The curves show the proportion of metabolites expected to be detected in a case control study as a function of effect size.

<p>Effect size is defined by the relative risk (RR, on the x-axis) of disease comparing individuals in the top and bottom quartiles of the “usual” metabolite level. The top axis shows the naïve relative risk that would be observed in the study without adjusting for measurement error. Each figure varies one parameter: sample size, α-level, or the number of samples/individual. (A) presents power curves according to different sample size (n of 500, 1,000 and 5,000) under a Bonferroni-adjusted α-levels (0.05/539); (B) presents power curves with different α-levels in a case-control study of 1,000 individuals; (C) presents power curves in a case-control study of 1,000 individuals, with different number of distinct urinary samples (1, 3, and 10, α-level = 0.05/539).