Improving visualisation and interpretation of metabolome-wide association studies (MWAS): an application in a population based cohort using untargeted 1H

: 1 H NMR spectroscopy of bio fl uids generates reproducible data allowing detection and quanti fi cation of small molecules in large population cohorts. Statistical models to analyze such data are now well-established, and the use of univariate metabolome wide association studies (MWAS) investigating the spectral features separately has emerged as a computationally e ffi cient and interpretable alternative to multivariate models. The MWAS rely on the accurate estimation of a metabolome wide signi fi cance level (MWSL) to be applied to control the family wise error rate. Subsequent interpretation requires e ffi cient visualization and formal feature annotation, which, in-turn, call for e ffi cient prioritization of spectral variables of interest. Using human serum 1 H NMR spectroscopic pro fi les from 3948 participants from the Multi-Ethnic Study of Atherosclerosis (MESA), we have performed a series of MWAS for serum levels of glucose. We fi rst propose an extension of the conventional MWSL that yields stable estimates of the MWSL across the di ff erent model parameterizations and distributional features of the outcome. We propose both e ffi cient visualization methods and a strategy based on subsampling and internal validation to prioritize the associations. Our work proposes and illustrates practical and scalable solutions to facilitate the implementation of the MWAS approach and improve interpretation in large cohort studies.