5 at the cell centre, X-axis). (C) Data from (B) were compiled into single distributions. The percentage of foci in each cell width window is given on the histograms. (D) Examples of simulated distributions. A quarter of a cell section is shown with five

cell slices (X-axis). Yellow areas show areas of permitted localisation of foci for each model. The corresponding distributions of foci in the five cell width slices are shown as histograms with the corresponding percentage of total foci. Figure 3 Distribution of ter locus foci along the cell diameter. (A) Distributions of foci along the cell diameter for the ter locus in the two cell classes. Legend as for Figure 2B. (B) Examples of simulated distributions. Legend as for Figure 2D. Figure 4 Distributions of foci along the cell diameter in Ndd-treated cells. (A) Micrographs of Ndd-treated

cells showing the relocation of chromosomal DNA towards the cell periphery. Dactolisib Legend as for Figure 1B (parS site inserted at the ori locus). (B) Distributions of foci of the indicated loci along the cell diameter. Legend as for Figure 2C. (C) Legend as for Figure 2D. Again, cells were classified into major Y-27632 chemical structure populations depending on the number of foci they contained. For ori, right and NS-right loci, the distributions of foci did not differ significantly between cell populations. Thus, there was no obvious correlation between positioning and cell cycle progression (Figure 2B and data not shown). We therefore combined the datasets of the different classes into a single distribution (Figure 2C). The ori and right loci appeared to be similarly distributed into four axial selleckchem sections, but were less frequently found in the most peripheral section (Figure

2C). Comparison of the observed and expected datasets using the χ2 test showed that the distribution of the ori and right loci was significantly different from all simulated distributions stiripentol except the 90% central model (Figure 2D; χ2 = 2.7 and 2.8, respectively; corresponding to p-values of 0.6). The 90% central model is consistent with the mean position of the nucleoid, which appears as a central DNA mass partly excluded from the extreme periphery of the cell (Figure 1B). The ori and right loci thus appeared randomly positioned across the width of the nucleoid. The NS-right locus clearly tended to localise closer to the cell centre than the ori and right loci without being completely excluded from the cell periphery. However, we failed to find a model that corresponded to this distribution, the best p-value value obtained being 0.003 with the 80% central model (not shown). In the case of the ter locus, only cell populations harbouring one or two foci were statistically relevant. In both populations, a large fraction of foci were located close to either the cell pole or the mid-cell position where the division septum forms (Additional file 1, Figure S1).