Atlas of plasma nuclear magnetic resonance biomarkers for health and disease in 118,461 individuals from the UK Biobank

Abstract

Blood lipids and metabolites are both markers of current health and indicators of risk for future disease. Here, we describe plasma nuclear magnetic resonance (NMR) biomarker data for 118,461 participants in the UK Biobank, an open resource for public health research with extensive clinical and genomic data. The biomarkers cover 249 measures of lipoprotein lipids, fatty acids, and small molecules such as amino acids, ketones, and glycolysis metabolites. We provide a systematic atlas of associations of these biomarkers to prevalence, incidence, and mortality of over 700 common diseases. The results reveal a plethora of biomarker associations, including susceptibility to infectious diseases and risk for onset of various cancers, joint disorders, and mental health outcomes, indicating that abundant circulating lipids and metabolites are risk markers well beyond cardiometabolic diseases. Clustering analyses indicate similar biomarker association patterns across different types of diseases, such as liver diseases and polyneuropathies, suggesting latent systemic connectivity in the susceptibility to a diverse set of diseases. The release of NMR biomarker data at scale in the UK Biobank highlights the promise of metabolic profiling in large cohorts for public health research and translation.

Quality control and removal of technical variation of NMR metabolic biomarker data in ~120,000 UK Biobank participants

“…Presently, NMR metabolic biomarkers have been quantified (Nightingale Health Plc.) and made available through UK Biobank for approximately one third of randomly selected participants ( Methods ) 4 . After sample QC ( Methods ), NMR metabolic biomarker data were available for 121,657 UK Biobank participants: 117,994 at baseline assessment (2006–2010) and 5,139 at first repeat assessment (2012–2013).…”

“…Technical variation is inherent to all laboratory measurements used for molecular phenotyping at biobank scale. Here, we comprehensively explored and document the sources and impacts of technical variation on the NMR metabolite biomarker data that has been made available for ~120,000 UK Biobank participants 4 . We found that the vast majority of biomarkers are relatively robust to technical variation, but that a small subset are significantly impacted by inter-spectrometer variation, within-spectrometer drift over time, and intra-plate variation.…”

“…Absolute concentrations of 168 biomarkers and 81 biomarker ratios were quantified by NMR spectroscopy (Nightingale Health Plc.) from non-fasting plasma samples (UK Biobank aliquot 3) of 121,695 randomly selected UK Biobank participants as previously described 4 . These included 117,981 participants at baseline assessment and 5,141 participants at repeat assessment, among which there were 1,427 participants with measurements at both time points.…”

“…To date, NMR biomarker quantification has been completed and made available by UK Biobank for 121,695 participants; showcased by Julkunen et al . 4 . Molecular phenotype data in large sample sizes are typically subject to the effects of unwanted technical variation (e.g.…”

Lipoprotein profile and metabolic fine-mapping of genetic lipid risk loci

“…Here we leverage the recently published Nightingale NMR data that is now available for ~100,00 participants of the UK Biobank to refine the metabolic architecture of these genetic lipid risk loci with a comprehensive set of 168 metabolite and lipoprotein related traits (Methods and Supplementary Tables 1&2) [18][19][20][21] . We hypothesize that this kind of deep molecular phenotyping will shed new light onto the biological functions of the Graham et al lipid risk genes and their role in regulating lipoprotein composition and turnover.…”