Open AccessArticle
Power Analysis for Human Melatonin Suppression Experiments
by
Manuel Spitschan, Parisa Vidafar, Sean W. Cain, Andrew J. K. Phillips and Ben C. Lambert
Viewed by 2002
Abstract
In humans, the nocturnal secretion of melatonin by the pineal gland is suppressed by ocular exposure to light. In the laboratory, melatonin suppression is a biomarker for this neuroendocrine pathway. Recent work has found that individuals differ substantially in their melatonin-suppressive response to
[...] Read more.
In humans, the nocturnal secretion of melatonin by the pineal gland is suppressed by ocular exposure to light. In the laboratory, melatonin suppression is a biomarker for this neuroendocrine pathway. Recent work has found that individuals differ substantially in their melatonin-suppressive response to light, with the most sensitive individuals being up to 60 times more sensitive than the least sensitive individuals. Planning experiments with melatonin suppression as an outcome needs to incorporate these individual differences, particularly in common resource-limited scenarios where running within-subjects studies at multiple light levels is costly and resource-intensive and may not be feasible with respect to participant compliance. Here, we present a novel framework for virtual laboratory melatonin suppression experiments, incorporating a Bayesian statistical model. We provide a Shiny web app for power analyses that allows users to modify various experimental parameters (sample size, individual-level heterogeneity, statistical significance threshold, light levels), and simulate a systematic shift in sensitivity (e.g., due to a pharmacological or other intervention). Our framework helps experimenters to design compelling and robust studies, offering novel insights into the underlying biological variability in melatonin suppression relevant for practical applications.
Full article
►▼
Show Figures
---
---
---
---
---
---
---
n = 41 participants generated using the virtual-experiment (n = 41) function from the melluxdrc R package. The assumptions underpinning each of these panels are described in Section 2.2.
Full article
">
2). The individual points show the estimates of
parameters provided by the authors of [
22]. The uncertainty ribbon indicates the 2.5–97.5% posterior predictive quantiles; the black line indicates the 50% posterior predictive quantile. (
B) Posterior predictive check:
model. Plot shows a graphical check of the model fit of Equation (
4). Each black line represents a gamma density function corresponding to particular posterior samples of the parameters. The blue bars indicate the values of
estimated by the root-finding algorithm. (
C) Assessing virtual individual generation: dose-response parameters. Each orange point represents a draw of
parameters (in Equation (
1)) obtained via Algorithm 1: here, we show 25,000 such estimates; each green point represents raw estimates from [
22]. (
D) Assessing virtual individual generation:
quantiles. Each orange point represents the
values correspond to a draw of
parameters (in Equation (
1)) obtained via Algorithm 1: here, we show 25,000 such estimates; each green point represents the
quantiles corresponding to the raw estimates from [
22].
Full article
">
<
5
%
) and upper (percentage of observations ) saturations. Each black line corresponds to saturations generated from a single virtual experiment of sample size . Each orange point corresponds to the real saturation.
Full article
">
n
=
41
participants generated using virtual-experiment (n = 41) function from the melluxdrc R package. Panel B shows the same but assuming a reduction in individual variance given by as given by virtual-experiment (n = 41, individual-variation-level = 0.2). The assumptions underpinning each of these panels are described in Section 5.
Full article
">