Investigation of `objective functions´ in S. cerevisiae

Explicitly or implicitly, much of the research in molecular biology over the last decades has been based on the theory of deterministic genetics, which assumes a direct path from gene to protein to function and the presence of responses of a system to external perturbations. So far little is known how cells integrate signals generated by different receptors into physical response, few biological systems have produced sufficient experimental data that allowed the generation of mathematical models that simulate the dynamic behavior of the system. Systems biology, with its focus on the dynamic networks offers a potential to overcome some or many of these limitations. Two principle aspects that appear to be inseparable are that systems biology involves the use of mathematical models and high-throughput `omics`-data. A contrived model, based on experimental data, should understand the complexity and interactions among various parts of the system and not merely organize and catalog them into arbitrary classifications. In this project we will examine the predictive capacity of linear and non-linear network objectives by comparison of FBA-based (constraint-based flux balance analysis) flux predictions, by means of metabolomics and 13C- labelled flux data from Saccharomyces cerevisiae under five different environmental conditions that enforce different modes of metabolic operation (four different carbon sources: glucose, galactose, ethanol, and succinate with 5g/L concentration; one of which, glucose, at 0.5 and 5g/L to enforce respiratory and respirofermentative metabolism). Main emphasis of the work will be the systematical testing of a set of `objective functions` and all their possible permutations with or without additional constraints, to spot pertinent combinations to predict in vivo fluxes by in silico fluxes (FBA). Due to the choice and combination of `objective functions` and constraints that commonly predefine the degree of freedom in terms of specific pathway usage, appropriate permutations might be potentially used to predict and estimate metabolic behavior. On the basis of the genome scale model of yeast it will be possible to calculate the theoretical fluxes and their distribution in the model-organism yeast FY strain for the growth under specific conditions. `Objective functions` that will be implemented in the constraint-based FBA are maximization of biomass yield, the maximization of ATP yield, and the minimization of the overall intracellular flux. Quantitative metabolomics and 13C-flux directions based on different growth conditions as described above on different carbon substrates (glucose, galactose, ethanol, and succinate) in batch cultures of yeast cells will be performed. The obtained data will be the basis for network embedded thermodynamic analysis (NET-analysis) that might indicate reactions that are possibly regulated. To gain new insights into the metabolic network of Saccharomyces cerevisiae, especially questions in terms of general principles how metabolism and its regulation will work, it will be necessary to vary growing conditions (different carbon sources and concentrations) for comparison of in vivo and in silico fluxes. Combination of these two computational and practical approaches is expected to result in advanced knowledge of general principles for the operation of metabolism. Although it was done once now in bacteria Saccharomyces cerevisiae (yeast) represents an organism with a significantly higher complexity and therefore it is proposed for the first time to compare, if such simple principles can be transferred from bacteria to eukaryotes.