Harnessing the power of more than 795,000 computers around the world, a new research project that will analyze human proteins in the fight against cancer begins Tuesday using the World Community Grid, which was built and is maintained by IBM.
By using the combined computing power of the grid, the Help Conquer Cancer project will allow cancer researchers to drastically shorten the amount of time it would take to analyze 90 million images of crystalized proteins, from 162 years using existing computing systems to between one to two years using the harnessed power of the grid.
"Even with the largest computers we have, it would not be possible to finish this task," said Dr. Igor Jurisica, who leads the research team at the Ontario Cancer Institute in Canada, where the work is being done. Also participating in the work are scientists at Princess Margaret Hospital and the University Health Network.
The researchers will analyze the results of experiments on proteins using data collected by other scientists at the Hauptman-Woodward Medical Research Institute in Buffalo, N.Y.
The World Community Grid was created by IBM about three years ago as a way to harness unused global computing power to help solve a variety of health and scientific issues. The project calls on home and corporate PC users to register with the grid, then download and install a small software program that allows their unused computer cycles to work on critical scientific research.
Robin Willner, vice president of global community initiatives for IBM, said the total number of grid participants so far is about 795,000 around the world and grows daily. The combined computer power so far would create a supercomputer that would be the fifth most powerful in the world if it were in one place, she said. The grid uses participants' computers when the systems are idle.
The results of the research will go into the public domain and will be used by cancer researchers around the world, she said.
Three levels of security are part of the grid system and security audits are done constantly, Willner said.
By using the grid to better understand the structure of human proteins, researchers are trying to understand disease-related proteins and how they function, Jurisica said.
Once the 90 million images of some 9,400 different proteins are analyzed, data mining techniques will be used to go through the results, he said. Previous experiments have looked at smaller groups of samples because the means didn't exist to analyze them all, he said.
"This will be important for future research," Jurisica said. "Hopefully it will shed light on the principles or mechanisms of the proteins."
"We know that most cancers are caused by defective proteins in our bodies, but we need to better understand the specific function of those proteins and how they interact in the body," he said. "We also have to find proteins that will enable us to diagnose cancer earlier, before symptoms appear, to have the best chance of treating the disease -- or potentially stopping it completely."
Eight other projects have been run so far on the World Community Grid, including protein folding and FightAIDS@Home, which completed five years of HIV/AIDS research in six months. Additional projects are also being scheduled.