The genome project is little more than reams of meaningless data

'Modern biology is like playing with the best and biggest Meccano set in the world'

Next week we can expect the announcement that the human genome project has largely been completed; and, paradoxical though this may sound, the moment will become a milestone in the slow process of understanding how little DNA matters to human beings. Why should that be?

Next week we can expect the announcement that the human genome project has largely been completed; and, paradoxical though this may sound, the moment will become a milestone in the slow process of understanding how little DNA matters to human beings. Why should that be?

We live at a time when "genes" are offered as the explanation for almost everything human, with varying degrees of irony - sometimes none at all. People who talk about the "shopping gene" do not really believe that there is a chunk of their chromosomes that determines that they would rather shop than go for country walks - at least I hope they don't. But some people who talk about "gay genes" certainly mean that: it's worth remembering that the reaction of the former Chief Rabbi Lord Jakobovits to the news that there might be a gay gene was to welcome it because it would allow parents selectively to abort children afflicted by that abomination. Talking about genes has become a way to talk about individuality; there seems to be a link between knowing your DNA and knowing yourself.

Genes of that sort are a long way from the stuff that the human genome project works with. At the Sanger Centre, near Cambridge, where much of the work is done, there is an LCD display above the receptionists' desk, across which a line of green figures is constantly rushing. As you watch, they resolve themselves into the four letters "ACGT", chasing each other across the screen in apparently inexhaustible combinations. That is the actual human genome unrolling as it is being sequenced: each letter stands for one of the four nucleotides that are strung together to make a molecule of DNA. String them together 3.25 billion times in the right order and you will have half a human genome and enough information to make the other half.

It is an awe-inspiring feat to have identified three billion of these base pairs, a feat that is expected to be announced next week. But it's just as obvious that the human genome is not a human being; nor do we understand the ways in which a human being - or any other animal - develops from a genome. That has been particularly clear since 1998, when the complete gene sequence of Caenorhabditis elegans was worked out. C elegans is one of the most boring and inconspicuous animals it is possible to imagine: a tiny, transparent hermaphrodite worm that lives in dungheaps and eats bacteria. It has only 959 cells, and almost everything it does is genetically programmed. Because it is so simple, it makes a wonderful place to study the way in which whole live animals assemble themselves from a fertilised egg.

Without the worm project, the human genome project might never have got under way. Yet the worm sequence was only a preliminary to the real dream, which is to understand how the particular proteins that a genome specifies interact to build the different cells that interact to build a living worm. That knowledge is nearly as far off as it was 30 years ago, when the worm was first intensively studied. "It is going to be extraordinarily - staggeringly - complex," says John Sulston, the director of the Sanger Centre. "The complexity of development is far beyond the complexity of the genome."

For the scientists who have done the work, such complexity is its own reward. Modern biology is like playing with the best and biggest Meccano set in the world. The analogy is remarkably accurate: one way of looking at the genome is as a set of microscopic bumps and wiggles whose exact physical conformation shapes the proteins out of which we and all other living things are built.

Or you can understand DNA as software. Sulston himself speaks about "hacking the genome". But it is not enough just to have the parts - or, for software, a listing of the source code. What is needed is a deep understanding of how it fits together and how it works; that will take decades at least.

The trouble is that it has been hugely expensive to get even this far, when much of the genome is known, even if what it does is not - knowing about the genome does not necessarily tell you much about genes. Genes consist of stretches of the genome that can be used by the machinery of the surrounding cell as templates for proteins. These stretches are seldom continuous, which is one complication; another is that some of the proteins that DNA specifies block the actions of other bits of DNA, so that in any given cell only some of the possible genes are read, or "expressed". Most important, there is as yet no generally agreed way to decide which bits of the genome are actually genes that might be expressed. The figure of 80,000 human genes is often bandied about, but the two most recent estimates for the number of genes on the human genome are 27,000 and 130,000.

Clearly, more - very expensive - research is needed. Most of the money has been raised in the expectation that it can be recouped by finding profitable uses for the knowledge of genome sequences. One trouble with that is that diseases with clear-cut genetic causes are rare. The majority of cases of such prevalent killers as heart disease are not caused by genes in any interesting sense. Perhaps if exercise were patented, people would believe more in its effects on health.

Of course, it is seldom put like that when public money is being raised. When the project had to be sold to the American government, Jim Watson explained that he "didn't want to miss out on learning how life works". The US Department of Energy, one of the main sponsors of the research, sells the project under the heading: "To know ourselves". But few, if any, of the things that we are interested in learning about our own, individual selves will be revealed by the human genome project. That would hardly need saying were it not for the way in which DNA has come to mean something like "soul" or "essence", so that even car advertisements talk about the DNA of their designs, where once they would have talked about "spirit" or "magic".

The sequence of the human genome is, if you like, a listing of all the bricks you would need to build all the houses in London. Such a list would certainly add something to our understanding of the city. But it wouldn't help you to get from Paddington to Waterloo nor even tell you how to recognise Buckingham Palace. Of course, London is made of bricks, by mechanical processes, and would not exist without them. But that is not why the city is interesting; and the fact that there wouldn't be any people without the human genome does not mean it is the thing that makes us human.

author@darwinwars.com The writer is the author of 'The Darwin Wars' (Touchstone, £7.99)