Technology for Turning the Promise of Personalized Medicine into a Reality

Steve Hamm in Smarter Planet / new intelligence, July 1st, 2010

This is the way innovation sometimes happens: Two people who work for the same organization but have different areas of expertise meet by chance, start talking, and come up with an idea that has the potential to make a big difference in the world.

In this case, the people are Gustavo Stolovitzky and Stanislav Polonsky, two scientists at IBM Research in Yorktown Heights, New York. Gustavo’s specialty is genomics and Stas’ is semiconductors. Three years ago they met in a corridor at the lab and began talking about some of the challenges in biology. Gustavo said great things could be accomplished in medical science if researchers had the ability to sequence an individual’s entire genome quickly and inexpensively. The two discussed the possibility of using semiconductor technology to help create a quick and inexpensive gene sequencing machine. Within a few weeks they came up with the idea for what we now call DNA Transistor–a new way of sequencing the genome.

That bold concept came a major step closer to the marketplace today when Roche and IBM announced a partnership to develop a gene sequencer based on the DNA Transistor that the companies hope will directly read and decode DNA quickly and efficiently. “The big potential payoff from such technology is the ability to do things cheaper, faster, and better,” says Ajay Royyuru, the senior manager at IBM Research’s computational biology center.

Cheaper, faster, and better could make a big difference in the effort to draw knowledge from genomic research that will dramatically improve physicians’ abilities to correctly diagnose and treat diseases. It required $3 billion in research spending to sequence the first individual genome, in the US-funded Human Genome Project, and, so far, only a handful of people worldwide have had their entire genomes read. As a result, the great promise of genomics research has yet to be realized. A June 12 article in the New York Times marking the 10-year anniversary of the first draft of the Human Genome Project concluded that “medicine has yet to see any large part of the promised benefits.” Roche and IBM hope that their joint research make it possible to analyze an individual’s genome for a cost of between $100 and $1,000. A cost that low would make it practical to sequence the entire genomes of many individuals–enough to draw reliable conclusions. Roche and IBM will jointly develop the technology and hope to produce a working prototype in three years.


Here’s how the DNA Transistor works: A silicon membrane is placed vertically in a solution-filled chamber, dividing it in two. Strands of genetic material are placed in one side of the chamber, and electrical charges are applied–negative in the part of the chamber where the genetic strands reside, and positive on the other side. The positive charge draws a strand of genetic material into a tiny opening, or nanopore, in the membrane. As the strand passes through the opening, the combination of the electrical charges and the composition of the membrane has the effect of ratcheting the strand through one bead of genetic material at a time. This mechanism makes it possible to accurately and quickly read the genetic makeup of the strand.

There are plenty of challenges remaining for the researchers. Gustavo and Stas have shown through simulations and experiments that the ratcheting effect should work. Now, working with Roche scientists, they have to do more inventing to put the concept into practice. In addition, they need to develop a sequence-reading mechanism.

The collaboration got off to a promising start on June 29, when a dozen Roche scientists met with a dozen IBM scientists for discussions and a brainstorming session at the Yorktown lab. It was an emotional experience for Gustavo: “I remembered when this was a pie-in-the-sky idea, and I suddenly had this great feeling that you rarely get–when you feel you’re doing something that’s very worthwhile.”

The two companies seem well matched–even to the point of having the code-name “Watson” in common. Roche’s 454 Life Sciences subsidiary demonstrated the effectiveness of its existing gene sequencing technology by reading the genome of James Watson, the scientist who along with collaborator Francis Crick defined the structure of DNA in 1953. Watson, of course, is also the name of IBM’s first CEO, T.J. Watson, and the code-name of our new question-answering computer, which is in training to take on champions of the Jeopardy! TV quiz show early next year.

The code-name for the joint Roche/IBM project is what you would expect: DNA Transistor.

For additional info, see:

https://researcher.ibm.com/researcher/view_project.php?id=1120
http://www.research.ibm.com/compsci/compbio

and links to videos
http://www.youtube.com/v/pKi30ai35mU&hl=en_US&fs=1&rel/=0&showsearch=0&showinfo=0
http://www.youtube.com/v/wvclP3GySUY&hl=en_US&fs=1&rel/=0&showsearch=0&showinfo=0

Technorati Tags: DNA Transistor, genome, IBM, personalized medicine, Roche