Bistability preserving model reduction in apoptosis

Abstract

Biological systems are typically very complex and need to be reduced before they are amenable to a thorough analysis. Also, they often possess functionally important dynamic features like bistability. In model reduction, it is sometimes more desirable to preserve the dynamic features only than to recover a good quantitative approximation. We present an approach to reduce the order of a bistable dynamical system significantly while preserving bistability and the switching threshold. These properties are important for the operation of the system in the context of a larger network. As an application example, a bistable model for caspase activation in apoptosis is considered.