Prof. Daniel L. Cook

PHYSIOLOGICAL KNOWLEDGE REPRESENTATION FOR MULTISCALE BIOSIMULATION: KEY TO THE VIRTUAL HUMAN
World-wide efforts are emerging to build multiscale biosimulations of human physiology and pathology that integrate human biological function from the molecular level to the whole organism. Such integrative "multiscale" models can be used to better understand the general human biology of health and disease and can also be used to predict pathological outcomes of individuals. The computational integration, reuse and generation of specific biosimulation models will depend on bioinformatical representation of human anatomical, physiological and pathological knowledge in the form of computational ontologies. In collaboration with investigators in the UW Dept. of Biological Structure we are developing the Physiological Reference Ontology (PRO) as a companion ontology to the available Foundational Model of Anatomy. As part of the DARPA-sponsored Virtual Soldier Project, we prototyped ontologies of physiology and pathology as the basis for the predicting post-wound diagnostic hypotheses that were used to automatically modify quantitative biosimulation models to predict clinical outcomes. These prototype studies are being followed up with more formal and comprehensive PRO components including an Ontology of Physics in Biology and an Ontology of Biological Processes.

At the molecular pathway level, we are continuing to develop a systems representation and qualitative analysis tool, Chalkboard, that is a Java application with a biologist-friendly user-interface for drawing molecular players and their interactions. Chalkboard models can be posted as annotated web-pages, queried qualitatively for cause-effect relations (PathTracing) and used to automatically generate quantitative biosimulation code in the JSim simulation language.