Biochemical mechanism studies often assume statistical models derived from Michaelis–Menten kinetics, which are used to approximate initial reaction rate data given the concentration level of a single substrate. In experiments dealing with industrial applications, however, there are typically a wide range of kinetic profiles where more than one factor is controlled. We focus on optimal design of such experiments requiring the use of multifactor hybrid nonlinear models, which presents a considerable computational challenge. We examine three different candidate models and search for tailor-made D- or weighted-A-optimal designs that can ensure the efficiency of nonlinear least squares estimation. We also study a compound design criterion for discriminating between two candidate models, which we recommend for design of advanced kinetic studies. Supplementary materials accompanying this paper appear on-line

中文翻译：

---

混合非线性模型实验的优化设计，在扩展 Michaelis-Menten 动力学中的应用

生化机制研究通常采用源自 Michaelis-Menten 动力学的统计模型，这些模型用于在给定单一底物浓度水平的情况下近似初始反应速率数据。然而，在处理工业应用的实验中，通常有多种动力学曲线，其中控制不止一个因素。我们专注于需要使用多因素混合非线性模型的此类实验的优化设计，这提出了相当大的计算挑战。我们检查了三种不同的候选模型，并寻找可以确保非线性最小二乘估计效率的定制 D 或加权 A 最优设计。我们还研究了用于区分两个候选模型的复合设计标准，我们建议将其用于高级动力学研究的设计。