Ariadne Genomics received a $100,000 SBIR phase I grant from NIH’s National Center for Research Resources for the development of MedScan technology. MedScan employs natural language processing to extract biomedical information from scientific text and put it into databases for computational analysis.
The SBIR grant will be used to expand MedScan from its traditional focus of pathway analysis into the area of toxicology, small molecule-enzyme interactions, and human metabolism.
MedScan utilizes a high-content linguistics approach to extract functional relationships between biological entities, such as proteins, cell processes, and small molecules, from Medline abstracts and full text articles. It recognizes types of regulatory mechanisms involved, the effects of regulation, and can be customized to extract other types of information and focus on different organisms.
“Our ultimate, long-term goal is to convert the entire available corpus of biomedical literature into a formal database suitable for advanced computational and logical analysis,” states Nikolai Daraselia, senior director of research.