The development of metabolomics research in population health science has been driven by mass spectrometry (MS) and proton nuclear magnetic resonance (NMR) spectroscopy as the two key experimental technologies. NMR spectroscopy is increasingly used because of its capability to simultaneously detect a wide range of metabolites, thus providing a "snapshot of the metabolite composition" in biofluids. NMR-based applications can also offer fully automated and highly reproducible high-throughput experimentation in a very cost-effective manner. During the last 5 years Prof Mika Ala-Korpela's Computational Medicine team has focused on developing an NMR metabolomics platform for analysing human serum - 1H NMR spectroscopy (Tukiainen et al 2008). This novel methodology has been used to analyse serum from over 100,000 individuals (in less than 4 years). The method uses three molecular windows, (two applied to native serum and one to serum lipid extracts requiring minimal preparation) to quantify the 216 metabolomic traits, including 14 lipoprotein subclasses with detailed molecular information on serum lipid extracts including free and esterified cholesterol, sphingomyelin, degree of saturation and omega-3 fatty acids, 117 metabolites (including 80 lipoproteins, 15 lipids and 22 low molecular weight metabolites), 99 derived metabolic measures indexing amino acid metabolism, gluconeogenesis, ketogenesis, kidney function, and fatty acid saturation, as well as measures such as apolipoprotein A-I and B (see list in Kettunen et.al 2012) . The majority of analyte data is reported as a molar concentration, with derived variables reported as ratios . 1H NMR spectroscopy-based quantification reduces measurement error for main blood lipid fractions, compared to the usual enzymatic methods, and also allows detection of associations with lipid and metabolic sub-phenotypes that may be proximal consequences of genetic variation and environmental risk factors.

Genome-wide associations with novel metabolic pathways are likely to be identified with greater statistical efficiency using phenotypes obtained from this method, and estimates from non-genetic association studies are also likely to be more precise. For example, in a GWAS of 8,330 adults using 216 serum metabolic phenotypes measured using 1H NMR spectroscopy 31 GWAS loci with p less than 10-10 were identified, 20 of which replicated previously known associations obtained in much larger sample sizes and 11 of which were novel(Kettunen et. al. 2012). In a second study of 9,179 adults, 6 novel metabolites associated with CVD mortality, which were independent of established risk factors, were identified. (al-hussaini et.al. 2012)

We plan to analyse all suitable plasma and serum samples collected from ALSPAC participants, their mothers and fathers at as many timepoints as possible. Methodology exists for analysis of EDTA plasma and serum samples from adults and children. Methodology will be developed for analysis of cord blood plasma and/or serum samples and may be developed for heparin plasma samples.

Data generated will become part of the ALSPAC resource and utalised for a numerous association studies which will be covered by separate applications to ALSPAC Executive.

Samples will initially be analysed in Prof Ala-Korpela's Computational Medicine laboratories,

University of Oulu & University of Eastern Finland (Kuopio), (details of first sets on samples in Appendix 1b below).

UoB has recently invested in core metabolomics facilities, including a new 500 MHz instrument NMR system 2, which will be overseen by Prof Ala-Korpela. In the future, analysis will be completed in these new UoB facilities.

Tukiainen T, Tynkkynen T, Makinen VP et al. A multi-metabolite analysis of serum by 1H NMR spectroscopy: early systemic signs of Alzheimer's disease. Biochem Biophys Res Commun 2008;375:356-361

Kettunen J, Tukiainen T, Sarin AP et al. Genome-wide association study identifies multiple loci influencing human serum metabolite levels. Nat Genet 2012;44:269-276

al-hussaini A, Sehmi J, Tan T, Ala-Korpela M, Kooner J, Chambers J. Identification of novel metabolic biomarkers for cardiovascular mortality. Journal of the American College of Cardiology 2012;59:E1649-doi:10.1016/S0735-1097(12)61650-7