Hybrid computational modeling methods for systems biology

Systems biology models are typically considered across a spectrum from mechanistic to abstracted description; however, the lines between these forms of modeling are increasingly blurred. Ever-increasing computational power is providing novel opportunities for bridging time and length scales. Furthermore, despite biological mechanisms or network topology often ill-defined, the acquisition of high-throughput data leaves modelers with the desire to leverage available measurements. This review surveys modeling tools in which two or more mathematical forms are blended to describe time-dependent processes in a multivariate system. While most commonly manifested as continuous/discrete description, other forms such as mechanistic/inference or deterministic/stochastic hybrid models can be generated. Recent innovations in hybrid modeling methodologies and new applications illustrate advantages for combining model formats to gaining biological systems level insight.

Automated summary

This review surveys hybrid modeling methodologies that blend two or more mathematical forms to describe time-dependent processes in a multivariate system. These models can bridge different scales and forms, providing new opportunities for studying biological systems.