Installation

NicheCompass is available for Python 3.9. It does yet not support Apple silicon.

We do not recommend installation on your system Python. Please set up a virtual
environment, e.g. via conda through the [Mambaforge] distribution, or create a
[Docker] image.

Additional Libraries

To use NicheCompass, you need to install some external libraries. These include:
- [PyTorch]
- [PyTorch Scatter]
- [PyTorch Sparse]
- [bedtools]

We recommend to install the PyTorch libraries with GPU support. If you have
CUDA, this can be done as:

pip install torch==${TORCH}+${CUDA} --extra-index-url https://download.pytorch.org/whl/${CUDA}
pip install pyg_lib torch_scatter torch_sparse -f https://data.pyg.org/whl/torch-${TORCH}+${CUDA}.html

where ${TORCH} and ${CUDA} should be replaced by the specific PyTorch and
CUDA versions, respectively.

For example, for PyTorch 2.0.0 and CUDA 11.7, type:

pip install torch==2.0.0+cu117 --extra-index-url https://download.pytorch.org/whl/cu117
pip install pyg_lib torch_scatter torch_sparse -f https://data.pyg.org/whl/torch-2.0.0+cu117.html

To install bedtools, you can use conda:

conda install bedtools=2.30.0 -c bioconda

Alternatively, we have provided a conda environment file with all required
external libraries, which you can use as:

conda env create -f environment.yaml

To enable GPU support for JAX, after the installation run:

pip install jaxlib==0.3.25+cuda${CUDA}.cudnn${CUDNN} -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

For example, for CUDA 11.7, type:

pip install jaxlib==0.4.7+cuda11.cudnn86 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

Installation via PyPi

Subsequently, install NicheCompass via pip:

pip install nichecompass

Install optional dependencies required for benchmarking, multimodal analysis, running tutorials etc. with:

pip install nichecompass[all]