The Johns Hopkins University

Whiting School of Engineering

Department of Electrical and Computer Engineering

Control and Information-theoretic Analysis of Biological Signaling Systems

A Dissertation Defense by

Burt Andrews

Graduate Research Assistant

Electrical and Computer Engineering

Abstract:

Because the molecular processes that govern the behavior of many cellular organisms are highly stochastic, the proper functioning of these biological systems requires the ability to cope with this. Using theoretical tools from engineering, this dissertation investigates the characteristics that enable cellular systems to cope with a variety of stochastic effects in the biochemical environment. We focus primarily on chemotaxis -- the process by which cells migrate toward chemical attractants. Optimal filtering theory from control is used to show that the signal transduction network of E. coli effectively acts as a low-pass filter with a bandwidth that balances the need to detect noisy chemical cues accurately on a timescale set by the effects of diffusion. From information theory, we use rate distortion theory to show that gradient-sensing cells such as amoeba respond to spatial chemo attractant profiles using as little information as possible to achieve tolerable chemo tactic performance levels. Our results suggest that many cellular organisms have evolved to respond to external chemicals in a manner that is consistent with optimal engineering designs. Moreover, the identification of key molecular components and processes that lead to this behavior provides biologists and experimentalists with further areas of study.

This work highlights the fact that many theoretical tools from engineering can be used to study the biological mechanisms used by cells to survive and perform essential physiological functions. However, fundamental differences between biological and engineered systems can inspire theoretical contributions to engineering.  We demonstrate this by concluding with a derivation of an extension to the internal model principle that is well-suited for models of biological systems.

Wednesday, August 29, 2007

10:00 a.m.

Barton Hall 225

FOR DISABILITY INFORMATION

CONTACT: Candace Abel (410) 516-7031 cabel@jhu.edu