HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kong, Junxiao, Tang, Chupei, Wang, Di, Zhai, Jixiu, He, Yi, Tang, Moyu, Lu, Tianchi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911622525616128
author Kong, Junxiao
Tang, Chupei
Wang, Di
Zhai, Jixiu
He, Yi
Tang, Moyu
Lu, Tianchi
author_facet Kong, Junxiao
Tang, Chupei
Wang, Di
Zhai, Jixiu
He, Yi
Tang, Moyu
Lu, Tianchi
contents Accurate prediction of drug-target binding affinity accelerates drug discovery by prioritizing compounds for experimental validation. Current methods face three limitations: sequence-based approaches discard spatial geometric constraints, structure-based methods fail to exploit hydrogen bond features, and conventional loss functions neglect prediction-target correlation, a key factor for identifying high-affinity compounds in virtual screening. We developed HBGSA (Hydrogen Bond Graph with Self-Attention), a 3.06M-parameter model that encodes hydrogen bond spatial features. HBGSA uses graph neural networks to model hydrogen bond spatial topology with self-attention enhancement and Pearson correlation loss. Experimental results on PDBbind Core Set and CSAR-HiQ dataset demonstrate that HBGSA outperforms baseline methods with strong generalization capability. Ablation studies confirm the effectiveness of hydrogen bond modeling and Pearson correlation loss.
format Preprint
id arxiv_https___arxiv_org_abs_2604_23115
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction
Kong, Junxiao
Tang, Chupei
Wang, Di
Zhai, Jixiu
He, Yi
Tang, Moyu
Lu, Tianchi
Machine Learning
Accurate prediction of drug-target binding affinity accelerates drug discovery by prioritizing compounds for experimental validation. Current methods face three limitations: sequence-based approaches discard spatial geometric constraints, structure-based methods fail to exploit hydrogen bond features, and conventional loss functions neglect prediction-target correlation, a key factor for identifying high-affinity compounds in virtual screening. We developed HBGSA (Hydrogen Bond Graph with Self-Attention), a 3.06M-parameter model that encodes hydrogen bond spatial features. HBGSA uses graph neural networks to model hydrogen bond spatial topology with self-attention enhancement and Pearson correlation loss. Experimental results on PDBbind Core Set and CSAR-HiQ dataset demonstrate that HBGSA outperforms baseline methods with strong generalization capability. Ablation studies confirm the effectiveness of hydrogen bond modeling and Pearson correlation loss.
title HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction
topic Machine Learning
url https://arxiv.org/abs/2604.23115