Knowledge Integration for Bio-Threat Response

Introduction......................................................................................................................... 2 The Need: Bio-Threat Response......................................................................................... 3 Current State of the Art....................................................................................................... 6 Database Integration ....................................................................................................... 6 Knowledge Systems in Biological Network Analysis .................................................... 7 Biomedical Ontologies.................................................................................................... 9 Ontological Technologies ............................................................................................. 11 Text and Natural Language Processing ........................................................................ 12 Technology Gaps .............................................................................................................. 14 References......................................................................................................................... 17 Abstract Bio-threats require rapid analysis and response to prevent widespread consequences to the population. The inability to easily link and exploit biological knowledge for both human and automated analysis is a major limitation on the speed of complex knowledge development and bio-threat response. Today, linkage among the vast array of biological knowledge repositories is primarily by hand. Key requirements for smooth linkage of knowledge sources include shared ontologies of concepts and semantic relations, capabilities for unifying terminology and extracting meaningful relations from text, and inference mechanisms to link and unify heterogeneous databases. While research in all these areas continues, the gaps to the needed knowledge integration technologies are immense, and would benefit from co-ordinated cross-agency funding.