The role of sleep in the human brain and body: insights from multi-organ imaging genetics

Abstract

Sleep is essential for the health of the brain and heart. Although sleep has been identified as a factor in a few specific clinical outcomes, a systematic analysis of the relationship between sleep and brain/heart and their genetic underpinnings is lacking. Medical images can provide useful clinical endophenotypes for organ structures and functions. Here we present a systematic genetic investigation of sleep-brain/heart connections using multi-modal brain and cardiac images from over 40,000 subjects in the UK Biobank. We identified novel phenotypic and genetic links between sleep and a wide range of imaging traits, such as brain structures, white matter integrity, brain activities, as well as cardiac structures and functions. We prioritized a number of imaging modalities and traits for specific sleep conditions, such as the resting brain function measures in the somatomotor network with narcolepsy. Sleep and imaging had overlapping genetic influences in 39 genomic loci, some of which showed evidence of shared causal genetic variants. In conclusion, large-scale imaging genetic data illuminate the implications of sleep on brain and cardiac health and their genetic links. An interactive web browser (www.ig4sleep.org) has been developed to facilitate exploring our results.

Sleep-insomnia superposition: opposing brain signatures of sleep in task-based and resting-state conditions

“…Superior parietal regions showed significant associations with the duration of 22 longest sleep bout, depressive symptoms, and cognition. Insomnia and dozing showed few significant associations, in line with previous univariate analyses on the same data 27 .…”

“…In addition, resting state results revealed widespread associations similar to Fan et al 2022 27 , especially for daytime dozing. We found that functional connectivity between the frontoparietal network and in particular the posterior inferior frontal junction (IFJp) and lateral occipital regions was positively associated with duration of longest sleep bout.…”

“…However, we believe that the first explanation is more likely given the phenomenon of contradictory subjective and objective sleep measure results was observed in polysomnography but not accelerometry measures [19][20][21][55][56][57] and that our results were reproduced in the HCP dataset where sleep duration was self-reported 24 . This pattern was also reproducible through analyzing publicly available coefficients from an independent analysis of UK Biobank 27 .…”

“…This pattern was similar, with the daytime dozing functional connectivity also showing negative correlations. To confirm the validity of these results, we retrieved independently modeled associations from Fan et al 27 between selfreported insomnia, daytime dozing, and sleep duration (data retrieved at http://www.ig4sleep.org/) and performed the same correlational analyses. Reassuringly, we found nearly identical patterns of correlations between effects (Table S4); self-reported 15 insomnia and insomnia had a correlation coefficient of -0.66 (compared to -0.59 in our analysis).…”

“…One limitation of this study was that they used the overall Pittsburgh Sleep Quality Index (PSQI) as their measure of sleep 26 and only assessed resting state data, thus missing the heterogeneity of sleep-related phenotypes manifest under different conditions. More recently, Fan et al (2022) performed a systematic analysis of multiple sleep phenotypes in a subset of white-British population of the UK Biobank 27 . They analyzed restingstate and task-based fMRI data, diffusion tensor imaging, and cardiac MRI data, testing their independent relationships with self-reported sleep data and other environment and mental health variables.…”