WORMS AND MOUSETRAPS

Dick Pountain/Thu 17 October 2002/11:11 am/Idealog 99

Sir John Sulston's winning of the Nobel Prize for physiology and medicine offers me a rare opportunity to write an entirely positive column, free from the carping that's become almost a reflex of late. Sulston's combination of scientific brilliance, beard-wearing and motorcycle riding makes it hard to resist a little hero-worship. Better still, it enables me to use the word 'worm' for once in a hacker-free context.

I won't bore you with a full summary of the work for which Sulston, Horvitz and Brenner won the prize, as the national dailies have done that pretty thoroughly already. In very brief, he and his colleagues for the first time worked out the semantics, as it were, in addition to the grammar of a chunk of DNA. They sequenced the genes of the tiny nematode worm Caenorhabditis elegans and then observed how each gene was expressed to produce the developing worm's organs, nerves and muscles: in computing terms they not only decompiled its DNA into source code, but then single-stepped the program to see how it works. Actually that's a rather poor metaphor because a single-step symbolic debugger was precisely what the biologists lacked. You can't slow-down or stop the expression of genes and still get a healthy organism at the end of it, so they had to observe the whole process in real time, at nature's speed and with lots of different processes happening at the same time.

Once the full DNA sequence of Caenorhabditis had been established, the problem became one of mapping the full history of each cell in its body as they developed in the embryo worm, to discover which genes controlled them. Another team working in this area had tried enlisting hi-tech aids, in the shape of high-speed video and computers running advanced pattern-recognition software, but this had been unsuccessful. Sulston therefore undertook the task manually, using a cross-hair microscope to watch embryos develop inside an adult worm (which is handily transparent). It took him 18 months, working two four-hour shifts per day, watching the birth and death of every cell that appeared during embryonic growth. This is pretty strong evidence for the argument that genius consists not of some romantic ability to perceive 'flashes of inspiration', but rather of a hugely increased capacity for concentration compared to the general population - an observation that has also been made about, for example, Beethoven and Gauss. (That's what makes it so hard to give up smoking or coffee, the concentration-aids of mere mortals). His feat also demonstrates that computers are still some way from being adequate for the toughest pattern-matching problems at which the human brain excels, though they were used very successfully in the DNA sequencing part of the project to identify the overlaps among 17,000 fragmentary sequences. 

Another lesson that could usefully be drawn from this triumph is the importance of choosing the right model: their task would probably have been impossible with an organism any simpler or any more complex than Caenorhabditis. It has just 300 nerve cells, which makes its nervous system small enough to understand but, unlike a single-celled bacterium, still gives it enough different behaviours to be interesting. This choice of the correct model is no fluke, but rather another aspect of scientific brilliance.

However what prompted me to choose Sulston as the subject for this column was not so much the science as the politics, every bit as fascinating and neatly illustrating some of the things I've been saying in recent columns about Intellectual Property. Sulston is a passionate advocate of freedom of information in science, and came into conflict with his superiors and funders over his policy of releasing gene sequences as soon as they were obtained, so that other teams elsewhere in the world could use them. This was in strong contrast to a rival US team at Craig Venter's Celera Corporation, which insisted on patenting every gene discovered to prevent others from catching up. I see parallels between the public-spirited Sulston and Tim Berners-Lee, who gave the WWW protocols to the world rather than exploit them for his personal enrichment. But, I hear you mutter, weren't you last month defending being paid for Intellectual Property (for example this column). Indeed I was and there is no contradiction. Some endeavours, most obviously science and medicine, require information to be free in order for the processes to work, and they therefore have to be supported from the public purse (or charity - the Wellcome Foundation supported Sulston). Sulston explained this issue perfectly in an interview when he said that patenting a discovery that's unique (like a human gene) merely creates a monopoly, while the purpose of patents is actually to encourage competition. A patent guarantees a mousetrap inventor a period of profitability, encouraging him to develop the invention, but because there are *other ways* to design a mousetrap it doesn't prevent others from trying to better him. With a gene there is no other way.

Much of the software that runs the Internet was developed within scientific institutions, hence the strong free-information ethos, the open source and free software movements. But someone, somewhere, paid for that software to be developed. Totally-planned economies where the public purse paid for everything gave us the Trabant and the cardboard shoe, and have been decisively rejected by their citizens almost everywhere. It's not at all unreasonable to expect the public purse to pay for your medicine but not for your magazines and CDs - they used to call it the 'mixed economy'.