Data Science & Analytics
Data Translator ARAGORN
Description
The NCATS Biomedical Data Translator program applies semantic integration strategies to share chemical, genetic, phenotypic, disease, ontological, and other ‘knowledge sources’. Taken together, the >300 knowledge sources form an integrated data ecosystem and technology platform—the Translator system—to support clinical and translational science. RENCI contributes to three Translator Projects, including the Ranking Agent project team. The Ranking Agent team has produced a tool called ARAGORN (Autonomous Relay Agent for Generation Of Ranked Networks), which represents the next iteration of its successful ROBOKOP tool. ARAGORN answers user-specified biomedical questions by reasoning over a federated knowledge graph and applying on-the-fly ranking to produce the most accurate results quickly. Additionally, ARAGORN attempts to bridge the precision mismatch between data representations and algorithms that require specificity and users who pose questions and prefer answers at a more abstract level. ARAGORN uses elements of specific answers to create gestalt explanations, clustering, and combining answers with similar content, thereby revealing the commonalities and contradictions in answers. As development continues, the Ranking Agent team will formulate and validate new graph-based methodologies for determining the best answers to user-asked biomedical questions.
RENCI's Role
Alex Tropsha is PI. Chris Bizon is also helping lead the RENCI team.
