The scientists in the field of micro cellular biology and mathematics managed to build a "computer" using bacteria. To be more precise E. coli bacteria.
There was a lot of fuss on the web about this subject but I found no real academic comment about this break through (as they call it).
As I see it, there is no real break through at all, it's genetic algorithms implemented in real genetic material. Clearly you must be a genius to think that genetic algorithms could work if implemented in real genetic material instead of a silicon based computer :)
The computer works like this: you encode the problem in the genetic material of bacteria, you encode the solution detector and then leave the bacteria to multiply. When the solution is found the solution detector will highlight the colony that found the solution. Then sequence the bacteria DNA and get the solution.
All these could take a while, so I don't think that this "computer" could provide you a valid solution faster than a real computer for a NP complete problem.
It's still unclear to me how they managed to realize the solution detector but undoubtedly this is a good start point for organic computers.
Imagine how a computer virus and anti-virus looks like for a future bacterial computer :)
You can read the abstract for more information about "
Solving a Hamiltonian Path Problem with a bacterial computer".