I say we. But when politicians and scientists, the representatives of our knowledge, appear together in public we must suspend belief. Politicians are untrustworthy as a class, and we know it. Scientists, who ought to know it too, can be all too gullible. Entranced by the media attention leaders attract, they surrender. Dr. Faustus, the original scientist, sold only his own soul to the devil. Today, his successors will sell ours as well, if they have the chance.
On 26 June last year, two scientists, Francis Collins and Craig Venter, flanked Bill Clinton to announce that all was sweetness and light in the world of genomics. The relationship between Collins and Venter had never been wonderful. Open war broke out between them in May 1998 when they became head-to-head rivals to determine the sequence of the human genome in media hype the blueprint of (human) life. Now we were supposed to believe that peace and co-operation between them had been brokered by William Jefferson Clinton.
The issue at stake could hardly be more momentous: who owns our knowledge about ourselves? Know thyself has long been a fundamental instruction for moral life. Now the issues are whether a scientific company can add provided you pay, and what follows if self-knowledge becomes part of the cash nexus.
Forget arguments about nature and nurture; at least for a few minutes. The form of all living organisms is determined by the sequence of their DNA. This sequence is a code of four letters, A, G, C and T. Apparently simple but often lethal bacteria may have a genome consisting of a few million of these letters. The foot and mouth virus responsible for the present UK epidemic has a sequence of only 7,733 letters. A small fly or simple weed has a sequence of some 100 million letters; that of mouse, elephant or humans a sequence of about three billion letters.
We have known this for years: the first sequence of a virus was determined in the late 1970s; of the first bacteria in 1995; of a fly and a weed last year. Since the early 1990s it has been clear that research technology could sequence the very large genomes of mouse and human. Nobody underestimated the magnitude of the task. Government agencies in the United States, and the Wellcome Trust, an independent UK biomedical charity, were persuaded to make the necessary investment. Slowly at first, the sequencing of the human genome began.
Factories had to be built, very different in scale from the biological research laboratories with which we were familiar. By 1998, less than 10 per cent of the human genome had been sequenced. With growing investment and mounting confidence the ramp-up in sequencing rate started. In the process two important principles about the endeavour were established.
The Bermuda Agreement
First, it would be truly international. Despite the domination of the USA and UK, groups from Germany, France, Japan, China and other countries were making significant contributions. Second, it would be truly public. At a historic meeting in February 1996 in Hamilton, Bermuda, the partners agreed not to patent any of the sequences that they determined, but instead immediately to deposit them in a public database, where they were freely available to all.
Bermuda was, and should have remained, a momentous agreement, on a par with the great treaties and conventions like those on human rights which punctuate modern history. It had three essential elements. First, that something as fundamental to our nature as the sequencing of our own genome should not be the property of any individual, group or government, however enlightened, but should belong freely to all humankind. Such knowledge, Bermuda ruled, should as a matter of principle be freely available to all: to simply hold in one''s hand, to ponder, to study and analyse or, within certain ethical limits, to exploit.
The second essential feature was more pragmatic. Everyone at Bermuda was aware that determining the sequence of genomes, whether human or not, is but the first step to understanding the complete meaning of the message they contain. This process could take decades of work. Progress in it will be much faster if the problem of interpreting sequences attracts the brightest interested minds, and most skilled and experienced geneticists, available everywhere. This possibility would be seriously impeded if access to a sequence is controlled in any way, either by legal or commercial constraints.
The standard argument for patents of inventions is that they encourage investment and therefore accelerate the development of the advances they embody. Here, the opposite holds true. The swift advancement of knowledge involves unimpeded scientific rivalry and collaboration.
The third core element at Bermuda was more technical but equally important. There are two opposing ways in which a genome may be analysed (I exaggerate for emphasis). These may be contrasted as the gene-centric view and the holistic view. There is intense interest among clinicians and scientists in understanding the genetic basis of much of human disease. For them a gene-centric view of the genome is sufficient: they need to be able to study in immense detail the properties of individual genes. But scientists also need to study the genome overall, to compare, for example, the entire genome of humans with that of the mouse, or to compare specific long sequences.
Building the database of life
Since the early 1980s scientists in Europe, the USA and Japan have collaborated to create a single database of all public DNA sequences. Since the mid-1980s it has become practice that no scientist could publish work concerning a DNA sequence unless that sequence had been deposited in this database, as the scientific journals made such deposition a precondition of publication. The Bermuda Agreement was a confirmation of this practice. Because the database is free, anyone can download it across the internet onto their own computer any time of the day or night without asking permission of anyone. Thanks to this it has become the central resource for our attempts to make sense of any DNA sequence.
In the last 20 years some 50,000 different scientists have contributed sequences to the databank. If I determine a new sequence in my laboratory it is the work of a moment to compare it with the entire databank, now some 12 billion bases. Consider the alternative: each of these 50,000 contributions would become sequences placed on different web sites of varying quality and accessibility. I would need to trawl through all of them to analyse my sequence. In practice, this would either be impossible or it would enormously delay research into an overall understanding of genomes. Without this public resource the achievement of sequencing the human genome, let alone its analysis, would have been far more difficult than it was.
Bermuda was, therefore, a hugely positive agreement in three ways: in establishing a fundamental principle, in improving and accelerating the growth of human knowledge, and in facilitating practical understanding of what whole sequences of the genome actually do.
Celera and the sequencing split
But in May 1998 Craig Venter announced he was forming a new company to sequence the human genome in three years and make money by selling access to the sequence to business, in total breach of Bermuda. This was not pure bravado. It coincided with a technological breakthrough in sequencing technology, achieved by Venters backers, PE Biosystems Inc. They wagered that their new machines could be coupled with novel computational methods to make the task much easier, cheaper and faster.
Reactions to Venters announcement varied from outrage to disbelief. Some said that it could not be done technically; others said that even if it could it was immoral, unethical and offended all of the laws of human nature other than those of capitalism. Those responsible for the public sequencing project had three choices: to lay down and let Venter do it; to collaborate with Venter; or to compete. They competed. The competition did, of course, have a bright side - despite their public statements to the contrary, the public sequencing group accelerated their progress many-fold in response to Celera. It was an ugly, not friendly, competition. Both sides took every opportunity to question the integrity of their opponent in public.
Until the famous meeting in the White House, when Clinton used what was still his credibility to mediate a truce. Hands were shaken; the cameras whirled and clicked, and wise heads hid themselves in their hands. Joint publication of the scientific papers was promised. It soon became an open secret that this would take place in the scientific magazine Science, a weekly owned by the American Association for the Advancement of Science, part of whose mission statement reads: Foster scientific freedom and responsibility.
Fight for scientific recognition
The rules for publication in Science demand submission of the data in the appropriate database. It was clear to everyone in the field that the phrase signified the international sequence database. Once deposited there, however, Celeras sequence would be open to all without restraint, even to Celeras commercial competitors.
Celera wanted to make a commercial return on its very considerable investment; were the data to be published on the public database, their potential revenues would be reduced. On the other hand it, or at least Craig Venter, desperately wanted the approval of the scientific community and his personal achievement to be recognised, perhaps even with a Nobel Prize. This meant the work had to be published in a respectable scientific magazine.
The major scientific weeklies Science in the US and Nature in the UK are not published for fun. They too are commercial ventures that compete for the hottest science. Science was keen to publish the two human sequence papers, the public one and Celeras that would be a considerable commercial coup. Science allowed itself to be persuaded to publish Celeras paper with neither public data deposition nor full disclosure of the methods by which the work was done. The fact that Celera is a major financial donor to the AAAS (Sciences charitable owners) had, of course, no influence on this decision.
Science magazine bows to Celera
It took some time for the full implications of this decision to sink in. In early December 2000, I wrote an open letter to all the directors of the AAAS and all the editorial board of Science urging them to reverse this decision. The reply of Don Kennedy, the editor of Science, was far from convincing and, in response, the publicly funded group pulled its manuscript and sent it to Nature, paradoxically a journal owned by a commercial conglomerate. Both were published in the week of 12 February, to considerable media attention.
Most, though not all, reputable scientists regard the Celera paper published in Science as little more that an expensive advertisement, a form of science by press release. One of the defining features of science is that it is public knowledge. Only if scientific knowledge and the methods used to acquire it are freely available, can the cardinal principle of the reproduction and testing of experiments take place. It is a tragedy that a great scientific magazine has seen fit to bow to the will of big business to compromise such a basic principle.
The consequences are only beginning to be felt. When Heather Kowalski of Celera Genomics announced at the end of April that the company had assembled the genetic code of the mouse, it was with a mixture of PR-speak (the Rosetta stone) and keep-off-the-grass. The denial of access to non-subscribers means that the scientific credibility of the findings cannot be tested where it matters in the public realm.
At a time when the sober assessment and understanding of scientific findings has never been more important, the scientific community needs to restore, not lessen, its legitimate standing with the public. Above all this means ensuring its independence from corporate interests and politicians. We have to work with them, clearly. The less we work for them, the better.