OpenWorm data-driven model validation hub

This repository represents a collection of SciUnit tests for various subprojects of OpenWorm.

Conventional Instructions:

• Set an environment variable for the root path of all of your openworm repositories. For example, you keep the ChannelWorm repository at /path/to/openworm/ChannelWorm, then you would set:

export OPENWORM_HOME=/path/to/openworm

• For each Open Worm subproject repository you wish to test (e.g. ChannelWorm, CElegansNeuroML):

cd $OPENWORM_HOME
git clone http://github.com/openworm/REPO_NAME # Replace REPO_NAME with e.g. ChannelWorm
cd REPO_NAME # Ditto
git pull # Retrieve all branches (in any recent version of the git client)
git checkout sciunit # Switch to the sciunit branch of the repo (which will contain updates for testing)
pip install -e . --process-dependency-links # Install as a developer

• Clone and install this repository:

cd $OPENWORM_HOME
git clone http://github.com/openworm/tests
cd tests
pip install -e . --process-dependency-links

• Launch and run any of the notebooks (owtests/\*.ipynb), or run:

cd $OPENWORM_HOME
python -m unittest owtests

to run all of them in batch from the command line.

Docker instructions:

We provide a Docker container for the same installation:

git clone http://github.com/openworm/tests
docker build -t openworm/owtests tests # Will build the container and run all the tests

docker run -it openworm/owtests # To explore test artifacts from the shell
#or
chmod 744 docker-interact
./docker-interact owtests # To explore test notebooks from the browser

To Do:

• Add a lot more tests
• Allow tests to be run and output to be logged using SciUnit command line tools, e.g. sciunit run.