Multiscale Computational Models of Complex Biological Systems

Abstract

Integration of data across spatial, temporal, and functional scales is a primary focus of biomedical engineering efforts. The advent of powerful computing platforms, coupled with quantitative data from high-throughput experimental platforms, has allowed multiscale modeling to expand as a means to more comprehensively investigate biological phenomena in experimentally relevant ways. This review aims to highlight recently published multiscale models of biological systems while using their successes to propose the best practices for future model development. We demonstrate that coupling continuous and discrete systems best captures biological information across spatial scales by selecting modeling techniques that are suited to the task. Further, we suggest how to best leverage these multiscale models to gain insight into biological systems using quantitative, biomedical engineering methods to analyze data in non-intuitive ways. These topics are discussed with a focus on the future of the field, the current challenges encountered, and opportunities yet to be realized.

A Green's function method for simulation of time-dependent solute transport and reaction in realistic microvascular geometries

“…Such applications arise naturally in the multi-scale modelling of biological systems (Walpole et al, 2013) and in studies addressing the goals of the Physiome project (Hunter, 2004). For example, in the development of cancer chemotherapies, transport characteristics influence the ability of drugs to penetrate all parts of a tumour (Beard, 2006;Baish et al, 2011).…”

Agent-based computational model investigates muscle-specific responses to disuse-induced atrophy

“…However, these previous models of muscle adaptation have been based on phenomenological equations that describe measurements of tissue responses to alterations in the mechanical environment and therefore are unable to capture and study the effects of molecular signals and cellular behaviors on muscle tissue adaptation. Agentbased computational modeling, an approach that has been applied to study the underlying mechanisms of vascular remodeling, is well suited for investigating how molecular signals and cell behaviors integrate to cause tissue-level adaptations (4,18,21,28,64,71). Agent-based models (ABMs) represent individual biological cells as computational agents and can simulate how collections of cells within a tissue will respond emergently to literature-derived rules.…”

How to deal with parameters for whole-cell modelling

“…Numerous models now face the challenge of being large (in terms of numbers of species and parameters represented), multi-scale and/or hybrid in nature (incorporating multiple different mathematical representations) [23,[28][29][30]. The most ambitious models to date-the WCMs-aim to provide faithful in silico representations of real biological cells, including all major cellular processes and components, and are both very large scale and hybrid (figure 1) [31][32][33].…”

“…Approaches and difficulties relating to submodel construction and parametrization are discussed in subsequent sections, but even once we have extensively characterized, carefully parametrized and validated models describing different aspects of a complete system, combining these will remain a formidable challenge [28,29,34]. Submodels were successfully integrated in the original WCM by assuming independence on short time scales, and defining a collection of cell variables that could be shared among the various distinct cellular processes [31].…”

“…Submodels were successfully integrated in the original WCM by assuming independence on short time scales, and defining a collection of cell variables that could be shared among the various distinct cellular processes [31]. However, we should bear in mind the difficulties faced in other scientific fields, where complex systems are frequently studied using multi-scale [29] and multi-physics modelling approaches [35]. In climate modelling, for example, multi-physics approaches combine models of atmospheric chemistry with models of ocean currents, which are then coupled, using, for example, partial differential equations.…”