While 'wet' biologists hold the line with drugs their 'dry' counterparts are painstakingly building the genetic models that offer our best hope against this group of deadly diseases. Paul Fisher reports
Find the key to cell division, whether uncontrolled as in cancer or regulated in a stable organism, and we find a key to life
Perhaps we'll have no choice but to believe in a free will we know not to exist; a dry paradox from a realm where dryness reigns
A damp formaldehyde smell permeates the headquarters of the Imperial Cancer Research Fund. Many of its rooms are lined with test tubes and phials oozing the conventional hospital odour and, equally conventional these days, every office resounds with a dry computer hum. Researchers and fund-raisers gaze with fixed focus at screens and, as with all computer work, it looks like nothing much is happening.
Ask what they're doing and what's whispered is a cure for cancer. Not tomorrow, but soon. And perhaps not a cure exactly, but definitely techniques allowing life with cancer for three score years and ten, plus a few extra to justify pension contributions.
It's the big promise. Every year in Britain, 160,000 people die from cancer - a quarter of all deaths - so self-interest as well as altruism has us giving £60 million a year to Imperial Cancer alone. That there is substantive research to justify the donations is beyond dispute and provides an article of faith to carry us into the next century.
Computer power underpins a new optimism in the power of science and researchers rely on it absolutely. They fall into two categories, the wet and the dry. Wet biologists concoct new drugs to halt cell division. Side-effects can be ghastly, for healthy cells are also destroyed during drastic drug regimes. That said, the work of the wets has had enormous successes, most notably in childhood leukaemia, where survival rates are up from near zero in 1965 to over 90 per cent today.
New drugs (some still derived from First World War poison gas) go through clinical trials to build databases where an agglomeration of patient reactions elicits most efficient prescriptions. Nationally networked computers reaching into research institutes and drug companies bring a level of automation to what medical research has always done: suck it and see.
Number-crunching the specific pain of thousands of people is a cruel process which could eventually be short-circuited by dry researchers who are into the greater precision of genetic research. This promises kindlier, technological treatments where genetically engineered viruses could be injected into host cells, or a patient's cells could be extracted, corrected and reinserted. But first, researchers must more closely identify genetic components.
The human genome consists of coiled threads of DNA and linked protein molecules packed in compact units, the chromosomes. DNA comprises two strands connected by four types of rung-like bases and total genome size is the number of these base pairs: roughly three billion.
At Imperial Cancer they're also number-crunching the generalities of life itself, sampling and sequencing up to 60,000 base pairs on a busy day. Mathematical definitions are dumped, via the Internet, to a computer in Cambridge near the labs where Francis Crick and John Watson unravelled the basic structure of DNA in the Fifties.
The emerging genome map is accessible to all who can use it and, in a display of cooperation to rival our organic coherence, the European Bioinfomatics Institute swaps data regularly with similar sites in America and Japan. (Contrast this with wet biologists who work with patent-obsessed drug companies and thus, for example, talk of "a new colo-rectal drug" and then clam up.)
Dermot Harris, head of EBI systems, reckons the first draft of the genome map will be finished by 2003 and that, when protein information has been encoded, it will occupy some 40 gigabytes of disk space.
John Sgouros is a computational biologist and one of 20,000 scientists contributing to the global database. A star Imperial Cancer researcher, he made his name in a European project for determining the base sequence of chromosome III from baker's yeast. He says: "I'm looking for similarities in the DNA and protein sequence databases that can provide hints about the function of newly discovered genes. The size of the databases and the complexity of the queries would make this work impossible without state-of-the-art computing."
When asked to assess likely results of his labours, Sgouros picks his words carefully, aware of the dangers of raising false hopes. "I'm optimistic we'll come up with a fairly complete view of how cell division is triggered and controlled. We won't get rid of cancer but maybe we'll live with it." The big promise. Find the key to cell division, whether uncontrolled as in cancer or regulated in a stable organism, and we find a key to life.
Putting aside, for a moment, the possibilities of delaying deaths from cancer, computational biology is refining a grand, mechanistic theory defining our soft, wet bodies as 40 gigabyte bundles of dry digits. Life defined as a species of information technology leads straight to the idea of a hard-wired physical and moral fate.
That has odd consequences. It is quite reasonable, at least by actuarial logic, to deny life or health insurance to those with cancerous genes. Should this happen, it will provide extra financial imperative to develop gene therapy from what is it at present still academic research.
Morality, choice, will be overridden in other ways for reductive logic based on a knowledge of genetic predetermination may find it equally reasonable to excuse a murdering drunk on the grounds that he's made that way. Perhaps we'll have no choice but to believe in a free will we know not to exist; a dry paradox from a realm where dryness is in the ascendant.
Harris reckons computational biologists are already occupying the wets' territory, saying: "Computer modelling of deep physical processes will lead to firm predictions. Instead of developing drugs by trial and error, we'll be able to accelerate research a millionfold with drugs designed and tested by computer."
He also foresees genomics applied to computer design, taking life forms that have evolved over two and half billion years to design new computers in two and a half years. As life becomes mechanical, so the mechanical will become lifelike, wet will be dry, dry will be wet: weird ideas for an article prompted by the death by cancer of my best friend.
He died distended beyond endurance by experimental drug treatments. Shortly before dying he explained his move toward Christianity, and why this seemed to him a progression in that he had always been suspicious of mechanistic interpretations of life. "The triumphalism of science (any minute and we'll show you the face of God) seemed quite as silly to me as the foam-flecked enthusiasm of fundamentalist religion," he wrote. He insisted on "open-ended multiplicity" and quoted Norman Mailer, one of his literary heroes: "A universe based on metaphor rather than measure."
With this phrase in mind I asked John Sgouros if genetic research was in any sense metaphorical. "It's not a metaphor," he replied without pause. "No, it describes an exact process just like describing how a car's components function. What you see are facts and facts are facts. This has nothing to do with being religious or not."
Moments later he was contradicted by Mike Mitchell, an Imperial Cancer bioinfomatics support analyst and the sort of man who announces his Christianity with a large cross round his neck. "Science is a form of faith," he said. "Evolution can explain why ducks have bills but it can't explain ducks."
Metaphor or measure? We must continue a project that started with the Renaissance and roll our faiths into one ball to accommodate a new explicitness which is establishing us as 40-gigabyte mathematical constructs. At the end of a century which has taught us that technological progress brings destruction as well as ease, cancer research is giving grounds for a less qualified optimism than we've grown used to. Lest anyone draw immediate hope from the research, it must be pointed out again that there are no magic cures yet. A 21st century medical arsenal will need be needed if cancer is to be beaten, for cancer is a single name for many diseases with mostly ill-defined causes, some that arrived with us in our bones, some picked up on the way.
My friend was a determined man who drew on gigabytes of courage. I can't, however, believe anything predetermined him to the strength of mind that had him writing to the bitter end. "Chemotherapy is too mad and crude, like using a machine gun to switch off the kitchen gas, but it can work. For systemic cancers such as my lymphoma it's also pretty much all they have. With all the progress in genetically engineered antigens there will probably be a pill in 20 years time which will get rid of lymphoma as septrin gets rid of spots. It's a nuisance that all I can do now is watch my hair fall out for the third time, but when the magic bullet arrives I would like to be around to take it."
For more information about the work of Imperial Cancer, call 0171-242 0200 or visit its Web site at www.icnet.uk/ where there are links to many other cancer-related organisations.
