Plenary Lecture

Synergy Analysis in Reaction Systems

Profesor Zelimir Kurtanjek
University of Zagreb
Faculty of Food Technology
CROATIA
E-mail: zkurt@pbf.hr

Abstract: This work introduces the concept of analysis of synergy interactions in chemical (biochemical) reaction systems. The systems are defined by mass balances for reacting species, reaction network given by a stoichiometric matrix, reaction (kinetic) parameters, and parameters of interaction between the reaction system and a surrounding. Systems analysis is based on statistical evaluation of numerical results of computer simulation of effects of a set of uncorrelated stochastic parameters, defined by corresponding ranges and probability distributions. From the simulation data are extracted parameter global sensitivities and synergy effect of each parameter in interaction with the whole system. The global parameter sensitivities are defined as ratios of conditional variances of expected values for each parameter and the total dispersion of simulation data. The synergy effects are defined as the relative difference between expected values of the conditional variances with exclusion of each parameter and the variance of the conditional expected value of the parameter. Numerical evaluations are performed by use of sampling the parametric space by Lissajous type of curves and the Fourier Amplitude Sensitivity Test (FAST) algorithm. The method is applied for analysis of propagation hepatitis B virus (HBV) in a stochastic single cell model, and for metabolic control analysis of a branched network. The results for virus propagation show dominant synergy effect of the system parameters, compared to negligible influence of individual parameters. The metabolic control analysis show that under homeostatic conditions influence of the individual enzymes is dominant and the key enzyme responsible for flux regulation is determined. However, under perturbed homeostatic conditions influence of synergy effects dominates over impact of individual parameters. Applicability of the proposed concept is discussed in view of model improvement and potential control of reaction systems.

Brief Biography of the Speaker:
Zelimir Kurtanjek was born in 1946 in Zagreb, Croatia. He graduated in 1971 with engineering degree in physics from the Department of Sciences, University of Zagreb. He completed postgraduate studies in Technical Cybernetics at Faculty of Technology, University of Zagreb. In 1975 he enrolled at the postgraduate study in chemical engineering at the Department of Chemical Engineering, The University of Houston, TX, USA. In 1979 he received doctoral degree under mentorship of Prof. Dan Luss in the Laboratory for Reaction Engineering. He completed his postdoctoral studies with Prof. G. Froment at the Department of Chemical Engineering, University of Gent, Belgium. During 1991. he was a visiting professor through EU project TEMPUS at the Department of Biological Sciences, University of Ulster, Coleraine, Northern Ireland. Since 1980 he is employed at the University of Zagreb, Faculty of Food Technology and Biotechnology, University of Zagreb as a professor of chemical engineering. He is teaching reactor engineering, mathematical modeling and process control to students of biochemical, chemical and food engineering. In his scientific work he is interested in modeling and control of reactors, modeling of bioprocesses and food processes, and application of AI methods in modeling and process control. He has published over 50 papers in international and national journals. Since 1976 he is a member of American Institute of Chemical Engineers, a delegate of Croatia in European Federation of Biotechnology, and also is a member of Croatian Society of Chemical Engineers, Croatian Society of Biotechnology, and Croatian Technical Academy. He is editor in chief of Chemical and Biochemical Engineering Quarterly and a member of the editorial board of Food Technology and Biotechnology journals.