EUVIMEDScopri EUVIMED
Open access

Learning the rules of peptide self-assembly through data mining with large language models

Zhenze Yang iD, Sarah K Yorke iD, Tuomas P J Knowles iD, Markus J Buehler iD

Science Advances · 2025-03-28 · 11 (13)

DOI: 10.1126/sciadv.adv1971 ↗

At a glance

Journal article in Science Advances, published on 2025-03-28. Written by Zhenze Yang and 3 co-authors. Institution: Laboratory for Atomistic and Molecular Mechanics, Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, 77 Massac…. Major MEDLINE topics: Data Mining, Peptides, Machine Learning. Cited 34 times so far (Crossref); 75 references. The full text is freely available in PubMed Central. Funded by: NIAMS NIH HHS, NIBIB NIH HHS.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Peptides are ubiquitous and important biomolecules that self-assemble into diverse structures. Although extensive research has explored the effects of chemical composition and exterior conditions on self-assembly, a systematic study consolidating these data to uncover global rules is lacking. In this work, we curate a peptide assembly database through a combination of manual processing by human experts and large language model–assisted literature mining. As a result, we collect over 1000 experimental data entries with information about peptide sequence, experimental conditions, and corresponding self-assembly phases. Using the data, machine learning models are developed, demonstrating excellent accuracy (>80%) in assembly phase classification. Moreover, we fine-tune a GPT model for peptide literature mining with the developed dataset, which markedly outperforms the pretrained model in extracting information from academic publications. This workflow can improve efficiency when exploring potential self-assembling peptide candidates, through guiding experimental work, while also deepening our understanding of the governing mechanisms.

Subjects

MeSH terms (MEDLINE)

Data MiningPeptidesMachine LearningHumansLarge Language Models

Substances

Peptides

Authors and institutions

  • Laboratory for Atomistic and Molecular Mechanics, Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Room 1-165, Cambridge, MA 02139, USA.
  • Department of Materials Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, USA.
  • Yusuf Hamied Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, UK.
  • Center for Computational Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, USA.
  • Center for Materials Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139, USA.

ORCID profiles: Zhenze Yang, Sarah K Yorke, Tuomas P J Knowles, Markus J Buehler

Funding

  • NIAMS NIH HHS
  • NIBIB NIH HHS

Cite

APA

Yang, Z., Yorke, S. K., Knowles, T. P. J., & Buehler, M. J. (2025). Learning the rules of peptide self-assembly through data mining with large language models. Science Advances, 11(13). https://doi.org/10.1126/sciadv.adv1971

Vancouver

Yang Z, Yorke SK, Knowles TPJ, Buehler MJ. Learning the rules of peptide self-assembly through data mining with large language models. Science Advances. 2025;11(13). doi:10.1126/sciadv.adv1971

BibTeX
@article{Yang2025,
  author = {Yang, Zhenze and Yorke, Sarah K and Knowles, Tuomas P J and Buehler, Markus J},
  title = {Learning the rules of peptide self-assembly through data mining with large language models},
  journal = {Science Advances},
  year = {2025},
  volume = {11},
  number = {13},
  doi = {10.1126/sciadv.adv1971}
}
Start for free →