An OWL knowledge base for classifying and querying collections of physiological models: A prototype human physiome

The human physiome is envisioned as the quantitative description of the whole of human physiology. Researchers actively working toward achieving this grand challenge have populated publicly available repositories of quantitative physiological models; however, no mechanism has been developed that integrates the information from these models into a single knowledge resource that can be said to represent a physiome. Here we present a mechanism for automatically generating a physiome knowledge base (SemPhysKB) from models collected from the CellML repository, the National Simulation Resource repository, and the BioModels database. Applying description logic axioms and automated classification, we demonstrate that the SemPhysKB can be queried to retrieve models, sub-models and mathematical formulations of interest and provides a set of logically defined physiological reference terms currently unavailable among biomedical ontologies.