Hafner Jasmin, Hatzimanikatis Vassily
Laboratory of Computational Systems Biotechnology (LCSB), Institute of Chemical Sciences and Engineering (ISIC), School of Basic Sciences (SB), Swiss Federal Institute of Technology (EPFL), 1015 Lausanne, Switzerland.
Bioinformatics. 2021 Oct 25;37(20):3560-3568. doi: 10.1093/bioinformatics/btab368.
Finding biosynthetic pathways is essential for metabolic engineering of organisms to produce chemicals, biodegradation prediction of pollutants and drugs, and for the elucidation of bioproduction pathways of secondary metabolites. A key step in biosynthetic pathway design is the extraction of novel metabolic pathways from big networks that integrate known biological, as well as novel, predicted biotransformations. However, the efficient analysis and the navigation of big biochemical networks remain a challenge.
Here, we propose the construction of searchable graph representations of metabolic networks. Each reaction is decomposed into pairs of reactants and products, and each pair is assigned a weight, which is calculated from the number of conserved atoms between the reactant and the product molecule. We test our method on a biochemical network that spans 6546 known enzymatic reactions to show how our approach elegantly extracts biologically relevant metabolic pathways from biochemical networks, and how the proposed network structure enables the application of efficient graph search algorithms that improve navigation and pathway identification in big metabolic networks. The weighted reactant-product pairs of an example network and the corresponding graph search algorithm are available online. The proposed method extracts metabolic pathways fast and reliably from big biochemical networks, which is inherently important for all applications involving the engineering of metabolic networks.
https://github.com/EPFL-LCSB/nicepath.
Supplementary data are available at Bioinformatics online.
寻找生物合成途径对于生物体代谢工程以生产化学品、污染物和药物的生物降解预测以及次生代谢产物生物合成途径的阐明至关重要。生物合成途径设计的关键步骤是从整合了已知生物学以及新的预测生物转化的大型网络中提取新的代谢途径。然而，对大型生化网络进行高效分析和导航仍然是一项挑战。
在此，我们提出构建代谢网络的可搜索图表示。每个反应被分解为反应物和产物对，并且为每对分配一个权重，该权重根据反应物和产物分子之间保守原子的数量计算得出。我们在一个跨越6546个已知酶促反应的生化网络上测试了我们的方法，以展示我们的方法如何从生化网络中优雅地提取生物学相关的代谢途径，以及所提出的网络结构如何能够应用高效的图搜索算法来改善大型代谢网络中的导航和途径识别。一个示例网络的加权反应物 - 产物对以及相应的图搜索算法可在线获取。所提出的方法能够快速且可靠地从大型生化网络中提取代谢途径，这对于所有涉及代谢网络工程的应用来说本质上都很重要。
https://github.com/EPFL-LCSB/nicepath。
补充数据可在《生物信息学》在线获取。
