EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TSEDTA: a transformer-based neural network with SMILES transformer and ESM2 embeddings for drug-target binding affinity prediction

Xu Sun, Xiaoying Liu, Juanjuan Huang, Jiageng Wu, Yuchen Sun, Jiwei Jia iD

Bioinformatics · 03.05.2026 · 42 (5)

DOI: 10.1093/bioinformatics/btag298 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 09.05.2026. Verfasst von Xu Sun und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computational Mathematics, School of Mathematics, Jilin University, Changchun, 130012, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Proteins, Computational Biology. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shenzhen Loop Area Institute focused project (Single-Cell Foundation Model and Applications), National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Drug-target binding affinity (DTA) prediction plays a vital role in drug repositioning. The emergence of large language models (LLMs) has introduced new perspectives for predicting DTA. Herein, we present TSEDTA, a Transformer-based neural network with SMILES Transformer and ESM2 embeddings for predicting DTA. It leverages pre-trained LLMs (SMILES Transformer and ESM2) to extract deep evolutionary representations from drug SMILES and protein sequences. The representations are directly fused with raw sequence embeddings and processed via dual Transformer encoders to capture complex local and global dependencies. Results The experiments demonstrate that TSEDTA outperforms ten advanced models on the Davis and KIBA datasets, and seven on the BindingDB dataset. Ablation studies show that incorporating LLM embeddings significantly improves the performance of TSEDTA. Furthermore, a practical case study demonstrates its real-world applicability. Ultimately, TSEDTA provides a highly accurate, robust tool for DTA prediction, offering new insights into the application of LLMs for DTA tasks. Availability The source code and data are available at: https://github.com/SunXu24Math/TSEDTA. The version of record is archived in Zenodo with the DOI: 10.5281/zenodo.19103249.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerProteinsComputational BiologyLarge Language ModelsProtein BindingPharmaceutical Preparations

Substanzen

ProteinsPharmaceutical Preparations

Autor:innen und Institutionen

  • Department of Computational Mathematics, School of Mathematics, Jilin University, Changchun, 130012, China.
  • Department of Laboratory Medicine, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, 450007, China.
  • School of Advanced Manufacturing and Robotics, Peking University, Beijing, 100871, China.
  • AI for Science and Engineering Center, Shenzhen Loop Area Institute, Shenzhen, 518048, China.

ORCID-Profile: Jiwei Jia

Förderung

  • Shenzhen Loop Area Institute focused project (Single-Cell Foundation Model and Applications)
  • National Natural Science Foundation of China

Zitieren

APA

Sun, X., Liu, X., Huang, J., Wu, J., Sun, Y., & Jia, J. (2026). TSEDTA: a transformer-based neural network with SMILES transformer and ESM2 embeddings for drug-target binding affinity prediction. Bioinformatics, 42(5). https://doi.org/10.1093/bioinformatics/btag298

Vancouver

Sun X, Liu X, Huang J, Wu J, Sun Y, Jia J. TSEDTA: a transformer-based neural network with SMILES transformer and ESM2 embeddings for drug-target binding affinity prediction. Bioinformatics. 2026;42(5). doi:10.1093/bioinformatics/btag298

BibTeX
@article{Sun2026,
  author = {Sun, Xu and Liu, Xiaoying and Huang, Juanjuan and Wu, Jiageng and Sun, Yuchen and Jia, Jiwei},
  title = {TSEDTA: a transformer-based neural network with SMILES transformer and ESM2 embeddings for drug-target binding affinity prediction},
  journal = {Bioinformatics},
  year = {2026},
  volume = {42},
  number = {5},
  doi = {10.1093/bioinformatics/btag298}
}
Kostenlos starten →