Prediction of co-regulated genes in prokaryotic genomes on the basis of upstream elements conserved across closely related species

摘要

To know biological phenomena on the molecular level means to understand how each gene is regulated by what kinds of elements or signals, how it is controlled and when expressed. For instance, E.coli has about 4,000 genes and each of them is not expressed regularly. All of them have their own expression level and some special conditions in which the expression occurs. Analysis of such transcriptional regulatory networks is required for the understanding of many biological phenomena, such as proliferation, cell cycle, development, differentiation. Experimental methods for recognition of co-regulated genes are mostly provided by the microarray technique. For the purpose of determination of the transcriptional factor binding sites nuclease protectionis widely used. Although experimental methods give the most precise knowledge about these biological phenomena, they have disadvantages when we have to maintain a large amount of uncharacterized sequence data. Therefore, we need powerful computational techniques to identify key elements such as transcription factor binding sites in the transcriptional regulatory networks. Here, we try to detect them by the comparison of upstream regions across evolutionally related species. We assume that not only theircoding regions but also most, if not all, of their transcriptional regulatory mechanisms should have been conserved during the evolution. A similar research on Bacillus genus was executed by our group (it will he also presented on another poster at GIW2001). In that work, it was concluded that comparative DMA sequence analysis enable identification of functionally conserved elements within closely related species, and some plausible co-regulated genes were presented. In this work, we applied this approach to other prokaryotic genomes and predicted the relationship of the co-regulated genes in them.