Related Experiment Video
Updated: Jun 3, 2025
07:35
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
Published on: October 13, 2023
1.6K
NRGCNMDA: Microbe-Drug Association Prediction Based on Residual Graph Convolutional Networks and Conditional Random
Xiaoxin Du1,2, Jingwei Li3,4, Bo Wang3,4
1Computer and Control Engineering College, Qiqihar University, Qiqihar, 161006, China. xiaoxindu@qqhru.edu.cn.
Interdisciplinary Sciences, Computational Life Sciences
|January 8, 2025
Summary
A new computational model, NRGCNMDA, accelerates microbe-drug association discovery. This method significantly outperforms existing techniques, offering a faster and more cost-effective approach to identifying potential drug candidates for microbial targets.
Area of Science:
- Microbiology
- Computational Biology
- Pharmacology
Background:
- Traditional methods for microbe-drug discovery are time-consuming and expensive.
- There is a need for efficient computational approaches to predict microbe-drug associations.
Purpose of the Study:
- To propose a novel computational model, NRGCNMDA, for predicting microbe-drug associations.
- To overcome the limitations of traditional drug discovery methods.
Main Methods:
- Constructing a heterogeneous network of microbes and drugs using Node2vec.
- Employing a Graph Convolutional Network with a residual mechanism (REGCN) for feature learning.
- Utilizing conditional random fields (CRF) for feature embedding alignment.
Main Results:
- NRGCNMDA achieved superior performance compared to existing deep learning methods.
- The model demonstrated high accuracy with AUC of 95.16% and AUPR of 93.02%.
- Case studies successfully predicted associations for specific microbes (e.g., Enterococcus faecalis) and drugs (e.g., ibuprofen).
Conclusions:
- NRGCNMDA offers an effective and efficient computational strategy for predicting microbe-drug associations.
- The model has the potential to accelerate the identification of novel antimicrobial drugs.
- This approach can significantly reduce the cost and time associated with drug discovery.
