Tweet
Physical shearing methods for double stranded DNA include sonication, nebulization and hydroshearing. We use all three. They work well for constructing a handful of libraries. But without specialized equipment/automation, constructing larger numbers of libraries really starts to become a bottleneck.
Just in the last week I have seen advertisements for two new fragmentation methods. One, from New England Biolabs (NEB), "dsDNA Fragmentase", and another from Epicentre Biotechnologies, "Nextera".
NEB's ds Fragmentase uses a mutant Vibrio vulnificus nuclease and a mutant T7 endonuclease (both fused with maltose binding protein for reasons unclear to me) and another from Epicentre Biotechnologies deploys an engineered transposase
More information available at the NEB web site:
http://www.neb.com/nebecomm/products/productM0348.asp
An aside here: As scientists I think we should demand that commercial entities wanting to sell us products reveal the compositions of those products. Lamentably this has become increasingly rare. But NEB displays a level of transparency in this regard that should be required of all companies wanting to transact with researchers using public funding. Just wanted to mention that. I have no connection with either NEB or Epicentre, other than I buy their products.
Epicenter's Nextera uses their "Transposome" technology to randomly bomb DNA. The ends of their engineered transposon apparently are not connected, so each insertion generates two ends (a break), each terminated with an appropriate adaptor. The trick in next gen library construction is always getting adaptor "A" on one end and "B" on the other. Since the "A" and "B" ends of the Transposome end up terminating different molecule, we would still expect half the fragments generated to be "A-A" and "B-B". But that is no worse than ligation methods deliver, and it cuts out the need for a ligation step. More information here:
http://www.epibio.com/nextera/nextera_tech_overview.asp
Just to be clear about the benefits, physical fragmentation methods generally require substantial "hands on" time for every library being constructed. That is, you take the DNA sample, you load it into the device (nebulization unit, sonicator, or hydroshear), perform the procedure, then unload. Even ignoring issues like the potential for cross-contamination and maintenance costs for these devices, scaling is an issue. Making 10 libraries is do-able. But 96?
So the allure of non-mechanical methods is obvious--you could probably set up either of these in a 96 well plate and process using multichannel pipettors, or even use laboratory robotics. But let us not forget why we flocked to mechanical methods in the first place: enzymes never seem to be as random as the mechanical methods. Being derived from living systems they come with their own agenda, so to speak. And that agenda usually results in bias. My advice would be to pay attention to the tests both companies did on differential depth of coverage for libraries generated using their methodologies.
--
Phillip
Just in the last week I have seen advertisements for two new fragmentation methods. One, from New England Biolabs (NEB), "dsDNA Fragmentase", and another from Epicentre Biotechnologies, "Nextera".
NEB's ds Fragmentase uses a mutant Vibrio vulnificus nuclease and a mutant T7 endonuclease (both fused with maltose binding protein for reasons unclear to me) and another from Epicentre Biotechnologies deploys an engineered transposase
one randomly generates nicks on dsDNA and the other recognizes the nicked site and cuts the opposite DNA strand across from the nick, producing dsDNA breaks. The resulting DNA fragments contain short overhangs, 5´-phosphates, and 3´-hydroxyl groups.
http://www.neb.com/nebecomm/products/productM0348.asp
An aside here: As scientists I think we should demand that commercial entities wanting to sell us products reveal the compositions of those products. Lamentably this has become increasingly rare. But NEB displays a level of transparency in this regard that should be required of all companies wanting to transact with researchers using public funding. Just wanted to mention that. I have no connection with either NEB or Epicentre, other than I buy their products.
Epicenter's Nextera uses their "Transposome" technology to randomly bomb DNA. The ends of their engineered transposon apparently are not connected, so each insertion generates two ends (a break), each terminated with an appropriate adaptor. The trick in next gen library construction is always getting adaptor "A" on one end and "B" on the other. Since the "A" and "B" ends of the Transposome end up terminating different molecule, we would still expect half the fragments generated to be "A-A" and "B-B". But that is no worse than ligation methods deliver, and it cuts out the need for a ligation step. More information here:
http://www.epibio.com/nextera/nextera_tech_overview.asp
Just to be clear about the benefits, physical fragmentation methods generally require substantial "hands on" time for every library being constructed. That is, you take the DNA sample, you load it into the device (nebulization unit, sonicator, or hydroshear), perform the procedure, then unload. Even ignoring issues like the potential for cross-contamination and maintenance costs for these devices, scaling is an issue. Making 10 libraries is do-able. But 96?
So the allure of non-mechanical methods is obvious--you could probably set up either of these in a 96 well plate and process using multichannel pipettors, or even use laboratory robotics. But let us not forget why we flocked to mechanical methods in the first place: enzymes never seem to be as random as the mechanical methods. Being derived from living systems they come with their own agenda, so to speak. And that agenda usually results in bias. My advice would be to pay attention to the tests both companies did on differential depth of coverage for libraries generated using their methodologies.
--
Phillip
Comment