This is an application for assembling, augmenting, combining, visualizing and analyzing gene sets: genesetcart.cfde.cloud
# go to app (imagineset) directory
cd imagineset
# prepare .env file & review
cp .env.example .env
# start database
# (NOTE: If you're running another postgres database on your system, you should turn it off as the ports will conflict)
docker-compose up -d g2sg-postgres
# install node modules
npm i
# initialize prisma
npx prisma migrate deploy
# run dev server
npm run dev
Open http://localhost:3000 with your browser to see the platform.
For the NEXTAUTH_SECRET
value
Run
openssl rand -base64 32
Set NEXTAUTH_URL=http://localhost:3000
To run the python backend. Run the command:
gunicorn --workers 2 --timeout 600 --bind 0.0.0.0:8000 app:app
And set PYTHON_API_BASE='http://0.0.0.0:8000' in the .env file
To use the site, you have to populate your local database with gene and crossing pairs data. Firstly, make sure apply migrations to create the tables
# go to app (imagineset) directory
cd imagineset
npx prisma migrate deploy
and then run ingest script in database directory to add the data to these tables
# go to database directory (use the following commands if in imagineset directory):
cd ..
cd database
# download requirements
pip install -r requirements.txt
# prepare .env file containing DATABASE_URL
cp .env.example .env
# run ingest script
python3 ingest.py