Installation guide

Install from PyPI

Note

Installing scSLAT within a new conda environment is recommended.

First, create a clean environment to install the scSLAT. We recommend you install on a machine equipped CUDA-enabled GPU, scSLAT will be 5x-10x faster than running on CPU.

conda create -n scSLAT python=3.11 -y && conda activate scSLAT
pip install scSLAT
install_pyg_dependencies

Install from Github

You can also install latest version of scSLAT from Github, clone the repo and install:

conda create -n scSLAT python=3.8 -y && conda activate scSLAT

git clone git@github.com:gao-lab/SLAT.git
cd SLAT
pip install -e ".[dev, docs]"

Docker

You can pull the docker image from here or build it from Dockerfile:

docker pull huhansan666666/slat:latest