<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets"^^ . "Identifying essential genes in pathogens facilitates the identification of the corresponding proteins as potential drug targets and is the basis for understanding the minimum requirements for a synthetic cell. However, the experimental assessment of gene essentiality is resource-intensive and not feasible for all organisms, especially pathogens. Thus, the computational identification of new drug targets has become an important pursuit in biomedical research. In particular, essential metabolic enzymes have been successfully targeted by specific drugs. For directed drug development, the prediction of essential genes, especially in metabolic networks, is needed. In this thesis, I describe our development of a graph-based investigation tool aimed at finding possible deviations in a mutated network by knocking out particular reactions, and examining its producibility with a breadth-first search algorithm. We showed that this approach performed well at predicting new targets for antimalarial drugs. In addition, we analyzed the metabolic networks of bacteria and developed a machine learning approach based on various graph-based descriptors, including our own developed descriptor, that were potentially associated with the robustness and stabilization of metabolic networks. These descriptors were related to gene essentiality and included flux deviations, centrality and shortest paths. Besides these network topological features, we also used genomic and transcriptomic features, such as sequence characteristics and co-expression properties, as descriptors. The machine learning technique was developed to identify drug targets in metabolism. The metabolic networks of Escherichia coli, Pseudomonas aeruginosa and Salmonella typhimurium were analyzed. The well-studied metabolic network of Escherichia coli was used because it was an ideal model for formulating and validating our method. With publicly available genome-wide knockout screens, it was shown that topological, genomic and transcriptomic features describing the network are sufficient for defining drug targets. Furthermore, we tested our method across bacterial species and strains by using the experimental data from the genome-wide knockout screens of one bacterial organism to infer essential genes for another related bacterial organism. Our method is generic, and it enables the prediction of essential genes from a bacterial reference organism to a related query organism without any knowledge about the essentiality of the genes of the query organism. In general, such a method is beneficial for inferring drug targets when experimental data about genome-wide knockout screens are not available for the investigated organism."^^ . "2011" . . . . . . . . "Kitiporn"^^ . "Plaimas"^^ . "Kitiporn Plaimas"^^ . . . . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (PDF)"^^ . . . "kplaimas_CompletePhDThesis.pdf"^^ . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (Other)"^^ . . . . . . "indexcodes.txt"^^ . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (Other)"^^ . . . . . . "lightbox.jpg"^^ . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (Other)"^^ . . . . . . "preview.jpg"^^ . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (Other)"^^ . . . . . . "medium.jpg"^^ . . . "Computational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets (Other)"^^ . . . . . . "small.jpg"^^ . . "HTML Summary of #12888 \n\nComputational Analysis of the Metabolic Network of Microorganisms to Detect Potential Drug Targets\n\n" . "text/html" . . . "004 Informatik"@de . "004 Data processing Computer science"@en . .