Breeding a better bean.
I'm Bob Karson with the discovery files--new advances in science and engineering from the National Science Foundation.
In the evolution of plants, many traits have come about as a result of competition with other plants for limited resources like water and nutrients. But agricultural crops don't face those same competitive challenges. Better for them to be bred "just get along" and share resources. Now, researchers at the University of Illinois at Urbana-Champaign have developed a computer model that can help breed more optimum crops: Soybeans for starters.
The team set out with multiple goals: To be able to breed plants that are more productive, use less water and have a structure that reflects more sunlight back into space.
Their technique let them try out large numbers of combinations of structural trait--things like leaf arrangement on the stalk, and leaf angle--to see which combinations would result in a plant that meets their three goals.
The computer model predicted a design for a soybean crop with 8.5 percent more productivity, using 13 percent less water, and reflecting 34 percent more sunlight.
Once the computer predicts an optimal structure, the crop could be bred from the many forms of soybeans already available. No need for genetic engineering! The team says the system could work with other crops too, and help boost the world's food supply.
A different kind of bean-counting.
"The discovery files" covers projects funded by the government's National Science Foundation. Federally sponsored research--brought to you, by you! Learn more at nsf.gov or on our podcast.