TITLE: Random Walks through the Gene Machine: the Ensemble Method of Biological Circuit Identification

SPEAKER: Professor Bernd Schüttler,

TIME: Thursday Oct. 23, 2003 at 4 PM

PLACE: George P. Williams, Jr. Lecture Hall, (Olin 101)

Biological circuit models are a convenient way to describe the complex dynamical properties of the intricate network of coupled gene regulatory and biochemical processes in the cell. I will first give an introduction to the modeling of biological circuits in terms of chemical reaction networks governed by coupled kinetic rate equations. I will then describe the recently developed "Ensemble Method" of biological circuit identification (Battogtokh et al., PNAS 99, p. 16904, 2002), within the context of such network kinetics models, applied to the analysis of time-dependent mRNA and protein profiling data. The approach is based on the notion that a statistical ensemble of model parameters, rather than a poorly defined unique parameter set, should be used to describe sparse, noisy data sets with parameter-rich models. An efficient Monte Carlo method for walking through the parameter space is employed to identify such ensembles of network kinetics models with rate constants and initial conditions consistent with mRNA and protein profiling data. Applications to the quinic acid (qa) gene cluster and to the biological clock in Neurospora crassa will be discussed.