EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Exploring homology detection via k-means clustering of proteins embedded with a large language model

Thomas Minotto iD, Antoine Claessens iD, Thomas D Otto iD

Bioinformatics · 2025

DOI: 10.1093/bioinformatics/btaf472 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2025. Verfasst von Thomas Minotto und 2 weiteren Autor:innen. Beteiligte Einrichtung: LPHI, CNRS, University of Montpellier, Montpellier 34095, France. Hauptthemen laut MEDLINE: Proteins, Sequence Analysis, Protein, Computational Biology, Sequence Homology, Amino Acid. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: ExposUM Institute of the University of Montpellier, French government, managed by the Agence Nationale de la Recherche (ANR), Occitanie Region.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Inferring protein homology from sequence information is essential for understanding species evolution and enabling functional annotation transfer. Besides similarity-based methods, several machine learning approaches have been developed using various ways of representing protein data. Results Here, we represent proteins with a biologically oriented large language model and apply k-means clustering to the embedded data to extract homology relationships. Although our approach lacks the sensitivity of other tools, we obtain better precision for the detection of n:m orthologs. Furthermore, we successfully reconstruct full orthologous groups from scratch, highlighting the growing potential of using large language models in combination with clustering algorithms for the analysis of protein data. Availability and implementation Datasets are available on OrthoMCL-DB as indicated in the Methods. Source code is available on GitHub at https://github.com/ThomasGTHB/OrthoLM and Zenodo at https://doi.org/10.5281/zenodo.16640170.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsSequence Analysis, ProteinComputational BiologySequence Homology, Amino AcidCluster AnalysisAlgorithmsDatabases, ProteinMachine LearningSoftwareLarge Language Models

Substanzen

Proteins

Autor:innen und Institutionen

  • LPHI, CNRS, University of Montpellier, Montpellier 34095, France.
  • IMAG, CNRS, University of Montpellier, Montpellier 34095, France.
  • School of Infection and Immunity, University of Glasgow, Glasgow G12 8TA, United Kingdom.

ORCID-Profile: Thomas Minotto, Antoine Claessens, Thomas D Otto

Förderung

  • ExposUM Institute of the University of Montpellier
  • French government, managed by the Agence Nationale de la Recherche (ANR)
  • Occitanie Region
  • Wellcome Trust

Zitieren

APA

Minotto, T., Claessens, A., & Otto, T. D. (2025). Exploring homology detection via k-means clustering of proteins embedded with a large language model. Bioinformatics. https://doi.org/10.1093/bioinformatics/btaf472

Vancouver

Minotto T, Claessens A, Otto TD. Exploring homology detection via k-means clustering of proteins embedded with a large language model. Bioinformatics. 2025. doi:10.1093/bioinformatics/btaf472

BibTeX
@article{Minotto2025,
  author = {Minotto, Thomas and Claessens, Antoine and Otto, Thomas D},
  title = {Exploring homology detection via k-means clustering of proteins embedded with a large language model},
  journal = {Bioinformatics},
  year = {2025},
  doi = {10.1093/bioinformatics/btaf472}
}
Kostenlos starten →