Metabolomics meets genetics - from an NMR metabolomics platform to the genetic architecture of serum metabolites

Metabolomics is a recently emerged field of science studying metabolites and how their levels change with biological perturbations. A key requirement for metabolomics analyses is a technology that can capture a multitude of metabolite information in a single measurement. As many of the available platforms have lacked automation in the metabolomics experimentation, including the data analysis and handling, the measurements have been costly and time-consuming, and thus metabolomics data had not been widely applied in large-scale studies. Metabolomics profiling, however, has great potential to provide further biological knowledge by, for example, elucidating in detail the mechanisms and pathways underlying disease. The first two publications of this thesis present a high-throughput proton nuclear magnetic resonance (NMR) -based serum metabolomics platform designed to facilitate the use of metabolomics data in large biomedical studies. The platform allows the highly-automated metabolomics profiling of tens of thousands of samples per year in a cost-effective manner and with the implemented models more than a hundred metabolites, including lipoprotein subclasses, other lipids and small molecules, can be quantified from the serum NMR data. The metabolomics profiling provided by the NMR-based platform has gained wide interest; the platform has run non-stop since it was set up in late 2008 as many Finnish and international cohorts have had their samples measured and used the data in several publications. In the two other publications included in this thesis, the quantitative metabolite data obtained through the platform was combined with detailed data on genetic variants in more than 8000 Finnish individuals. This unique data set was used a) to comprehensively characterize, in terms of metabolite and genetic associations, the genomic regions known to associate with blood lipid levels, and b) to dissect genetic components associated with the changes in the metabolite levels. A wealth of biological information was uncovered in these studies including new metabolic associations for the known genetic regions and several new genetic regions associated with the metabolites. These findings can help to understand the links between the genes and clinical conditions. Together the results of this thesis show how detailed metabolomics data greatly complements the conventional laboratory measurements and support the use of this data in biomedical studies as means to provide valuable biological knowledge.