I would like to study the molecular genetic profiling of the bacterial composition of dairy starter cultures by pulsed-field gel electrophoresis (PFGE)
Any recommendation?
Bio-Rad CHEF Mapper XA System or Bio-Rad CHEF-DR III System??
It doesn't really matter for your purposes what machine you take. Nearly everything will work, as your molecules are much smaller than for instance the fungal chromosomes I used to analyse with this technique. The good Thing about pulsed fields is that you see immediately if you have more than one genophor in your bacterial cells or maybe also larger plasmids.
But don't underestimate the technical efforts. Many bacteria need many different lysis conditions, and you cannot use brute-force methods, otherwise your molecules will break. It maybe that your approach is not the very best, as long as you don't know anything about your isolates. It could work nicely for different isolates of a single species, but I would personally not like to enter into a Screening programme for many different species. My recommendation, if you really want to follow this approach, is to find somebody in your vicinity who has such a machine and use it a few times, before you make decisions to buy one.
In the meantime I recommend to think about Werners suggestion to start with sequencing. It's pretty cheap today. However, it is also not trivial. It is not easy to recognise the existence of several genophors or megaplasmids just by sequencing. It's clear why. All programs have been instructed to assume a becterial genophor as a sngle crircular molecule. Both assumptions need not necessarily be true.
Unfortunately, I am not familiar with dairy starter cultures. Could it be that you essentially are interested in genetic variability of very few species? In that case, it could be rewarding just to restrict isolated DNAs with one or two of the endonucleases that recognise only very few sites in a given genome. You can then simply look for differing bands on normal agarose gels. Take big slab gels (> 20 cm) and run them very slowly and you will have a quite reasonable resolution also in the size region around 20 kb or even a bit higher.
Think about what you want to know and what you need to know and please go on discussing it, with your colleagues in the laboratory and here. It helps.