Machine learning-based prediction of cross-immunity
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag484 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Vivien Erzsébet Resch und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicinal Chemistry, University of Szeged, Dóm tér 8, H-6720 Szeged, Hungary. Hauptthemen laut MEDLINE: Machine Learning. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Molecular Modelling and Artificial Intelligence, University of Pécs, Hungarian Academy of Sciences: János Bolyai Research Scholarship.
Abstract (Originalsprache)
Cross-immunity, defined as the ability of T-cells to recognize multiple antigen peptide-major histocompatibility complexes, is a fundamental feature of adaptive immunity. However, the prediction of different peptide epitopes that can be recognized by the same T-cell receptor remains challenging. Currently, artificial intelligent (AI)-based machine learning (ML) methods can be successfully used for pattern recognition in epitope molecular space by detecting the functional similarity between peptide sequences. In this study, using literature-based experimental data, we examined ML-based binary classification models trained on small datasets to predict the activity of nine-amino-acid-long peptides. Our results suggest that the consensus function of well-established similarity matrix-based representations and structural-based descriptors of epitopes yields better performance because representation-specific noises are reduced and individual model weaknesses are partially compensated. We also sought to determine the extent to which the predictive power of the applied AIs procedure depended on the physicochemical content of the descriptor set during the training process. In addition, challenging the models, we applied them to an independent experimental dataset to examine the effects of diverse laboratory conditions on a regulated biological measurement. In summary, applying a consensus function can capture the biological complexity of cross-reactivity at the binary classification level, even when applied to relatively small datasets.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Medicinal Chemistry, University of Szeged, Dóm tér 8, H-6720 Szeged, Hungary.
- Institute of Physics, University of Pécs, Ifjúság útja 6, H-7625 Pécs, Hungary.
- Institute of Informatics, University of Szeged, Árpád tér 2, 6720 Szeged, Hungary.
- Plasma Chemistry Research Group, HUN-REN Research Centre for Natural Sciences, Magyar Tudósok Körútja 2, H-1117 Budapest, Hungary.
- Department of Medical Microbiology, Albert Szent-Györgyi Health Center and Albert Szent-Györgyi Medical School, University of Szeged, Semmelweis Str. 6, H-6725 Szeged, Hungary.
- Department of Theoretical Physics, University of Szeged, Tisza L. krt. 84-86, H-7625 Pécs, Hungary.
ORCID-Profile: Anita Rácz, Dezső Virok, Gábor Paragi
Förderung
- Molecular Modelling and Artificial Intelligence
- University of Pécs
- Hungarian Academy of Sciences: János Bolyai Research Scholarship
Zitieren
APA
Resch, V. E., Tóth, L., Rácz, A., Virok, D., & Paragi, G. (2026). Machine learning-based prediction of cross-immunity. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag484
Vancouver
Resch VE, Tóth L, Rácz A, Virok D, Paragi G. Machine learning-based prediction of cross-immunity. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag484
BibTeX
@article{Resch2026,
author = {Resch, Vivien Erzsébet and Tóth, László and Rácz, Anita and Virok, Dezső and Paragi, Gábor},
title = {Machine learning-based prediction of cross-immunity},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag484}
}
EUVIMED