MolHFCNet: Enhancing Molecular Graph Representations with Hierarchical Feature Combining and Hybrid Pretraining

Nguyen, Duy-Long; Ho-Viet, Duc-Luong; Ngo-Tran, Anh-Thu; Nguyen, Quang H.; Nguyen, Binh P.

doi:10.24963/ijcai.2025/1035

MolHFCNet: Enhancing Molecular Graph Representations with Hierarchical Feature Combining and Hybrid Pretraining

Duy-Long Nguyen, Duc-Luong Ho-Viet, Anh-Thu Ngo-Tran, Quang H. Nguyen, Binh P. Nguyen

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. AI4Tech: AI Enabling Technologies. Pages 9313-9321.

https://doi.org/10.24963/ijcai.2025/1035

Keywords

Advanced AI4Tech: Deep AI4Tech
Domain-specific AI4Tech: AI4Medicine and AI4Drug
Domain-specific AI4Tech: AI4Biotech

Abstract

Efficient molecular property prediction is crucial in bioinformatics and cheminformatics, with applications in drug discovery, materials science, and chemical engineering. This paper introduces MolHFCNet, a graph neural network designed to enhance molecular representation learning. At its core, the n-Hierarchical Features Combining (n-HFC) module aggregates information across multiple hierarchical feature spaces, effectively capturing both local and global graph structures. Unlike conventional models, n-HFC maintains computational complexity comparable to a single full-dimensional graph layer while supporting either 2D or 3D molecular graphs, ensuring flexibility across tasks. Furthermore, we propose a novel graph pretraining strategy that integrates predictive and contrastive learning, enabling the model to capture local chemical interactions and global molecular contexts for robust embeddings. Experimental results on benchmark datasets demonstrate MolHFCNet’s superior accuracy and efficiency compared to state-of-the-art methods, highlighting the potential of high-order hierarchical feature learning for advancing molecular graph analysis. Our code is available at https://github.com/ndlongvn/MolHFCNet.

Duy-Long Nguyen, Duc-Luong Ho-Viet, Anh-Thu Ngo-Tran, Quang H. Nguyen, and Binh P. Nguyen. 2025. MolHFCNet: Enhancing Molecular Graph Representations with Hierarchical Feature Combining and Hybrid Pretraining. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 9313–9321. https://doi.org/10.24963/ijcai.2025/1035

@inproceedings{ijcai2025p1035,
  title     = {MolHFCNet: Enhancing Molecular Graph Representations with Hierarchical Feature Combining and Hybrid Pretraining},
  author    = {Nguyen, Duy-Long and Ho-Viet, Duc-Luong and Ngo-Tran, Anh-Thu and Nguyen, Quang H. and Nguyen, Binh P.},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {9313--9321},
  year      = {2025},
  month     = {8},
  note      = {AI4Tech: AI Enabling Technologies},
  doi       = {10.24963/ijcai.2025/1035},
  url       = {https://doi.org/10.24963/ijcai.2025/1035},
}

TY  - CPAPER
TI  - MolHFCNet: Enhancing Molecular Graph Representations with Hierarchical Feature Combining and Hybrid Pretraining
AU  - Nguyen, Duy-Long
AU  - Ho-Viet, Duc-Luong
AU  - Ngo-Tran, Anh-Thu
AU  - Nguyen, Quang H.
AU  - Nguyen, Binh P.
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 9313
EP  - 9321
DO  - 10.24963/ijcai.2025/1035
UR  - https://doi.org/10.24963/ijcai.2025/1035
ER  -