Department of Mathematics, University of Michigan, Ann Arbor, MI, United States of America.
PLoS One. 2023 Feb 13;18(2):e0281672. doi: 10.1371/journal.pone.0281672. eCollection 2023.
Agent-based models (ABMs) are an increasingly important tool for understanding the complexities presented by phenotypic and spatial heterogeneity in biological tissue. The resolution a modeler can achieve in these regards is unrivaled by other approaches. However, this comes at a steep computational cost limiting either the scale of such models or the ability to explore, parameterize, analyze, and apply them. When the models involve molecular-level dynamics, especially cell-specific dynamics, the limitations are compounded. We have developed a global method for solving these computationally expensive dynamics significantly decreases the computational time without altering the behavior of the system. Here, we extend this method to the case where cells can switch phenotypes in response to signals in the microenvironment. We find that the global method in this context preserves the temporal population dynamics and the spatial arrangements of the cells while requiring markedly less simulation time. We thus add a tool for efficiently simulating ABMs that captures key facets of the molecular and cellular dynamics in heterogeneous tissue.
基于代理的模型 (ABM) 是理解生物组织中表型和空间异质性所带来的复杂性的一种越来越重要的工具。在这些方面，建模者可以达到的分辨率是其他方法无法比拟的。然而，这需要付出高昂的计算成本，限制了模型的规模或探索、参数化、分析和应用它们的能力。当模型涉及分子水平的动力学，特别是细胞特异性动力学时，限制就更加严重了。我们已经开发出一种全局方法来解决这些计算成本高昂的问题，这种方法可以大大减少计算时间，而不会改变系统的行为。在这里，我们将这种方法扩展到细胞可以根据微环境中的信号改变表型的情况。我们发现，在这种情况下，全局方法可以保留细胞的时间群体动态和空间排列，同时需要明显更少的模拟时间。因此，我们增加了一个工具，用于有效地模拟 ABM，该工具可以捕捉异质组织中分子和细胞动力学的关键方面。
