Recent research suggests that there are several subtypes of diabetes with different risks for complications like kidney disease, nerve damage, eye damage, dementia, and fatty liver disease. It is important to identify which patients are at risk for which complications as early as possible so that we can tailor prevention and treatment strategies.  

This case study will leverage advances in multiomic and computational technologies, including machine learning and AI, to analyze data from a cohort of individuals with diabetes. We will measure thousands of blood markers, genetic risk profiles, gut microbiome profiles, sleep and physical activity measures from wearables, and self-reported health measures from questionnaires.  

After defining subtype-specific molecular signatures, we will map those signatures onto individuals with diabetes who have been tracked for 15 years. This will allow us to identify subtype-specific, pre-diagnostic molecular markers or molecular signatures to inform the development of tailored preventative and risk-mitigating intervention strategies.