The experimental results than Leonard Adleman obtained in 1994 while using DNA to compute the direct Hamiltonian path problem with just 7 nodes do not seem encouraging. It took him several days of lab work to complete the experiment. Furthermore, although DNA computation offers a potential in terms of data storage, there is also the issue of molecular volatility.
Can biological computation provide a better alternative to classical computation?
Thank you, Joachim, for your reply.
I wonder whether there is a way to avoid the redundancy of DNA material by making sure that molecules do not "get lost", as you said. For instance, could the deamination process that converts cytosine to uracil be prevented by some quantum mechanical mechanisms?
Renata
Yes, I was thinking about a more uniform way to prevent unwanted changes in the DNA molecular material. A series of measurements to keep the molecules arrested, just like in the Zeno effect, would be such a mechanism. It seems to be a better alternative to a combination of error correction codes, which you mentioned, the use of restriction enzymes, and repeated computations etc.
I was also thinking about the destructive interference. There is an established technique for correcting noise in sound waves called “active noise control”. But, as you pointed out, it probably won’t work for classical states.
Regarding data retrieval, I came across a paper introducing analog biological computing titled “Analog Computation by DNA Strand Displacement Circuits” (link attached). The claim is that reading can be done by means of measuring the concentration in the solution rather than drawing samples. I guess this could resolve the issue of running out of material during computation.
Renata
http://pubs.acs.org/doi/abs/10.1021/acssynbio.6b00144
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Nucleic Acids
- DNA