Content area
Named entities, such as chemicals/drugs, genes/proteins, and diseases, and their associations are not only important components of biomedical literature, but also the foundation of creating biomedical knowledgebases and knowledge graphs. This work addresses the challenges of expressing co-occurrence associations between named entities extracted from a biomedical literature corpus in a machine-readable format. We developed a Resource Description Framework (RDF) data model and integrated it into the PubChemRDF resource, which is freely accessible and publicly available. The developed co-occurrence data model was populated into a triplestore with named entities and their associations derived from text mining of millions of biomedical references found in PubMed. The utility of the data model was demonstrated through multiple use cases. Together with meta-data modeling of the references including the information about the author, journal, grant, and funding agency, this data model allows researchers to address pertinent biomedical questions through SPARQL queries and helps to exploit biomedical knowledge in various user perspectives and use cases.
Scientific contribution
The RDF data model developed in this work encodes co-occurrence associations among chemicals, genes, and diseases, derived from biomedical literature. The developed model enables researchers to use SPARQL queries to semantically explore biomedical knowledge and make new discoveries. It also seamlessly links to scientific data in other information resources, improving the usability and accessibility of biomedical data in the Semantic Web.