Emergent Models in Complex System Simulations of Genetic and Biochemical NetworksHenk Stolk1, Kevin Gates2, Jim Hanan
firstname.lastname@example.org, University of Queensland; email@example.com, University of Queensland
We are exploring simulation methods for complex systems consisting of many interacting entities. A high level architecture for a software environment to simulate complex systems is proposed. The simulation environment mimics a complex systemís processes and structure by modeling it as interacting and concurrently executing software agents and processes. These agents and processes exist at various hierarchical levels. Properties and behavior of agents at each level depend on those on a lower level, and mechanisms modeling the emergence of higher level properties and behavior from lower level properties and behavior are studied. As an example of such mechanisms, genetic programming algorithms can discover macro-level emergent models from micro-level properties and behavior. These algorithms can also be applied to the inverse problem of deriving micro-level models from known macro-level behavior. Applications in computational molecular biology include on the one hand discovering macro-level gene expression patterns and behavior from micro-level genetic and biochemical networks, and on the other hand reverse engineering micro-level interactions in these networks from macro-level observed gene expression data.