Nettime mailing list archives

<nettime> DNA and Computers
eduardo on Wed, 3 Sep 2003 06:47:43 +0200 (CEST)

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

<nettime> DNA and Computers

Below is an interesting article on DNA and computers... one day we may have
little software working on our flue viruses. 

Eduardo Navas


DNA forms building block for next breed of computer

By Jonathan Sidener 

September 1, 2003 

For years, researchers have taken advantage of the ever-increasing power
of computers to crack the genetic code.

But Scripps Research Institute chemist Ehud Keinan and a handful of scientists
around the world are going in the opposite direction, using DNA ? the blueprint
for cellular life ? to crunch numbers inside a new breed of computer.

If the research is successful, children one day might operate a computer
powered not by silicon chips, but by "biochips" that run software and store
data using the same double-helix of DNA that determines whether our eyes
are blue or peas are green.

And it is conceivable that future generations may have microscopic DNA computers
coursing through their veins, capable of diagnosing and perhaps treating
a variety of ills at the molecular level.

"You can't compare and say a DNA computer is or isn't faster than this,"
Keinan said, gesturing toward a sleek, metallic Macintosh PowerBook in his
Scripps office. "It's a different type of computer."

DNA computers don't have keyboards and monitors. The computing takes place
on the laboratory bench as complex molecular-chemical reactions.

The first calculations, nearly 10 years ago, took place inside beakers and
test tubes.

A new generation of DNA computing uses biochips, devices built using semiconductor
manufacturing technology. A biochip has millions of pieces of DNA on its
surface instead of the millions of electronic circuits on a computer chip.

At the moment, DNA computers are laboratory curiosities. One computer can
play a respectable game of tick-tack-toe. Another can solve chess riddles.
The computations they handle would make the least powerful of today's computers

But DNA computers show some intriguing qualities.

DNA is extremely efficient, both in storing data and in its use of energy.
One gram of DNA, which would take up about as much space as an ice cube,
can hold as much information as 1 trillion compact discs.

With today's computer chips, energy consumption and the heat produced as
a byproduct can cause malfunctions. But the chemical reactions that make
a DNA computer work require little energy.

Most significantly, the biomolecular computers operate on different underlying

Electronic computers make their calculations by processing a series of zeroes
and ones, or binary code, one character at a time in a rapid sequence, like
a machine gun that fires a series of bullets from a single barrel in succession.

Not so with DNA computers. Because millions of DNA snippets can fit into
a drop of water, DNA computers can make many parallel calculations at once,
more comparable to a shrapnel grenade that launches many projectiles at
the same instant.

To Keinan and other researchers, this parallel processing provides much
of the allure of DNA computing, the idea that machines built with a fundamentally
different computing engine will be able to tackle fundamentally different

A young science
DNA computing was born in 1994 when University of Southern California researcher
Leonard Adleman devised a way to solve a mathematical brain teaser using
fragments of DNA.

Several research teams around the globe have taken different approaches
to this computing via biology, and it remains unclear which approach, if
any, will emerge from the laboratory.

Electronic computers read a language consisting of two characters, 0 and
1. DNA computers read a language of four characters, the four molecules
that make up DNA, known by their initials A, T, G and C.

Like its electronic brother, a DNA computer has software and hardware. The
software is the DNA string. The hardware consists of enzymes that "read"
the DNA strand and snip it at precise, known locations.

After each enzyme snip, four unbonded molecules remain at the end of the
DNA snippet. DNA does not like to have unbonded molecules, so a second enzyme
repairs the strand according to a set of rules hard-wired into DNA's biology.

DNA computers use the predictability of these repairs to perform their calculations.

The enzymes make a series of snips. At the final snip, the four letters
of the unbonded molecules provide the answer or output of the calculation.

'Language of biology'
Keinan is part of a team from the Technion, Israel's Institute of Technology,
that built the first autonomous DNA computer.

In addition to his work at Scripps, Keinan is the founder and head of the
Institute of Catalysis Science and Technology at the Technion.

Before the Technion's work, molecular computers had required human guidance
through a series of chemical reactions.

Keinan, an organic chemist who drifted into bio-molecular computing after
a chance meeting with another Israeli researcher, won't guarantee that DNA
computers will ever get beyond the laboratory.

"Maybe in 10, 20, 30 years, maybe not at all," he said. "The parameters
are so attractive that I believe something wonderful will come out of this.
But it would be foolish to make a prediction."

Pavel Pevzner, a computer science and engineering professor at the University
of California San Diego Jacobs School of Engineering, agrees that it is
too early to divine the future of DNA computers.

"The problem is how to prove that this is not just an intellectual game,"
Pevzner said.

So far, researchers are missing a breakthrough, he said.

"They have not found a problem DNA computers can solve that cannot be solved
with conventional computers. They need that killer application if they're
going to be taken seriously."

Keinan and his colleagues in Israel are about to publish the results of
research in which the output from a calculation turns on or off the pigment
gene in a colony of bacteria.

The answer "Yes" might produce a beaker full of blue bacteria, while "No"
would produce a colorless fluid.

While this is a simple example, it is a first step toward devices that can
interact with biology.

DNA researchers hope that one day DNA computers will deliver drugs or repair
cells from inside the bloodstream.

"DNA computers talk molecules, which is the language of biology, and therefore
they can talk directly to cells and tell them what to do," Keinan said.

"It is true that DNA computers still lag behind silicon computers. Nevertheless,
even today, DNA devices have already been able to do things in my lab, and
go to places, that silicon computers cannot."

Jonathan Sidener: (619) 293-1239; jonathan.sidener {AT} uniontrib.com

#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo {AT} bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime {AT} bbs.thing.net