The hard drive for humankind

John Sulston is about to reveal the secrets of self to the world. Martin Ince meets the modest man leading the Human Genome Project.

John Sulston, director of the Sanger Centre in Cambridge, has been used to the limelight since the media caught on to the immense potential of the Human Genome Project. But this week his press cuttings are setting new records as he and colleagues announced that their task is approaching completion. They have described 90 per cent of the human genome, a level that allows the data to be of value for all but a few applications. And they have done it well before the 2003 deadline originally set.

Sulston is a modest man with the air, and the beard, of a scientist from central casting. But he is sure of the importance of his work and that of the other participants in the United States, China, Japan, France and Germany who have divided up and sequenced the 60,000 genes that encode the secrets of human development.

As he sees it, completing the project is a "milestone in the longest sense," a key point in the human understanding of ourselves and of the universe.

More immediately, it is a milestone in molecular biology, the next step from Darwin and from the discovery of the structure of DNA.

The human genome consists of about 3 billion "base pairs" of the structural elements of DNA. There are so many of them that even if the biology had been in place, the genome project would have been impossible without the information technology revolution.

As it is, the 17 centres doing the sequencing have been linked in a database that produces and verifies data at an unprecedented rate and publishes it, free and without patent restrictions, each day on the worldwide web.

One early result is that the number of genes in the genome turns out to be smaller than expected, at about 60,000 rather than the 100,000 frequently mentioned. This has cost some genome researchers money in the global sweepstake run on the result.

And according to Ian Dunham, who has led the Sanger Centre's work on one of the human chromosomes, number 22, on which the genes are arranged, this smaller number increases confidence that few new genes remain to be discovered and also means that we share more with other species.

One way to think of it is that our instruction manual is about three times as thick as that for a worm, or four times as thick as that for a fly. The difference as one looks at more complex animals is that they have more "command and control". Only about 3 per cent of human genes are there to initiate the production of proteins directly. Large expanses of the genome are duplicated (about 30 per cent), large areas appear to have no function, but many genes have a function that relates to turning others on and off. This ensures the smooth development of most humans and when disrupted, can be disastrous.

Stephan Beck, head of human sequencing at the Sanger Centre, points to the example of a group of genes on chromosome six whose function has to do with "histocompatibility", - the ability of the body to recognise and accept or to reject other tissue it encounters.

These genes, says Beck, directly affect the body's machinery for rejecting or accepting transplants, and genome project knowledge is likely to be in use in hospitals "within a few weeks" in this area.

His point is reinforced by Mike Stratton, who recently joined Sanger (part of the Wellcome Trust) from the Institute of Cancer Research. Making the case that the project will soon affect medical practice, he points out that "a third of people get cancer and it kills one person in five". Cancer, he says, is "the consummate disease of DNA". Some cancers may implicate as few as five genes, he thinks, and others as many as 50. In the case of breast cancer, a comparatively small number of genes seems to be associated with the disease, and it could be that about 1,000 genes, 2 per cent of the total, are involved in cancers of all types. Stratton says: "This means that we need the fine anatomy of the genome to see the genes that code for particular cancers. This will lead us to a better knowledge of the environmental causes of cancers and to insights into cancer biology." He adds that cancer genes may also be targeted for gene therapy.

David Bentley, head of human genetics at the Sanger Centre, says that the present draft of the human genome has been drawn up from material from fewer than a dozen people. In time there will be deep analyses of genetic variations between groups and populations, but Bentley points out that 99.9 per cent of genes are shared by all humans. The differences are known as single nucleotide polymorphisms, and two people could differ by perhaps 3 million SNPs.

A consortium of academic institutions, 12 pharmaceuticals firms and others are identifying an initial 300,000 SNPs, of which 149,000 were identified in a single year. They will all be freely published and in the context of the whole genome will form the basis for future studies of human variation. Among early uses may be the detection of genes involved in drug breakdown in the body.

This week's announcement is being played as the end of the first stage of human genomics, but the very fact that only 90 per cent completeness is being claimed shows that the project (whose main funders have been the National Institutes of Health in the US and Wellcome) cannot shut up shop yet.

The sequences are determined by cloning comparatively short lengths of DNA - about 2,000 base pairs long - and analysing them. Called shotgun sequencing, this method was invented in Cambridge by Sir Fred Sanger, the double Nobel prizewinner, after whom the centre is named.

Some parts of the genome are more resistant than others to being cloned in this way. The last parts may not be determined until after 2003 and new strategies may be needed to unravel them. Michael Morgan, director of Wellcome's genome campus in Cambridge, points out that the first 90 per cent of the genome has been sequenced five times over to give an acceptable level of confidence in its accuracy, but the last and trickiest parts may need a tenfold treatment.

Morgan says that the campus will concentrate on functional genomics - finding out what the genes mean - but will also carry on sequencing. It is involved in mouse genome work, now well advanced, and the rat and zebrafish are also on the agenda. So are pathogens, such as those responsible for plague, tuberculosis, leishmaniasis and malaria, the battle against which has been a constant theme of the Wellcome Trust throughout its life.

Wellcome has been the main backer of the programme outside the US and is sure that it is a major step forward in its health research mission.

But the trust is not setting itself up as a moral authority on the use of genetic data. Sulston takes the old-fashioned view that discoveries and their application must be separated as firmly as possible.

"Experts have a responsibility to uncover the truth," he says. But "the advancement of knowledge is different from the way we use it. The responsibility for that lies with all of us as part of the democratic process. We need informed discussion involving all citizens."