Nelson, Alisa2022-11-142022-11-142022-08https://hdl.handle.net/11299/243134University of Minnesota Ph.D. dissertation. 2022. Major: Biomedical Informatics and Computational Biology. Advisor: Peter Crawford. 1 computer file (PDF); 177 pages.Metabolomics technologies hold significant potential for aiding research in the underlying drivers of metabolic disease. However, the complexity of metabolomics data sets can limit their impact. This thesis combines computational biology with analytical chemistry and metabolic physiology to deploy metabolomics pipelines directed toward two significant research problems: (i) distinguishing the serum metabolome of normal weight and overweight or obese trained runners; (ii) developing bioinformatic pipelines using stable isotopes to improve feature selection and elucidate metabolite-metabolite relationships.Previously, metabolomics tools have been applied to measure the differences between metabolically healthy (MHO) and metabolically unhealthy obesity (MUO), or untrained versus trained groups. However, how well-trained MHO groups respond to acute exercise is yet unknown. Using liquid chromatography-high resolution mass spectrometry- based untargeted metabolomics, multidimensional mass spectrometry-based shotgun lipidomics, correlation analysis and machine learning models, the effect of increased adiposity and acute exercise on trained runners are evaluated. Fatty acid esters of hydroxy fatty acids (FAHFAs), an anti-diabetic and anti-inflammatory lipid class, distinguish normal weight (NWT) and overweight/obese trained (OWT) runners who differ in body composition but have similar levels of cardiovascular fitness. Many diverse FAHFA species are elevated at baseline and decrease with acute running in NWT, but not OWT. These exercise-induced changes are inversely associated with changes in IL-6, a known myokine involved in inflammatory pathways, and are influenced by visceral fat mass. Baseline concentrations of FAHFAs are negatively associated with visceral fat mass and together with free fatty acids and purine nucleosides account for 53% of the variation in VO2max in this cohort. Finally, the formation of isobaric FA dimers in untargeted metabolomics pipelines when fatty acids are abundant necessitate the use of targeted measurement of FAHFAs in serum for accurate quantitation. This work reveals a novel role of FAHFAs in acute exercise adaptation that is affected by increased adiposity. These results may direct future research in the treatment of obesity and its related complications. Moreover, this thesis provides a rigorous approach to validating biologically significant features the broader metabolomics community can mimic in future untargeted studies.enaerobic exerciseFAHFAsmetabolomicsThesis or Dissertation