Verbeeck, R., Schultz, T., Alquier, L. and Stephens, S. 2010. Relational to RDF mapping using D2R for translational research in neuroscience. In Proceedings of the annual international conference on Intelligent Systems for Molecular Biology (ISMB2010), Bio-Ontologies SIG (Boston, Massachussetts, July 9, 2010) – paper
Motivation: To support translational research and external innovation, we are evaluating the potential of the semantic web to integrate data from discovery research through to the clinical environment. This paper describes our experiences in mapping relational databases to RDF for data sets relating to neuroscience.
Implementation: We describe how classes were identified in the original data sets and mapped to RDF, and how connections were made to public ontologies. Special attention was paid to the mapping of experimental measures to RDF and how it was impacted by the relational schemata.
Results: Mapping from relational databases to RDF can benefit from techniques borrowed from dimensional modeling. However, current tools like D2R are still evolving. Nevertheless, mapping data in RDF, if done properly and consistently, facilitates data integration efforts.