"DNA computers have opened up an entirely new vista in computing. Instead of silicon chips and electrical currents, DNA computers use deoxyribonucleic acids--A (adenine), C (cytosine), G (guanine) and T (thymine)--as the memory units, and recombinant DNA techniques carry out the fundamental operations. In a DNA computer, computation takes place in test tubes. The input and output are both strands of DNA, whose genetic sequences are considered to encode certain information. A program on a DNA computer is executed as a series of biochemical operations, which have the effect of synthesizing, extracting, modifying and cloning the DNA strands.
"Broadly speaking, the only fundamental difference between conventional computers and DNA computers is the capacity of memory units: electronic computers have two positions (on or off), whereas DNA has four (C, G, A or T). So, in theory, DNA computers can compute whatever kinds of problems that conventional computers can. But DNA computers also have some special merits.
"First, DNA memory is compact. One cubic centimeter of DNA soup could store as much as 10^21 bits of information, whereas, with the current technology, conventional computers have a memory of at most 10^14 bits. Second, whereas most electronic computers operate linearly--they manipulate one block of data after another--biochemical reactions are highly in parallel: a single step of biochemical operations can be set up so that it affects trillions of DNA strands in the test tube.
"These properties suggest that DNA computers might be useful for solving some problems that have been traditionally regarded as intractable. DNA computers are not without drawbacks, however; the biochemical operations involved are subject to errors and are often slow. Rigorous tests of the accuracy of the operations and further technological development are needed in order to clarify the real potential of DNA computers."
Carter Bancroft, a professor in the department of physiology and biophysics at the Mount Sinai School of Medicine, offers some additional insights:
"Workers in the field of DNA-based computation (of which I am one) hope that, at least for some applications, DNA-based computers will be far more powerful and efficient than silicon-based electronic computers. This hope is based mainly on the following properties of DNA:
"First, it is the major information storage molecule in living cells, and billions of years of evolution have tested and refined both this wonderful informational molecule and highly specific enzymes that can either duplicate the information in DNA molecules or transmit this information to other DNA molecules.
"Second, each molecule of DNA is roughly equivalent to a little computer chip. But DNA molecules are much smaller than computer chips, so you can get lots of them into a little space--roughly 10 trillion of them would fit in a space the size of a marble.
"And third, all these molecules can work together at once, so you could theoretically have 10 trillion calculations going on at the same time in this little space (or the number of simultaneous operations could be far greater still if your computer were a lot larger than a marble).
"To answer the question about the kind of computations we can do now with DNA: The field is still very new, and there are at present really only two types of computation we know how to do. One comes from the work of Leonard Adleman of the University of Southern California and involves solving combinatorial problems such as the famous 'Traveling Salesman Problem' (roughly defined, it is the task of finding the most efficient way to visit a number of cities and get back to your starting point). The other comes from our recent work in which we showed how DNA can be employed to carry out a very fundamental computer operation: addition of two numbers expressed in binary. We believe these two types of computation represent just the tip of the iceberg of the power of DNA as a computational molecule