What are the long-term goals of this research and what is its potential?

1) What is the eventual potential?
Our long term goal is to construct nanocomputers that solve biological problems by operating in biochemical environment in a useful way (for medical purposes, for example). Hopefully one day they will operate in vivo, detecting anomalies and synthesizing drugs to fix them. In the medium term (hopefully within a decade), one might be able to use such computers to analyze DNA in vitro. Instead of sequencing DNA, and sending the info to an electronic computer for processing, one might be able to "throw the computer to the testtube", and analyze the DNA then and there.

2) How much potential do DNA computers have in terms of being faster and smaller than silicon computers? Can you quantify this in some way?
While they are much much smaller ("a trillion computers in a drop of water"), each computer on its own is slow. A trillion computers do a billion operations per second in 120 micro liters solution. One computer performs one operation per 1000 seconds on average. We believe we can get this down to a small number of seconds. However, competing head to head with silicon computers is not the goal. The goal is operation in a biochemical environment.

3) What would it take to develop this computer into something commercially viable?
There are two dimensions along which this research progresses -- improving the computer, and moving from in vitro to in vivo. They can be explored somewhat independently, but both could take quite a long time before a commercially viable product comes out.

4) Can you give me an estimate of when the research could be technically ready to be applied practically?
For laboratory applications less than a decade, for medical applications decades.

5) Could DNA computing replace silicon based processing in the longer term as the density of electronic operation on silicon reaches its physical limit?
Not in our line of research, where the goal is not to construct faster computers that compete with electronic computers, but to construct smaller computers that can operate within a biochemical environment and interact with it in a program-controlled way via molecular input and output

6) What would be the equivalent on an electronic computer? Would it be costly?
A "standard" electronic computer is a general purpose computer, unlike the special purpose finite automaton we constructed. Normally a finite automaton is realized by a simple software program on such a computer. Building the electronic equivalent of our finite automaton, with input, output and software capabilities would be quite costly, I suppose. However such a computer would not be able to input biomolecules and output biomolecules, which is the main purpose of our biomolecular computing approach.


Back