UC San Diego

Cells live with fluctuations arising from various sources and occurring at a broad spectrum of scales. Stochasticity plays an amazing role in cellular functions and physiological behaviors through biological networks that compose a living cell. Like two faces of a coin, noise may be destructive in many biological systems but can also be constructive on the other hand. In this work, I combined computational, theoretical and experimental approaches to explore stochasticity in biological networks. The origins, consequences and significance of stochasticity were investigated through the developments of methodological techniques as well as studies of specific yet important networks. Effective temperature was proposed as a measurement of noise in gene networks. It serves as an alternative to noise classification by "intrinsic'' and "extrinsic'' contributions and is in some sense a more fundamental approach. A generalized Gillespie algorithm was derived for stochastic simulation of biochemical reactions that allows one to simulate biological systems with time-dependent reaction rates and system volumes. In addition to these developments, an important network topology abstracted from the multi-site phosphorylation networks of nuclear factors of activated T-cells was studied. Signal transduction of the network was mapped onto a random walker problem in nonequilibrium statistical mechanics and an optimal enzyme concentration was found that favors fast transduction. Noise at the cellular population level was also studied. A generalized variation index was proposed to measure variability and diversity of cellular populations. We found that cellular population variability may depend on its initial conditions and environments. Finally we turned to stochastic recombinant events in a gene circuit. A synthetic switch with both phenotypic and genotypic transitions was studied using combined experimental and theoretical approaches. This led to the result that there is always a bias of cellular population to one specific fate. These studies show the two sides of stochasticity and help us to better understand noise in biological systems and to aid in better design strategy of genetic circuits