EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction

Kaili Wang iD, Yaoxiang Yin, Yanglan Gan, Wentao Dai, Zhijun Fang

Molecules · 2026

DOI: 10.3390/molecules31213856 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Molecules, 2026. Verfasst von Kaili Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Information and Intelligent Science, Donghua University, Shanghai 201620, China. Hauptthemen laut MEDLINE: Drug Discovery, Proteins. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Shanghai Sailing Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of drug–target binding affinity is essential for drug discovery. However, existing methods face limitations in capturing local and global molecular features and in effectively integrating protein and ligand representations for interaction modeling. Here, we present HierL2GDTA, a sequence-based hierarchical framework for predicting binding affinity. Pretrained protein and molecular representations are processed by multi-scale convolutions to extract local features, followed by a bidirectional gated recurrent unit and dilated convolutions to capture broader protein and ligand context, respectively. A modified Transformer encoder integrates global protein context and ligand features through sequential cross-attention to model intramolecular and intermolecular dependencies. HierL2GDTA outperformed the evaluated baseline methods on both BindingDB and Davis. Further evaluations demonstrated the effectiveness of the model components and favorable predictive performance across selected kinase families. Attention visualization highlighted molecular substructures associated with protein binding, while retrospective screening ranked known inhibitors of disease-associated targets among the leading candidates. These results indicate that HierL2GDTA is an effective deep learning method for binding affinity prediction, with potential application for target-based drug screening.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DiscoveryProteinsLigandsProtein BindingDeep LearningModels, Molecular

Stichwörter der Autor:innen

deep learningdrug discoverydrug–target affinity predictionintramolecular and intermolecular dependenciesmodified transformer encodersequence-based hierarchical frameworktarget-based drug screening

Substanzen

LigandsProteins

Autor:innen und Institutionen

  • School of Information and Intelligent Science, Donghua University, Shanghai 201620, China.
  • NHC Key Laboratory of Reproduction Regulation, Shanghai Institute for Biomedical and Pharmaceutical Technologies (SIBPT), Fudan University, Shanghai 200237, China.
  • Institute of Trustworthy Embodied AI, Fudan University, Shanghai 200433, China.

ORCID-Profile: Kaili Wang

Förderung

  • National Natural Science Foundation of China
  • Shanghai Sailing Program

Zitieren

APA

Wang, K., Yin, Y., Gan, Y., Dai, W., & Fang, Z. (2026). HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction. Molecules. https://doi.org/10.3390/molecules31213856

Vancouver

Wang K, Yin Y, Gan Y, Dai W, Fang Z. HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction. Molecules. 2026. doi:10.3390/molecules31213856

BibTeX
@article{Wang2026,
  author = {Wang, Kaili and Yin, Yaoxiang and Gan, Yanglan and Dai, Wentao and Fang, Zhijun},
  title = {HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction},
  journal = {Molecules},
  year = {2026},
  doi = {10.3390/molecules31213856}
}
Kostenlos starten →