Mining Complex Biomedical Literature for Actionable Knowledge on Rare Diseases

Complex scientific phenomena and processes underpin drug discovery and development that have historically been addressed through iterative and statistical strategies to derive knowledge from data using labor intensive, inefficient, and costly practices. Complicating the task of data analysis even further, a lot of useful information about drug activities has been historically described in publications and scientific reports in writing. This naturally required reading by the experts to understand the reported facts and extract useful knowledge from publications, a manual, and therefore, non-scalable process. The opportunity now exists to extract knowledge from reading sources using modern text mining to rapidly and affordably identify and develop new or repurposed drug candidates. Nowhere could this be more important than addressing the unmet need in rare diseases.