Purpose & context This is a public dataset to be used as the community wants. The dataset was contributed for the purpose of running the University of Copenhagen course entitled, "Hackathon – application of machine learning in biomedical research". The course is identified by its activity number, 3927-23-00-00. The first part of the course included introductory training into machine learning for those with a minimal understanding of Python this course can be found here along with teaching materials on Github.

Abstract This data is coming from a multi-omics study (Tebani et al. 2020) where the purpose of the study was to monitor the stability in molecular profiles for healthy individuals. In the study, 100 healthy individuals in the age range of 50-65 years were followed over a year with four sampling time points. Blood molecular profiles including proteomics, transcriptomics, lipidomics, metabolomics, autoantibodies and immune cell profiling, complemented with gut microbiota composition and routine clinical chemistry. During the course, only the metabolic profiles are studied, and from another view than in the published study. The aim is to explore the metabolomics data with machine learning methods and identify the metabolites that associated with:

1. Metabolic differences between gender
2. Metabolic differences based on reported improvement on health during the study period
3. Body fat measured by bioimpedance