Related Experiment Video
Updated: Jul 18, 2026
07:35
A Knowledge Graph Approach to Elucidate the Role of Organellar Pathways in Disease via Biomedical Reports
Published on: October 13, 2023
2.0K
Identifying Microbe-Disease Association Based on a Novel Back-Propagation Neural Network Model
Summary
This study introduces BPNNHMDA, a novel computational method using neural networks to predict microbe-disease associations. BPNNHMDA offers a faster and more accurate approach than traditional experiments for identifying links between microbes and diseases.
Area of Science:
- Bioinformatics
- Computational Biology
- Microbiome Research
Background:
- Microbe-human body associations are crucial for health and disease.
- Traditional experimental methods for identifying microbe-disease links are costly and time-consuming.
Purpose of the Study:
- To develop and validate a novel computational method, BPNNHMDA, for predicting potential microbe-disease associations.
- To improve the efficiency and accuracy of identifying microbe-disease relationships.
Main Methods:
- A novel neural network model (BPNNHMDA) was designed using known microbe-disease associations as input.
- A new activation function based on the hyperbolic tangent function was developed.
- Initial connection weights were optimized using Gaussian Interaction Profile (GIP) kernel similarity for microbes.
Main Results:
- BPNNHMDA achieved high Area Under the Curve (AUC) values: 0.9242 (LOOCV), 0.9127 ± 0.0009 (5-Fold CV), and 0.8955 ± 0.0018 (2-Fold CV).
- The model demonstrated superior performance compared to existing state-of-the-art methods.
- Case studies on inflammatory bowel disease, asthma, and obesity confirmed its excellent predictive ability in practical scenarios.
Conclusions:
- BPNNHMDA is an effective and efficient computational tool for predicting microbe-disease associations.
- The method holds significant potential for advancing research in microbiome-related diseases.
- BPNNHMDA offers a valuable alternative to traditional experimental approaches.
