Whole-genome expression experiments provide the building blocks forunderstanding cellular control mechanisms underlying health anddisease, from cardiac function to learning and memory to oncogenesis.To put post-genomic information to work requires buildingbiophysically detailed, quantitative models for the biochemicalinteractions controlling cell behavior. Such biochemical signalingnetworks form a class of high-dimensional, stochastic nonlineardynamical systems, analytical and numerical techniques for the studyof which are only in the early stages of development. I will presentsome approaches to reducing the complexity of cellular signalingnetworks and to extending the classical deterministic approximationsfor these systems to include fluctuation effects. I will focus onapplications to biophysical systems currently under investigation at UCSD, Scripps and Salk Institute.