As the success of the Human Genome Project has made clear, DNA stores information at very high density – 15 atoms per bit of stored information. But, while biology has evolved some very sophisticated and compact ways of reading that information, we’re stuck with some clunky and expensive methods of sequencing DNA. Of course, driven by the Human Genome Project, the techniques have improved hugely, but it still costs about ten million dollars to sequence a mammal-sized genome (according to this recent press release from the National Institutes of Health). This needs to get much cheaper, not only to unlock the potential of personalised genomic medicine, but also if we are going to use DNA or analogous molecules as stores of information for more general purposes. One thousand dollars a genome is a sum that is often mentioned as a target.
Clearly, it would be great if we could simply manipulate a single DNA molecule and directly read out its sequence. One of the most promising approaches to doing this envisages threading the molecule through a nanoscale hole and measuring some property which changes according to which base is blocking the pore. A recent experiment shows that it is possible, in principle, to do this. The experiment is reported by Ashkenasy, Sanchez-Quesada, and M. Reza Ghadiri, from Scripps, and Bayley from Oxford, in a recent edition of Angewandte Chemie (Angew Chemie Int Ed 44 p1401 (2005)) – the full paper can be downloaded as a PDF here. In this case the pore is formed by a natural pore forming protein in a lipid membrane, and what is measured is the ion current across the membrane.
This approach isn’t new; it originated with David Deamer at Santa Cruz and Dan Branton at Harvard (Branton’s website in particular is an excellent resource). A number of groups around the world are trying to do something similar; there are various variations possible, such as using an artificially engineered nanopore instead of a membrane protein, and using a different probe than the ion current. It feels to me like this ought to work, and this latest demonstration is an important step along the path.
Hi Richard,
I am reading your book soft machines.
Now, I am trying to understand application of reading DNA. What I mean is that while it would be cool to cheaply sequence DNA, outside of knowing that you will die on dd-mt-yr, what is the point?
If it is a stepping stone, where can I get information regarding a future history?
Eleggua
The Human Genome Project website I link to above is a good place to start looking for applications of sequencing DNA. I’m not such a genetic determinist to think that your date of death is encoded in your DNA, though you will presumably find a lot of not necessarily welcome news about your propensity for various diseases. Aside from the direct, and very important, applications to biology, this interests me simply as a way of reading information at very high data density. You can imagine future histories in which this aspect is important as well as I can!
this is future of us!!!!!!!