Derek Sherry

Signal processing is vital to many biological systems. The signal transduction network in E. coli chemotaxis is an important test case for such biochemical signal processing. In this thesis, I introduce a new model for this signaling, driven from equilibrium reactions between neighboring units in an extended signaling array. The model can achieve arbitrarily high gain through time-scale separation, without the need for fine-tuning, and it also captures qualitative experimental results which are difficult to reconcile with existing models. I also begin the development of some possible molecular implementations of the model where long flexible linkers could potentially mediate interactions between neighboring core units in the signaling lattice. Finally, I determine the critical exponents of the model and describe its relationship to the voter model and the Ising universality.