First principal models, neural networks and functional graphical models for Defining metabolic capacity as a Tool for Personalized nutrition (FOODTOP)
The project aims to understand individualised human response to food intake which could potentially lead to individually tailored dietary advice.
People digest and metabolize nutrients differently and such differences have been linked to disease. Better understanding of individualized dietary responses and their links to health outcomes thus holds great potential for predicting, preventing and treating certain diseases. To do so requires integration of many different types of large, complex data sets to get the full picture and understand cause and effect relationships, and new methods will be required to fully achieve this.
This project will develop new computational methods for analysing and integrating different types of data available from an existing cohort of young adults, including continuous glucose measurements, images of meals, metabolomics data, gut microbiome samples, etc.
The methods will be shared with the scientific community and could later be used to guide and execute clinical studies on personalized nutrition for symptom relief in diseases such as asthma, allergy, obesity, and metabolic syndrome.
University of Copenhagen has published a news story on the project (external link): https://food.ku.dk/english/news/food-news-2021/food-scientist-wants-to-create-data-model-for-personalised-dietary-recommendations/.