INDUSTRY

Techniques developed under the UK e-Science Programme led to the identification of the anti-MRSA drugs that received widespread publicity in the UK national media last month. E-Therapeutics, a spin-out company from Newcastle University, announced the discovery of three drugs that are effective against antibiotic-resistant superbugs, such as that scourge of hospitals, methicillin-resistant Staphylococcus aureus (MRSA). The drugs enter clinical trials this month and should be available for use within 2-3 years. They're the first antibiotics employing a truly novel mode of action to be discovered for decades. Pharmaceutical companies and others have put a lot of effort into the search for drugs to beat the superbugs. e-Therapeutics has succeeded where others have failed by using grid computing and e-science techniques to trawl through the portfolio of existing, licensed drugs for any that showed action against the superbugs. "With these techniques we can search through all the data, warts and all, very rapidly and identify candidates within a fraction of the time it would take using conventional drug discovery methods," says Professor Malcolm Young, chairman of e-Therapeutics and pro-vice chancellor at Newcastle University. Research funded by EPSRC and the DTI under the e-Science Programme demonstrated the reliability of the new method by showing that it could predict accurately the action and side effects of all 103 previously known antibiotics and many other drugs. Network security may seem to have little to do with drug discovery, but it was research to find out where networks are most vulnerable that led to these latest discoveries. Inside a cell, proteins interact with other proteins forming a network which makes the cell function. In most networks - and cells are no exception - some nodes are more important to the network's integrity than others. The e-science techniques pioneered at Newcastle enable the discovery of these key proteins. The next step is to find out which chemicals target the proteins that are vital to the invading superbug while leaving those of the person alone. All of this is a very big task, as it involves working with large amounts of data on proteins and drugs held in different databases. Gaining access to and integrating this data requires e-science techniques to get over obstacles, such as the data being held in many different formats and with different names. Many calculations need to be performed very rapidly to test out all the possibilities, exhausting local computing resources. So grid computing is required to access sufficient computing power remotely. Even creating a new database to contain the results of searches is a big task, requiring e-science techniques. In theory, some of this could have been done without e-science and grid computing, but it would have been so big a task and taken so long that no-one has ever bothered. These powerful new analytical and computing techniques, however, make it possible to do this rapidly. e-Therapeutics have identified some candidate drugs for clinical trials in less than two weeks. As well as the three antibiotics, the company has identified more than 60 "lead" compounds in many different medical areas. The techniques are also being taken up in the search for new drugs from the Amazon rain forest. Last year, e-Therapeutics formed relationships with pharmaceutical companies in north-east Brazil to test substances extracted from rain forest plants for their efficacy against a range of diseases. The techniques reduce the time taken to analyse a substance from, typically, two years to two months. "Drug discovery is a search problem. Conventional drug discovery is running out of steam because the search methods employed to date are so inefficient. What we're trying to do is industrialise drug discovery by implementing all these processes in computers, powered by e-science," says Young.