2
$\begingroup$

I am writing the section about history of DNA sequencing in the introduction chapter and after reading quite a few research papers, I am still confused about them. Here I compile some questions to make sure that I understand it correctly.

  1. Is second generation sequencing the same as next generation sequencing?
  2. Is Sanger sequencing the first generation?
  3. As of today, is there any commercial 3rd generation sequencing technology in use (or is it still in development)?

If there is any paper reference that could explain all of the above then it would be really great.

$\endgroup$
7
$\begingroup$

I'm assuming you mean DNA sequencing (excluding things like RNA-seq).

Is Sanger sequencing the first generation?

From Metzker 2010:

The automated Sanger method is considered as a "first-generation" technology

Some of the technology that was in development when this review was written is no longer in development, but this is still an excellent review and a great place to start.

Is second generation sequencing the same as next generation sequencing?

Also from Metzker 2010:

and newer methods are referred to as next-generation sequencing (NGS).

The "newer methods" in the review are:

  • Emulsion PCR based like Roche/454
  • Solid phase amplification like Illumina/Solexa
  • Single molecule like Helicos BioSciences and Pacific Biosciences

As of today, is there any commercial 3rd generation sequencing technology in use (or is it still in development)?

The development of single molecule methods (third item in the above list) has historically taken longer than the other two, so sometimes they are referred to as third generation. However, note how Metzker above considers them all "next-gen".

A recent paper from PacBio itself makes the distinction of halting:

Second-generation sequencing (SGS): Sequencing of an ensemble of DNA molecules with wash-and-scan techniques.

Third-generation sequencing (TGS): Sequencing single DNA molecules without the need to halt between read steps (whether enzymatic or otherwise).

Ultimately, I think the "generations" are more useful to make the "Sanger vs. Not Sanger" distinction. Sanger sequencing is very different than next-gen in terms of what kind of experiment you can do and what sort of data you obtain.

Past that, the number of a generation is more of a marketing term. For instance, at the bottom of the PacBio paper:

Conflict of Interest statement. All authors are employees of Pacific Biosciences and own stock in the company.

In practice, PacBio competes with second generation platforms. It is in their interest to say their technology is on "another level" compared to what's available on the market, but in practice, this is debatable.

Anyhow, to answer your question:

  • There is a technology in use and commercially available which has been referred to as third gen (PacBio)
  • There was a technology in development that falls under most definitions of third gen (Helicos), which went bankrupt in 2012
  • There is a technology in development (ON MinION) that is single-molecule based and was introduced after PacBio, so if PacBio is third generations, so is MinION (although the MinION is apparently very different from PacBio)
$\endgroup$
  • $\begingroup$ Oxford Nanopore shipped MinIONs to a some researchers a few weeks ago. $\endgroup$ – 5heikki Sep 7 '14 at 1:01
  • $\begingroup$ @5heikki I was actually unfamiliar with MinION. Does it count as third-gen? Feel free to edit my answer and add it (or submit your own). $\endgroup$ – Superbest Sep 7 '14 at 7:41
  • $\begingroup$ In my opinion referring to Sanger sequencing as 'first generation' is historically incorrect. Maxam-Gilbert sequencing preceded it and produced lots of useful results - see Addendum in my answer here: biology.stackexchange.com/questions/17598/… $\endgroup$ – Alan Boyd Sep 7 '14 at 8:12
  • $\begingroup$ Ultimately we, as scientists, shouldn't get hung up on the exact use of this terminology since it is clearly now just part of marketing the latest advance in sequencing technology. $\endgroup$ – Alan Boyd Sep 7 '14 at 8:15
  • $\begingroup$ @AlanBoyd Strictly saying, it is not about precedence: the distinction between the first generation and the next generation is in terms of throughput. In this respect both Maxam-Gilbert and Sanger belong to the same class of methods. $\endgroup$ – har-wradim Sep 7 '14 at 19:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.