Identifying pathogenicity islands in bacterial pathogenomics using computational approaches

Abstract

High-throughput sequencing technologies have made it possible to study bacteria through analyzing their genome sequences. For instance, comparative genome sequence analyses can reveal the phenomenon such as gene loss, gene gain, or gene exchange in a genome. By analyzing pathogenic bacterial genomes, we can discover that pathogenic genomic regions in many pathogenic bacteria are horizontally transferred from other bacteria, and these regions are also known as pathogenicity islands (PAIs). PAIs have some detectable properties, such as having different genomic signatures than the rest of the host genomes, and containing mobility genes so that they can be integrated into the host genome. In this review, we will discuss various pathogenicity island-associated features and current computational approaches for the identification of PAIs. Existing pathogenicity island databases and related computational resources will also be discussed, so that researchers may find it to be useful for the studies of bacterial evolution and pathogenicity mechanisms.

Figure 1

A schematic view of a pathogenicity island with associated features. The PAI region has biased sequence composition. The PAI regions are associated with virulence genes (vir1, vir2, vir3, and vir4), phage-related genes (phag1 and phag2), mobile genes (int and trans), hypothetic protein genes (hypo1, hypo2, and hypo3), insertion sequence elements, direct repeats, and tRNA gene.

Figure 2

A schematic view of genomic region alignment in the comparative genomic based approach for island prediction. Three phylogenetically closely-related reference genomes (G1, G2, and G3) are shown here for the detection of island region in the query genome (G4).