Conceptual Biology Research Supporting Platform: Current Design and Future Issues Ying Xie The Department of Computer Science and Information Systems Kennesaw State University yxie2@kennesaw.edu Jayasimha Katukuri, Vijay V. Raghavan, The Center for Advance Computer Studies University of Louisiana at Lafayette {jrk8907, Raghavan}@cacs.louisiana.edu Anthony Prestigiacomo Araicom Research, LLP tony.presti@araicom.com 1. Introduction With the exponential growth of the accumulated biomedical facts in various databases, the rapid advance of data mining techniques, and the further development of biomedical ontology, conceptual biology is anticipated to take its place as an essential component in biomedical research. By following the hypothesis-driven, experimental research paradigm, conceptual biology utilizes vast amount of published data as sources to generate and test hypotheses at the conceptual level. Compared with labor-based biological research, conceptual biology is expected to be more efficient, cost-effective, and knowing no bounds across fields. Current research on conceptual biology focuses on hypothesis generation from biomedical literature. Most of these algorithms are dedicated to produce one type of hypothesis called pairwise relation by interacting with certain search engines such as Pubmed [12]. However, in order to fully implement its potential, we see the need of constructing a comprehensive conceptual biology research supporting platform, which supports generating and conceptually testing multiple types of biomedical hypotheses. Guided by this vision, we developed a novel integrated architecture that encapsulates a suit of interrelated data structures and algorithms which support 1) automatically revealing multiple types of potential relations among biomedical entities embedded in enormous amount of literature data sets without any user’s input; 2) visualizing multiple types of hypotheses; 3) facilitating validation of generated hypotheses based on heterogeneous repositories; 4) tracking research history of a biological discovery; 5) providing APIs that allow users to discover their own types of hypothesis and testing approaches. In this paper, we will discuss the architecture and some major components/algorithms of the conceptual research support platform. Future research and design issues will also be discussed. 10.1007/978-3-540-78534-7 1 2008