HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction
- 1School of Information and Intelligent Science, Donghua University, Shanghai 201620, China
- 2NHC Key Laboratory of Reproduction Regulation, Shanghai Institute for Biomedical and Pharmaceutical Technologies (SIBPT), Fudan University, Shanghai 200237, China
- 3Institute of Trustworthy Embodied AI, Fudan University, Shanghai 200433, China
Abstract
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.
Related
- PCIM-DTA: pairwise conditional interaction modeling for drug–target affinity prediction under cold-start scenarios
- Hierarchical Surrogate Modeling and Optimization of the SO2 Recovery–Water Loss Trade-Off in Basic Aluminum Sulfate Regeneration
- Hierarchical generative modeling for the design of multi-component systems
- Hierarchical universal open structure dataset for atomistic modeling
- Hierarchical crystal structure prediction of zeolitic imidazolate frameworks using DFT and machine-learned interatomic potentials
BibTeX
@article{wang2026hierl2gdta,
title = {{HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction}},
author = {Kaili Wang and Yaoxiang Yin and Yanglan Gan and Wentao Dai and Zhijun Fang},
journal = {Molecules},
year = {2026},
month = oct,
doi = {10.3390/molecules31213856},
url = {https://www.mdpi.com/1420-3049/31/21/3856},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction
AU - Kaili Wang
AU - Yaoxiang Yin
AU - Yanglan Gan
AU - Wentao Dai
AU - Zhijun Fang
PY - 2026
DA - 2026/10/26
JO - Molecules
DO - 10.3390/molecules31213856
UR - https://www.mdpi.com/1420-3049/31/21/3856
AB - 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.
SN - 1420-3049
ER -