Nearly half of Americans have tried to lose weight in the past year, according to a CDC study. Many adults in the United States also live with diet-related chronic diseases such as cardiovascular disease, high blood pressure, or type 2 diabetes. Despite the widespread focus on dieting and nutrition, much about the chemical makeup of food remains unknown.
Researchers at the USDA-NIFA AI Institute for Next Generation Food Systems (AIFS) at UC Davis are working to close these gaps using artificial intelligence. AIFS is one of several AI institutes funded by the National Science Foundation since 2020 and involves collaboration among UC Berkeley, UC Davis, and UC Agriculture and Natural Resources.
The institute’s project FoodAtlas aims to compile scientifically validated information about food from decades of published research. This knowledge graph connects data on foods, their chemical compounds, and potential health effects. “Knowing what chemicals are in our food and what they do to our bodies is essential for public health,” said Ilias Tagkopoulos, director of AIFS. “FoodAtlas has been created with this mission in mind: an AI system that reviews the published literature and databases to map foods, their molecular composition, their health effects, and other parameters that are important for decision support in creating healthier, more sustainable food.”
Although consumers recognize basic nutritional elements like calories or vitamins listed on packaging labels, scientists estimate that a typical food contains between 20,000 and 50,000 different compounds—far more than those tracked by agencies like the U.S. Department of Agriculture. Researchers refer to this vast area of unknowns as “nutritional dark matter.” Efforts such as the Periodic Table of Food Initiative (PTFI) in the U.S., led by UC Davis, and FooDB in Canada are helping expand scientific understanding of these compounds.
One example where science has advanced understanding is the Mediterranean diet. Studies have documented its benefits for heart health; researchers found that a compound called DMB (3,3-dimethyl-1-butanol), present in olive oil and red wine among other foods common to this diet, may help lower risk factors for heart disease.
Tagkopoulos explained that FoodAtlas synthesizes published findings from reputable journals into an accessible database. Nearly half of the information gathered had not previously been included in any database. The integration of these studies can accelerate ongoing research into how various food compounds affect human health.
To make this growing body of knowledge useful for everyday decisions about eating habits, UC Davis is developing an app called “Swap it Smart.” The app will use insights from FoodAtlas and PTFI to help users select meals based on quality, nutrition content, environmental impact, affordability, flavor profile—and personal health goals. Ingredient substitutions recommended by the app will draw upon current research linking specific foods with both individual wellness outcomes and sustainability considerations.
“Swap it Smart is building on the strengths of FoodAtlas and PTFI to create a platform for meal development,” Tagkopoulos said. “It is designed to help nutritionists in schools and chefs in restaurants make delicious meals and dishes that are good for our health and the environment.”
As research continues into thousands of unstudied or little-understood compounds present in common foods—some with no effect on human biology—tools like FoodAtlas may eventually enable personalized dietary recommendations tailored to individual biomarkers or medical needs.
