# RGFMDA

**Repository Path**: Tomhappy/RGFMDA

## Basic Information

- **Project Name**: RGFMDA
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-04-11
- **Last Updated**: 2026-04-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# RGFMDA
RGFMDA: a novel residual graphSAGE-based model with nonlinear adaptive feature fusion and triplet contrastive learning for miRNA-disease association prediction

## Requirements
  * python==3.7
  * dgl==0.6.1
  * networkx==2.5
  * numpy==1.16.6
  * scikit-learn==0.20.3
  * pytorch==1.5.0
  * tqdm==4.15.0

## File
### data
  The data files needed to run the model, which contain HMDDv2.0 and HMDDv3.2.
  * disease semantic similarity matrix 1.txt and disease semantic similarity matrix 2.txt: Two kinds of disease semantic similarity
  * miRNA functional similarity matrix.txt: MiRNA functional similarity
  * known disease-miRNA association number.txt:Validated mirNA-disease associations
  * disease number.txt: Disease id and name
  * miRNA number.txt: MiRNA id and name
### code
  * eval.py: The startup code of the program
  * train.py: Train the model
  * model.py: Structure of the model
  * utils.py: Methods of data processing
 
## Usage
  * download code and data
  * execute ```python eval.py```