Applying an inverse homeostasis perspective to simplify the design and implemention of robustly-nearly-homeostatic biological networks

Abstract A nearly-homeostatic system is able to maintain the steady state system output close to a fixed set point, despite varying levels of environmental perturbation. Studying such systems enables the de- sign of synthetic cellular systems able to function consistently under a wide range of environmental conditions. Here we present the inverse homeostasis perspective, a novel approach to studying and de- signing nearly-homeostatic systems. It represents a graphical approach allowing visualization of how each regulatory parameter affects near-homeostatic performance while being more accessible than situation- specific, mathematically complex approaches. Given the difficulty of precise parameter measurement in biology, we focus on adjusting experimentally-determined steady state response curves to implement near- homeostatic systems without requiring explicit parameter measurements. One implication of the inverse homeostasis perspective is an illustration of the much stronger dependence of nearly-perfect integral con- trollers on the properties of the existing system being controlled, compared to the system-independence of perfect integral controllers.