EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

UAMRL: multi-granularity uncertainty-aware multimodal representation learning for drug-target affinity prediction

Wenzhe Xu iD, Xiaorong Liu, Jie Wang, Fan Zhang, Dongfeng Hu, Liansong Zong

Bioinformatics · 02.10.2025 · 41 (10)

DOI: 10.1093/bioinformatics/btaf512 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 30.09.2025. Verfasst von Wenzhe Xu und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer and Software Engineering, Xihua University, Chengdu 610039, China. Hauptthemen laut MEDLINE: Drug Discovery, Machine Learning, Deep Learning, Computational Biology. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: China Postdoctoral Science Foundation, the Postdoctoral Fellowship Program of CPSF, Fundamental Research Funds for the Central Universities.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Computational prediction of drug–target affinity (DTA) plays a critical role in modern drug discovery. However, the limited interpretability of traditional deep learning models and the heterogeneity of multimodal data from compounds and proteins hinder their reliability in practical drug development applications. Results We propose a novel Uncertainty-aware Multimodal Representation Learning (UAMRL) framework to address these challenges. UAMRL employs a dual-stream encoder to learn cross-modal association mappings between drugs and targets in a latent space and integrates heterogeneous information from different modalities. Moreover, an uncertainty quantification mechanism based on the Normal-Inverse-Gamma distribution is introduced to model the reliability of heterogeneous information and suppress less trustworthy contributions during fusion. Experiments show that UAMRL achieves superior predictive accuracy on multiple public DTA datasets, improving both prediction performance and decision transparency. Availability and implementation The source code is available at https://github.com/Astraea2xu/UAMRL.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DiscoveryMachine LearningDeep LearningComputational BiologyUncertaintyHumansProteins

Substanzen

Proteins

Autor:innen und Institutionen

  • School of Computer and Software Engineering, Xihua University, Chengdu 610039, China.
  • School of Computer Science and Software Engineering, Southwest Petroleum University, Chengdu 610500, China.
  • School of Computing and Artificial Intelligence, Southwest Jiaotong University, Chengdu 611756, China.
  • School of Design, Southwest Jiaotong University, Chengdu 611756, China.

ORCID-Profile: Wenzhe Xu

Förderung

  • China Postdoctoral Science Foundation
  • the Postdoctoral Fellowship Program of CPSF
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Xu, W., Liu, X., Wang, J., Zhang, F., Hu, D., & Zong, L. (2025). UAMRL: multi-granularity uncertainty-aware multimodal representation learning for drug-target affinity prediction. Bioinformatics, 41(10). https://doi.org/10.1093/bioinformatics/btaf512

Vancouver

Xu W, Liu X, Wang J, Zhang F, Hu D, Zong L. UAMRL: multi-granularity uncertainty-aware multimodal representation learning for drug-target affinity prediction. Bioinformatics. 2025;41(10). doi:10.1093/bioinformatics/btaf512

BibTeX
@article{Xu2025,
  author = {Xu, Wenzhe and Liu, Xiaorong and Wang, Jie and Zhang, Fan and Hu, Dongfeng and Zong, Liansong},
  title = {UAMRL: multi-granularity uncertainty-aware multimodal representation learning for drug-target affinity prediction},
  journal = {Bioinformatics},
  year = {2025},
  volume = {41},
  number = {10},
  doi = {10.1093/bioinformatics/btaf512}
}
Kostenlos starten →