Grass genes tapped to breed better crops

A diverse array of Andropogoneae tribe grasses grown in a greenhouse at the Donald Danforth Plant Sciences Center in St. Louis, Missouri.

Researchers from Cornell and the United States Department of Agriculture will tap into genetic information found in more than 700 species of related grasses, in hopes of making maize and sorghum more productive and resilient to extreme weather brought about by climate change.

The investigation of the Andropogonae tribe of grasses – which include maize, sorghum and sugarcane – is made possible thanks to a four-year, $5 million National Science Foundation grant.

By studying the genetics of grass species so closely related to these major crops, the researchers will be able to mine genes that encompass roughly 1.5 billion years of evolutionary history.

“Each generation of plants out in the field experiences various types of weather and environments, and the ones that succeed, pass on those genes to the next generation,” said project principal investigator Ed Buckler, a research geneticist at the U.S. Department of Agriculture-Agricultural Research Service (USDA–ARS) and adjunct professor of plant breeding and genetics with Cornell’s Institute of Biotechnology.

“As we try to breed crops that are better adapted to climate change, we’re now able to tap into this massive amount of evolutionary time and genetic history that we haven’t been able to do by just looking at one species,” Buckler said.

Commercial and public-sector plant breeders will be able to use this information to breed more productive and weather-resilient maize and sorghum.

The researchers will use advanced genomic techniques to sequence the genomes of the Andropogonae grasses, generating huge amounts of data. Once the genomes of more than 700 species are sequenced, each species will be compared with one another and to maize and sorghum. The researchers plan to identify functionally important base pairs (basic units of DNA double-helix) in the genomes that may be mutated in maize and sorghum and could be preventing these crops from being as well-adapted or high-yielding as possible, Buckler said. Processing all those data will be possible through machine-learning technologies.

“Machine learning speeds up and deals with the incredible complexity of working with more than 700 different species,” Buckler said.

Part of the grant will be used to train the next generation of scientists in computational biology. The group will hold bioinformatics training workshops and “hackathons,” where computer scientists and biologists will work together to develop prototype solutions to problems presented at the outset. “It trains computer scientists to learn biology and biologists to learn how to code,” Buckler said.

There will also be a traveling museum exhibit that educates people on how plant diversity is used to design better crops.

Collaborators include researchers from University of California, Davis; Cold Spring Harbor Laboratory; Donald Danforth Plant Sciences Center and Iowa State University.