Each strain chromosome is depicted as a series of ordered LCBs (Locally Collinear Blocks) with the putative origin of replication indicated by a black rectangle on the left side of each alignment. Vertical lines connect homologous LCBs across the genomes. LCBs identically present in the eleven genomes are given the same colors and horizontally flipped LCBs identify chromosomal inversions. Genomic locations of tandem repeats [12] (pink), rRNAs (red), tRNAs (green) and IS elements (blue) are depicted as short vertical lines below the LCBs. IS elements were identified using the ISfinder database [13]. Gaps or white spaces in LCB order represent strain-specific regions. Three large-scale inversions (dark orange) occurring in 4 strains are shown at the bottom of the alignment.

A) Venn diagram showing the number of predicted genes either shared (black) or uniquely found in either the 2009 Bp annotation (red) or the original 2004 K96243 annotation (green). B) Bp K96243 genomic tracks showing novel genes. Row 1: Genomic locations of 9 Bp genes on Chr1 on both the positive (+) and the negative (−) strand in the 2009 genome annotation including a novel gene BPSL2192.1 (red arrow) on the positive (+) strand. Row 2: Genomic locations of 8 Bp genes on the same region of Chr 1 in the 2004 genome annotation. C) mRNA transcripts associated with novel genes. (left) Row 1: Genomic locations of 5 Chr 2 Bp genes (blue bars) and 1 novel gene (BPSS1202.1: red bar) in the 2009 annotation. The novel gene lies on the negative (−) strand. Row 2: Genomic locations of the same 5 Chr 2 genes on Chr 2 (green bars) in the 2004 annotation. Rows 3–7: Transcript expression under six growth conditions. BPSS1202.1 is expressed in all conditions except early stationary phase in minimal media. Row 8: Probe coverage associated with these genes. (right) A second example of a novel gene (BPSS0279.1) expressed primarily in both early and late stationary phase in rich media. D) Evolutionary conservation of a novel gene. Cross species comparison by BLAST of a predicted novel gene (BPSL3348.1) across five Burkholderia species. Multiple sequence alignments were generated using ClustalX [21]. Organism names are indicated at the beginning of the alignment. Identical residues are indicated as black dots. The conserved domain (identified by CDD search) is shown at the top of the sequence alignment in pink. The red block below the alignment indicates the level of conservation. The sequence logo of the alignment is shown at the bottom. E) Comparison of gene length between the 282 novel genes supported by two lines of evidence (red), and all 2004 genes (green) for chromosomes 1 and 2. The graph plots cumulative gene frequency against gene length in amino acids.

A) Depletion curves for the Bp core genome (blue). Vertical bars represent standard deviation values based upon one hundred randomized input orders of the Bp genomes (http://www.rproject.org/, The R Project for Statistical Computing) [30]. B) Accumulation curves for the Bp pan genome (green). C) Distribution of orthologous genes between the Bp K96243 core genome (4908 core genes + K96243 paralogs  = 5063 genes), B. mallei ATCC23344 and B. thailandensis E264. The Venn diagram depicts the number of genes either shared or unique between one or more Burkholderia species. Figures in brackets indicate the total number of genes compared.

A) Distribution of SNPs across 11 Bp genomes. Core genome genes in all strains were compared against Bp K926243 to identify SNPs. For primary and relapse strain pairs (1106a/b and 1710a/b), only the primary strain is depicted. Geographical origins of the strains are depicted as different column colors. B) Chromosomal patterns of synonymous and nonsynonymous SNPs. Rates of synonymous (K_s) and nonsynonymous (K_a) substitution were estimated for Chromosomes 1 and 2 using a set of 4673 high-quality orthologous genes covering the 11 Bp strains. Each hourglass plot (interquartile range, IQR) represents the 25% to 75% range for that chromosome, with the bottleneck placed at the sample median. Horizontal tick marks show the range of all elements within Quartile 1–1.5 X IQR and Quartile 3+1.5 X IQR (equivalent to the 99.3% interval of a normal distribution). Open circles represent outliers (data points outside this range). The width of the bottleneck (i.e., the length of the V-shaped notch) is an indication of the confidence of the median; a lack of overlap of the bottleneck between samples implies that the samples are statistically different. Chromosomes differ significantly in K_a values (P = 2.42×10⁻²¹, t-test) but not in K_s values. C) Functional enrichments in the Bp core genome. COG functional categories are indicated on the x axis, and the percentage of genes in each COG category is shown on the y axis. Dark blue bars represent Bp core genes. Light blue bars indicate genes under positive selection in Bp strains. COG categories that are significantly enriched, (P<0.05, binomial test; bonferoni correction applied) in positively selected genes relative to the core genes are indicated by an asterisk.

A) Relative Virulence of TFP4 Deletion Mutants: Graphs show survival curves of BALB/c mice following intranasal challenge with varying dosages of Bp (left – K96243 wild-type, right – TFP4 deletion strains). See Methods for infection assay details. The TFP4 deletion strain is significantly less virulent compared to Bp K96243 parental controls (p = 0.048, Mantel-Haenszel log rank test). Units in the color bar refer to Bp colony forming units (CFU). B) Transfection of HeLa cells using i) vector, ii) BPSS0415, iii) BPSS1525 (BopE) and iv) BPSL1057F1. Cells were stained with rhodamine-phalloidin and DAPI to identify actin filaments and nuclei respectively. All genes were tagged with GFP at the N-terminus. Cells transfected with either empty vector or BPSS0415 exhibited normal actin structures and filaments (arrowheads). Cells transfected with either BopE or BPSL1057F1 exhibited dissolution of normal actin filaments with the presence of actin stress fibers (arrowheads); Bar 10 µm. C) Population analyses of transfected cells. Values were presented as percentage of the total number of transfected cells (n = 40), obtained from four independent experiments. D) Protein sequence relationships of nine taurine dioxygenase (tauD) genes from Burkholderia pseudomallei K96243. Bootstrapped Neighbor-Joining trees [52] were constructed using ClustalX [21], and drawn using NJplot [53]. Each branch was compared against 1000 resamplings of the alignment data. Bootstrap values and branch lengths are shown at the branch points, and distance units are shown in the lower right hand corners of the tree. The blue bar indicates the presence of a tauD homolog in B. thailandensis, red bar in B. mallei. BPSS0665 is the tauD gene located in GI14. The tauD gene BPSS0161 exhibits a signature of positive selection. E) Taurine utilization in Bp and Bt. Strains were grown in medium containing either taurine or free sulfate (Na₂SO₄-). To normalize differences in intrinsic growth rate, each strain was plotted as a ratio between its growth in taurine compared to free sulphate (y-axis) during 0 hrs or log-phase growth (x-axis). No growth was observed in taurine or Na₂SO₄-deficient media. Error bars represent the standard deviations between replicate cultures. F) Identification of taurine-regulated genes. Transcriptome profiles of taurine-exposed Bp were compared to Bp grown in laboratory rich media. Both populations were isolated at stationary phase. Rows represent Bp Chr 1 and Chr 2. Y-axes represent levels of transcriptional up-regulation of Bp genes in the presence of taurine relative to rich media (LB). Arrows depict gene clusters related to flagella (blue), iron metabolism (black), fimbrae/pili (green), and taurine metabolism (pink, BPSS1572-1575).