In this repository we are creating a prototype for a novel biomedical journal for publishing data and analysis resources. This work was started during the Japan 2017 Biohackathon (September 2017).

Key goals the data journal are:

1. Make it easy for researchers to publish data with a short paper
2. Provide a review mechanism
3. Make the paper citeable
4. Provide usable FAIR-data mechanisms

FAIR means the data should be

1. Findable: provide useful metadata
2. Accessible: provide a stable URL
3. Interoperable: use common data formats
4. Reproducible: provide analysis with data (at least as an example)

All versions of papers, review and comments are stored into a public git repository which means everything can we recovered. We will also use a database that acts as a cache of git checkouts.

Rather than rolling our own ontologies we'll use whatever is available.

Wikidata/wikicite and open citations have semantic web representations which can be used to handle citations.

To build in speed and robustness authors are encouraged to host data on the interplanetary file system (IPFS) which provides download URLs.

We opted for Python/flask for the web server. Mostly because Python has become a lingua franca in bioinformatics and Flask is a well tested and simple web development framework. For some functionalities and (perhaps) performance we will use plug-ins which can be written in any language. The REST API may be written in Elixir.

See INSTALL.md.

For further reading see development docs

All code is published under the AGPL. See LICENSE.

Pjotr Prins, Raoul Bonnal, Francesco Strozzi (C) 2017