| Metabolic networks are complex network systems composed of all metabolic reactions or pathways in vivo and its material flow,i.e.the flux,and the associated regulatory processes.As an abstract description of the interaction of metabolic reactions,in-depth study of metabolic networks will help people better understand and use the process of cell metabolism,which greatly promote the development of metabolic engineering,fermentation engineering,bio-manufacturing and other industries,and ultimately move towards synthetic biology,the ultimate goal of the biological industrialization.In addition,as a typical complex network,the research of metabolic networks will have dramatically referential significance for other complex networks,such as social network and literature network.Metabolic network flux describes the metabolic state of a cell at a given time and is the ultimate identifier of a cell's functional state.Metabolic network flux reflects species properties and is determined by the biological adaptation of the organism to the environment.Metabolic flux has strong stability under steady state environment,which embodies the evolutionary behavior of organisms and can be used as phylogenetic analysis data.The research of metabolic network flux has important theory significance for the study of species evolution and understanding of the intrinsic process of life,and also has broad application prospects in disease diagnosis and treatment,drug design and so on.In parallel with other-omics,13C-fluxomics has documented large amounts of information regarding the flux distribution of numerous organisms.Comparative analysis and theoretical simulation of these results prove to be an extraordinarily powerful method for exploring the hidden meanings among numerous fluxomics results.In recent years,with the gradual completion of whole-genome sequencing work for thousands of species and the rapid development of high-throughput technology,the massive metabolic network data of different scales of various species have been produced.How to effectively share,manage and analyze these data is a great challenge for metabolomics and is highly valued by the researchers.In this paper,metabolic network flux sharing and storage,flux comparison,network alignment and reconstruction are further studied.The main research work and innovation of this paper are as follows.(1)Acquisition and standardization of metabolic network flux data.At present metabolic network flux data are mostly published in the form of research papers and scattered,which mean that data acquisition is difficult.The metabolites,catalytic enzymes,reactions and flux units in these papers are variously expressed in different formats resulting in different experimental data cannot be horizontally compared.This paper uses manual search and network submission to acquire metabolic network flux data.Then the data are scanned and verified to filter out pollution reactions and convert into a unified definition format of metabolic network referring to KEGG.(2)Alignment of metabolic network flux distribution similarity.Algorithms based on vector and stoichiometric matrix are proposed in this paper to calculate a similarity score of metabolic network flux distribution.Currently,metabolic network alignment algorithms mostly transform metabolic network to the form of graph,but rarely consider the flux as the weight.Based on this method,flux is added in this paper.The metabolic network is then transformed into a weighted directed graph and the flux is taken as the computing factor of the weighted adjacent matrix.Through this transformation,considering the matching of vertices and edges,the metabolic network flux distribution alignment is formalized into an integer programming problem.A topology-based algorithm is proposed to align the metabolic network flux distribution.The result is visualized and the enzymes matched are given in detail.The conservative modules and GAPS in the metabolic network are also analyzed.(3)Metabolic network alignment based on hypergraph.Previous metabolic network alignment algorithms represent metabolic networks in the form of graph.But such transformation results in incomplete information representation,parallel edges,and increased computational cost.In this paper,the metabolic network is represented by a hypergraph.By adding null vertices,the matching of vertices and hyperedges in the hypergraph is represented uniformly by a high order tensor.The hypergraph alignment is transformed into an optimization problem of maximizing the score objective function.Then BNHOP(BiNormalized Higher-order Power Method)is proposed to solve the problem.(4)An integrated platform for storage,analysis and visualization of metabolic network flux big data.Although there are lots of biological network databases,metabolic network flux database is not available to share and manage flux data.To this end,we present the CeCaFDB(The Central Carbon Metabolic Flux Database,url:http://www.cecafdb.org),which,to the best of our knowledge,is the first professionally designed online database for fluxomics data regarding the central carbon metabolic systems of microbes and animal cells.CeCaFDB is published in the Nucleic Acids Research(2015 IF=9.202,doi: 10.1093/nar/gku1137)entitled “CeCaFDB: a curated database for the documentation,visualization and comparative analysis of central carbon metabolic flux distributions explored by 13C-fluxomics”.The CeCaFDB encompasses 581 cases of flux distribution in 118 literatures among 36 organisms and integrates tools for metabolic network flux alignment and visualization.CeCaFDB is an open platform where all data can be freely downloaded and shared.Researchers can also submit data to CeCaFDB,use the tools provided by CeCaFDB for data analysis,and share data with others.(5)Reconstruction and analysis of tea metabolic network.The quality of tea is determined by the various functional compounds present in it,which may have been of nutrition and health functions.The functional compounds present in tea are produced by the metabolic reactions catalyzed by enzymes encoded by genes.These metabolic reactions form the metabolic network.Reconstruction and analysis of the tea metabolic network is important for understanding metabolic reactions and nutrients in tea and improving tea quality.This paper constructs a local enzyme database utilizing the data from BRENDA.On this basis,an algorithm was designed to construct and visualize metabolic network from EST sequence of tea.The topological characteristic and biological implication of the constructed network are computed and extracted by a series of analytical methods. |
PDF Full Text Request