KGG : Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions
Journal of Chemical Information and Modeling · 2025
DOI: 10.1021/acs.jcim.5c01068 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, 2025. Verfasst von Van-Thinh To und 7 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Pharmacy, University of Medicine and Pharmacy at Ho Chi Minh City, 41 Dinh Tien Hoang, District 1, Ho Chi Minh City 700000, Vietnam. Hauptthemen laut MEDLINE: Supervised Machine Learning. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea International Cooperation Agency.
Abstract (Originalsprache)
Molecular property prediction has become essential in accelerating advancements in drug discovery and materials science. Graph Neural Networks have recently demonstrated remarkable success in molecular representation learning; however, their broader adoption is impeded by two significant challenges: (1) data scarcity and constrained model generalization due to the expensive and time-consuming task of acquiring labeled data and (2) inadequate initial node and edge features that fail to incorporate comprehensive chemical domain knowledge, notably orbital information. To address these limitations, we introduce a Knowledge-Guided Graph (KGG) framework employing self-supervised learning to pretrain models using orbital-level features in order to mitigate reliance on extensive labeled data sets. In addition, we propose novel representations for atomic hybridization and bond types that explicitly consider orbital engagement. Our pretraining strategy is cost efficient, utilizing approximately 250,000 molecules from the ZINC15 data set, in contrast to contemporary approaches that typically require between two and ten million molecules, consequently reducing the risk of potential data contamination. Extensive evaluations on diverse downstream molecular property data sets demonstrate that our method significantly outperforms state-of-the-art baselines. Complementary analyses, including t-SNE visualizations and comparisons with traditional molecular fingerprints, further validate the effectiveness and robustness of our proposed KGG approach. The key advantages of KGG are its data efficiency and architectural versatility, driven by orbital-informed representations. By distilling essential chemical knowledge from modest corpora, it avoids extensive pretraining and excels in low-data fine-tuning, providing a robust and chemically meaningful foundation for diverse GNN architectures.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Faculty of Pharmacy, University of Medicine and Pharmacy at Ho Chi Minh City, 41 Dinh Tien Hoang, District 1, Ho Chi Minh City 700000, Vietnam.
- Department of Mathematics and Computer Science, University of Southern Denmark, DK-5230 Odense M, Denmark.
- Bioinformatics Group, Department of Computer Science, Interdisciplinary Center for Bioinformatics, and School for Embedded and Composite Artificial Intelligence (SECAI), Leipzig University, Härtelstraße 16-18, D-04107 L…
- Max Planck Institute for Mathematics in the Sciences, Inselstraße 22, D-04103 Leipzig, Germany.
- Department of Theoretical Chemistry, University of Vienna, Währingerstraße 17, A-1090 Vienna, Austria.
- Facultad de Ciencias, Universidad Nacional de Colombia, Bogotá D.C. 111321, Colombia.
ORCID-Profile: Rolf Fagerberg, Tuyen Ngoc Truong
Förderung
- Korea International Cooperation Agency
Zitieren
APA
To, V. T., Van Nguyen, P. C., Truong, G. B., Phan, T. M., Phan, T. L., Fagerberg, R., Stadler, P. F., & Truong, T. N. (2025). KGG : Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.5c01068
Vancouver
To VT, Van Nguyen PC, Truong GB, Phan TM, Phan TL, Fagerberg R, et al. KGG : Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions. Journal of Chemical Information and Modeling. 2025. doi:10.1021/acs.jcim.5c01068
BibTeX
@article{To2025,
author = {To, Van-Thinh and Van Nguyen, Phuoc-Chung and Truong, Gia-Bao and Phan, Tuyet-Minh and Phan, Tieu-Long and Fagerberg, Rolf and Stadler, Peter F and Truong, Tuyen Ngoc},
title = {KGG : Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions},
journal = {Journal of Chemical Information and Modeling},
year = {2025},
doi = {10.1021/acs.jcim.5c01068}
}
EUVIMED