Skip to main content

ORIGINAL RESEARCH article

Front. Pharmacol.
Sec. Ethnopharmacology
Volume 16 - 2025 | doi: 10.3389/fphar.2025.1543966
This article is part of the Research Topic Artificial Intelligence in Traditional Medicine Research and Application View all 4 articles

AMFGNN:An adaptive multi-view fusion graph neural network model for drug prediction

Provisionally accepted
Fang He Fang He *Lian Duan Lian Duan *Guodong Xing Guodong Xing *Xiaojing Chang Xiaojing Chang *Huixia Zhou Huixia Zhou *Mengnan Yu Mengnan Yu *
  • Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China

The final, formatted version of the article will be published soon.

    Drug development is a complex and lengthy process, and drug-disease association prediction aims to significantly improve research efficiency and success rates by precisely identifying potential associations. However, existing methods for drug-disease association prediction still face limitations in feature representation, feature integration, and generalization capabilities. To address these challenges, we propose a novel model named AMFGNN (Adaptive Multi-View Fusion Graph Neural Network). This model leverages an adaptive graph neural network and a graph attention network to extract drug features and disease features, respectively. These features are then used as the initial representations of nodes in the drug-disease association network to enable efficient information fusion. Additionally, the model incorporates a contrastive learning mechanism, which enhances the similarity and differentiation between drugs and diseases through cross-view contrastive learning, thereby improving the accuracy of association prediction. Furthermore, a Kolmogorov-Arnold network is employed to perform weighted fusion of various final features, optimizing prediction performance. Cross-validation results on three datasets demonstrate that AMFGNN exhibits significant advantages in prediction performance. Moreover, a case study on asthma further validates the model's effectiveness and potential value in practical applications.

    Keywords: Drug prediction, Drug-disease association prediction, Graph attention network, Contrastive learning, Kolmogorov-Arnold Network

    Received: 12 Dec 2024; Accepted: 10 Jan 2025.

    Copyright: © 2025 He, Duan, Xing, Chang, Zhou and Yu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence:
    Fang He, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China
    Lian Duan, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China
    Guodong Xing, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China
    Xiaojing Chang, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China
    Huixia Zhou, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China
    Mengnan Yu, Faculty of Pediatrics, the Chinese PLA General Hospital, Beijing, China, Beijing, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.