Nonlinear Design of Stimulus Experiments for Optimal Discrimination of Biochemical Systems

摘要

Biochemical reaction networks in the form of coupled ordinary differential equations (ODEs) provide a powerful modeling tool to understand the dynamics of biochemical processes. During the modeling process a pool of competing nonlinear models is generated, from which the most plausible set has to be selected, given distributed model parameters and hence distributed model prediction. At this point, robust (=taking distributed model responses into account) model-based stimulus experiments can be used, to find experimental conditions at which models show maximal dissimilarities to focus experimental efforts. Response variabilities are typically obtained from linearization. Here we compare this method to the nonlinear Sigma-Point approach for a nonlinear, multi-stable model and show its advantage for model discrimination, especially for large parameter variances.