Multi-Scale Temporal Flows for Peptide Trajectory Generation
Abstract
Peptides occupy a particularly challenging regime of biomolecular dynamics. Short peptides lack a stable folded core and populate broad conformational ensembles, while cyclization adds ring closure, non-local residue coupling, and stereochemical diversity. Deep generative models have made remarkable progress in emulating molecular-dynamics trajectories directly, yet each is trained on windows cut at a fixed interval and therefore observes the process at a fixed temporal resolution. More fundamentally, none takes that interval as an input, so the physical time a window spans is never represented, leaving the sparse frames on which a peptide crosses between metastable states difficult to learn. Here, we introduce PepTIDE, a multi-scale temporal framework for peptide trajectory generation. The stride of each training window is drawn from a continuous range, so one model observes the same dynamics at many temporal resolutions, and an Adaptive Physical-Time Embedding encodes the relative interval of a window together with the absolute time of each frame, making the shared velocity field interval-aware. All frames are generated jointly through a stochastic-interpolant flow, and pair-aware invariant point attention captures the geometry and topology of both linear and cyclic peptides. On PepMD and a 50-system cyclic-peptide benchmark, PepTIDE reaches state-of-the-art distribution agreement and structural validity. Read in temporal order, its trajectories recover these sparse transition frames and reproduce inter-state fluxes more faithfully, confirming that modeling multiple temporal scales is what makes these rare events accessible.
Related
- Sortase-mediated generation of various macrocyclic peptide scaffolds from low-reactivity electrophilic substrates
- Pep2Net: an integrated peptide knowledge base and cross-attention model for multi-task prediction of peptides
- CP-GFM-SGX: Topology-Aware Cyclic Peptide Generation and Design with Receptor-Conditioned Structural Evaluation and Experimental KRAS Affinity Optimization
- Multi-Peptide Prompting Enables In-Context Learning in Protein Language
- Coarse composition suffices: tabular in-context learning for multi-activity antimicrobial peptide profiling
BibTeX
@misc{sun2026multiscale,
title = {{Multi-Scale Temporal Flows for Peptide Trajectory Generation}},
author = {Xichen Sun and Wentao Wei and Xiaoxi Zhang and Yuedong Yang and Jiahua Rao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.01086},
eprint = {2610.01086},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.01086},
}
RIS
TY - UNPB
TI - Multi-Scale Temporal Flows for Peptide Trajectory Generation
AU - Xichen Sun
AU - Wentao Wei
AU - Xiaoxi Zhang
AU - Yuedong Yang
AU - Jiahua Rao
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.01086
UR - https://arxiv.org/abs/2610.01086
AB - Peptides occupy a particularly challenging regime of biomolecular dynamics. Short peptides lack a stable folded core and populate broad conformational ensembles, while cyclization adds ring closure, non-local residue coupling, and stereochemical diversity. Deep generative models have made remarkable progress in emulating molecular-dynamics trajectories directly, yet each is trained on windows cut at a fixed interval and therefore observes the process at a fixed temporal resolution. More fundamentally, none takes that interval as an input, so the physical time a window spans is never represented, leaving the sparse frames on which a peptide crosses between metastable states difficult to learn. Here, we introduce PepTIDE, a multi-scale temporal framework for peptide trajectory generation. The stride of each training window is drawn from a continuous range, so one model observes the same dynamics at many temporal resolutions, and an Adaptive Physical-Time Embedding encodes the relative interval of a window together with the absolute time of each frame, making the shared velocity field interval-aware. All frames are generated jointly through a stochastic-interpolant flow, and pair-aware invariant point attention captures the geometry and topology of both linear and cyclic peptides. On PepMD and a 50-system cyclic-peptide benchmark, PepTIDE reaches state-of-the-art distribution agreement and structural validity. Read in temporal order, its trajectories recover these sparse transition frames and reproduce inter-state fluxes more faithfully, confirming that modeling multiple temporal scales is what makes these rare events accessible.
ER -