Grant: CIHR Fellowship

High fasting blood glucose (FG) doubles the risk of death from heart disease. A healthy diet can lower FG, but people are affected differently even after eating the same foods. This variation among people shows that "one size fits all" dietary guidelines may not be the best; there is a need for personalized nutrition. Factors like the gut microbiome, genetics, age, and lifestyle may influence different FG responses to diet, but how this happens is unclear. The relationship between diet and FG levels is inconsistent, because of limited data, small and similar samples, single time-point studies that ignore the time and age effect, and methods that do not capture complex interactions between different variables. My proposed research aims to address these issues by using data measured in over 10,000 adults in the "Nutrition for Precision Health" (United States) cohort and 13,000 children in the "Applied Research Group for Kids!" (Canada) cohort. Both datasets include multiple measures over time of FG and other blood markers, diet, gut microbiome, metabolomics, genetics, and over 2,000 lifestyle and environmental factors. Crucially, I will use advanced artificial intelligence (AI) methods to dig deeper into the complex interactions between these factors and FG levels, while examining how these relationships vary across different populations and age groups. Using advanced AI techniques, this project will integrate complex data and provide a deeper understanding of the reasons why we see variation in FG responses to diet. I will identify patterns and predictors of FG responses with greater precision for the first time in both children and adults. Using cutting-edge data analysis and big datasets, this project will inform personalized nutrition guidelines to lower FG levels based on each individual's needs and optimize disease prevention and treatment.