EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Graph Neural Network-Based Toxicity Prediction by Integrating Molecular Fingerprints and Knowledge Graph Features

Junjie Xie, Wei Liu, Wei Hu, Mei Ouyang iD, Tingting Huang

Toxics · 05.11.2025 · 13 (11) · 953

DOI: 10.3390/toxics13110953 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Toxics, online erschienen am 05.11.2025. Verfasst von Junjie Xie und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Informatics, Hunan University of Chinese Medicine, Changsha 410208, China. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Molecular toxicity prediction plays a crucial role in drug screening and environmental health risk assessment. Traditional toxicity prediction models primarily rely on molecular fingerprints and other structural features, while neglecting the complex biological mechanisms underlying compound toxicity, resulting in limited predictive accuracy, poor interpretability, and reduced generalizability. To address this challenge, this study proposes a novel molecular toxicity prediction framework that integrates knowledge graphs with Graph Neural Networks (GNNs). Specifically, we constructed a heterogeneous toxicological knowledge graph (ToxKG) based on ComptoxAI. ToxKG incorporates data from authoritative databases such as PubChem, Reactome, and ChEMBL, and covers multiple entities and relationships including chemicals, genes, signaling pathways, and bioassays. We then systematically evaluated six representative GNN models (GCN, GAT, R-GCN, HRAN, HGT, and GPS) on the Tox21 dataset. Experimental results demonstrate that heterogeneous graph models enriched with ToxKG information significantly outperform traditional models relying solely on structural features across multiple metrics including AUC, F1-score, ACC, and balanced accuracy (BAC). Notably, the GPS model achieved the highest AUC value (0.956) for key receptor tasks such as NR-AR, highlighting the critical role of biological mechanism information and heterogeneous graph structures in toxicity prediction. This study provides a promising pathway toward the development of interpretable and efficient intelligent models for toxicological risk assessment.

Schlagwörter

Stichwörter der Autor:innen

Tox21graph neural networksknowledge graphmolecular toxicity prediction

Autor:innen und Institutionen

  • School of Informatics, Hunan University of Chinese Medicine, Changsha 410208, China.

ORCID-Profile: Mei Ouyang

Zitieren

APA

Xie, J., Liu, W., Hu, W., Ouyang, M., & Huang, T. (2025). Graph Neural Network-Based Toxicity Prediction by Integrating Molecular Fingerprints and Knowledge Graph Features. Toxics, 13(11), 953. https://doi.org/10.3390/toxics13110953

Vancouver

Xie J, Liu W, Hu W, Ouyang M, Huang T. Graph Neural Network-Based Toxicity Prediction by Integrating Molecular Fingerprints and Knowledge Graph Features. Toxics. 2025;13(11):953. doi:10.3390/toxics13110953

BibTeX
@article{Xie2025,
  author = {Xie, Junjie and Liu, Wei and Hu, Wei and Ouyang, Mei and Huang, Tingting},
  title = {Graph Neural Network-Based Toxicity Prediction by Integrating Molecular Fingerprints and Knowledge Graph Features},
  journal = {Toxics},
  year = {2025},
  volume = {13},
  number = {11},
  pages = {953},
  doi = {10.3390/toxics13110953}
}
Kostenlos starten →