liudongliangHI/metabolic-ninja

Predict heterologous pathways for the given organism and chemical of interest

Metabolic Ninja

Development

Run make setup first when initializing the project for the first time. Type make to see all commands.

Environment

Specify environment variables in a .env file. See docker-compose.yml for the possible variables and their default values.

• Set ENVIRONMENT to either
  • development,
  • testing, or
  • production.
• SECRET_KEY Flask secret key. Will be randomly generated in development and testing environments.
• SENTRY_DSN DSN for reporting exceptions to Sentry.
• ALLOWED_ORIGINS: Comma-seperated list of CORS allowed origins.

Updating Python dependencies

To compile a new requirements file and then re-build the service with the new requirements, run:

make pip-compile build

Database Migrations

It is important to migrate the database before starting your work and also to create migrations whenever you change database models.

make databases

The following commands will be useful. Please also read the full documentation online.

docker-compose run --rm web flask db init
docker-compose run --rm web flask db migrate
docker-compose run --rm web flask db upgrade

You also need to create the database and tables like so:

dock
>>> from .app import app, init_app
>>> from .models import db
>>> init_app(app, db)
>>> db.create_all()

Before deploying the service you should run those commands with the -e ENVIRONMENT=production or staging flag in order to create the required tables on our production/staging database.

Products

The Products API resource lists all the metabolites in the universal model cameo.models.universal.metanetx_universal_model_bigg_rhea. To allow the web service to avoid loading the large universal models into memory, the list of metabolites is written to data/products.json by the script src/scripts/dump_products.py. This dump must be updated whenever the list of metabolites in the universal model changes.