Install the system dependencies needed to install and run the wrapper (and materializing its tests' datasets):
- The latest version of the VRE Nextflow wrapper should be fetched through git. Also, the wrapper itself uses
git
command line to materialize workflows, sogit
command must be reachable through PATH environment variable:
sudo apt-get -y install git
- Install Python's'
venv
module, which also fetchespip
:
# This first one is needed to have venv module
sudo apt-get -y install python3-venv
- Docker must be installed and running in the machine, as it is expected that Nextflow workflows use docker containers. Next docker installation instructions are valid for Ubuntu / Debian:
# These are pre-requisites for docker, described at
# https://docs.docker.com/install/linux/docker-ce/ubuntu/#install-using-the-repository
sudo apt update
sudo apt install apt-transport-https ca-certificates curl software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository \
"deb [arch=amd64] https://download.docker.com/linux/ubuntu \
$(lsb_release -cs) \
stable"
# This is the installation of the docker package as such
sudo apt update
sudo apt install docker-ce
Remember to add the username to the docker
group, using for instance next recipe.
sudo usermod -a -G docker $USER
Once this last step is done, it is recommended to restart the Linux graphical session, so the belonging to the docker group is visible to all the processes of the user.
- jq command line is used to ensure reproducible tests, as it helps peeking the JSON config files from the tests:
sudo apt install jq
- The repo latest version can be materialized through next command:
git clone https://github.com/inab/vre-process_nextflow-executor.git
- Install the wrapper dependencies
cd vre-process_nextflow-executor
python3 -m venv .py3Env
source .py3Env/bin/activate
pip install --upgrade pip wheel
pip install -r requirements.txt
-
Create a
VRE_NF_RUNNER.py.ini
file, and customize it accordingly to your machine. You can use VRE_NF_RUNNER.py.ini.template as an starting point:cp VRE_NF_RUNNER.py.ini.template VRE_NF_RUNNER.py.ini
-
Docker image with Nextflow is fetched on first wrapper invocation. The specific version is determined both by the constraints within the workflow being run and the content of
VRE_NF_RUNNER.py.ini
.