Design And Application Research Of A Multi-Cell Metabolic Network Reconstruction System

Metabolic Network Reconstruction is a process when all the genes, proteins, enzymes, reactions and metabolites that participate in the metabolism are identified, categorized and interconnected to form a metabolic network. It can be used to reconstruct and analyze the metabolic network of organisms from an integral perspective.According to present situation and deficiency of current tools for metabolic network reconstruction, we have designed and implemented a Multi-Cell Metabolic Network Reconstruction System (MCMNRS) based on the formal metabolic network reconstruction system for microorganisms. This system can not only reconstruct the metabolic networks of microogranisms but can also reconstrut the metabolic networks of multi-cell organisms. The metabolic networks of organisms can be reconstructed efficiently and automatically from gene annotation files, protein expression data or gene chip data. The main work of this paper is as follows:(1) According to the former work in our lab and after the feature analysis of the enzyme names and gene products, we designed and implemented a Hybrid Participle Enzyme Identification Algorithm. We find the optimal arguments for this algorithm and validated the efficiency and utility of this algorithm by comparing it to the PathoLogic algorithm.(2) We designed the framework for MCMNRS and implemented the Input Data Process Module, the Large-Scale Network Reconstruction Module and the metabolic Network Analysis Module. We also improved the DB Information Extraction Module and the Standard Network Reconstruction Module. After that we integrated all the modules into a whole Multi-Cell metabolic Network Reconstruction System.(3) We reconstruted the metabolic networks of Ratus norvegicus, Mus musculus, Canis familiaris and Equus caballus based on the gene annotation files for these organisms, and analyzed the gene identificatioin ratio, the enzyme classificatioin and the substrate specificity of enzymes in those networks.(4) We reconstructed the metabolic networks for the 12 tissues of Mus musculus based on the gene chip data of these tissues and we analyzed the specificity of these networks to find out the special reactions in these networks. We also performed the network clustering in three different ways: network distance, the network reaction matrix and the specific network reaction matrix. At present, MCMNRS has been used to reconstruct the metabolic networks of more than 60 microorganisms and many kinds of mammalians. It has also been used to reconstruct the metabolic networks of tissues based on the protein expression data or gene chip data. All this has gained good research results.