# scGTN

scGTN: Deep Siamese Graph Transformer Network for Single-cell RNA Sequencing Clustering

## Requirements

```bash
pip install -r requirements.txt
```

## Usage

```bash
python train_scGTN.py --dataset Muraro_human_Pancreas_cell --n_top_genes 2000
```

## Data Format

Place `.h5` files in the `datasets/` folder. 

## Project Structure

```
scGTN/
├── config.py          # Model configuration
├── model.py           # scGTN model
├── layer.py           # Loss functions
├── preprocess.py      # Data preprocessing
├── utils.py           # Utility functions
├── train_scGTN.py     # Training script
├── datasets/          # Data folder
└── README.md
```