May 14, 2008, 12:01 am
Join the Hunt for Super-Rice

By Steve Lohr

There is no quick fix to the world food crisis, but a project getting underway Wednesday could make a difference in the long run.

A team of researchers at the University of Washington are putting a genomics project on the World Community Grid in the computational search for strains of rice that have traits like higher yields, disease resistance and a wider range of nutrients.

Ram Samudrala, the principal investigator at the University of Washington, describes the goal of the project as the pursuit of “super hybrids.”

The purpose is to hasten the pace of modern rice genetics, which since the 1960s has delivered a series of new strains, starting with higher-yielding semidwarf varieties, a breakthrough that was hailed as the Green Revolution.

But the demand — all those mouths to feed — keeps rising. Rice is the main staple food for more than half the world’s population. In Asia alone, more than two billion people get up to 70 percent of their dietary energy from rice.

Climate change, says Mr. Samudrala, is complicating things. “Places that didn’t used to have a lot of water now do and vice versa,” he said. “So we want to create a hardier rice.”

The World Community Grid, begun in 2004, gives selected humanitarian scientific projects access to massive computing resources. It taps the unused computing cycles of nearly one million computers around the world — much like SETI@home, the best-known distributed computing effort, which claims it has harnessed more than 3 million PCs in the search for extraterrestrial life.

The World Community Grid places a small piece of software on your PC that taps your unused computing cycles and combines them with others to create a virtual supercomputer. Its equivalent computing power would make it the world’s third-largest supercomputer, according to I.B.M., which has donated the hardware, software and technical expertise for the project.

Like so many sciences, molecular biology and genetics are being transformed by the use of computational tools. “This project is an excellent fit for the World Community Grid, both the science and the mission,” said Joseph M. Jasinski, director of healthcare and life sciences research at I.B.M.

The grid will run a three-dimensional modeling program created by the computational biologists at the University of Washington to study the structures of the proteins that make up the building blocks of rice. Understanding the structures provides clues to their functions, interactions between the molecular parts and how certain desired traits are expressed.

But the computing, which should last a year or two, takes the search only so far, noted Mr. Samudrala. It speeds along the study of 30,000 to 60,000 protein structures and the selection of rice strains to breed. But the super-hybrids must still be developed in greenhouses in places like the famed International Rice Research Institute in the Philippines.

“In the end, it’s still all breeding, but what we’re doing should make it more targeted and productive,” said Mr. Samudrala.