Back

The Diurnal Rhythmicity of Untargeted Metabolomic Profiles in Humans

Nagarajan, P.; Faquih, T.; Yu, B.; Dashti, H. S.; Noordam, R.; Xiao, Q.; Kaplan, R.; Qi, Q.; Isasi, C. R.; Saxena, R.; Scheer, F. A. J. L.; Redline, S.; Sofer, T.; Wang, H.

2026-04-29 epidemiology
10.64898/2026.04.28.26347027 medRxiv
Show abstract

It is well established that the circadian rhythm system plays a crucial role in metabolic regulation. Yet, the influence of time of sample collection in cross-sectional, untargeted metabolomics datasets remains underexplored. We therefore investigated the temporal oscillations of 853 serum metabolites from 5954 fasting samples in the Hispanic Community Health Study / Study of Latinos cohort, where blood samples were drawn within a short daytime window (7:45 am - 1:38 pm). In total, 84 serum metabolites were significantly associated (p<5.9e-05) with blood draw time. Time-varying metabolites exhibit enrichment in pathways of incretin hormones and light signaling, mechanistic connections to neurological and circadian control (dopamine response, melatonin activity), and implications for therapeutics (serotonergic drugs) and health disorder understanding (sleep, depression). Unsupervised analysis revealed latent, nonlinear patterns in five metabolite clusters according to their temporal profiles. Impact on false and missed discovery, and altered effect size estimates for metabolite-health disorder associations underscore the importance of accounting for time. Results fundamentally show that serum metabolites exhibit non-negligible temporal variation even in a restricted short daytime window. Accounting for this variance appears important for precise disease understanding from metabolomics data.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.