Genome evolution reveals biochemical networks and functional modules.

摘要: The analysis of completely sequenced genomes uncovers an astonishing variability between species in terms gene content and order. During genome history, the genes are frequently rear-ranged, duplicated, lost, or transferred horizontally genomes. These events appear to be stochastic, yet they under selective constraints resulting from functional interactions genes. genomic form basis for a variety techniques that employ systematic comparisons predict associations among most powerful date based on conserved neighborhood, fusion events, common phylogenetic distributions families. Here we show these techniques, if integrated quantitatively applied sufficiently large number genomes, have reached resolution which allows characterization function at higher level than individual gene: global modularity becomes detectable protein network. In Escherichia coli, predicted modules can bench-marked by comparison known metabolic pathways. We found as many 74% enzymes clustering together modules, with average pathway specificity least 84%. extend beyond metabolism, led hundreds reliable predictions both level. results indicate networks is intrinsically encoded present-day