Big Blue goes after holy grail

YORKTOWN HEIGHTS, N.Y.—Although it hasn't claimed victory yetin the race toward a $1,000 genome, computer technology giant IBM has boldlyentered the fray in the hunt to create technology that will reach that goal—andit's already thinking down the line to something as low as a $100 genome if itsDNA Transistor works as well as hoped.

"What is the next big thing in biotechnology?" asks Dr.Gustavo Stolovitzky, manager of functional genomics and systems biology of theComputational Biology Center at IBM's T.J. Watson Research Center. "The answeris kind of simple if you are in the field, and what we need to know is how tosequence DNA, fast and cheap."

The basic premise of the work that Stolovitzky and hiscolleagues are doing at IBM Research is to use something very much like atypical silicon microprocessor-style chip, with a 3-nanometer-diameter holethrough which a DNA molecule would be passed. Using electrical charges turnedon and off at intervals, the DNA would progress through the DNA Transistor anda sensor would read the DNA step-by-step.

"This device [has a] few tiny electrodes that allow us tomake the DNA go through this little pore and be trapped by the electric fieldcreated by these electrodes," Stolovitzky explains. "We can [also] get rid ofthis electric field and the DNA will traverse through the pore a little bit,and then we trap it again, and each time we trap the DNA, we can do somethingwith it. For example, interrogate each base and ask, 'Are you a C, a G, an A ora T, and that will allow us to sequence the DNA."

"What we are doing is using an electric field to control theflow of DNA strands in the nanopore," adds Dr. Stas Polonsky, another IBMresearcher on the project. "By applying voltages to the metal layers lining thenanopore, we create potential wells, which interact with the charges along thebackbone of the DNA strand, moving it along one base at a time."

Using this technology, Stolovitzky says, it is very possiblethat IBM could make the $1,000 genome milestone achievable. That is the "holygrail—the gold medal—of the DNA sequencers," he says, and the point at whichpersonalized medicine will truly become a feasible reality.

In the past several years, the cost of genetic sequencinghas been falling at a rate of tenfold annually, George M. Church, a Harvardgeneticist, told The New York Timesrecently, and he anticipates the industry will continue with impressive gainsalong those lines for the foreseeable future.

Reporting on the IBM research, The New York Times noted that at least 17 startup and existingcompanies are in the sequencing race, pursuing a range of "third-generation"technologies, including United Kingdom-based startup Oxford Nanopore (seesidebar below). The newspaper also notes that human genome sequencingcosts currently run between $5,000 and $50,000, although none of those effortshave yet been "completely successful" and no one has yet produced an entiregenomic sequencing of a single individual.

Writing at ScienceBlogs.com, Daniel MacArthur, who writes about the genetic testing and genomicsindustries, was intrigued by IBM's work, but he expressed much greaterconfidence that Oxford Nanopore would produce a mature sequencing platformbefore IBM would.

"I am trying to keep my eyes in the sky but my feet on theground," Stolovitzky says of the work he and his colleagues are doing at IBMResearch. "We have a number of steps before we can overcome all thelimitations. Having said all that, we expect that with the DNA Transistor, inprinciple, in a few hours you should be able to sequence a human genome."

One of the great advantages of the DNA Transistor, if itworks as intended, would also be that it might eliminate the "complicated andtime-consuming and cost-consuming manipulations of the DNA at the input of thisdevice," Stolovitzky notes, and make the entire process of retrieving DNAinformation an almost fully electronic one.

One of the key goals right now, IBM notes, is to optimize aprocess for controlling the rate at which a DNA strand moves through thenanoscale aperture—not surprising, since completely controlling DNA's movementshas been a challenge for every company pursuing nanopore sequencingtechnologies.

"Slowing the speed is critical to being able to read the DNAstrand," notes an IBM news release announcing the DNA Transistor research. "IBMscientists believe they have a unique approach that could tackle thischallenge."

The work that IBM is doing brings together a team ofscientists from four fields in the Yorktown Height, N.Y.-based Watson ResearchCenter—nanofabrication, microelectronics, physics and biology. Also, the IBMResearch division is one of several companies to have been awarded grantstotaling almost $6 million by the National Human Genome Research Institute toinvestigate technologies able to provide personalized genome sequencing forless than $1,000—about a half million dollars in IBM's case.

"The beauty of [what we are doing] is that [the DNATransistor] uses technology that IBM is very good at doing," Stolovitzky notes,"which is a mixture of things like nanotechnology, very sophisticated atomiclayer deposition and chemical vapor deposition—all kinds of technology that hasbeen very much at the forefront of the thinking and innovation at IBM."

Oxford Nanopore pursues protein-based nanopore sequencing

OXFORD, U.K.—Existing genomic sequencing methods tend torely on expensive optical technologies, fluorescent labels and in some casescomplex sample preparation, and that is why a label-free, electrical,single-molecule method is needed to make sequencing truly affordable to themasses, notes Oxford Nanopore. To that end, the United Kingdom-based companyhas partnered with the likes of Illumina, Harvard University and the Universityof California, Santa Cruz, between late last year and early this year onnanopore sequencing technologies.

In late February, Oxford Nanopore also published a paper in NatureNanotechnology

demonstrating that it can detect unlabelled DNA bases andmethylated cytosine using a protein nanopore covalently attached to an adaptermolecule.

According to the company, this "validated the feasibilityand accuracy of the nanopore sensing component of Oxford Nanopore's sequencingsystem … As such, the work represents a step toward the company's goal ofdeveloping the first label-free, single-molecule DNA sequencing technology."

Each of the DNA bases reportedly causes a characteristiccurrent disruption as it moves through the nanopore that Oxford Nanopore hasdeveloped, thus allowing continuous sequencing without fluorescent labeling.

In contrast to Oxford Nanopore's technique—using aring-shaped protein at the top of the pore through which individual base can becut and passed through—the recently announced DNA Transistor research from IBMis said to used a "naked" nanopore. Oxford Nanopore does expect to move to anaked nanopore system ultimately, but probably not until after it has releaseda commercial platform that can actually handle continuous strands of DNA usingthe protein-based system.