DynaTCR: dynamic hard-negative ensemble graph learning improves TCR-epitope binding prediction
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag544 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Xiangzheng Fu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology, Shenzhen 518107, China. Hauptthemen laut MEDLINE: Receptors, Antigen, T-Cell, Computational Biology, Software, Epitopes, T-Lymphocyte. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Natural Science Foundation of China, The Natural Science Foundation of Hunan Province, the Educational Commission of Hunan Province.
Abstract (Originalsprache)
Motivation T-cell receptors (TCRs) recognize antigenic peptides presented by major histocompatibility complex (MHC) molecules and are central to adaptive immunity. Computational prediction of TCR–epitope binding (TEB) can accelerate immunotherapy development, yet remains hampered by limited labeled data, false-negative noise in unobserved pairs, and over-smoothing in graph-based models. Results We present DynaTCR, a dynamic graph ensemble learning framework for TEB prediction. DynaTCR encodes TCR and epitope sequences with protein language model embeddings and organizes them into a bipartite interaction graph. A graph regularization-variance-preserving aggregation (GR-VPA) encoder stabilizes message propagation and alleviates over-smoothing, while a global attention layer captures long-range dependencies. Multiple base learners are trained with iteratively updated hard-negative samples to reduce false-negative predictions. Under the StrictTCR evaluation protocol on four public datasets, DynaTCR achieves AUC improvements of 4.0–8.2 percentage points over the strongest existing method and up to 15.8 percentage points in AUPR. On the most stringently curated dataset, DynaTCR attains an AUC of 95.1%. Furthermore, on an independent structure-derived test set, DynaTCR achieves the highest AUC (72.6%) among all compared methods, demonstrating its robustness and effectiveness for TEB prediction and candidate prioritization. Availability Source code and data can be downloaded from: https://github.com/2014402680/TEB/.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology, Shenzhen 518107, China.
- School of Data Science and Artificial Intelligence, Wenzhou University of Technology, Wenzhou 325000, China.
- College of Computer and Mathematics, Central South University of Forestry and Technology, Changsha, Hunan 410004, China.
- School of Information Engineering, Changsha Medical University, Changsha 410219, China.
- Xiangya School of Pharmaceutical Sciences, Central South University, Changsha 410003, China.
- Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 610054, China.
ORCID-Profile: Quan Zou
Förderung
- the National Natural Science Foundation of China
- The Natural Science Foundation of Hunan Province
- the Educational Commission of Hunan Province
Zitieren
APA
Fu, X., Zhang, X., Zhuo, L., Chen, Y., Cao, D., & Zou, Q. (2026). DynaTCR: dynamic hard-negative ensemble graph learning improves TCR-epitope binding prediction. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag544
Vancouver
Fu X, Zhang X, Zhuo L, Chen Y, Cao D, Zou Q. DynaTCR: dynamic hard-negative ensemble graph learning improves TCR-epitope binding prediction. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag544
BibTeX
@article{Fu2026,
author = {Fu, Xiangzheng and Zhang, Xinyu and Zhuo, Linlin and Chen, Yifan and Cao, Dongsheng and Zou, Quan},
title = {DynaTCR: dynamic hard-negative ensemble graph learning improves TCR-epitope binding prediction},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag544}
}
EUVIMED