16 February 2007
Stanislav Shvartsman
Department of Chemical Engineering
Princeton University
Modeling of pattern formation and morphogenesis is extremely nontrivial due to the high levels of structural and parametric uncertainty and a truly multiscale nature of development. At the same time, an expanding arsenal of computational and experimental tools can constrain models and directly test their predictions, making the modeling efforts not only necessary, but feasible. I will try to support this statement using three examples from our recent work on MAPK signaling in development. The first example will show how to handle parametric uncertainty in large biochemical models of signaling networks. The second example, from our experimental work on the terminal patterning system in the Drosophila embryo, will demonstrate how the dynamics of MAPK signaling can be driven by nucleocytoplasmic shuttling. Finally, the third example, from our modeling and experimental work on Drosophila oogenesis, will emphasize the spatiotemporal signal processing capabilities of networks formed by multiple signaling pathways.