Kaznessis Yiannis N
Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Ave SE, Minneapolis, MN 55112, USA.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:6408-11. doi: 10.1109/IEMBS.2009.5333516.
Reacting systems away from the thermodynamic limit cannot be accurately modeled with ordinary differential equations. These continuous-deterministic modeling formalisms, traditionally developed and used by chemical engineers can be distinctly false if the number of molecules of reacting chemical species is very small, or if reaction events are very rare. Then stochastic-discrete representations are appropriate. Importantly, in cases where in a network of reactions there are some parts that must be modeled discretely and stochastically, yet others can be modeled continuously and deterministically, the need for development of multiscale models emerges naturally. In computational synthetic biology, such cases arise often. In this work we present the development of multiscale models for synthetic biology applications, demonstrating accuracy, computational efficiency and utility.
远离热力学极限的反应系统无法用常微分方程进行精确建模。这些传统上由化学工程师开发和使用的连续确定性建模形式，如果反应化学物种的分子数量非常少，或者反应事件非常罕见，可能会明显错误。此时随机离散表示法就适用了。重要的是，在一个反应网络中，有些部分必须用离散和随机的方式建模，而其他部分可以用连续和确定性的方式建模，那么自然就出现了开发多尺度模型的需求。在计算合成生物学中，这种情况经常出现。在这项工作中，我们展示了用于合成生物学应用的多尺度模型的开发，证明了其准确性、计算效率和实用性。
