Complex Systems, Big Data and Informatics Initiative (CSBII), Lincoln University, Christchurch, New Zealand.
Biosystems. 2020 May;191-192:104128. doi: 10.1016/j.biosystems.2020.104128. Epub 2020 Mar 9.
Biological systems are difficult to understand complex systems. Scientists continue to create models to simulate biological systems but these models are complex too; for this reason, new reduction methods to simplify complex biological models into simpler ones are increasingly needed. In this paper, we present a way of reducing complex quantitative (continuous) models into logical models based on time windows of system activity and logical (Boolean) models. Time windows were used to define slow and fast activity areas. We use the proposed approach to reduce a continuous ODE model into a logical model describing the G1/S checkpoint with and without DNA damage as a case study. We show that the temporal unfolding of this signalling system can be broken down into three time windows where only two display high level of activity and the other shows little or no activity. The two active windows represent a cell committing to cell cycle and making the G1/S transition, respectively, the two most important high level functions of cell cycle in the G1 phase. Therefore, we developed two models to represent these time windows to reduce time complexity and used Boolean approach to reduce interaction complexity in the ODE model in the respective time windows. The developed reduced models correctly produced the commitment to cell cycle and G1/S transfer through the expected behavior of signalling molecules involved in these processes. As most biological models have a large number of fast reactions and a relatively smaller number of slow reactions, we believe that the proposed approach could be suitable for representing many, if not all biological signalling networks. The approach presented in this study greatly helps in simplifying complex continuous models (ODE models) into simpler models. Moreover, it will also assist scientists build models concentrating on understanding and representing system behavior rather than setting values for a large number of kinetic parameters.
生物系统是难以理解的复杂系统。科学家们继续创建模型来模拟生物系统，但这些模型也很复杂；因此，越来越需要新的简化方法将复杂的生物模型简化为更简单的模型。在本文中，我们提出了一种基于系统活动时间窗口和逻辑（布尔）模型将复杂的定量（连续）模型简化为逻辑模型的方法。时间窗口用于定义慢活动区和快活动区。我们使用所提出的方法将连续的 ODE 模型简化为描述 G1/S 检查点的逻辑模型，有无 DNA 损伤作为案例研究。我们表明，这个信号系统的时间展开可以分解为三个时间窗口，其中只有两个显示高水平的活动，而另一个显示很少或没有活动。两个活跃的窗口分别代表细胞承诺进入细胞周期并进行 G1/S 转换，这是 G1 期细胞周期的两个最重要的高级功能。因此，我们开发了两个模型来表示这些时间窗口，以降低时间复杂度，并在各自的时间窗口中使用布尔方法来降低 ODE 模型中的相互作用复杂度。所开发的简化模型通过参与这些过程的信号分子的预期行为正确地产生了对细胞周期的承诺和 G1/S 转移。由于大多数生物模型具有大量的快速反应和相对较少的缓慢反应，我们相信所提出的方法可能适用于表示许多（如果不是全部）生物信号网络。本研究提出的方法极大地帮助将复杂的连续模型（ODE 模型）简化为更简单的模型。此外，它还将帮助科学家构建专注于理解和表示系统行为的模型，而不是为大量动力学参数设置值。
