Abstract This project aims to develop and implement multiple but integrated computational tools for drug repurposing by exploiting complimentary big data streams, i.e., unstructured texts (social media networks, published biomedical literature, and electronic health records), electronic databases of chemical-biological interactions and pathways, and laboratory data (biological screening of chemical libraries). We expect that tools to be developed in this project will be useful for repurposing observational textual data for research projects (addressing the second challenge of the underlying RFA). In addition, the envisioned translation of this data into a format amenable to quantitative modeling of drug effects will also enable integration of textual and laboratory data to create minable metadata (cf. the third challenge).

Project Narrative This project aims to develop and implement computational tools for drug repurposing by exploiting two complimentary big data streams, i.e., unstructured texts (social media networks, published biomedical literature, and electronic health records) and laboratory data (biological screening of chemical libraries). The proposed tools and underlying computational framework will employ observational human data together with experimental bioactivity data to discover and validate novel therapeutic applications of existing drugs, i.e., following the knowledge discovery path from man to molecules to man.