A team of Eberly College researchers is the recipient of a $749,990 U.S. Department of Energy Genesis Mission grant that will be used to develop an artificial intelligence platform that could transform scientists’ ability to predict and engineer microbial communities.
Leading the team is Ember Morrissey, associate biology professor in the West Virginia University Eberly College of Arts and Sciences, along with Edward Brzostek, biology professor, and Aldo Romero, Eberly Distinguished Professor of Physics and Astronomy.
Scientists know how to read the genetic code of microbes, but they still can’t reliably predict what those microbes will actually do, according to Morrissey.
That’s because microbes don’t live or work alone, she explained. They compete and cooperate with other microbes in their environment, and those interactions can change how they behave in ways that are difficult to predict.
“Many microorganisms have valuable metabolic capabilities, like the ability to pull nitrogen from the air to support plant growth,” Morrissey said. “In order to leverage these functions, we must be able to understand and predict how microbes interact with one another in the wild.”
With the funding, Morrissey and other team members will work to develop the Integrated Microbiome Analytic Generator of Interaction Network Evolution-AI (IMAGINE-AI) project with the goal of building an AI system to reliably predict microbe behavior.
If the team’s efforts are successful, it could solve a longstanding problem that’s beleaguered scientists and create a framework that could lead to advances in agriculture, bioenergy, and biotechnology.
Morrissey was in Washington Wednesday (July 22) for the announcement of The Genesis Mission grants.
Along with WVU, the project includes researchers with Johns Hopkins University, Pacific Northwest National Laboratory, and Lawrence Livermore National Laboratory.
The Genesis Mission is a collaborative federal effort to unite DOE national labs, industry, academia, and more to harness AI for scientific breakthroughs.