Sunday, October 21, 2012

Nanopores: Fission or Fusion?

It's fall, and the foliage in New England is putting on its usual spectacular show.  What isn't showing is any of the progress in nanopore sequencing that I got caught up in last February.  Oxford Nanopore made quite a splash, and I wrote a quite breathless account based on a phone conversation with them.  Since then, other than raising some serious cash, Oxford has been quite quiet, and has neither released any data the community (or at least this blogger!) is clamoring for nor is there any sign that alpha units have been placed.  Genia was another nanopore company making noises about unveiling this year, but now they have licensed a new nanopore-based sequencing chemistry and promised boxes in 2014.   When will nanopores actually hit the market?

In the LinkedIn NGS board, someone posted an item about a review that either examines the remaining difficulties of nanopore approaches or makes them sound insurmountable; I'm not sure as I haven't broken through the paywall (a cardinal rule of this blog is it must not consume personal funds!).  Did the publisher tart up a teaser to get more business?  Is the author a creditable expert on the field (disputed in another thread)?  Both really outside my scope.  But, I do have to mildly object to a comment made in the LI board that "it just may be that nanopore sequencers are the perpetual motion machines or our age".  We know perpetual motions machines just cannot ever work, but I think I know what the commenter (Tim Hunkapillar, who clearly knows a thing or two about DNA sequencing technology) was trying to get at: the risk that nanopores are a technology that can never quite deliver.

To me, a more apt comparison would be to the world of nuclear power, and the question is whether nanopores are more akin to nuclear fission or nuclear fusion.  With the Manhattan Project, the United States pushed from the cutting edge of laboratory research to demonstrating the ability to execute either controlled or uncontrolled nuclear fission chain reactions.  Less than a decade was required to go from the first controlled fission reaction to a prototype nuclear reactor (according to Wikipedia).  Since then, nuclear fission has been an important, if troublesome, contributor to the electric power requirements of many nations.

On the other hand, despite a significant investment over more than half a century by multiple nations, no working fusion reactor prototype yet exists.  Since I was a young lad watching Nova (and before!), the promise of energetic break-even has always been just around the corner.  But that has never quite happened.

So nuclear fission and fusion represent two very different technological stories.  Both looked like primarily engineering projects, with the science mostly solved.  Fission (and Project Apollo) largely played out in the manner, though not without a lot of nasty (and sometimes fatal) surprises en route.   Fusion's story is not over, but is in danger of never being realized.  There is always a risk that a fusion facility would become so expensive to build that it would never make economic sense.

Will nanopores ever deliver fast, cheap sequencing?  Are the seemingly constant shifts in direction, chemistries and such really necessary tacks required on the course for victory, or are they random wanderings that will never reach home?  Or, perhaps nanopores will succeed, but only after becoming economically obsolete due to the advent of another sequencing technology?  As is attributed to either Yogi Berra or Niels Bohr (or Nielsi Bohra?) , prediction is hard, especially about the future.


Jacob Rosenstein said...

It seems worth pointing out that contrary to the impressions of some of threads linked above, this review doesn’t have any particularly radical opinions that need to be disputed. Prof. Wanunu is a very well-regarded researcher and most of this paper highlights interesting biophysical experiments that have been done with nanopores, despite the fact that nanopore DNA sequencing has been slower to arrive than everyone might have hoped.

Anonymous said...

Keith -
I asked them to lecture to a class on their basic technology, and received polite refusals, as they were working "on the final development of GridION and MinION in preparation for commercialization, which has us all fully committed."

Nanoporous said...

Few comments on this:

1. The Wanunu paper is a few months old (I read it back in June I think). I wonder why the publisher chose to highlight it now?

As it is, the review is nowhere close to being as dismissive of nanopore sequencing as suggested in the press release. As pointed out in the first comment, the review mostly deals with the biophysical aspects of nanopores. The paper also describes some of the drawbacks that could be associated with the enzymatic slowing of DNA - ie similar to the approach taken by ONT (though ONT is supposedly using a motor protein). These drawbacks have been pointed out by a number of people over the years at conferences and forums.

2. While ONT has not released any data, but has received all the hype, very few people paid attention to the work of the Gundlach lab. They are using polymerase to slow down the DNA -which indeed has some disadvantages - but have been very successful in demonstrating single base discrimination for long DNA molecules. This is not actual sequencing, but the closest published result out there validating the nanopore approach.

(I can send you pdfs of both the papers if you are interested. please email at nanoporous at omespeak dot com).

Overall I agree with your post about wrongly dismissing the technology as 'perpetual motion machines of our age'.

[Full disclosure: I work for a company developing nanopore sequencing that is not named Genia, ONT, or Nabsys; nor am I associated with the Gundlach group in nay manner.]

sciengr said...

There's a company in Seattle called Stratos Genomics. I saw they announced a financing based on achieving some level of sequencing on WT alpha hemolysin pores and claim to have a solution for homopolymers. If the animation on their site is accurate this could be the path to commercial nanopore sequencing. I hadn't heard anything about these guys until their press release, anyone have more on them?

Brian Krueger said...

If you read the paper based on the new Genia licensed tech, it's a lot of proof of concept hand waving. They use PEG-bases with varying pegylation. They show that the polymerase can incorporate the base and spit off the PEG, but then say that the leftover triphosphate is too charged. So they replace it with an NH2 using alkaline phosphatase and pass THAT through the pore. But instead of linking all of these steps together, they chemically synthesize the PEG by-products and pass those through the pore. Of course, they say in the future they can link all of these steps together, but the technical hurdles seem much greater than a 2 year solution even for this technology.

The Wanunu review is pretty good, although, of the 30 pages, only about a paragraph is spent on nanopore DNA sequencing. The review wasn't dismissive at all, but said that even after decades of research the tech is still in its infancy and needs to overcome many hurdles.

Anonymous said...

Wanunu paper is available from his web site: