Deriving Simulation Models from a Molecular Biology Knowledge Base

Qualitative reasoning has been proven a powerful technique to represent the relations between components and objects of physical processes. This technique is of great interest for simulation of biological processes where detailed quantitative information is not available and only qualitative or fuzzy statements about the nature of interactions are known. Model building is an inevitable and time consuming task that even can turn the building of a putatively simple model into a major task. To facilitate this task, we present a new Model Description Language, MDL, to describe templates that can be used to translate a selected item of interest from a molecular biology knowledge base into simulation objects and processes. This is done by use of the presented Model Description Interpreter, MDI. We show that qualitative reasoning can be combined with automatic transformation of contents of genomic databases into simulation models to give an interactive modelling system that reasons about the relations and interactions of biological entities.