It's in the genes: DNA technology could change the way we compute. (Inside Technology).
Sure, computer-chip manufacturers are burning the mid-night oil developing the next microprocessor to topple current speeds. But if you subscribe to Moore's Law--Intel founder Gordon Moore's assertion that personal computer speed doubles every year--then microprocessors made of silicon will eventually reach their limits in speed and miniaturization. That means producers will eventually need new materials to power PCs. In addition to DNA, scientists are also considering "quantum computing" and other such Star Trek-like solutions. And while DNA is perhaps the last substance that comes to mind when you think of your desktop or laptop, it does have the potential to perform calculations faster than today's most robust computers while storing colossal amounts more data.
So how did we get from PCs to DNA? The movement began about eight years ago when Leonard Adleman, a computer scientist at the University of Southern California, introduced the idea of using DNA to solve complex mathematical problems. Adleman's findings and past research show that DNA is similar to computers in the way it stores permanent information about our genes. Helping give credence to the movement, he also discovered that one gram of DNA can hold as much data as one trillion CDs--that's a lot of MP3s. Adleman used his DNA computer to solve the Hamiltonian Path problem that most of us likely encountered in junior high or high school math class. Also known as the "traveling salesman" problem, the goal is to find the shortest route between seven cities going through each only once.
So how does this Star Trek-meets-Fantastic Voyage technology work? First, because it's in its infancy, most existing DNA computers consist of only synthetic, made-to-order DNA strands attached to gold plates on one end, with the other end floating freely in test tubes or petri dishes that are linked to myriad scientific devices in university labs. Second, the most rudimentary explanation of operation is, just as current hardware or software is programmed, so are made-to-order, synthesized, single DNA strands. "DNA is composed of four basic building blocks: A, C, G, and T," says professor Lila Kari, research chair in bio computing at the University of Western Ontario in London, Canada. "DNA is just like an alphabet. In the same way you can use the alphabet to write, say, communist propaganda or Walt Whitman poems, you can use it to write human genes or to write numbers."
When these single strands are placed in the proper solution and environment, they seek out their complementary counterpart, thus helping to perform the required calculation or task in what amounts to "contents addressable storage." In a task such as finding the balance in a banking account, one strand with the account holder's personal information seeks out its complementary strand, which possesses the balance information.
If all this sounds like science fiction, just remember a few decades ago it seemed impossible for a chip made of silicon--ones we use in computers today--to quickly and accurately spell check a document or enable you to play solitaire on your PC. Another potential benefit of DNA computing is performing calculations in parallel vs. taking on one task at a time, as most computers today do. With many DNA molecules in a test tube or on a chip, you're doing many computations simultaneously. "If you're using a milligram of DNA, you're processing, all at the same time,10 to the 17th power independent bytes of data," says David Harlan Wood, a research professor at the University of Delaware. That's in comparison to some of today's conventional parallel computers that do tens, hundreds, or maybe thousands of computations in parallel, he says.
ARE WE READY FOR DNA TECH?
Despite the current research, it's unlikely that the technology will make its way to the average American household. For now, the main issue is where is DNA computing most suitable. One thing for sure is you can bet these "wetware," or biological systems, won't be available in your local Best Buy or Circuit City anytime soon--if ever. Some experts believe home-based DNA computers are possible but for now think they'll be better suited in corporate or government settings, solving voluminous calculations, cracking secret codes, or helping the government with its current war on terrorism. "You might be happy if you can go to Circuit City and purchase something that would have thousands of CDs worth of music, but would it be worth it to you to buy something with the equivalent of millions or trillions of CDs?" asks Wood. Probably not.
It's also likely that DNA computers will be better at human-related tasks. "Perhaps DNA computers will be better at problems at which humans are better," says Kari. "Electronic computers are better than humans at adding, and they will always be. But if you want to talk about face recognition, humans are much better than the best software program. Maybe there will be a niche for some kind of specialized problem for which DNA computing might be better."
Whether the average Joe will someday get his hands on a DNA-powered PC of some kind is debatable. Not many people thought the PCs we use today would be so pervasive, considering their precursor was the simple calculator, a device developed from vacuum tubes and used to measure the trajectory of artillery shells during World War II. But even as DNA technology gains its legs, many in the industry don't seem to be worried about it replacing the silicon chip. "The use of silicon is going to carry us for the foreseeable future," asserts Drew Prairie, a spokesperson for Sunnyvale, California-based AMD, the industry's second-largest chip supplier for PCs. "DNA computing and other ideas are far down the road; we still have plenty of life left in transitioning our current solutions."
Adds Colin Hill, founder and CEO of Gene Network Sciences in Ithaca, New York, "That's a field that's yet to do anything substantial even though it holds a lot of promise" Hill's company simulates life in silico (on computers) to advance drug discovery. "There are a number of limitations as to how far DNA computing can get in the near term."
Despite its naysayers, DNA computing scientists and biologists are marching forward. Simple experiments such as teaching DNA molecules to learn effective strategies to play a game of poker, like the one Wood's group is working on, is part of the necessary evolution of the technology.
|Printer friendly Cite/link Email Feedback|
|Author:||Mckay, Jason P.|
|Date:||Nov 1, 2002|
|Previous Article:||Don't blow it. (Savvy Solutions).|
|Next Article:||Back to business: Ken Coleman says a weak tech market is a good time to start a company. (Tech Buzz).|