Toward a complete dataset of drug–drug interaction information from publicly available sources

• DOI: 10.1016/j.jbi.2015.04.006
• Publication Date: 2015-04-26T02:45:17Z
• Authors: Serkan Ayvaz; John Horn; Oktie Hassanzadeh; Qian Zhu; Johann Stan; Nicholas P. Taton...; Santiago Vilar; Mathias Brochhausen; Matthias Samwald; Majid Rastegar-Mo...; Michel Dumontier; Richard D. Boyce
• Keywords: DrugBank

Although potential drug–drug interactions (PDDIs) are a significant source of preventable drug-related harm, there is currently no single complete PDDI information. In the current study, all publically available sources information that could be identified using comprehensive and broad search were combined into dataset. The dataset merged fourteen different including 5 clinically-oriented sources, 4 Natural Language Processing (NLP) Corpora, Bioinformatics/Pharmacovigilance sources. As source, might benefit pharmacovigilance text mining community by making it possible to compare representativeness NLP corpora for extraction tasks, specifying elements can useful future purposes. An analysis overlap between across data showed was little overlap. Even lists such as DrugBank, KEGG, NDF-RT had less than 50% with each other. Moreover, incomplete coverage two focus on PDDIs interest in most clinical settings. Based this information, we think systems provide access lists, APIs RxNorm, should careful inform users may respect drug experts suggest clinicians aware of. spite low degree overlap, several dozen cases where provided product labeling augmented also shown improve performance an existing pipeline recently published protocol. Future work will improvement methods mapping identifying use algorithms, integrating high-quality from Wikidata, accessible Semantic Web Linked Data.