| Content | |
|---|---|
| Description | Bio2RDF |
| Contact | |
| Research center | Laval University |
| Laboratory | Centre de Recherche du CHUL, |
| Authors | François Belleau |
| Primary citation | PMID 18472304 |
| Release date | 2008 |
| Access | |
| Website | http://bio2rdf.org |
Bio2RDF is a biological database that uses semantic web technologies to provide interlinked life science data.
In its first form, described in 2008, Bio2RDF provided an RDF warehouse of human and mouse genome data with resources identified by URIs normalized around the bio2rdf.org namespace. A second coordinated release in January 2013 expanded coverage to 19 resources with over one billion triples and added mappings to the Semanticscience Integrated Ontology which allowed ontology-based querying of the data.
The third coordinated release in July 2014, comprised about 11 billion triples across 30 datasets, ten of them new, which the authors described as a tenfold increase over release 2. The largest datasets were PubMed (about 5 billion triples), iProClass (about 3 billion) and NCBI Gene (about 1 billion). The release added clinically relevant datasets, extended dataset statistics and a refactored linked data platform.
Bio2RDF scripts convert heterogeneously formatted data, such as flat files, tab-delimited files, SQL databases and XML into RDF. Entities are identified by URIs that can be resolved through a web application which answers requests by running SPARQL queries against the appropriate endpoints. Because releases are numbered, a dataset can be cited as a specific version in a scientific article.