Mechanistic mathematical models of biochemical systems are important tools in the study of cellular processes. Such models make explicit current assumptions about the structure and dynamics of the systems of interest whilst statistical methodology enables their evaluation against experimental observation. This talk will present Bayesian statistical methods for system identification, that is, for the estimation of unmeasured parameters and the dynamics of unobserved species in biochemical models described by systems of nonlinear Ordinary Differential Equations (ODE). In addition a means of objectively ranking a number of plausible mathematical models based on their evidential support as assessed by Bayes factors will be presented. A large scale study of the Extra-Cellular Regulated Kinase (ERK) pathway will be discussed where recent Small Interfering RNA (siRNA) experimental validation of the structural predictions made using the computed Bayes factors is presented.