Adaptive drug repositioning method based on feature-driven isomorphic graph learning

By using a feature-driven isomorphic graph learning method, adaptively adjusting the edge weights of the isomorphic graph, and combining dual-path MLP and multivariate loss function, the problems of noise interference and insufficient fusion of heterogeneous and isomorphic graphs in drug relocation are solved, thereby improving the accuracy and efficiency of drug relocation.

CN122201504APending Publication Date: 2026-06-12GUIZHOU NORMAL UNIVERSITY
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUIZHOU NORMAL UNIVERSITY
Filing Date
2026-01-14
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing drug relocation methods based on graph neural networks suffer from problems such as noise interference from isomorphic graphs and insufficient fusion of heteromorphic and isomorphic graphs, making it difficult to adaptively learn the complex nonlinear relationship between drugs and diseases, resulting in limited prediction accuracy.

⚗Method used

We employ a feature-driven isomorphic graph learning method, which adaptively weights isomorphic graph edges using a learnable parameter mask, combines attention mechanisms and L1 regularization, performs heteromorphic graph convolution operations, and deepens feature learning through a dual-path MLP to generate predicted probabilities of drug-disease associations. We then use a multivariate loss function for multi-objective optimization.

🎯Benefits of technology

It effectively reduces isomorphic noise interference, improves the accuracy and robustness of drug relocation, and enhances the accuracy and efficiency of drug-disease association prediction, making it suitable for large-scale drug relocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

The application discloses an adaptive drug repositioning method based on feature-driven isomorphic graph learning, relates to the technical fields of drug repositioning and graph neural networks, and comprises the following steps: acquiring drug-disease association data and corresponding graph network data information; initializing feature embedding of disease nodes and drug nodes according to the drug-disease association data and the corresponding graph network data information, and constructing a heterogeneous graph and an isomorphic graph; performing a heterogeneous graph convolution operation through disease-to-drug direction and drug-to-disease direction propagation, and calculating the importance weight between the disease nodes and the drug nodes by using an attention mechanism to generate heterogeneous information; adaptively weighting the edges in the isomorphic graph through a learnable parameter mask, controlling the binarization degree of the weight through L1 regularization and entropy regularization, performing an isomorphic graph convolution operation, and generating isomorphic information. The application effectively reduces the noise interference in the pre-defined graph by adaptively weighting the edges in the isomorphic graph through the learnable parameter mask.
Need to check novelty before this filing date? Find Prior Art