Parameter Set Selection for Signal Transduction Pathway Models including Uncertainties

Abstract

It is common that only a subset of the parameters of dynamic models can be accurately estimated. One approach for identifying a subset of parameters for estimation is to perform clustering of the parameters into groups based upon their sensitivity vectors. However, this approach has the drawback that uncertainty cannot be directly incorporated as the sensitivity vectors are based upon the nominal values of the parameters. One technique to address this deficiency is to define sensitivity cones, where a sensitivity cone includes all possible sensitivity vectors of one parameter for different values resulting from the uncertainty. Parameter clustering can then be performed based upon the sensitivity cones, instead of the sensitivity vectors. This paper applies this new approach to a signal transduction pathway model with a large number of uncertain parameters.