Installation¶

GraphiC depends on a number of other libraries for constructing Hi-C graphs as well as for using Hi-C specific message-passing schemes. These should be installed in advance.

Note

We recommend installing GrapHiC in a virtual environment. ..

Note

Some of these packages have more involved setup depending on your requirements (i.e. CUDA). Please refer to the original packages for more detailed information

Creating Conda Environment¶

conda create -n graphic python=3.7

Installing PyTorch¶

# Install PyTorch: MacOS
$ conda install pytorch torchvision -c pytorch                      # Only CPU Build

# Install PyTorch: Linux
$ conda install pytorch torchvision cpuonly -c pytorch              # For CPU Build
$ conda install pytorch torchvision cudatoolkit=9.2 -c pytorch      # For CUDA 9.2 Build
$ conda install pytorch torchvision cudatoolkit=10.1 -c pytorch     # For CUDA 10.1 Build
$ conda install pytorch torchvision cudatoolkit=10.2 -c pytorch     # For CUDA 10.2 Build

Installing Pytorch Geometric¶

$ pip install torch-scatter==latest+${CUDA} -f https://pytorch-geometric.com/whl/torch-${TORCH}.html
$ pip install torch-sparse==latest+${CUDA} -f https://pytorch-geometric.com/whl/torch-${TORCH}.html
$ pip install torch-cluster==latest+${CUDA} -f https://pytorch-geometric.com/whl/torch-${TORCH}.html
$ pip install torch-spline-conv==latest+${CUDA} -f https://pytorch-geometric.com/whl/torch-${TORCH}.html
$ pip install torch-geometric

Install all needed packages with ${CUDA} replaced by either cpu, cu92, cu100 or cu101 depending on your PyTorch installation.

Note

Follow the instructions in the Torch-Geometric Docs (https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html) to install the versions appropriate to your CUDA version.

Install Cython & git-lfs¶

$pip install Cython
$conda install git-lfs

Clone the git repo¶

git clone https://github.com/dhall1995/GrapHiC
cd GrapHiC
pip install -e .

Setup Notebook graphic kernel (optional)¶

Optionally, to run the tutorial notebooks, run the following from within the conda environment:

conda install -c anaconda ipykernel
python -m ipykernel install --user --name=graphic

Then when starting a jupyter notebook choose the graphic kernel