Alex- 7 is more related to being the number that can give good discrimination. The use of 7 genes for Campylobacter (where there are hypervariable regions in the genome) has been relatively successful, but the extended 10 gene system is now increasingly used. Of course for a small genome (1.8Mb) like Campylobacter WGS is easier-I guess Xanthomonas is much bigger (about 6 Mb?). As far as I can see there may be more of an issue for you unless the recombination is impacting on the target genes, then MLST may not give the differences you are looking for.

Paul, thank you. I guess we have here the same fenomen as in use of different DNA length: for instance, if we take 400 bp of gyrB gene we can discriminate species of Xanthomonas (5 000 000 bp), if we use 800 bp - some pathovars, but using full gene (over 2000 bp) gives separated strains. Additional gene or nucleotides will not improve much the phylogenetic tree, only bootstrep.

I told about recombination because we had this idea and data long time ago, but were severely critisized for it, and then our colleagues published a couple of papers, we were encouraged to present such hypothesis for our population. And as you see, it was not supported later.