Cassava, banana and plantain, staple foods for millions of the world's poorest people, are notoriously difficult to breed. But an international team of scientists aims to change that, using a revolutionary new approach to plant breeding developed at the University of California, Davis.

The project is supported by a grant of $1.2 million from the NSF-BREAD (Basic Research to Enable Agricultural Development) program, a joint initiative of the Bill & Melinda Gates Foundation and the National Science Foundation.

"These are very important food security crops, but they take a long time to reproduce and it's difficult to create new varieties," said Simon Chan, assistant professor of plant biology at UC Davis.

Recently Chan and the other team members -- Hernan Ceballos, of the International Center for Tropical Agriculture in Cali,  Colombia; Jim Lorenzen, from the International Institute of Tropical Agriculture in Tanzania; and Leena Tripathi of the International Institute for Tropical Agriculture in Nairobi, Kenya -- were invited, with other recipients of NSF-BREAD grants, to present their work to Bill Gates at the foundation's headquarters in Seattle.

"He was very interested in the science and had good questions for everyone," Chan said.

Most successful crop varieties are hybrids created by crossing two inbred varieties. While this is relatively easy to do in well-established annual crops like maize or wheat, it is much harder with slower-growing crops like cassava, banana and plantain. As a result, cassava, banana and plantain growers are currently forced to create new varieties by crossing two hybrid parents -- a highly unpredictable process.

New crop varieties allow farmers to cope with pests, disease, drought and other problems.

Working with the small laboratory plant Arabidopsis thaliana, Chan's lab recently discovered a method to create plant seeds that carry the DNA from only one of their parents, allowing breeders to immediately create a hybrid that "breeds true," dramatically cutting the time required to create new crops with traits such as disease- or drought-resistance.

"What's exciting is that what started with Arabidopsis is already jumping from a little plant to big crops," Ceballos said.

With the NSF-BREAD grant, the research team will develop Chan's technology for use with these major crop plants.