# GraphHSA

**Repository Path**: Tomhappy/GraphHSA

## Basic Information

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

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-08-08
- **Last Updated**: 2026-08-08

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

### GraphHSA

Code and datasets for "Revealing Herb-Symptom Associations and Mechanisms of Action in Protein Networks Using Subgraph Matching Learning"

**Tuturial**

To get the experiment results by running the scripts (scenario is warm start or cold start for herb/symptom/pair):

```
python main_cv.py --save_dir results_cv_{scenario}/ --data_dir ../data/{scenario}/   ###5-fold cross-validation, .pickle file is obtained from save_train_val_test_data.py
```

```
python main_indep.py --save_dir results_indep_{scenario}/ --data_dir ../data/{scenario}/   ###independent test
```

Users can use their **own data** to train prediction models.

```
python save_train_val_test_data.py   ###splitting the input data by different strategies
```

**Requirements**

- numpy 1.25.0
- pandas 1.5.3
- networkx 3.1
- scipy 1.10.1
- scikit-learn 1.2.2
- pytorch 2.0.1
- torch-geometric 2.3.1
- python 3.9.18

**Contact**

Please feel free to contact us if you need any help ([mengluli@webmail.hzau.edu.cn](mailto:mengluli@webmail.hzau.edu.cn) or [zhangwen@mail.hzau.edu.cn](mailto:zhangwen@mail.hzau.edu.cn)).