By Weilong Chen · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.