Lagergren earned his bachelor's degree in applied mathematics from ETSU and his master's and doctorate in the same field from North Carolina State University. His interest in machine learning developed during his graduate studies and continued during an internship helping expand AI capabilities at a private company.
When it came time to start his postdoctoral position, Lagergren had already noticed that ORNL was a leader in the fields of computational science and supercomputing, and was intrigued by the lab's biological sciences research. “When you talk about science and Tennessee, ORNL is the epicenter,” said Lagergren, who began his postdoctoral position in 2021 and took up his current position two years later.
Lagergren credits the laboratory's resources and expertise for driving his scientific success. “The data that APPL continues to generate, and the computing that Frontier does to ingest and process big data and learn patterns, is absolutely inspiring,” he says. “Everyone I've had the opportunity to work with here is amazing. There are so many interesting stories to tell, and I get to speak to people with incredible expertise in so many different fields. It's different from working in academia or industry. ORNL is a place with incredible resources and the ability to work with purpose to solve important problems that impact the country and the world.”
Lagergren is excited about future efforts with his colleagues to leverage APPL’s capabilities to create new paradigms in plant science.
Slicing Massive APPL Data for a Better Plant
“I'm excited about the time series aspect, and that's the potential that APPL has. You can go out and do a field survey or work in a greenhouse and get a snapshot at one point in time, but APPL is collecting data every day,” Lagergren says. “You can look at plants at different stages of growth and in different conditions over time.”
“We can extract high-dimensional image features into closed-form equations that can then be used to do time series predictions or estimate parameters. Only these parameters can be targeted for genomic analysis. We are essentially driving new mathematical models that don't exist yet: how plants change, how roots change, how they interact, the role of the environment, different nutritional conditions, etc. We will gain a deeper understanding of the dynamics that drive response and resilience.”
One ultimate goal, Lagergren said, is “to adapt bioenergy crops so that they can grow in places they don't want to grow, because they can't take over corn fields or other food crop land. That's not an option. So we want plants that can grow in harsh environments and places where there are stresses like disease and pests. The challenge is to identify the genetic changes needed to make the plants more resilient, to grow faster using fewer resources and have a better biomass composition. We want to have a bioenergy future that can also offset the carbon we're releasing into the atmosphere.”
Advice for young scientists? “Focus on understanding rather than just memorization,” says Lagergren. “As an undergraduate, it's easy to memorize content and repeat it on tests, but the real world is not that simple. You need to understand the basics, so that when the parameters change and you're faced with a new challenge, you can use that fundamental understanding to solve the problem at hand.”
This approach is particularly effective when applying machine learning to solve problems, he says: “It's best to take a problem that you're passionate about, learn as much as you can, and dig deep to get to the nuances. You'll learn a lot more from that than a broad, shallow understanding.”
Lagergren said his job is fundamentally motivated by two things: solving complex puzzles with many variables and that don't follow standard rules, and working on solutions to real-world problems. “What's great about working at ORNL is that we're trying to solve world-class challenges in areas like bioenergy, plant resilience and zoonotic disease spills. I'm not just solving vague math problems, I'm solving problems that have real consequences if done right.”
UT-Battelle manages ORNL for DOE's Office of Science, the largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.
