Living Cells Turned Into Computers

ananyo writes "Synthetic biologists have developed DNA modules that perform logic operations in bacteria. These 'genetic circuits' could, for example, be used by scientists to track key moments in a cell's life or, in biotechnology, to turn on production of a drug at the flick of a chemical switch. The researchers have encoded 16 logic gates in modules of DNA and stored the results of logical operations. The different logic gates can be assembled into a wide variety of circuits."

Re:

[Ol Biscuitbarrel]

Don't forget Greg Bear. [sfsite.com] That book is a trip-and-a-half.

Re:

[brantondaveperson]

Now there's a film adaptation I would pay money to see. Not that anyone has ever even mentioned the remotest possibility of making a film of Blood Music, but man, it needs to be done.

Re:

[bmo]

Forge of God
Eon
Blood Music
Dead Lines (his one ghost story).

I'd pay to see all those. I just mentioned it and the discussion was that a lot of Blood Music is internal dialog and it would be hard to represent without making it boring.

I think Dead Lines could be done in a Poltergeist-esque manner and is probably the easiest to do out of all of them.

--
BMO

multicellular cluster computing

[spazdor]

Imagine a Beowulf cluster of these babies.

Wait, I guess that's like basically just a person.

Re:

[neoshroom]

Haha. Mod this up. :)

Re:

[sourcerror]

Except that your neurons don't use boolean logic. There's a good reason why humans are really bad at numerical computation.

Also, in the article they're using DNA, not neurons. The funny thing is that the DNA much more resembles the classic Turing machine than any practical computer ever built.

Re:

[spazdor]

Who's talking about neurons? If the 'computation' being performed is that of 'generating a phenotype' rather than something mundane like cognition, then every multicellular organism is a compute cluster, whether it has a nervous system or not. ;)

Re:

[sourcerror]

Just because the DNA (the machine) is Turing equivalent doesn't mean that every organism (the program) is Turing complete.

" then every multicellular organism is a compute cluster, whether it has a nervous system or not. ;)"

A bunch of identical cells can just sit next to each other without communicating (see variuos algae), and I wouldn't call that a computational cluster.

Re:

[spazdor]

But they did communicate, by virtue of having come from the same mitosis process. Lots of parallel compute systems don't require inter-node communication after the nodes have received their initial work packet. Remember those "distributed.net" RC5-cracking competition clients? All they needed was to be told what section of the keyspace they were responsible for searching, and then they ran autonomously after that.

Re:

[davester666]

Just like popping out a baby...

Re:

[Ceriel Nosforit]

Except that your neurons don't use boolean logic.

This is not True.

AAGCCCAGACACAA

[Anonymous Coward]

Hello, world. Stop. Boobs!

Re:

[shaitand]

I'd be more excited if they could dead cells into live ones.

Re:

[fisted]

I'd be even more excited if there was a laser involved somewhere.

Re:

[game kid]

Laser zombie coprocessor-cats.

(Like regular laser cats, except with less SNL and more lifespan and bra[aaaa]ins for Folding@home.)

Why wetware ?

[Turminder Xuss]

I suppose the ability to store data and program instructions in DNA would enable a Von Neumann architecture. The possibility of simulataneous "operations" on different parts of the genome might even make common bus based bottlenecks (where data cannot be fetched simultaneously with an instruction) less of a limit. But the speed of the thing would be agonisingly slow compared to silicon. Massively parallel perhaps but slow as a wet week.

Re:

[loufoque]

But it could regenerate itself!

Diagnosis please...

[Anonymous Coward]

Combine this with the previous article and one could have a built in doctor.

Re:

[ikaruga]

I guess you mean IBM Watson medical libraries entirely coded in DNA and working inside a cell. I'd mod you interesting or insightful if I had points. I Just hope, assuming this technology actually becomes practical, that the cells inside my body coded with an advanced A.I. don't try to take over my stupid mind.

Great.

[bogidu]

We all know what happens when bio-gelpacks on our starships catch a cold!

Synthetic biologists

[a_hanso]

I think I'll wait until these results can be verified by real biologists.

Rebugging the Debugger

[VortexCortex]

These 'genetic circuits' could, for example, be used by scientists to track key moments in a cell's life or, in biotechnology, to turn on production of a drug at the flick of a chemical switch.

Code-Monkey Translation:
Scientists, lacking a good debugger for living organisms, have made a breakthrough: They're now able to employ the tried and true tradition of adding
printf( "Made it here and didn't crash!" );
and/or
if ( DEBUG && VK_LSHIFT_DN ) { ... }
code into bacteria.

Despite the platform being in open beta for as long as anyone can remember and its undeniable popularity the world over, professional coders experienced with situations that require resorting to this technique in undocumented code, badly supported 3rd party plug-ins, and poorly understood niche embedded systems, are advising the scientists to wait for the more mature 1.0 release of the DNA API specification before implementing their own domain specific language on the platform.

We are Borg...

[AlienSexist]

Getting closer all the time. Cybernetic enhancements, nanobots/nanoprobes, and a few transhumanist cults.