【作者】 齐存森；

【导师】 滕虎；

【作者基本信息】 大连理工大学 ， 生物化学与分子生物学， 2011， 硕士

【摘要】 代谢工程20年来的发展表明,从基因组水平对其代谢网络进行建模与分析十分必要。如今随着系统生物学等相关学科的发展,虽然已经建立了87个基因组尺度代谢网络模型,但是每个模型的建立都很耗时耗力,而且对这些模型的更新同样是十分不便。分析原因主要是由于生物自身数据的复杂性以及由此造成的KEGG对构建代谢网络模型的不适应性,所以有必要建立一个更适合代谢网络建模的数据库,实现代谢网络建模的系统化与规范化操作。克雷伯氏菌作为一种兼性厌氧菌,被广泛地应用在各种发酵过程中,特别是在发酵甘油生产1,3-丙二醇的过程中,克雷伯氏菌以其较高的转化率和1,3-丙二醇生产能力而最受关注。但是截至目前,尚未能找到稳定地提高其1,3-丙二醇生产能力的有效改造策略,对菌株的改造工作往往是顾此失彼,究其原因是对克雷伯氏菌的代谢调控机制缺乏系统水平的理解,所以对其基因组尺度代谢网络模型进行建立与分析显得尤为重要。但是,迄今还未见有关克雷伯氏菌基因组尺度代谢网络模型的报道。为此,本文首先以KEGG为基础,综合其它数据库的数据和文献资料,对已有代谢数据进行重新整理,建立起了更加适合代谢网络建模与分析的数据库,其中包含七个独立表,2471个生物的8601条反应记录。然后利用建立的数据库对K. pneumoniae 342的基因组尺度代谢网络进行了建模：首次提出以基因作为边,以反应物作为节点构建的结构分析模型,使其代谢网络层次能够反映基因信息,所建立的模型共包含7735条记录,涉及1209个基因,1998个反应的共计2143个反应对。最后,通过对其应用复杂网络理论进行分析发现,K. pneumoniae 342菌株的基因组尺度代谢网络的聚合系数为0.038,度分布符合幂率分布,属于无标度网络,对随机节点删除具有较高的鲁棒性,但对目的性攻击比较脆弱。对巨强连通体的分析表明最大路径长度为30,半径为16,特征路径长度为8.155,符合小世界网络特征,这使得环境因素造成的影响能够迅速传遍网络,对菌体快速适应复杂多变的环境具有重要意义。此外,本文利用建立的本地数据库,建立了K. pneumoniae 342菌株用于动力学分析的基因组尺度代谢网络模型。但由于调控信息等的缺乏,暂未进行动力学模拟与分析。对K. pneumoniae 342基因组尺度代谢网络模型的建立与分析,填补了克雷伯氏菌在这方面研究的空白,为今后对其进行基因组尺度的代谢调控动力学研究奠定了基础。更多还原

【Abstract】 It is necessary to construct and analyze the genome-scale metabolic network after the 20 years development of metabolic engineering. Although, following the development of system biology and its related study, it has been constructed over 87 genome-scale network now, each one is very time-consuming, labor-intensive, and troublesome to update. Since the complexity of biological data led to the KEGG could not directly be used to construct the metabolic network, it’s necessary to build a new database, more suitable to construction of metabolic network.Klebsiella pneumoniae as a facultative anaerobe is widely used in various fermentation processes, especially in the bioconversion of glycerol to produce 1,3-propanediol due to its high conversion rate and the 1,3-propanediol production capacity. However, it has not been developed with an available way to steadily increase the yield of 1,3-propanediol through gene engineering heretofore, and it is particularly important to model and analyze its genome-scale metabolic network. Regretfully, there was no publication about the genome-scale metabolic network of Klebsiella pneumoniae.Therefore, we firstly establish a database from databases such as KEGG and Brenda. This new metabolic network database contains seven separate tables and 8601 chemical reaction records of 2471 organism, which actually suit for the modeling and analysis of metabolic network. And then, we constructed the genome-scale metabolic network of K. pneumoniae 342 based on the new database. This network totally contains 7735 records,1209 genes,2143 reaction pairs comes from 1998 reactions. The metabolic network is constructed by choosing the gene as edge and the compounds as node, respectively, which is advantage in reflecting gene information at the level of metabolic network. Moreover, it is found that the genome scale metabolic network of K. pneumoniae 342 has clustering coefficient of 0.038 which accords with power law connection degree distribution so it belongs to scale-free network which has a higher robustness against random node deletion and weakly purposeful attacking according to the theory of complex network. It also shows that the maximum path length of giant strongly connected component is 30, the radius is 16 and its characteristic path length is 8.155 which is consistent with the characteristics of small-world network. The small-word network could response quickly to the variation of the environmental factors and made the organism to adapt to the complex environment. It shows that the construction of metabolic network is convenient, fast and accurate according to our database. Therefore, the construction of genome-scale metabolic network of K. pneumoniae 342 filled the blank of this area and contributed to its dynamic research in future.更多还原