epaaso / comp-oncology

Example notebook for the recommended analyses for single cell cancer data

Comp-oncology

Example notebook for the recommended analyses for single cell cancer data. We use the lung cancer data from Laughney et. al for immune development in lung cancer.

Pre-reqs

We highly recommend using a computer or server with cuda enabled. Docker must be a given to run the containers. Another advantage would be to have docker installed with the cuda package for transfering your cuda install to the containers.

Nevertheless, we added a list of the package versions for R and python in requirements.txt and r_packages.txt. Or you can base yourself on the commands in the Dockerfile if you don't want to use docker.

The notebook has commands to download the .h5ad file for scanpy. It contains the raw count matrix in Anndata form.

Running

We have designed a docker image that has all the necessary libs. It is however very big, because it has all the R and python packages including the ones for ML.

To be able to run the notebook, you should run the container with this notebook repo inserted as a volume. In the next command, $HOME/2021-SC-HCA-LATAM/CONTAINER is the path of the repo, and the other one is where the large data files would be stored.

docker run --interactive --runtime=nvidia --gpus all --tty --name comp_onco --publish 8888-8892:8888-8892 --volume $HOME/2021-SC-HCA-LATAM/CONTAINER:/root/host_home --volume /datos:/root/datos --workdir /root/host_home/ netopaas/comp-onco:paga /bin/bash

We publish some ports to use the jupyter server.

After that just run the command jl inside the container and a jupyter lab server will be launched.

Running in servers

It is recommended to run the notebook on a server with lots of RAM. Another advantage of having a container is that you can run everything without giving sudo access. If you are doing so, you can launch a reverse proxy to access this server from your computer with a command like this:

ssh -p 5265 -N -f -L 5432:localhost:8888 sefirot.inmegen.gob.mx

This forwards the port 8888 in the server to 5432 in your localhost. The -p is for when the ssh port on your server is different from the default.

Troubleshooting

If you keep your repo inside the container, the user might change. We recommend configuring the ssh keys inside the container. These are the steps:

git config --global user.email [e-mail]
git config --global user.name "Ernesto Paas"

We suggest saving a key pair that can be generated with the command ssh-keygen -t ed25519 -C "[e-mail]". in the folder that contains the repo, and then copying them to ~/.ssh/id_ed25519 etc.