# scSHARP_tool

**Repository Path**: duangao/scSHARP_tool

## Basic Information

- **Project Name**: scSHARP_tool
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2024-01-22
- **Last Updated**: 2024-01-22

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# scSHARP

Repository for "Consensus Label Propagation with Graph Convolutional Networks for Single-Cell RNA Sequencing Cell Type Annotation" extended abstract submission for Learning on Graphs Conference and full paper available at https://doi.org/10.1101/2022.11.23.517739.

## R tools Installation
You will need one R dependency to run scSHARP. Repository and install instructions can be found [here](https://github.com/W-Holtz/R4scSHARP)

## scSHARP Installation

### Create a new conda environment
```
conda create -n <env name> python=3.9
conda activate <env name>
```
### Install pip to conda environment
```
conda install pip
```
### Install torch

Linux with GPU:
```
conda install pytorch pytorch-cuda=11.7 -c pytorch -c nvidia
``` 
Mac OS:
```
conda install pytorch -c pytorch
```

### Install torch geometric
```
conda install pyg -c pyg
```

### Install scSHARP
You will need to use the version of pip installed to your new conda environemt. In order to find the path to your conda environment, you can use:
```
conda config --show envs_dirs
```
Run the pip install of scSHARP directly from the binary in yout conda directory
```
./anaconda/envs/<env_name>/bin/pip install scSHARP
```

Installation of torch and torch geometric required prior to pip install.

See demo.ipynb for example work flow.
Please input raw counts.