Fig. 2: Deconvolution of transcription in living embryos reveals gene-specific behavior at repression onset.
Heatmap showing the number of polymerase initiation events in nc14 for snaMS2/+ (a) and sogMS2/+ (b) as a function of time. Each row represents one nucleus, and the number of Pol II initiation events per 30 s bin is indicated by the bin color. c Deconvolution of the transcriptional site intensities into RNA polymerase II initiation events over time. The average waiting time (<τ > ) between polymerase initiation events is calculated for all nuclei within a sliding time window (∆t). The inverse of <τ> is the product of the probability to be active, denoted pON, and the polymerase initiation rate (kini) for the given time window. The inverse value is plotted over time as a proxy for the stability of the underlying transcriptional kinetic regime. Stationarity is denoted by a slope ≈ 0. Kinetic parameter stability as a function of time for snaMS2/+ (d) and sogMS2/+ (e) transcription expressed as the product of the probability to be active (pON) and the RNA polymerase II initiation rate (kini). Error represents the upper and lower bounds of the 95% confidence interval from all movies. False-colored projections from live imaging of snaMS2/+ (f) and sogMS2/+ (g) embryos, with active nuclei indicated in teal and inactive in gray (Supplementary Movies 1, 2). scale bar is 20 μm. Distribution of switching times for initiation of stable repression in nuclear cycle 14 for snaMS2 (h) and sogMS2 (i) as determined using Bayesian Change Point Detection. The p value for the difference in distributions is 0.0012 (Kolmogorov-Smirnov test). Statistics: snaMS2/+: N = 6 embryos, n = 448 nuclei. sogMS2/+: N = 3 embryos, n = 141 nuclei.
