Genesis Machines by Martyn Amos
A nice introduction to several topics. I know next to nothing about genetic engineering – which is odd seeing as it was a topic I was extremely interested in before I turned to Physics, and I’d like to know more… so it’s nice to read about the way that bioscientists actually manipulate strands of DNA and the tools that they use in wet-world. A nice introduction to the Polymerase Chain Reaction (PCR), and the early work of Watson and Crick, the book is written from a historical perspective that gives you a feeling of really ‘following the field’ as it emerged, by explaining the key players and early proceedings of conferences, as well as both disappointments and breakthroughs in the key lab experiments. It also had a nice section on SAT and graph theory from a DNA computing perspective. Graph theory deserves more appreciation; it’s just cool.
The book spent a fair bit of time labouring the point, especially describing how the experiments were performed using analogies from popular culture, but I find this happens with most pop-sci books. I guess it depends how good you are at data-mining literature as to whether or not you need to have important concepts repeated.
This book however did make me smile; I could tie in lots of links from the world of Quantum Computing (although the author only mentions this once, and very briefly). Essentially DNA computing is massively parallel from the point of view that large numbers of molecules can compute all possible solutions to a problem at once, as the strands ‘stick together randomly in different configurations’ in a test tube, and the strands encoding the correct (or optimal) result are kept. Which means you can have a good go at solving small Hamiltonian Path Problems fairly easily. However, it does not mean that it is anymore computationally powerful than a classical computer – just that you have more resources available (space resources in this case – equivalent to having a silicon chip with an enormous transistor density). For example the time taken to solve NP-complete problems still grows exponentially with the problem size.
I’m not convinced that it’s quite as cool (or scalable, or viable) as Quantum Computing. And is having a robot (or set of mechanically activated valves) that mixes together tubes of solutions together inside your PC any less far fetched than having a computer with an onboard dil-fridge cryocooler? I think not. Although it would look awesome if the DNA container cells were all backlit with LEDs 🙂
Whether or not this field will continue to grow, I do not know, but it is nice to read about alternate implementations of computation. I think DNA computing will have important ramifications in medicine and molecular biology. For example, one of the ideas reported in the book was to use DNA computers to control drug production factories within the body itself.
As ‘wetware’ technology and engineering improves, one area that I am very interested in following is bio-electronics research (particularly the interfacing of conventional electronic circuitry with the human nervous system). I can envisage a fruitful collboration between biotechnology and conventional silicon processing, which I hope I live to see (and perhaps participate in).
… and I wish we’d had LEDs to light the glassware in organic chemistry lessons.