An Integrated Coarse-Grained Simulation and Graph-Transformer Framework for the Discovery of Self-Assembling Cyclic Peptides

Abstract Self-assembling cyclic peptides are promising building blocks for functional biomaterials owing to their enhanced structural stability, proteolytic resistance, and diverse supramolecular architectures. However, how cyclization regulates peptide self-assembly remains poorly understood, and computational frameworks specifically developed for cyclic peptides are still lacking. Here, we demonstrate that disulfide-mediated cyclization fundamentally reshapes peptide self-assembly by significantly enhancing aggregation propensity (AP) and promoting the formation of diverse nanostructures compared with their linear counterparts, as confirmed by coarse-grained molecular dynamics (CGMD) simulations and experimental characterization. Motivated by these findings, we established an integrated computational-experimental framework for the prediction and rational design of self-assembling disulfide-cyclized peptides. A data set comprising 8,000 cyclic pentapeptides was generated using CGMD simulations, and multiple machine learning and deep learning models were systematically evaluated, among which a Graph-Transformer achieved the best predictive performance (R2 = 0.971). The optimized model was subsequently applied to high-throughput screening of 160,000 cyclic hexapeptides using a ranking strategy integrating AP and hydrophobicity, leading to the identification of multiple previously unreported self-assembling candidates. Experimental validation showed excellent agreement with model predictions. Furthermore, systematic sequence analysis revealed residue- and position-dependent assembly rules unique to disulfide-cyclized peptide systems. This work establishes a computational framework for the rapid discovery of self-assembling disulfide-cyclized peptides and provides new mechanistic insights and design principles for peptide-based functional materials.

Authors

Institutions

Publication Details

Journal
Journal of Chemical Information and Modeling
Published
2026-11-17
DOI
https://doi.org/10.1021/acs.jcim.6c03241
Primary Topic
Supramolecular Self-Assembly in Materials
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

An Integrated Coarse-Grained Simulation and Graph-Transformer Framework for the Discovery of Self-Assembling Cyclic Peptides

Jia Jin, Shuyao Zhang, Fei Ye, Kui Zhang et al.
3 links
Journal of Chemical Information and Modeling
Supramolecular Self-Assembly in Materials
article

An Integrated Coarse-Grained Simulation and Graph-Transformer Framework for the Discovery of Self-Assembling Cyclic Peptides

Jia Jin, Shuyao Zhang, Fei Ye, Kui Zhang, Siqi Guo, Runqiu Ma, Xinyu Fan, Kelin Li, Hao Liu
article en

Abstract

Abstract Self-assembling cyclic peptides are promising building blocks for functional biomaterials owing to their enhanced structural stability, proteolytic resistance, and diverse supramolecular architectures. However, how cyclization regulates peptide self-assembly remains poorly understood, and computational frameworks specifically developed for cyclic peptides are still lacking. Here, we demonstrate that disulfide-mediated cyclization fundamentally reshapes peptide self-assembly by significantly enhancing aggregation propensity (AP) and promoting the formation of diverse nanostructures compared with their linear counterparts, as confirmed by coarse-grained molecular dynamics (CGMD) simulations and experimental characterization. Motivated by these findings, we established an integrated computational-experimental framework for the prediction and rational design of self-assembling disulfide-cyclized peptides. A data set comprising 8,000 cyclic pentapeptides was generated using CGMD simulations, and multiple machine learning and deep learning models were systematically evaluated, among which a Graph-Transformer achieved the best predictive performance (R2 = 0.971). The optimized model was subsequently applied to high-throughput screening of 160,000 cyclic hexapeptides using a ranking strategy integrating AP and hydrophobicity, leading to the identification of multiple previously unreported self-assembling candidates. Experimental validation showed excellent agreement with model predictions. Furthermore, systematic sequence analysis revealed residue- and position-dependent assembly rules unique to disulfide-cyclized peptide systems. This work establishes a computational framework for the rapid discovery of self-assembling disulfide-cyclized peptides and provides new mechanistic insights and design principles for peptide-based functional materials.

Journal of Chemical Information and Modeling
Zhejiang Sci-Tech University (CN), Nanjing Forestry University (CN)
National Natural Science Foundation of China, Zhejiang Sci-Tech University, Natural Science Foundation of Zhejiang Province, Key Research and Development Program of Zhejiang Province
Supramolecular Self-Assembly in Materials
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon