Regions of unusual statistical properties as tools in the search for horizontally transferred genes in Escherichia coli

The observed diversity of statistical characteristics along genomic sequences is the result of the influences of a variety of ongoing processes including horizontal gene transfer, gene loss, genome rearrangements, and evolution. The rate at which various processes affect the genome typically varies between different genomic regions. Thus, variations in statistical properties seen in different regions of a genome are often associated with its evolution and functional organization. Analysis of such properties is therefore relevant to many ongoing biomedical research efforts. Similarity Plot or S-plot is a Windows-based application for large-scale comparisons and 2D visualization of similarities between genomic sequences. This application combines two approaches wildly used in genomics: window analysis of statistical characteristics along genomes and dot-plot visual representation. S-plot is effective in detecting highly similar regions between two genomes. Within a single genome, S-plot has the ability to identify highly dissimilar regions displaying unusual compositional properties. The application was used to perform a comparative analysis of 50+ microbial genomes as well as many eukaryote genomes including human, rat, mouse, and drosophila. We illustrate the uses of S-Plot in a comparison involving Escherichia coli K12 and E. coli O157:H7.