URBANA — Computer scientists and plant biologists combined their talents to create an elaborate computer model showing how soybean plants respond to higher levels of carbon dioxide in the atmosphere.
Now, the University of Illinois-led project has been awarded $5 million to expand the effort to more plants and more scenarios, to help scientists understand how crops will respond to climate change.
The grant from the Foundation for Food and Agriculture Research will go to the UI's "Crops in silico" project, a global scientific collaboration that pulls together multiple computer models on plant growth and development to study the whole plant virtually.
With global population and the demand for food rising, and the climate continuing to change, researchers need to understand how crops respond and adapt to environmental changes to address food insecurity, officials said in announcing the grant Friday.
"Four crops — corn, soybean, sorghum and wheat — account directly or indirectly for about 60 percent of human calories. Yet they are susceptible to declining yields due to the impending stresses of climate change, including water shortages, elevated carbon-dioxide levels and soil degradation," said Amy Marshall-Colon, UI assistant professor of plant biology, who is the principal investigator for the new four-year grant.
Developing crops using traditional growing methods is costly and labor-intensive. Computer modeling allows researchers to quickly determine and test characteristics that help crops thrive in specific environments.
They can also conduct more experiments than can be realistically achieved in a field and test billions of possible changes to develop more productive and sustainable crops in different environments, researchers said.
Puzzle pieces fit
The idea of using integrative computer modeling to build an entire virtual crop is "something that's really ambitious" and a fairly new concept in biology generally, said Marshall-Colon, a West Virginia native in her fifth year on the UI faculty.
Astrophysicists have long used computer modeling to study the universe, on a much different scale, laying the groundwork for biologists who pioneered the concept by modeling bacterial systems, she said.
Plants are much more complex, she said. While many researchers have developed models for different components of a plant's biology — what's happening inside a cell at the molecular level "all the way up to what we see, like a plant leaf" — the challenge was pulling those puzzle pieces together to model an entire plant, she said.
So UI plant scientists partnered with computer scientists at the National Center for Supercomputing Applications to overcome the biggest hurdle — "getting these different models to talk to each other," she said.
The Institute for Sustainability, Energy and Environment at Illinois provided $350,000 in seed funding to establish the original Crops in silico project in 2015 in collaboration with NCSA, which provided $212,000 in seed funding.
NCSA scientists built the framework for the different models to communicate, essentially "the glue between the puzzle pieces," she said.
They also developed many of the tools used to visualize crops and simulate environmental conditions.
Marshall-Colon and Matthew Turk, assistant professor of astronomy and a research scientist at NCSA, also received a $274,000 grant from FFAR in 2017 to extend their work.
During the "proof of concept" phase, researchers built a model for soybeans that could essentially mimic biology, Marshall-Colon said.
"We already had data from the field about how soybean plants respond to high levels of carbon dioxide in the atmosphere. What our model was able to do was simulate what was seen in the field," she said, proving their approach worked.
A global effort
Now that the researchers have developed the platform, they can expand the computer modeling to other plants and other scenarios, to see how they will respond to rising temperatures or changes in gene expression from climate change. They plan to create models for corn, sorghum and wheat, and continue the work on soybeans, "covering four of the staple crops that we rely on for food or bioenergy," she said.
The new grant will allow researchers to quickly and accurately test how a plant responds to a combination of changes, skipping many generations of in-field experiments to figure out how to produce more food in a changing environment, officials said. It will also make the entire platform available to the public.
"By scaling up our work to whole plants and fields, we can move years ahead in optimizing plants for different growing conditions," Marshall-Colon said.
The Crops in silico project includes researchers from around the world involved in modeling, she said.
Co-investigators on this grant are Stephen Long, UI professor of plant biology and crop sciences; Kaiyu Guan, UI assistant professor of resources and environmental sciences; NCSA research scientist Meagan Lang; and scientists from the University of Nebraska, Purdue, Penn State and Oxford University in England.
"This project is a perfect example of using technological advances to identify how crops will respond to environmental stressors and how to help the crops thrive despite environmental changes — all while saving time, money and making this platform publicly available," Sally Rockey, executive director of the Foundation for Food and Agriculture Research, said in a release.