Rev-erbα basal expression and circadian oscillations depend on cell density. (A) Schematics of the systematic computational analysis pipeline used to calculate the circadian power fraction (cpf) of a population of cells expressing RevVNP and H2B-mCherry and imaged during 72 h via time-lapse confocal microscopy. (B) Confocal microscopy (top) and phase contrast (PC; bottom) images of RevVNP-expressing cells grown at low (left) and high (right) density. Scale bar, 50 µm. (C) Violin plots representing the distribution of the single-cell RevVNP intensities of low- and high-density populations of a typical experiment of 14; n = 112 cells and n = 622 for low density and high density, respectively; medians and interquartile ranges are depicted as white circles and black bars, respectively. Two-sided Wilcoxon rank sum test; **** indicates a P value < 0.0001. Full P values are reported in Table S1. (D) Raw data of the RevVNP intensities over time, represented in kymograph style, of 325 cells grown and tracked under low- or high-density conditions, from a representative experiment of 14. The single tracks are ordered from lower (top) to higher (bottom) circadian power fraction and aligned along the time axis according to maximum cross-correlation with the median track. (E) Violin plots representing the distribution of the single-cell RevVNP circadian power fraction of low- and high-density populations; n = 2,726 and 2,389 cells, respectively, from four experiments; medians and interquartile ranges are depicted as white circles and black bars, respectively. Two-sided Wilcoxon rank sum test; **** indicates a P value <0.0001. Full P values are reported in Table S1.