What actually happens in the body after a meal, and why do people respond differently to the same foods? In his doctoral research, Nikolai B. Aunbakk applied modern statistical methods that capture changes in metabolism over time. The findings show that analysing entire response trajectories can reveal important differences between individuals and provide new insights into the effects of dietary interventions.
Below, he answers three questions about his research:
Why is this research important?
The body's metabolism is constantly changing, including after we eat and when the body is exposed to physical exertion. Nevertheless, such processes are often studied using only a few blood samples collected at specific time points. The aim of this PhD project was to investigate how statistical methods can be used to describe these changes over time in a more detailed way, even when some measurements are missing.
What are your most important findings?
With my colleagues, I investigated, among other things, how blood glucose, C-peptide (a marker of insulin secretion), and blood lipids change during the hours following a meal in individuals with obesity. The results showed substantial variation between participants, not only in the levels reached, but also in when the responses peaked and how they developed over time. Men and women also exhibited different response patterns.
The data came from the CARBFUNC dietary intervention study, which also examines the long-term effects of three different diets for obesity over a one-year period. The diets differed in both the amount and quality of carbohydrates but had the same energy and protein content. The researchers found that blood glucose and C-peptide responses decreased following all three dietary interventions, whereas reductions in blood lipids varied more between the diets.
To uncover these patterns, the researchers used functional data analysis, a statistical approach in which a series of measurements is analysed as a continuous curve rather than as separate individual time points. They also investigated how sporadically missing measurements can be imputed in a way that takes into account how each participant's measurements change over time.
This is important because sporadically missing data are common in studies that follow people over extended periods. If such missing values are not handled appropriately, they can lead to a misleading representation of the biological processes being studied.
What impact could your research have on the field and in a broader context?
The research demonstrates that analysing complete trajectories as curves over time can provide information that is lost when only individual measurements or simple summary measures are considered. These methods may contribute to more detailed analyses of metabolism and other biological processes that change over time.
In a broader context, this approach can help researchers make better use of data from human studies and increase our understanding of why people respond differently to dietary interventions and other physiological challenges.
FAct box:
Nikolai B. Aunbakk
- Previous education: Master's degree in Food Science and Nutrition, specialisation in Food, Nutrition and Health, from NMBU
- From: Lillestrøm, Norway
- Has completed a PhD at: the Faculty of Chemistry, Biotechnology and Food Science, NMBU
- Main supervisor: Kathrine Frey Frøslie (KBM, NMBU)
- Co-supervisors: Simon E. Nitter Dankel (University of Bergen), Siv Kjølsrud Bøhn (NMBU)
- Thesis title (Norwegian and English)
- Metabolisme i bevegelse: Funksjonell dataanalyse av longitudinelle og postprandiale data
- Metabolism in Motion: Functional Data Analysis of Sparse Longitudinal and Postprandial Data