Functional integration of a metabolic network model and expression data without arbitrary thresholding
Functional integration of a metabolic network model and expression data without arbitrary thresholding
复制标题
DOI：
10.1093/bioinformatics/btq702
复制
发表时间：
2011-02-15
期刊：
影响因子：
5.8
通讯作者：
Papin, Jason A.
中科院分区：
文献类型：
--
作者：
Jensen, Paul A.;Papin, Jason A.
动机：通量平衡分析 (FBA) 已被广泛用于分析各种生物体中基因组规模、基于约束的代谢模型。最近，通过整合代谢基因和蛋白质的高通量表达谱，此类模型的预测准确性得到了提高。然而，FBA 的扩展通常需要将此类数据先验离散成“开”或“关”的基因或蛋白质组。这个过程需要选择相对主观的表达阈值，通常需要多次迭代和细化来捕获表达动态并保留模型功能。结果：我们提出了一种将表达数据从一组环境、遗传或时间条件映射到代谢网络模型上的方法，而不需要任意的表达阈值。差异表达代谢调整 (MADE) 利用基因或蛋白质表达变化的统计显着性来创建最准确地概括表达动态的功能代谢模型。 MADE 用于生成一系列模型，反映酿酒酵母从发酵呼吸向甘油呼吸转变过程中的代谢调整。计算出的基因状态与 98.7% 可能的表达变化相匹配，所得模型捕获了代谢转变的功能特征。
Motivation: Flux balance analysis (FBA) has been used extensively to analyze genome-scale, constraint-based models of metabolism in a variety of organisms. The predictive accuracy of such models has recently been improved through the integration of high-throughput expression profiles of metabolic genes and proteins. However, extensions of FBA often require that such data be discretized a priori into sets of genes or proteins that are either 'on' or 'off'. This procedure requires selecting relatively subjective expression thresholds, often requiring several iterations and refinements to capture the expression dynamics and retain model functionality.Results: We present a method for mapping expression data from a set of environmental, genetic or temporal conditions onto a metabolic network model without the need for arbitrary expression thresholds. Metabolic Adjustment by Differential Expression (MADE) uses the statistical significance of changes in gene or protein expression to create a functional metabolic model that most accurately recapitulates the expression dynamics. MADE was used to generate a series of models that reflect the metabolic adjustments seen in the transition from fermentative-to glycerol-based respiration in Saccharomyces cerevisiae. The calculated gene states match 98.7% of possible changes in expression, and the resulting models capture functional characteristics of the metabolic shift.