DNA computers? Silicon chip death?

[UFgatorSean]

Hello All!

This is my first post to the forums. I read today that scientists in Israel have created a computer that works on DNA. WOW! Does anybody have any more information on this? I am an Electrical Engineering stundent at the University of Florida and I also have an interest in human genetics. Do you think that one day our PC's and cell phones will run on DNA instead of silicon? Perhaps we could implant computers into peoples cells to cure diseases. The possibilities sure seem endless. I am really interested in things of this sort and hope that one day i might help engineer some of the pioneering devices that work on such principles.

 

[MustPost]

I don't think DNA computing is going to replace Si anytime soon. What about it would make it so good?
It seems like the moleculor computing people are doing, putting a few atoms of carbon on a board to act as a transister would be more effiecient.

 

[UFgatorSean]

I have also read about new ideas about using individual atoms as transistors. Apparently, it would be possible to manipulate the electron configuration about an atom in order to designate an on or an off (1 & 0). This would make for the ultimate in removable storage! You could carry a terrabyte device that's roughly the size of a coin! I just think all of these innovations in microelectronics are amazing.

 

[Shalmanese]

It's not really a "Computer" per se but rather a very highly specialised machine for doing certain things Very Very well. It can do some tasks hugely faster than a computer but it has no logic AFAIK.

 

[alpha88]

DNA computers can only do one thing really (in theory):

The path example, you have N number of points and different one way paths connecting various points (can be visualized by thinking of many US cities and many - but limited - different flights)

The goal is to find the path that travels through every point.

This problem isnt solveable for classical computers because it goes up exponentially with N (along with number of routes)

The way a DNA computer solves it is as such, you make a batch unique DNA strands for each path (or leg) and you then mix them together. Every possible solotion to the problem "forms" within a few seconds. The trick is to find the correct solotion, in many cases it is the longest strand. The set up for this takes days, the calculation takes seconds and the analyzation takes a week or too. It has only been successful where N=6, but it is theroretically possible for very large values of N.