Shiraishi F, Savageau M A
Department of Microbiology and Immunology, University of Michigan Medical School, Ann Arbor 48109-0620.
J Biol Chem. 1992 Nov 15;267(32):22919-25.
When kinetic models of complex biochemical systems are reconstructed from knowledge of the component reactions that have been characterized in vitro, or when values must be assumed for some of the parameters, errors are invariably encountered, and, as a consequence, the resulting model is frequently internally inconsistent. The simplest and most basic manifestations of such logical inconsistency are the failure of the model to exhibit a steady state or to yield a steady state that is in agreement with the actual steady state of the integrated system, or to yield a steady state that is dynamically stable. Models that are consistent may nonetheless be lacking in robustness, which is manifested as a pathological sensitivity to small changes in the values of their parameters. In this paper, we examine the current model of the tricarboxylic acid cycle in Dictyostelium discoideum (see Shiraishi, F., and Savageau, M. A. (1992) J. Biol. Chem. 267, 22912-22918) with regard to these basic indicators of model quality. This may be viewed as a preliminary analysis; the object is to determine whether or not the model is reasonable and worthy of a more refined analysis and, if not, to diagnose the areas in need of modification before further analysis is undertaken. The results demonstrate that the current model of the tricarboxylic acid cycle is self-consistent and possesses a steady state that is in agreement with experimental evidence. However, the results also suggest that this model is not very robust. The high sensitivities of parameters influencing pyruvate metabolism indicate that the experimental characterization of these reactions might be fruitfully re-examined. These high sensitivities lead us to predict that this model of the tricarboxylic acid cycle should be accurate only over a very narrow range in variation of the independent variables. This is verified by the results presented in the following paper (Shiraishi, F., and Savageau, M. A. (1992) J. Biol. Chem. 267, 22926-22933).
当根据体外已表征的组成反应的知识重建复杂生化系统的动力学模型时,或者当必须为某些参数假定值时,总是会遇到误差,结果,所得模型常常在内部是不一致的。这种逻辑不一致的最简单和最基本的表现是模型未能展现出稳态,或者产生与整合系统的实际稳态一致的稳态,或者产生动态稳定的稳态。尽管一致的模型可能缺乏稳健性,这表现为对其参数值的微小变化具有病态敏感性。在本文中,我们根据这些模型质量的基本指标,研究了盘基网柄菌中三羧酸循环的当前模型(见Shiraishi,F.和Savageau,M. A.(1992)J. Biol. Chem. 267,22912 - 22918)。这可以被视为初步分析;目的是确定该模型是否合理且值得进行更精细的分析,如果不合理,则在进行进一步分析之前诊断需要修改的领域。结果表明,三羧酸循环的当前模型是自洽的,并且具有与实验证据一致的稳态。然而,结果也表明该模型不是非常稳健。影响丙酮酸代谢的参数的高敏感性表明,这些反应的实验表征可能值得重新审视。这些高敏感性使我们预测,三羧酸循环的这个模型仅在自变量变化的非常窄的范围内才应该是准确的。这在下一篇论文(Shiraishi,F.和Savageau,M. A.(1992)J. Biol. Chem. 267,22926 - 22933)中给出的结果得到了验证。