IBM's announcement that it will be entering the DNA sequencing technology race (which I mentioned briefly earlier today) has created a tremendous amount of mainstream media interest. That's understandable given IBM's size and history of innovation, but how likely are they to represent serious contenders for the potentially lucrative sequencing market?
Firstly, it's important to note that this announcement represents more of an expression of interest in the field rather than anything close to a mature technology; IBM will not be rolling out a machine to sequence your genome at any stage in the near future. Rather, the press release seems to have been triggered by yesterday's announcement that IBM had been awarded a $557,000 grant from the National Human Genome Research Institute to help develop the sequencing technology.
IBM's proposed technology (which would make use of solid state nanopores, basically tiny holes in a silicon membrane) would certainly offer substantial advantages over existing sequencing platforms, but there remain some serious challenges to developing this tech. A friend working in the nanopore area proposed these as the most serious obstacles ahead:
- Slowing down the DNA for long enough to read each base;
- Manufacturing consistent holes in a solid state medium - something that is extremely hard at a nanoscale (other current nanopore approaches instead use biological nanopores that are sturdy, cheap to produce using bacteria, and highly consistent);
- Reading each base to the accuracy required for an industrial sequencing technology.
It's important to remember that nanopore technology has already been the subject of intense research by other companies in the field, most notably Oxford Nanopore. Oxford has maintained a powerful stealth field over progress with their technology (in contrast to other sequencing companies), but everything I hear about them so far has been impressive; I certainly expect to see a mature nanopore sequencing platform from Oxford long before I see anything from IBM.
So: an interesting development, and certainly a platform to watch, but I'm not particularly excited about IBM sequencing yet. We'll see what happens over the next few months.
Subscribe to Genetic Future.
Life Science
Medicine & Health
Comments
You might want to add that one aspect of this is the bioinformatics tools needed to process and analyse the data. This may also be an aspect that IBM may be aiming to address, too. In many ways, the algorithms needed to process and analyse the data are just as much of a problem as the hardware issues. Just a thought.
Posted by: BioinfoTools | October 6, 2009 3:59 PM
Bioinfotools,
The assembly tools for nanopore sequencing are far simpler than for pyrosequencing approaches, as nanopore sequencers are expected to generate very long (100s to 1000s of kbp) reads...the annotation tools are the same used for other approaches, so I don't understand your point.
Posted by: MarkEE | October 7, 2009 3:13 PM
IBM was one of the first companies in the world to foray into “Project Checkmate,” in which IBM and The Scripps Research Institute proposed to conducted advanced biological research on influenza viruses in 2006! The effort involved using a powerful IBM Blue Gene Supercomputer to model the molecule hemagglutinin—the H in H1N1—in an attempt to predict the way flu viruses mutate.
Far as I know, the H1N1 pandemic began around April 2009.
Just makes ya go hmmmmmm.
Posted by: ahrcanum | October 8, 2009 8:47 AM