Abstract
Discovering molecules with desirable molecular properties, including ADMET profiles, is of great importance in drug discovery. Existing approaches typically employ deep learning models, such as Graph Neural Networks and Transformers, to predict these molecular properties by learning from diverse chemical information. However, these models often lack mechanisms for effective interaction among multi-level features. To address these limitations, we propose a Hierarchical Interaction Message Passing Mechanism, which serves as the foundation of our model, the Hierarchical Interaction Message Net (HimNet). Our method enables interaction-aware representation learning across atomic, motif, and molecular levels via hierarchical attention-guided message passing. This design allows HimNet to effectively balance global and local information, ensuring rich and task-relevant feature extraction for downstream property prediction tasks. We systematically evaluate HimNet on eleven datasets, including eight widely-used MoleculeNet benchmarks and three challenging, high-value datasets for metabolic stability, malaria activity, and liver microsomal clearance, covering a broad range of pharmacologically relevant properties. Extensive experiments demonstrate that HimNet achieves the best or near-best performance in most molecular property prediction tasks. We believe that HimNet offers an accurate and efficient solution for molecular activity and ADMET property prediction, contributing significantly to advanced decision-making in the early stages of drug discovery.
Data availability
The datasets used in this study are publicly available and can be accessed at https://github.com/Hugh415/HimNet. All data generated or analysed during this study are included in this published article and its Supplementary Data, and please see Supplementary Data 1.
Code availability
The custom code for the Hierarchical Interaction Message Net (HimNet) is deposited in the Zenodo repository at https://doi.org/10.5281/zenodo.18030100. The code is also available on GitHub at https://github.com/Hugh415/HimNet under an MIT license.
References
DiMasi, J. A., Grabowski, H. G. & Hansen, R. W. Innovation in the pharmaceutical industry: new estimates of r&d costs. J. Health Econ. 47, 20–37 (2016).
Ferreira, L. L. G. & Andricopulo, A. D. Admet modeling approaches in drug discovery. Drug Discov. Today 24, 1157–1165 (2019).
Wu, Z. et al. Moleculenet: a benchmark for molecular machine learning. Chem. Sci. 9, 513–530 (2018).
Xu, K., Hu, W., Leskovec, J. & Jegelka, S. How powerful are graph neural networks? In Proc. International Conference on Learning Representations (ICLR, 2019).
Velickovic, P. et al. Graph attention networks. In Proc. International Conference on Learning Representations (ICLR, 2018).
Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O. & Dahl, G. E. Neural message passing for quantum chemistry. In Proc. 34th International Conference on Machine Learning, Vol 70, 1263–1272 (PMLR, 2017).
Yang, K. et al. Analyzing learned molecular representations for property prediction. J. Chem. Inf. Model. 59, 3370–3388 (2019).
Song, L., Zhan, X. & Peng, R. Contextual message passing neural network for molecular property prediction. Bioinformatics 37, 4753–4759 (2021).
Stienstra, C. M. K. et al. Graphormer-ir: Graph transformers predict experimental IR spectra using highly specialized attention. J. Chem. Inf. Model. 64, 4613–4629 (2024).
Liu, H. et al. Advancing admet prediction through multiscale fragment-aware pretraining with msformer-admet. Brief. Bioinform. 26, bbaf506 (2025).
Tan, Z., Zhao, Y., Zhou, T. & Lin, K. Hi-mgt: a hybrid molecule graph transformer for toxicity identification. J. Hazard. Mater. 457, 131808 (2023).
Li, M. et al. Vidta: enhanced drug-target affinity prediction via virtual graph nodes and attention-based feature fusion. In Proc. IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 42–47 (IEEE, 2024).
Zang, X., Zhao, X. & Tang, B. Hierarchical molecular graph self-supervised learning for property prediction. Commun. Chem. 6, 34 (2023).
Liu, S., Chen, M., Yao, X. & Liu, H. Fingerprint-enhanced hierarchical molecular graph neural networks for property prediction. J. Pharm. Anal. 15, 101242 (2025).
Costa, B. D. P. et al. Hydrazone-functionalized trans-a2b-corroles: effective synergy in photodynamic therapy of lung cancer. J. Med. Chem. 67, 21934–21951 (2024).
Dawood, K. M. & Gomha, S. M. Synthesis and anti-cancer activity of 1, 3, 4-thiadiazole and 1, 3-thiazole derivatives having 1, 3, 4-oxadiazole moiety. J. Heterocycl. Chem. 52, 1400–1405 (2015).
Degen, J., Wegscheid-Gerlach, C., Zaliani, A. & Rarey, M. On the art of compiling and using ‘drug-like’ chemical fragment spaces. ChemMedChem 3, 1503–1507 (2008).
Rogers, D. & Hahn, M. Extended-connectivity fingerprints. J. Chem. Inf. Model. 50, 742–754 (2010).
Vaswani, A. et al. Attention is all you need. in Advances in Neural Information Processing Systems, Vol 30, 5998–6008 (NIPS, 2017).
Xiong, Z. et al. Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism. J. Med. Chem. 63, 8749–8760 (2019).
Jeffrey, G. A. An introduction to hydrogen bonding. https://doi.org/10.1021/ed076p759.1 (1997).
Hunter, C. A. & Sanders, J. K. M. The nature of π–π interactions. J. Am. Chem. Soc. 112, 5525–5534 (1990).
Chandler, D. Interfaces and the driving force of hydrophobic assembly. Nature 437, 640–647 (2005).
Faulon, J.-L., Churchwell, C. J. & Visco, D. P. The signature molecular descriptor. 1. extending to generic chemical reactions. J. Chem. Inf. Comput. Sci. 43, 707–720 (2003).
Podlewska, S. & Kafel, R. Metstabon—online platform for metabolic stability predictions. Int. J. Mol. Sci. 19, 1040 (2018).
Gamo, F. J. et al. Thousands of chemical starting points for antimalarial lead identification. Nature 465, 305–310 (2010).
Houston, J. B. & Carlile, D. J. Prediction of hepatic clearance from microsomes, hepatocytes, and liver slices. Drug Metab. Rev. 29, 891–922 (1997).
Weininger, D. Smiles, a chemical language and information system. 1. Introduction to methodology and encoding rules. J. Chem. Inf. Comput. Sci. 28, 31–36 (1988).
Carhart, R. E., Smith, D. H. & Venkataraghavan, R. Atom pairs as molecular features in structure-activity studies: definition and application. J. Chem. Inf. Comput. Sci. 25, 64–73 (1985).
Durant, J. L., Leland, B. A., Henry, D. R. & Nourse, J. G. Reoptimization of mdl keys for use in drug discovery. J. Chem. Inf. Comput. Sci. 42, 1273–1280 (2002).
Campbell, J. W. & Vogiatzis, K. D. Augmenting maccs keys with persistent homology fingerprints for protein–ligand binding classification. J. Chem. Inf. Model. 65, 8113–8126 (2025).
Wood, D. J., de Vlieg, J., Wagener, M. & Ritschel, T. Pharmacophore fingerprint-based approach to binding site subpocket similarity and its application to bioisostere replacement. J. Chem. Inf. Model. 52, 2031–2043 (2012).
Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. Int. Conf. Mach. Learn. 119, 1597–1607 (2020).
Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proc. International Conference on Learning Representations (ICLR, 2015).
Wang, Y., Wang, J., Cao, Z. & Barati Farimani, A. Molecular contrastive learning of representations via graph neural networks. Nat. Mach. Intell. 4, 279–287 (2022).
Han, S. et al. Himgnn: a novel hierarchical molecular graph representation learning framework for property prediction. Brief. Bioinform. 24, https://doi.org/10.1093/bib/bbad305 (2023).
Fang, X. et al. Geometry-enhanced molecular representation learning for property prediction. Nat. Mach. Intell. 4, 127–134 (2022).
Zhou, G. et al. Uni-mol: a universal 3d molecular representation learning framework. https://doi.org/10.26434/chemrxiv-2022-jjm0j-v4 (2023).
Zeng, X. et al. Accurate prediction of molecular properties and drug targets using a self-supervised image representation learning framework. Nat. Mach. Intell. 4, 1004–1016 (2022).
Fawcett, T. An introduction to ROC analysis. Pattern Recognit. Lett. 27, 861–874 (2006).
Chai, T. & Draxler, R. R. Root mean square error (rmse) or mean absolute error (mae)?—arguments against avoiding rmse in the literature. Geosci. Model Dev. 7, 1247–1250 (2014).
Chawla, N. V., Bowyer, K. W., Hall, L. O. & Kegelmeyer, W. P. Smote: synthetic minority over-sampling technique. J. Artif. Intell. Res. 16, 321–357 (2002).
Pardridge, W. M. The blood-brain barrier: bottleneck in brain drug development. NeuroRx 2, 3–14 (2005).
Lipinski, C. A. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 46, 3–26 (2001).
Ertl, P., Rohde, B. & Selzer, P. Fast calculation of molecular polar surface area as a sum of fragment-based contributions. J. Med. Chem. 43, 3714–3717 (2000).
Ekins, S., Mestres, J. & Testa, B. In silico pharmacology for drug discovery: methods for virtual ligand screening and profiling. Br. J. Pharmacol. 152, 9–20 (2002).
Li, Y., Tarlow, D., Brockschmidt, M. & Zemel, R. Gated graph sequence neural networks. In Proc. International Conference on Learning Representations (ICLR, 2016).
Paszke, A. et al. Pytorch: an imperative style, high-performance deep learning library. https://pytorch.org (2019).
Fey, M. & Lenssen, J. E. Fast graph representation learning with PyTorch Geometric. In Proc. ICLR Workshop on Representation Learning on Graphs and Manifolds (ICLR, 2019).
Landrum, G. Rdkit: open-source cheminformatics. https://www.rdkit.org (2013).
Acknowledgements
This research was supported by the National Natural Science Foundation of China (No. 32560689), Scientific and Technological Innovation Platform Research Project of Guizhou Province(CXPTXM[2025]024, CXPTXM[2025]026), Guizhou Provincial Science andTechnology Projects ([2024]002, CXTD[2023]027), Guizhou ProvinceYouth Science and Technology Talent Project ([2024]317), Guiyang GuianScience and Technology Talent Training Project ([2024] 2–15), Natural Science Special Fund of Guizhou University (No. 202409). We also thank the Public Big Data Supercomputing Center and the National Key Laboratory of Green Pesticide of Guizhou University for providing high-performance computing resources.
Author information
Authors and Affiliations
Contributions
Huiyang Hong conceived and designed the study, performed experiments, and drafted the manuscript. Xinkai Wu assisted with experiments and manuscript writing. Hongyu Sun assisted with experiments and revised the model code. Chaoyang Xie contributed to the initial model design and discussions on innovative ideas. Yuquan Li and Qi Wang supervised the project and contributed to all aspects of the research and manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Hong, H., Wu, X., Sun, H. et al. A hierarchical interaction message net for accurate molecular property prediction. Commun Chem (2026). https://doi.org/10.1038/s42004-026-01922-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42004-026-01922-x
