Does anyone have a solution

[NathalieA]

I have to do 4 libraries preparation (Paired End 8kb library) on Francisella tularensis genome.

Usually for these kind of library, I used the Hydroshear to DNA fragmentation at the good size. But unfortunatly, our Hydroshear is out of order, and I have no other one.

I quickly read the forum but I didn't find any answer. The dsFragmentase will give me too small fragment.

I have the intention to do the fragmentation by an enzyamatic digestion.
My genome is a 1.89Mb, if I choose a 6base restriction enzyme, the probability of occurence is 1/4^6 (=1 on 4100 bases).
If I choose a 8bp, I have a probability of 1/4^8 (=1/65536).

do you have any advice to help me

[ajthomas]

I don't think a restriction enzyme is the way to go because you really need random fragmentation. If you use a 6-base cutter, for example, your average fragment size will be ~4000 bp, but there will be plenty of places in the genome where two sites for that enzyme might be several times that, and you will get no coverage in those locations.

If you must use an enzyme, I would suggest using a general DNAse, but limit the time to get the right average fragment size. You will need to figure that out yourself. Use a constant amount of enzyme in a digest, and remove aliquots at time intervals to look at the size distribution. You might want to try a few different enzyme concentrations as well.

A general DNAse will probably have some site preference, so the coverage won't be very even. The coverage might be more even if you used a 4-base cutter, or better yet, a mixture of several 4-base cutters (again titrating the digestion time). I don't know whether it would be better to mix the enzymes or use each one on separate aliquots of DNA and mix the DNA later.

*I haven't used this approach for PE libraries, but you asked for suggestions and this is just one idea that might work.

[RCJK]

I'd check out these if I were you:

http://covarisinc.com/g-tube.html