THE laborious process of deciphering the human genome could be accelerated by decades thanks to software that vastly improves researchers鈥� ability to collect and analyse data. A prototype system is now being developed at universities in the UK.
The amount of genome data available is expanding at breathtaking speed, but with molecular biologists generating information at about twice the rate at which computer processing power is rising, it is increasingly difficult to keep up. 鈥淭here鈥檚 a massive amount of information and it鈥檚 increasing daily,鈥� says Carol Goble, project leader at the University of Manchester, UK.
Worse, much of this data is stored in conflicting formats all over the world. 鈥淎t the moment everything is completely non-integrated and non-standardised,鈥� says Bruno Sobral, scientific director of the Virginia Bioinformatics Institute in Blacksburg. So researchers who develop a program to analyse information in one database cannot use it in another. This format war is a serious threat to future progress.
Advertisement
MyGrid, the new system, will allow biologists to analyse information in hundreds of databases much as one type of word processor can read the files of many others. Until now, to do this across different databases required many types of custom-built software. With MyGrid, biologists won鈥檛 need to become programmers, says Michael Luck, who is involved in designing the system at the University of Southampton, UK. The team is using software agents to help translate and standardise the contents of conflicting formats.
Claire Jennings and Simon Pearce at the University of Newcastle, UK, will be the first to try the prototype system. They are looking for genes associated with Graves鈥� disease, an autoimmune illness affecting the thyroid. They plan to feed the results of 70 microarray chips into MyGrid to analyse any candidate genes they find for the disease. 鈥淣ormally this would take us an age,鈥� Pearce says. MyGrid could analyse this in a matter of hours, possibly even minutes.
Other software tools they have tried are of only limited use, Pearce says. They don鈥檛 鈥渢ake you any further than giving you a list of genes鈥�, he says. MyGrid should automatically find any information relevant to the study, searching for genomic and proteomic data, regulatory networks and any other relevant facts.
Biologists waste a lot of time trying to extract data and reshape it in a usable format. 鈥淚n a generation or two biologists won鈥檛 be worried about it in the same way that we aren鈥檛 concerned about how the internet runs, we just use it,鈥� Sobral says.
The project is part of a 拢2.4-million e-Science project funded by the British government and spanning several universities. 鈥淢yGrid is the first wave,鈥� says Goble. 鈥淚t鈥檚 one of the pioneers in data intensive computing.鈥�
