The main target of our laboratory is stochastic phenomena observed in intracellular and multi-cellular systems. Cell is the basic building block of our body and all organisms on the earth. The complex combination of cells constitutes our metabolism, immunity, development, and brain. Even though these macroscopic phenomena of our body usually operate very robustly, it has been known experimentally that the behavior of individual cells seems to be highly stochastic. This fact means that the robust functions of biological systems are implemented by stochastic components. This property contrasts sharply with the man-made engineering systems whose building blocks are designed to operate deterministically. We are interested in the underlying design principle of the robust functions with unreliable noisy components. Toward the goal, we mainly have two different approaches. One is data-driven approach in which we investigate quantitative data of stochastic cellular phenomena. Specifically, we are working on stochastic dynamics of early mouse embryogenesis, circadian rhythm, polarity formation, and cell migration. To characterize these stochastic phenomena, we develop original algorithms for bio-image analysis, data analysis and mathematical models by employing the knowledge on nonlinear science, statistics, and machine learning. The other one is a theory-driven approach in which we create new theories for analyzing and describing the stochastic dynamics of cellular phenomena. The theory for Langevin equation, stochastic differential equation, Fokker-Plank equation, and master equation has been employed to analyze various stochastic dynamics in the field of theoretical systems biology. But, we think that these theories are not sufficient to understand how a cellular system can operate robustly with noise in its components. Actually, these theories can predict what kind of behavior will be generated by a given a reaction network. But they provide little information on why a system becomes robust to noise. We think that it is crucial to combine these theories with those of information and statistics that has been used in engineering. We have already marked the first concrete step on this topic.


Update of our Web Page
Laboratory for Quantitative Biologyブログ上