Slashdot Mirror
Switchable Net Woven from DNA
virtualXTC writes "A team of US researchers have managed to make a woven DNA mesh that expands and contracts. By coating the net in silver they can make nano-wire lattice in which individual wires can be lengthened or shortened using a short snippet of ssDNA to create a nano-switch."
I guess now they can say, "Every dark nucleotide has a silver lining."
In relation to another recent weird-science happening:
from Penny Arcade
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
Eric Drexler is rolling in his bunk-bed, muttering something about how carbon should have been there first.
At our research institute there is also a lot of research going on in the use of DNA in nanotechnology. I always get the feeling this is mainly because adding the word 'DNA' in the proposal increases the chances that your proposal gets approved and you get the money you want enormously. Of course there are a lot of buzzwords in this /. post about how this invention can lead us to a better future for all, but I wonder whether this could be done simpler with cheap standard polymers? On the other hand, DNA almost is a cheap standard chemical already.
-- Cheers!
We must remember that DNA-switches are radically different from the semiconductor switches used in current computers. The biggest difference is that DNA switches are not addressable by location -- you can't easily build a trace that carries a signal to or from a particular chunk of DNA. Thus you cannot layout a complex switching circuit in the same way you build an electronic CPU.
On the other hand, DNA is code-addressable and innately parallel in solutions. Create a liquid with molecules of one code and it will automatically find its match in the solution or on a substrate. This is an intensely parallel process with trillions of molecules bumping and matching simultaneously. The degree of parallelism is only limited by the permitted reaction time, total amount of reactants, and the relative concentrations of the matching components. (Imagine a computer that increases in power just by pouring it into a bigger beaker.) The result is that DNA-based systems can be massively parallel machines, but the likely clock rates will be a few Hertz at best.
The point is that DNA-based computers have totally different design principles from their semiconductor brethern. Until we get good at compiling algorithms into a sequence of code-match chemical reactions and create the substrate and chemical systems to reliably carry-out very long chains of DNA-based computations, we won't have a practical DNA-based computer.
Two wrongs don't make a right, but three lefts do.
Although "DNA" is a buzzword designed to excite grant reveiwers, in this example it is required.
The DNA switch described in the article uses base-pair code matching to do its magic. Thus the invention requires a polymer that contains a controllable sequence of monomers. At this time, I think DNA and RNA are the only polymers that have a well developed technology for controlling the exact sequence at the molecular level and for mass producing polymers with an exact sequence.
For this work, at least, DNA is more than a buzzword, it is the only practical way to do what they want to do.
Two wrongs don't make a right, but three lefts do.
Well, at least one of them got modded funny. ;)
You're right, I wouldn't steal a car. But if it were possible, I sure as hell would download one!
You're forgetting peptides/proteins. They too can be readily made with exact sequence at the molecular level, and can expand/contract controllably. Proteins might even be better in some situations due to their relative ease of addition of specifically placed functional groups.