New Department "Computational Precision Nutrition" at DIfE

On July 1, 2026, the new department "Computational Precision Nutrition" commenced its research activities at the German Institute of Human Nutrition Potsdam-Rehbruecke (DIfE), led by Prof. Stefan Konigorski. The team’s core objective is to advance the development of reliable and personalized nutritional recommendations using methodologies such as N-of-1 trials, coupled with advanced analysis techniques from artificial intelligence (AI) and causal inference.

 

Nutritional trends such as intermittent fasting or low-carb diets are ubiquitous. However, what works for one person sometimes fails to manifest in others. This is exactly where the research of the new department "Computational Precision Nutrition" (CPN) at DIfE is positioned. Under Prof. Stefan Konigorski’s leadership, the team is promoting a paradigm shift from "one-size-fits-all" thinking towards precision nutrition. Their goal is to analyze complex health datasets and develop strategies for personalized nutritional and behavioral recommendations for the prevention and treatment of chronic diseases, such as obesity or type 2 diabetes. 

Moving beyond generic advice – towards individual Precision Nutrition

For Stefan Konigorski, the solution is no longer just a diet, but a scientifically provable plan that accounts for every individual's unique metabolic profile. " We combine computational approaches with apps, large cohort study data, and personalized trials to establish causal relationships. The database is multimodal, meaning it integrates molecular markers, wearable device outputs, and detailed lifestyle records. This depth allows us to achieve real precision when it is needed," he explains.

N-of-1 Trials: Evidence built for the individual

Konigorski’s team develops innovative methods at the intersection of statistics and AI, integrating them with data from large population studies, such as the European Prospective Investigation into Cancer and Nutrition (EPIC) Potsdam Study or the NAKO Health Study. Crucially, they also plan individualized studies, for which the CPN team employs the rigorous principles of N-of-1 trial methodology. This statistical approach is essential for fully personalizing clinical research.

 

Central to studying how specific nutritional factors affect health is the expansion of the digital open-source study platform, StudyU. By combining data from this platform with causal inference methodology, it will be possible to isolate true cause-and-effect relationships between lifestyle habits, diet, and disease risks.

This approach helps answer highly personal questions such as: “Can intermittent fasting make me feel better and help me sustain lower blood sugar levels?” Konigorski elaborates: “In N-of-1 trials, participants would undergo phases with and without intermittent fasting, while continuously recording biomarkers of interest. Statistical analyses can then identify individual effects and provide personalized lifestyle recommendations. The participants thus learn directly which dietary patterns are beneficial for them.”

 

Combining AI and human biology

The unique combination of large databases, artificial intelligence, digital tools and N-of-1 trials enables more than just connecting micro-approaches with macro-data. The massive data diversity promises to yield precision medical insights into the development of cardiometabolic and age-related illnesses. “From previous research, we know which dietary patterns can be protective or therapeutically supportive. But what does that mean for a single individual? Who benefits from general recommendations? When is precision truly necessary?” Prof. Tilman Grune, Scientific Director at DIfE, emphasizes the importance of the new department: “Our new Computational Precision Nutrition department will provide answers to these questions and generate tools for personalized nutritional advice ranging from detailed data analysis to digital implementation. This will redefine the standard in nutritional science.”

 

Background information

N-of-1 Trials

This methodology has its roots in general medicine and clinical care, where measuring efficacy on an individual level is critical. In nutrition, this approach is highly valuable because human metabolism is extremely personal, making generic guideline-based recommendations often inappropriate.

Causal Inference

Causal inference is a scientific process designed to identify true cause-and-effect relationships in data. While correlation merely describes statistical association, causal inference investigates whether one variable directly causes another and quantifies the strength of that specific effect.

 


Scientific Contact

Prof. Dr. Stefan Konigorski
Head of the Department of Computational Precision Nutrition
phone: +49 33 200 88 - 2500
e-mail: stefan.konigorski@dife.de

 

Press contact

Press and Public Relations at DIfE
phone: +49 33200 88 - 2335
e-mail: presse@dife.de

 

 

 

 

 

Back