Restriction enzymes for genomic DNA library prep (e.g. NlaIII)

[454newbie]

Given the cutting frequency of NlaIII and DpnII, why aren't they used for generating libraries from genomic DNA? I realize that they were originally used to obtain sequence tags for expression, but haven't seen them used for preparing genomic libraries. why not?

in general, why would you not use restriction enzymes for preparing genomic libraries? Has anyone looked at this systematically recently?

[SeqAA]

There are enzymatic fragmentation kits....
From several commercial providers.

Very handy for bulk library prep.

[454newbie]

yes-we've used fragmentase, DNAseI, Nextera. My question is specifically about why restriction enzymes aren't used more, particularly now that we have decent reference genomes.

[pmiguel]

Illumina sequencers seem to have a peculiar Achilles heel with respect to sequencing libraries that largely start with the same sequence of bases. This can be worked around in a number of ways, but it is an issue.

--
Phillip

[kmcarr]

Given the cutting frequency of NlaIII and DpnII, why aren't they used for generating libraries from genomic DNA? I realize that they were originally used to obtain sequence tags for expression, but haven't seen them used for preparing genomic libraries. why not?

in general, why would you not use restriction enzymes for preparing genomic libraries? Has anyone looked at this systematically recently?

No matter how frequently a restriction enzyme cleaves a template it will never achieve the frequency required for preparing a proper shotgun library. The ideal for a shotgun sequencing library is fragments of uniform size with a perfectly random distribution of fragment ends. To achieve a random distribution of ends the fragmentation method must have equal probability of cleavage between any two bases in the substrate. Restriction nuclease cleavage is the antithesis of randomness.

We have sequenced reduced representation libraries prepared by restriction digestion of genomic DNA. After digestion a narrow range of fragment sizes were selected from the total pool. The express purpose here was to select a non-random subset of the total genomic DNA.

[flxlex]

This is RAD-tag sequencing, right?

[kmcarr]

This is RAD-tag sequencing, right?

We had a researcher approach us about performing some RAD sequencing but have not done it yet. It's definitely cool, but again, it's a method for selecting only a subset of the overall genome. It will produce a non-random, reduced representation of your genome.

If I understand the OP correctly they were asking about using restriction enzyme digestion as an alternative to mechanical breakage, chemical or non-specific enzymatic cleavage for preparation of general genomic sequencing libraries. Restriction enzymes are not appropriate for this application.

[454newbie]

The original question was really targeted at why restriction enzymes aren't used more frequently for preparing libraries for resequencing genomes for which decent reference genomes exist. In particular, for high frequency cutters like NlaIII, while a library generated would not be truly "shotgun" it might still be useful to answer a good number of questions, with very simple library prep using overhangs for specific end-adaptor ligation.

update- Okay- just read the RAD-tag paper-this is exactly what I was thinking. thanks for the reference. very helpful.