HierDTA: Hierarchical Structure-Aware Representation Learning for Drug–Target Affinity Prediction

Abstract

Drug–target affinity (DTA) prediction plays an important role in computer-aided drug discovery, where reliable computational models can help prioritize compound–protein pairs before experimental validation. A key challenge in DTA prediction is to effectively mine multilevel molecular information from drug structures while accurately modeling the binding-site environment of target proteins. To address these issues, we introduce HierDTA, a hierarchical structure-aware framework for DTA prediction. HierDTA represents each drug as an atom–motif heterogeneous graph and introduces an augmentation module that propagates functional-motif semantics to the corresponding atoms. For proteins, kinase domain structures are converted into residue interaction graphs. The resulting atom and residue representations are integrated using cross-attention to capture pair-specific relationships, while molecule fingerprints and pretrained ESM-2 embeddings provide complementary global information. Compared with state-of-the-art methods, HierDTA achieves superior performance on the Davis data set across all metrics and attains the highestrm2of 0.791 on KIBA. It also exhibits strong generalization under cold-start settings, particularly on KIBA, where MSE is reduced by 21.7% in the unseen-protein scenario. Ablation study and interpretability analysis supported the contributions of the hierarchical augmentation and interaction components. Case studies further showed that HierDTA provided early enrichment of candidate bioactive compounds. These results indicate that HierDTA is an effective framework for affinity prediction and target-specific virtual screening.

HierL2GDTA: Hierarchical Interaction Modeling with Local and Global Features for Drug–Target Affinity Prediction
Kaili Wang, Yaoxiang Yin, Yanglan Gan +2
Molecules26 Oct 2026Peer-reviewedCC BY 4.0General chemistryCrossref
Accelerating Proton Affinity Prediction with Multi-Fidelity Machine Learning
Debjyoti Bhattacharya, Yifan Liu, Valentino R. Cooper +1
ChemRxiv10 Sep 2026PreprintLicence not statedGeneral chemistryChemRxiv
FLOWR.ROOT – A flow matching-based foundation model for joint multi-purpose structure-aware 3D ligand generation and affinity prediction
Julian Cremer, Tuan Le, Mohammad Mehdi Ghahremanpour +3
Nature Communications6 Jul 2026Peer-reviewedCC BYFull text9 citedMedicinal chemistry & drug discoveryOpenAlex
Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction
Shunzhou Wan, Xibei Zhang, Xue Xiao +1
Journal of Chemical Theory and Computation29 Jul 2026Peer-reviewedCC BYFull text20 citedMedicinal chemistry & drug discoveryOpenAlex
Beyond the Ligand Applicability Domain: Generalization and Local Resolution in Structure-Aware Foundation Models for Affinity Prediction
Pieter B. Burger, Thomas M. Kaiser, Parker Dryja +7
ChemRxiv7 Oct 2026PreprintCC BYFull textGeneral chemistryChemRxiv

Cite

BibTeX RIS

BibTeX

@article{xu2026hierdta,
  title = {{HierDTA: Hierarchical Structure-Aware Representation Learning for Drug–Target Affinity Prediction}},
  author = {Huangchao Xu and Baohua Zhang and Qian Liu and Zhong Jin and Yi Du},
  journal = {ACS Omega},
  year = {2026},
  month = oct,
  doi = {10.1021/acsomega.6c07956},
  url = {https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c07956/5454289/HierDTA-Hierarchical-Structure-Aware},
  note = {CC BY-NC-ND 4.0},
}

RIS

TY  - JOUR
TI  - HierDTA: Hierarchical Structure-Aware Representation Learning for Drug–Target Affinity Prediction
AU  - Huangchao Xu
AU  - Baohua Zhang
AU  - Qian Liu
AU  - Zhong Jin
AU  - Yi Du
PY  - 2026
DA  - 2026/10/20
JO  - ACS Omega
DO  - 10.1021/acsomega.6c07956
UR  - https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c07956/5454289/HierDTA-Hierarchical-Structure-Aware
AB  - Drug–target affinity (DTA) prediction plays an important role in computer-aided drug discovery, where reliable computational models can help prioritize compound–protein pairs before experimental validation. A key challenge in DTA prediction is to effectively mine multilevel molecular information from drug structures while accurately modeling the binding-site environment of target proteins. To address these issues, we introduce HierDTA, a hierarchical structure-aware framework for DTA prediction. HierDTA represents each drug as an atom–motif heterogeneous graph and introduces an augmentation module that propagates functional-motif semantics to the corresponding atoms. For proteins, kinase domain structures are converted into residue interaction graphs. The resulting atom and residue representations are integrated using cross-attention to capture pair-specific relationships, while molecule fingerprints and pretrained ESM-2 embeddings provide complementary global information. Compared with state-of-the-art methods, HierDTA achieves superior performance on the Davis data set across all metrics and attains the highestrm2of 0.791 on KIBA. It also exhibits strong generalization under cold-start settings, particularly on KIBA, where MSE is reduced by 21.7% in the unseen-protein scenario. Ablation study and interpretability analysis supported the contributions of the hierarchical augmentation and interaction components. Case studies further showed that HierDTA provided early enrichment of candidate bioactive compounds. These results indicate that HierDTA is an effective framework for affinity prediction and target-specific virtual screening.
SN  - 2470-1343
ER  -

Abstract

Related

Cite