Can I use sheared genomic DNA as starting material of bacterial genome sequencing?

[omnivore]

I am planning to acquire draft genome sequence of a bacterial strain by using GS FLX titanium sequencer. My question is whether initially sheared genomic dna can be safely used for genome sequencing and assembly by 454 technology. By 'initially', I mean genomic DNA was sheared during extraction process. I attach gel image of my genomic dna samples (marker is lambda dna digested with EcoRI and HindIII).

Although I know that dna fragmentation step is involved in standard genome sequencing by GS FLX titanium, I am concerned because I think it might be possible that shearing during extraction can affect randomness of standard fragmentation step and so on.

Thanks in advance.

[pmiguel]

What size is the smear? I don't know what your size standard is.

--
Phillip

[omnivore]

Thanks for your interest.

I attach a new gel image (marker sizes are indicated).

[RCJK]

The recommendation is that the DNA to be nebulized is in > 1.5KB fragments as I believe this helps with ensuring "random" shearing of the DNA. Fragments smaller than that size do not nebulize as well, at least this is what I've been told by Roche before.

[pmiguel]

Are you sure about those sizes? Is this a normal gel or some sort of pulsed-field gel?

I ask because it is not easy to shear DNA that small without really trying.

Were you planning on constructing a paired-end library?


Anyway, if that really is the size and it is the result of mechanical shearing (not enzymatic -- from a DNAse in the DNA prep), then it doesn't really matter and should be fine. If it is from a DNAse contaminating the DNA prep, there are two issues:

(1) Most importantly, you must further extract the DNA to remove this DNAse prior to proceeding. Otherwise it will degrade your adaptors during ligation and you will not be able to construct a library. Also, the DNA will continue to be degraded.

(2) A DNAse may have sequence-specific biases in its action. This would lead to less than random distribution of your shear sites. In turn, this will make the genome harder to assemble. As long as the bias is not too pronounced, you would be okay. (In fact some protocols call for the use of a DNAse to fragment DNA during library construction. But these protocols are generally careful to choose enzymes/conditions that favor a low bias of cleavage.)

--
Phillip

[omnivore]

Thanks for replies.

I'm sure about DNA sizes (0.7% agarose gel was used).

I think that bead beating step (introduced for some reasons) was the cause of DNA shearing.

[pmiguel]

Actually, as long as you don't need to make large paired-end libraries, you may be fine.

It depends on the nature of the break that the bead beating introduced. As long the breaks predominantly occurred in the diphospho-ester parts of the DNA backbone (as seems to be case with hydroshearing and probably nebulization) then most of the ends should be repairable by typical (T4 poly/T4 PNK) reactions.

Of course it is possible that other chemical damage to the DNA also occurred and this may interfere with subsequent library construction steps. But, if it were me, I would give it a shot, as is. Unless it is fairly easy to come by more genomic DNA.

--
Phillip