Item

Abstract

The data available about signaling networks grows steadily, boosting the development of increasingly detailed models. However, the holistic properties of the signaling systems are difficult to grasp. A decomposition into modules has been proposed as a useful tool to untangle biological complexity. Once modules are defined, they can be thoroughly analysed, and later re-wired into the whole networks. However, a clear, unique definition of module is still lacking [1]. We try to demarcate modules from a system-theoretical point of view. For example, the absence of retroactivity in the connections between the modules might be an interesting criterion for decomposing signaling networks [2]. Such units have properties which are independent of downstream elements and can be relatively straightforwardly analyzed by means of system theory s tools [2-3]. Additionally, a model reduction based on this modular approach can lead to a significant simplification of the models [4]. We have applied this approach to several cases, ranging from the relatively simple two-component system in E. coli to the complex EGF signaling in mammal cells [2-4]. References: [1] Wolf D.M. and Arkin, A.P. Curr. Opin. Microbiol. 6(2): 125-34, 2003. [2] J. Saez-Rodriguez, A. Kremling and E.D. Gilles, Comput. Chem. Eng., in press. [3] J. Saez-Rodriguez, A. Kremling, H. Conzelmann, K. Bettenbrock and E.D. Gilles, IEEE CSM, 24(4): 35-52, 2004. [4] H. Conzelmann, J. Saez-Rodriguez, T. Sauter, E. Bullinger, F. Allgöwer and E.D. Gilles, IEE Systems Biology, in press.