Systems biology will revolutionise medicine

Remember the Human Genome Project? Conceived over 20 years ago, it was supposed to lead to breakthroughs in the treatment of a host of diseases by understanding every gene in our cells. It hasn't turned out that way. The genetic blueprint of even one human has proved bafflingly complex, and the links between genes and disease much hazier than the project's cheerleaders predicted.

Their response has been to embark on another heroic quest for genetic enlightenment, in the form of the 1,000 Genomes Project, launched last week. The hope is that by scouring the genetic blueprints of a thousand people, scientists will get the insights that have eluded them thus far.

Yet after two decades of bluster and hype, it's clear that these gene-centred spectaculars will never deliver. As sceptics have said for years, a far more sophisticated approach is needed - one that takes into account not just genes, but the biochemistry they control, the cells they inhabit and the interactions between those cells.

For years, this sounded like a counsel of despair, as there was little hope of capturing the full complexity of even a single cell. Now that's changing, thanks to smart scientists packing awesome amounts of computer firepower.

They describe themselves as systems biologists, a low-key name for a decidedly high-brow calling. Their goal is to create simulations reflecting as much of the complexity of living systems as available computing-power allows. And that means venturing far beyond the level of genes, using the world's most powerful computers.

It's a tough job, but scientists are already doing it. As the media gawped at the bright, shiny future promised by the 1,000 Genomes Project, scientists at the University of Montreal quietly unveiled the world's most detailed simulation of a living, beating heart.

The team captured key features of heart cells using billions of mathematical equations, and studied their behaviour. Despite performing more than a thousand billion operations a second, the computer took two hours to simulate just 1/200th second of heart activity. Even so, the results will allow the team to test ideas about how the heart works - and why, sometimes, it doesn't.

Such simulations have already helped debunk claims about inherited heart disorders based on simplistic genetic grounds. Now they are being used to identify new treatments. At Oxford University, a team led by Prof Denis Noble (left) has created a computer model of the heart so sophisticated it shows the action of simulated versions of real drugs, revealing new ways of preventing heart attacks. It is also being used to design drugs with minimal side-effects - the holy grail of the pharmaceutical industry.

But this is just the start. Systems biologists are building computer models of other organs, with the ultimate aim of linking them all together into a single 'virtual human'. That will take a lot of computing power: Noble is negotiating for time on a Japanese super-computer capable of a million billion operations a second.

In the end, the proof of systems biology will be in the lives saved by its insights. On current form, it looks a far better bet than the busted flush of the human genome. ·