Companies across the globe are using sophisticated computing techniques in an attempt to develop personalised medicine that can both pinpoint which patients are more susceptible to disease and tell you what drugs will work best to fight it.

By using bioinformatics, which applies analytical theory combined with mathematics and computer science to the study of the structure of biological data and structures, these companies hope to come up with answers to questions like "Who gets heart disease?" and "Who will this medicine help?"

According to Kari Stefansson, founder and CEO of deCODE Genetics in Iceland, finding the answers to these questions is essentially "an information flow problem", which is where computing techniques like knowledge management and data mining come in.

Millennium Pharmaceuticals is one of the pioneers in the area of personalised medicine, and uses such technology in its quest to link the right drugs to the right patients. It is currently working with Roche to develop a diagnomic test for rheumatoid arthritis.

Diagnomics are molecular diagnostics that can describe a patient's current medical condition and provide prognostic and therapeutic information to give doctors an idea of how the disease will progress and what drugs to prescribe.

Millennium is also investing in pharmacogenomics, which is the study of how a patient's genes determine his or her response to specific drugs.

The company is also working with partners on identifying genomic markers for oncology (cancer). "This could be predictive either for drug therapy response or disease risk," explains John Maraganore, senior vice president of strategic product development at Millennium.

So far, despite pouring millions of dollars into genomics, the company has only come up with one prototype genomics-derived drug, which targets obesity. But it is working hard to assimilate the data it has acquired with a view to expand its reach to other diseases.

DeCODE is working on complex conditions like asthma, diabetes, heart disease and obesity, all of which are caused by a mix of genetic and environmental factors, but could also prove massive potential markets as they affect such a huge proportion of the population.

In its quest to track down the genes that cause these conditions, deCODE has a secret weapon – data on 1,000 years of Icelandic genealogy. If an ancient mutation that increases the susceptibility for a particular disease arose in a member of this small, isolated population, the genome linkage studies should reveal the location of the errant gene.

"Genetic mutations are really just scrambled data, so we put the emphasis on bioinformatics and statistics," explains Stefansson. The result of deCODE's work is the discovery of a common asthma susceptibility gene, as well as the location of two genes linked to respitory allergies. It is also building a growing list of genes linked to hypertension and schizophrenia.

The next step is to create tailored diagnostics and drugs to treat these disorders, should these genes prove equally prevalent in other populations.

Genome Therapeutics is another company using these high-end computing techniques to identify a gene that causes asthma. It hopes that its finding will eventually allow patients to be identified even before they start to suffer any symptoms.

"Finding this gene opens the door to elucidating a new pathway," says Tim Keith, senior director of human genetics at Genome Therapeutics. "This could lead to therapeutics that actually target the disease rather than the symptoms, as well as tests to pinpoint asthma sufferers before they even have symptoms."

While these breakthroughs are encouraging, the aim is to create a process by which more and more genes can be identified. "One or two hits are not that hard to generate," says Millennium's Maraganore. "But building a pipeline that continues to deliver several new candidates year after year is something that proved elusive to everyone. And that is what we are going to have by 2005, thanks to genomics".