To understand biological networks, our lab has defined network motifs: basic interaction patterns that recur throughout biological networks, much more often than in random networks. The same small set of network motifs appears to serve as the building blocks of transcription networks from bacteria to mammals. Specific network motifs are also found in signal transduction networks, neuronal networks and other biological and non-biological networks.
We experimentally studied the function of each network motif in the transcription network of E. coli. Each network motif can serve as an elementary circuit with a defined function: filters, pulse generators, response accelerators, temporal-pattern-generators and more. Evolution seems to have converged on the same motifs again and again, perhaps because they are the simplest and most robust circuits that perform these information-processing functions.
To carry out these experiments we developed automated systems for measuring the behavior of gene circuits in living cells. We developed a library of 2000 E. coli strains in which green fluorescent protein reports for the activity of the vast majority of the organisms' promoters [Zaslaver 2006] . Using robotic liquid handling, we obtain accurate dynamics of E coli promoters under desired conditons.