Information-theoretical Model for Spatio-temporal Sensing of Gradient

Gradient sensing is one of the basic cellular information processing. Experimental analysis has revealed that various gradient sensing kinetics from yeast to human lymphocyte have extremely high efficiency to detect very small gradient even though sensory processes involve strong stochasticity. In addition, each organism seems to use different strategies and mechanisms of gradient sensing, which can be categorized at least into spatial and temporal sensing. Because of this diversity in mechanisms, mathematical models have been developed for individual sensing phenomena even though they share universal property of high efficiency. In this talk, I propose that information-theoretical approach enables us to characterize and analyze the high efficiency of gradient sensing in a way that integrates both spatial and temporal sensing. On the basis of this approach, the relation of robustness and adaptation in gradient sensing may be discussed.