A robustness measure for the stationary behavior of qualitative gene regulation networks

In this paper the stationary behavior of uncertain and possibly multistable gene regulation networks is considered. We first introduce a modeling framework which is able to represent the qualitative knowledge which is typically available for these systems. Then we turn to the problem of model discrimination: Given several alternative model structures that can all reproduce the experimental observations, is it possible to decide which structure may be the most appropriate description of the real system. To this end, a robustness measure for qualitative multistable gene regulation networks is introduced and also a method for the computation of this measure is presented. The benefit of the developed method is twofold: On the one hand it allows to compare the robustness properties of different model structures, on the other hand also the most fragile interconnections of a network can be detected. Finally, an example network is analyzed with this method.