Pathogenicity islands (PAIs) are major factors contributing to the pathogenicity of bacteria and to their resistance to antibiotics. In general, genomic islands (GIs), of which PAIs are a subset, increase the fitness of their hosts by providing new functions. With the number of available whole genome sequences growing exponentially, in silico methods have been developed to detect putative PAIs and GIs within them. Compositional methods rely on G+C content differences, codon usage and oligonucleotide biases. Other methods detect the presence of functional elements such as tRNA and mobility genes. Future availability of fast, high-throughput, inexpensive genome sequencing emphasizes the need for user-friendly applications able to detect, characterize and analyze putative GIs and PAIs. It may uncover new aspects of pathogenicity and provide better understanding of the evolution of pathogenic bacteria. These methods will be highly requested when whole genome sequencing techologies will be used by physicians for personal diagnosis.