Lipidx is a Flask application with a number of graphics dependencies including the Anaconda bokeh library
$ conda create -n lipidx bokeh=0.12.6 scikit-learn flask pillow pandas
$ conda install -c conda-forge phantomjs selenium
$ pip install flask_bootstrap flask_wtf
$ git clone https://github.com/harvardinformatics/lipidx.git
$ cd lipidx
$ export PYTHONPATH=`pwd`
$ export FLASK_APP=lipidx
$ export FLASK_DEBUG=1
$ flask run
You should be able to see the application at http://localhost:5000/lipidx/lipid_analysis/
Lipidx can be deployed as a Docker container.
$ git clone https://github.com/harvardinformatics/lipidx.git
$ cd lipidx
$ docker build -t lipidx .
$ docker run -d -p 8000:80 --name lipidx lipidx
Once running, navigate to http://localhost:8000/lipid_analysis
For production, you should set the following environment variables:
$ export LIPIDX_ADMIN_EMAILS='one@harvard.edu,two@harvard.edu'
$ export LIPIDX_KEY='someunintelligiblestring'
$ export WTF_CSRF_KEY='someotherunintelligiblestring'