search

hivdb / covid-drdb

MIT License
2 stars 0 forks source link

readme

COVID-DRDB Database

COVID-DRDB is created by the HIVDB team of Stanford University. It includes resistance data of SARS-CoV-2 for convalescent plasma, vaccinee plasma and monoclonal antibodies collected from published peer-reviewed/pre-print studies. The database also provides meta-data including but not limited to:

Access DRDB

The access to DRDB database is open to the public. An end user can access this database and summarize data via the web interface. This interface is open source at repository hivdb/chiro-frontend.

A SQLite dump file is created (almost) daily and can be retrieved from Release page of this repository. Raw CSV data are available in the hivdb/covid-drdb-payload repository.

This repository includes the code to validate the raw CSV data and generate SQLite dump files.

Create & release DB files

  1. Update CSV files in payload/tables/ directory.
  2. Use command make autofill to complete the CSV files. The command is incomprehensive, you may need to debug the code all the way.
  3. Use command make devdb log-devdb to create a Postgres instance and verify the payload CSVs.
  4. If #4 passed, use command make release or make pre-release to build and release SQLite db file. Credential required
  5. DB files will be stored in build folder, GitHub releases page, and s3 repository s3://cms.hivdb.org/covid-drdb.

How it works?

The Covid-DRDB is a SQL database built from CSV files stored in a Git repository hosted on GitHub. The rationale of storing CSV files in Git repository is to utilize Git's abilities such as version control and author/date tracking.

The latest database schema is maintained in this repository using the format of DBML. This allows us to visualize the schema from the same source. This repository also includes necessary scripts for creating, generating, and validating tables. All these scripts are automated into an easy-to-use Makefile described above. The first half of this pipeline is also automatically running on GitHub CI as our continuous integration.

Pipeline

Contribute to DRDB

Any contributions are welcome here. Currently, we accept feature requests, new study submissions and bug/error reports. We are still working on to provide a pull-request pipeline for outside contributors to add data into our database directly.