The Research Of Optimal Design Are Based On The Data Of ¹³C Metabolic Flux Analysis

With the progress of science and technology, fermentation technology was a great development, it has been to produce the desired product by artificially controlling the human transformation of microorganisms, while, in order to further meet the specific needs of the human organism, the researchers analyzed the study of cell metabolism regulate, the development of a new field: Construction of a new production of specific metabolic pathways desired product.While developing countries in the fermentation engineering and metabolic flux analysis in the metabolic engineering becomes more important, fermentation engineering time, substrate material, the substrate concentration is now able to control the application of computer technology to manage, but for deep level cell metabolic flux analysis, but few computer technology research, especially on metabolic model improvement and optimization of metabolic flux data, domestic research is relatively rare.With the objective of increasing demand for metabolic engineering, metabolic flux analysis system more and more urgent, for this use of ¹³C-labeled metabolic flux model system to study, metabolic parameters and optimize the pathway, so that bacteria body metabolites in the direction of the desired output.The main work done as follows:（1） build great ¹³ C metabolic flux analysis model and use the program.The method of using ¹³C-labeled labeled metabolic system to study information obtained as well as the model parameters, to find a more excellent metabolic model, so-throughput analysis more convenient, efficient and easier to highlight the hidden metabolic system in flux function, which is easier to obtain flow information.（2） An improved optimization algorithm, optimized metabolic model building.According to determine metabolic model, linear programming problem into a mathematical problem, this problem using a dimensionality reduction algorithm.Dimensionality reduction algorithm to make improvements, add gradient search algorithm to accelerate the convergence rate.（3） using the training set model for improved algorithm for testing.And particle swarm optimization algorithm to optimize performance is compared, and the convergence of the algorithm is to be verified, show that this metabolic flux analysis model constructed, improved dimension reduction algorithms have better performance, faster determination of metabolic flux information, increase the metabolic performance of the cell, to optimize the effect of the metabolic system.（4） Construction of Saccharomyces cerevisiae producing L- lactate metabolism model to analyze the metabolic pathway model, changing metabolic pathway flux, the purpose of improving the metabolic rate of the property.