Clinical Pharmacogenetics Implementation Consortium (CPIC) has released the first version of its database and API to the public. The CPIC Database is a relational database containing data from CPIC’s guidelines in structured formats. This data includes the guideline manuscript recommendations and information from all of the supporting guideline tables, including gene variant and allele frequencies, function assignments, definitions, diplotype-to-phenotype mappings and example CDS language, as well as example drug-based pre- and post-test alerts and CDS flow charts. Mappings of CPIC gene and drug names to multiple vocabularies/terminologies are also available. We encourage users to please read the extensive documentation about the data models and formats that is available. Different guideline gene-drug pairs require slightly different models which is explained in the documentation. Understanding these differences is critical to successful database queries. Additionally, each guideline has unique caveats and nuances that can only be fully appreciated by reading the guideline themselves, so we encourage users to read the guidelines when accessing and using data from the CPIC database.
The CPIC database can be accessed from the CPIC API or via whole database exports. The API is a RESTful interface and allows access to all parts of the currently defined data model. Documentation for how to use the API with examples can be found in the database documentation. Versioned, whole-database exports can be found in each release on GitHub.
The beta versions of the database and API have undergone several rounds of testing from CPIC volunteers, the CPIC informatics working group and the CPIC membership. However, we anticipate and encourage user feedback from the community. Please contact us a contact@cpicpgx.org with questions and comments.