Naslov
Dynamic Metabolic Flux Anbalysis by Decom-position and Principal Components

Autori
Kurtanjek, Želimir

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
XX. skup kemičara i kemijskih inženjera : knjiga sažetaka / Vasić-Rački, Đurđa - Zagreb : Hrvatsko društvo kemijskih inženjera i tehnologa (HDKI), 2007

Skup
Skup kemičara i kemijskih inženjera (20 ; 2007)

Mjesto i datum
Zagreb, Hrvatska, 26.02.2007. - 01.03.2007

Vrsta sudjelovanja
Poster

Vrsta recenzije
Domaća recenzija

Ključne riječi
metabolic networks; principal components

Sažetak
Metabolic networks of biological cells are very complex and selfregulated reaction systems which enable goal oriented and adaptive responses of living organisms on perturbations in surroundings. Modeling and mathematical analysis enables advance of control of biotec-nological processes on molecular level by genetic engineering, but also on macroscopic level by improvement of processes in a bioreactor. Metabolic flux analysis is commonly performed under steady state assumption which enables evaluation of the fluxes from knowledge of stoichiometric matrix and fundamental constraints, but analysis of the fluxes under dynamic conditions reveals mechanisms of imbedded enzyme control and is unreli-able due to uncertainties of kinetic models. Here are applied methods of decomposition and principal components with aim of essential simplification of the analysis. The method is based on the numerical procedure of singular value decomposition (SVD) applied to the model of dynamic balances. Analyzed are dynamic responses of the metabolite concentra-tions of E. coli central metabolism during 20 seconds of impulse perturbation of extracellu-lar glucose(1). Decomposition of the original model(2) based on 14 mass balances with 160 kinetic parameters given as a system of strongly coupled and stiff ordinary differential equations gave four principal components which account for 85 % of dynamic variations of the model metabolites. Statistical analysis of errors between the model and experimental data shows significant improvement of the predictions in comparison with the original model. Results obtained by decomposition enable additional simplification in the postponed kinetic modeling and more reliable insight into enzyme control mechanisms leading to more reliable predictions of effects of genetic engineering modifications.

Izvorni jezik
Engleski

Znanstvena područja
Biotehnologija

POVEZANOST RADA