Big Blue’s DNA-Reading Chips
IBM researchers are developing a chip for cheaper, faster DNA sequencing using fabrication techniques refined through semiconducting manufacturing. The chip uses layered electrodes to control the movement of individual DNA molecules and exploits a technique called nanopore sequencing. The approach could allow DNA to be passed through a sensor that would rapidly read off its genetic code.
Nanopore sequencing is attractive because, unlike existing sequencing methods, it could read long stretches of DNA without the need for labels or chopping and amplifying enzymes. “If this works, you should be able to read tens of thousands of bases with no labels, making it cheap and fast,” says Jeffery Schloss, program director for technology development at the National Human Genome Research Institute. Being able to read long stretches of DNA without chopping it up would also make the data-processing side of genome sequencing simpler. “If you can do long reads, you don’t have to make assumptions about the sequence or match it to existing sequences” in order to put it back together, says Schloss.
Several research groups are developing their own approach to nanopore sequencing. All involve the movement of DNA molecules through a tiny pore one base at a time; as the bases move through the pore, they can be read using various techniques. But one of the biggest obstacles to making a practical nanopore sequencer has been controlling the rate of the movement of the DNA. This is the problem the IBM group is working on. “The DNA goes through the pore too fast,” says Gustavo Stolovitzky, manager of functional genomics and systems biology at IBM’s T. J. Watson Research Center in Yorktown Heights, NY.
For the past two years, Stolovitzky’s group at IBM has been developing chips arrayed with “DNA transistors” that use layered electrodes to control the movement of the DNA. The electrodes are built on the company’s research fabrication line using the same technology employed to make silicon integrated circuits.
The IBM researchers first deposit conducting and semiconducting materials that will act as electrodes onto silicon wafer layers each about three nanometers thick. Then they use a transmission-electron microscope to blast a hole as small as one nanometer in diameter in the stack. A chip is cut from the wafer and placed in the middle of a container of potassium chloride, like a partition. DNA molecules are placed on one side of the solution, and a voltage is applied across the chip. Because DNA has an electrical charge, the IBM researchers can control its movement through the pore by using the electrodes to create electrical fields.
The IBM researchers are now performing simulated experiments to refine the chip’s design. The properties of the system can be varied by, for example, changing the thickness of the layers that make up the electrodes and the size of the pore. The movement of the DNA can also be altered using different voltages in the electrodes. Instead of fabricating every potential chip design and testing every voltage, however, the researchers are modeling the nanopore system using an IBM Blue Gene supercomputer. The software running on this machine can calculate the physics of tens of thousands of atoms in the DNA molecule and in the chip every picosecond. A version under development will enable them to model 200,000 atoms at this rate, says Stolovitzky.
The IBM group is working on methods for sensing each base as it passes through the pore. With modifications, Stolovitzky says, the same electronics used to control the movement of the DNA could also be used to measure electrical properties that distinguish the bases making up the genetic code.
“We look at this as a data problem,” says Stephen Rossnagel, a researcher at IBM Watson. Sequencing a genome today, Rossnagel says, involves making sense of three gigabits of data that’s “mixed up” and has to be put back together. Directly reading pieces of DNA without chopping them up simplifies this problem, and the DNA transistors could be made in large arrays, each reading the same sequence. The more times the same stretch of the genome is read, the better the quality of the resulting sequence. Rossnagel says the approach IBM is pursuing should be simpler to integrate with the microelectronics needed to crunch the resulting data.
According to Schloss, the IBM nanopores, which could be fabricated in large arrays, could prove more practical than previous efforts. “The ways this has been done before don’t lend themselves to sequencing,” he says. Some groups have slowed the movement of the DNA across a pore by attaching a bulky molecule to it that must be pushed down the strand as it passes through the nanopore. Others have stationed an enzyme at the pore that cuts the strand and passes the bases through individually. Controlling the movement of the DNA with microelectronics might prove more practical, and it seems to allow for better control, says Schloss.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
Meet the people who use Notion to plan their whole lives
The workplace tool’s appeal extends far beyond organizing work projects. Many users find it’s just as useful for managing their free time.
Learning to code isn’t enough
Historically, learn-to-code efforts have provided opportunities for the few, but new efforts are aiming to be inclusive.
Deep learning pioneer Geoffrey Hinton has quit Google
Hinton will be speaking at EmTech Digital on Wednesday.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.