Metabolic networks govern the flow of chemical reactions in living cells, which
is one of the most important aspects of their behavior.  By considering such
networks, we can hope not only to better understand their function but also to
devise designs to engineer organisms with compelling desired properties, such
as the ability for efficient fuel or drug production.

In this talk, we discuss methods we have developed for analyzing and optimizing
metabolic flow.  We discuss, in particular, GDLS (Genetic Design through Local
Search), a computational method that allows large-scale analysis of metabolic
networks to yield genetic designs for metabolic engineering purposes.  We
further discuss methods we have developed to understand large-scale metabolic
flow.