Robust experimental design and feature selection in signal transduction pathway modeling

Due to the general lack of experimental data for biochemical pathway model identification, cell-level time series experimental design is particularly important in current systems biology research. This paper investigates the problem of experimental design for signal transduction pathway modeling, and in particular, focuses on methods for parametric feature selection. An important problem is the estimation of parametric uncertainty which is a function of the true (but unknown) parameters. In this paper, two "robust" feature selection strategies are investigated. The first is a mini-max robust experimental design approach, the second is a sampled experimental design method inspired by the Morris global sensitivity analysis. The two approaches are analyzed and interpreted in terms of a generalized optimal experimental design criterion, and their performance has been compared via simulation on the IκB-NF-κB pathway feature selection problem.