A Data-Driven Measure of REM Sleep Propensity for Human and Rodent Sleep

Abstract

Mammalian sleep is characterized by multiple alternations between episodes of rapid-eye-movement sleep (REMS) and non-REM sleep (NREMS). While the mechanisms governing the timing of these ultradian NREMS-REMS cycles remain poorly understood, the phenomenon of REMS pressure, namely a drive for REMS that builds up between REMS episodes, is thought to be a contributing factor. Prior analyses of NREMS-REMS cycles in mice has suggested that time in NREMS is a primary contributor to REMS pressure. Building on that finding, we previously introduced a REMS propensity measure defined as the probability to enter REMS before the accumulation of an additional amount of NREMS. Analyzing mouse ultradian cycle data, we showed that REMS propensity at REMS onset was positively correlated with REMS bout duration and with the probability of the occurrence of a REMS bout followed by a short inter-REMS interval, called a sequential REMS cycle. In this paper, we extend our analyses of REMS propensity to human and rat ultradian NREMS-REMS cycle data. We show that, as in mice, human and rat sleep contain both short NREMS-REMS sequential cycles and longer single NREMS-REMS cycles, though there are some differences in the relative distributions of cycle durations. Although rodents exhibit polyphasic sleep in contrast with the consolidated sleep of humans, the calculated REMS propensity measures in all three species show similar profiles as functions of time spent in NREMS: specifically, REMS propensity increases with time spent in NREMS until it reaches a peak value, and then it decays with additional time in NREMS. Positive correlations of REMS propensity at REMS onset with REMS bout duration were present in both human and rat data as in mouse data, suggesting that time spent in NREMS also influences REMS duration in these species.

Figures (15)

• Figure 1: Example hypnograms of sleep behavior over 4 h in human (A), mouse (B, light period) and rat (C, light period). The inset highlights a representative REMS cycle that is initiated at the onset of a REMS bout, REMpre (green), and terminates at the onset of the next REMS bout, REMpost (magenta). The total time spent in NREMS during the inter-REMS interval is defined as $|N|$ (blue). Note that the full inter-REMS interval contains both NREMS and wake bouts shorter than 2 minutes.
• Figure 2: Inter-REMS interval duration increases with the duration of the preceding REMS bout across species. Positive correlations are exhibited between REMS bout duration $|\mathrm{REMpre}|$ and the subsequent inter-REMS interval duration $|\mathrm{IREM}|$ in all species: (a) human (green), (b) mouse (light phase, purple), and (c) rat (light phase, orange). Each point represents one REMS cycle; solid lines are least-squares fits with slopes $1.085, 3.339$, and $1.262$ for human, mouse (light), and rat (light), respectively. The reported significance levels were obtained from a two-sided test of zero Pearson correlation using the Fisher $z$-transform normal approximation, with null hypothesis $H_0: \rho=0$ and alternative hypothesis $H_A:\rho \neq 0$, where $\rho$ denotes the population correlation between $|\mathrm{REMpre}|$ and $|\mathrm{IREM}|$. The corresponding $p$-values are $p = 6 \times 10^{-94}$ (human), $p \ll 0.001$ (mouse), and $p=1.68 \times 10^{-67}$ (rat). The mouse and rat dark phases are shown in Supplementary Figure \ref{['fig: S1_dark']}.
• Figure 3: Empirical inter-REMS $|N|$ distributions across species suggest multiple characteristic timescales. Histograms show the empirical distribution of $|N|$, where $|N|$ is the cumulative duration of NREMS in the inter--REMS interval, for (a) human, (b) mouse (light phase), and (c) rat (light phase) data (see Supplementary Figure \ref{['fig: S2_dark']} for dark phases of rodent data). Across species, $|N|$ spans a wide range and exhibits structured, non-Gaussian profiles.
• Figure 4: Mixture model fits of inter-REMS $|N|$ distributions partitioned by $|\mathrm{REMpre}|$ duration. Mixture models fitted to $|N|$ distributions for human data (a) and $\log(|N|)$ distributions for rodent data (b: mouse (light phase), c: rat (light phase)) for REMS cycles with similar durations of the preceding REMS bout $|\mathrm{REMpre}|$; dark phase results for mouse and rat data are provided in Supplementary Figure \ref{['fig: S3_dark']}. Each subpanel corresponds to a distinct $|\mathrm{REMpre}|$ bin (ranges in seconds or minutes shown in panel titles). Histograms show empirical pdfs, where blue and red curves denote the short- and long-interval components of the mixture model.
• Figure 5: REMS propensity functions grouped by $|\mathrm{REMpre}|$ durations. Propensity functions $P(t,\Delta)$ (Eq. \ref{['eq:propensity']}) are shown for REMS cycles grouped by similar $|\mathrm{REMpre}|$ durations, corresponding to the binned inter--REMS $|N|$ distributions in Figure \ref{['fig:gmm_bin']}, for (a) human, (b) mouse (light phase), and (c) rat (light phase) data. For human data, $P(t, \Delta)$ was computed from the three part atom + E1-short + truncated-normal mixture model fit to $|N|$ (in minutes); for rodents, $P(t,\Delta)$ was computed from the two-component Gaussian mixture model (GMM) fit to $\log(|N|)$ (in seconds). In each panel, $P(t,\Delta)$ is plotted as a function of the accumulated NREMS duration $t$ since the prior REMS bout. Red markers denote local minima (troughs) and maxima (peaks) with the correponding $|N|$ values annotated. Dark-phase mouse and rat results are shown in Supplementary Figure \ref{['fig: S5_dark']}.