We wish to understand the evolution of BXW and the genetic basis for the differing host specificities between Xcm, a pathogen of banana, and its close relative Xvv, a pathogen of sugarcane. Genetic differences may also reflect adaptations for epiphytic fitness and for dispersal, perhaps via insect vectors (Tinzaara et al., 2006; Mwangi et al., 2007; Shimelash et al., 2008) rather than virulence factors per se. Recent developments in DNA-sequencing technology provide opportunities

for rapid and cost-effective comparative genomics studies PLX4032 in vitro (MacLean et al., 2009). We used the ‘Next Generation’ Illumina Solexa GA technology (Bentley et al., 2008) to generate draft genome sequences for banana-pathogenic Xcm NCPPB 4381 (Xcm 4381) and for the closely related Xvv NCPPB 702 (Xvv 702), which is not pathogenic on banana. Sequence analysis revealed several genetic differences between these two strains

that might be important for host specificity, virulence and EPZ-6438 order epiphytic fitness, including differences in the repertoires of secreted and translocated effector proteins, type IV pili (TFP) and enzymes for lipopolysaccharide biosynthesis. Xcm 4381 was originally isolated from banana in Uganda 2005. Xvv 702 was originally isolated from sugarcane in Zimbabwe in 1959. We obtained both strains from the National Collection of Plant Pathogenic Bacteria (NCPPB) at the Food and Environmental Research Agency (FERA, York, UK). Genomic DNA was prepared from cultures grown in NYGB (Nitrogen Yeast Glycerol Broth) medium using a Puregene Genomic DNA Purification Kit (Gentra Systems Inc., Minneapolis) and sequenced using the Illumina GA sequencing system. Illumina

sequencing of genomic DNA and sequence assembly were performed as described previously (Studholme et al., 2009). We used maq (Li et al., 2008), blast (Altschul et al., 1990) and mummer (Delcher et al., 2002) for sequence alignment of short reads, contigs and whole genomes, respectively, and cgview (Stothard & Wishart, 2005) for visualizing the alignments. splitstree (Huson, 1998) was used to build and draw phylogenetic trees. We deposited the draft genome data in GenBank under accession numbers ACHT00000000 (Xcm 4381) and ACHS00000000 (Xvv 702). We generated 5 052 905 pairs of 36-nucleotide reads from Xcm 4381; that is a Sclareol total of 363 809 160 nucleotides, representing approximately 72 times 5 megabases, the typical genome size for Xanthomonas species. From Xvv 702, we generated 2 913 785 pairs of 36-nucleotide reads; that is a total of 209 792 520 nucleotides, representing approximately 42 times the expected genome size of 5 megabases. We assembled the Illumina data using the velvet short-read assembly software (Table 1). The genome sequences of Xcm 4381 and Xvv 702 shared significantly greater nucleotide identity with each other than with the genome of any other sequenced Xanthomonas genome (Table 2).