Understanding biology at the system level has gained much more interest recently due to the rapid development in genome sequencing and high-throughput measurements. Mathematical descriptions of biological systems are normally formalized using two different approaches. The deterministic method is very efficient at predicting the overall behavior of the system but ignores the inherent fluctuations and correlations at lower concentration. The stochastic method, on the other hand, captures the intrinsic randomness but is often mathematically intractable and computationally expensive. Our group has developed a system based on a deterministic-stochastic crossover method. Biological models are formalized into a standard format using the eXtensible Markup Language (XML), allowing the system to perform robust simulation on more complicated models. Simulation studies have been performed on biological systems like auto-regulatory gene networks and glycolysis systems. The new system retains the high efficiency of the deterministic method while still reflecting random fluctuations at lower concentration. The ability to reveal stochastic properties with high efficiency makes this system very useful for research and applications based on system biology.

Committee
Dr. Robert Harrison (chair)
Dr. Yanqing Zhang