Primary Submission Category: Health behaviors
Morningness Polygenic Scores Interact with Social Schedules to Predict Sleep Timing in U.S. Young Adults
Authors: Jess M. Meyer Brandt Levitt Allison E. Aiello Kathleen Mullan Harris
Presenting Author: Jess M. Meyer*
Prior research suggests both social environments and genetic inheritance affect sleep schedules. Polygenic scores (PGSs) derived from genome-wide association studies (GWASs) provide the opportunity to explore in greater depth how genes and environment interact to affect sleep. A recent GWAS (Jones et al. 2019) of morningness—the extent to which someone identifies as a “morning” (vs. “evening”) person—included >400,000 UK Biobank participants of European ancestry. Using Add Health data, we examined how a morningness PGS constructed from those GWAS results predicts sleep timing in a nationally representative sample of U.S. young adults. We tested the expectation that genetic predictors of circadian preference operate more strongly in the absence of social constraints on sleep by examining how the association between this PGS and sleep midpoint, the halfway point between bedtime and wake time, differs between work and free days. Because the basis GWAS used a European ancestry sample, we also examined whether associations are heterogenous across European vs. non-European genetic ancestry. Preliminary results show that for European, African, and Hispanic genetic ancestry groups, higher values on morningness PGS were associated with earlier sleep midpoint on free days. However, within each of these groups, morningness PGS was less predictive of sleep midpoint on workdays. As expected, the PGS showed the strongest association with sleep timing in the European genetic-ancestry group, though this may reflect this group’s larger size (and statistical power) in our sample. These results boost confidence in the validity of this PGS for predicting sleep timing in U.S. young adults, though highlight the unfortunately ubiquitous issue of disparities in PGS predictiveness by genetic ancestry. Our results align with the expectation that genetic correlates of circadian preference are stronger predictors of sleep timing on free days compared to days with work or school activities.