Chem2Bio2RDF Dashboard: Ranking Semantic Associations in Systems Chemical Biology Space
Xiao Ping Dong, Bin Chen, Ying Ding, David J. Wild, Huijun Wang · 2010
Web technology has had a significant impact in scientific collaboration as it provides a common platform to integrate heterogeneous data sources and reasoning capabilities for knowledge discovery. In the biomedical science domain, more and more data providers are providing data in formats that are readily converted to Semantic Web formats, and this has resulted in some early initiatives to collate data in unified Semantic Web repositories such as Linked Open Drug Data (LODD) and Bio2RDF. Many critical problems in biomedical science can be phrased in terms of finding the necessary associations between individual entities (such as explaining drug mechanisms through associations between drugs and metabolic pathways). The networks are necessarily very large, and many association paths may exist between two given entities; therefore an effective and scalable framework for semantic association ranking is needed. In this paper, we describe Chem2Bio2RDF Dashboard, a prototype system for automatic collecting semantic associations within the systems chemical biology space and apply a series of ranking metrics to select the most relevant associations.