We study how the cell controls information flow from genes into proteins. In bacteria, primary control is at the transcriptional level, orchestrated by multiple sigmas (?s). Housekeeping ?s direct transcription from 1000's of promoters; alternative ?s direct transcription from specific genes to cope with environmental change. Our goal is to understand the interplay between ?s and RNA polymerase (RNAP), and the network properties that govern RNAP output. We bring to this work a long history of providing insight into the roles of ?s and a new genomic technology. In the current granting period, we will investigate three fundamental and unsolved issues: (1) The alternative "stress" ?s increase transcription from a limited gene set necessary to cope with the stress; we will determine features of alternative ?s that result in stringent promoter selection. (2) We have been developing new paradigms to predict alternative ?s. Using combinatorial libraries and high throughput approaches, we will critically test the ability of current algorithms to predict promoters of the ?E stress ?, and develop new algorithms that function better. (3) Transcription initiation has been studied largely in isolation. Using our new genomic methodology for high throughput, quantitative determination of genetic interactions, we will determine how RNAP is integrated into the broader workings of the bacterial cell. These benchmark studies will be made available immediately in a public database, thereby moving the entire field forward by identifying new proteins and functional relationships in transcription and new relationships among cellular processes. We perform these studies primarily in E. coli, where a wealth of genetic, biochemical and physiological data facilitate analysis, but our results are directly applicable to all bacteria, including pathogens and environmentally beneficial microbes. Moreover, the questions we are asking and the methods we are developing allow us to directly extend our studies to other bacteria that may be difficult or even impossible to study experimentally. Finally, our studies have important direct applications. The family of circuits we are dissecting may provide building blocks for synthetic circuits with diverse output properties. Additionally, the observation that many RNAP mutants have enhanced growth capabilities under particular conditions or increased resistance to environmental stress indicates that RNAP is a hotspot for adaptive evolution; understanding the mechanism behind these effects could be of use in drug discovery efforts, metabolic engineering, biofuel production and bioremediation. We study the molecular machineries and circuitries that govern gene expression in bacteria and enable them to proliferate under diverse and challenging environments. This information allows us to manipulate and control bacteria, including pathogens and environmentally beneficial microorganisms. Additionally, our work is of direct relevance to drug discovery, metabolic engineering, biofuel production and bioremediation.