Extended Data Fig. 5: Time averaged distance measurements and co-initiation controls.

a, Shown are the X,Y,Z distances (mean ± STD, N=3, n > 1.2x10⁴) between the MS2 and PP7 based foci in a control reporter line, with interlaced stem loops, measured in the same imaging conditions as the scyl/chrb / chrb/CG11652 data (see methods). Spot localization errors are the presented STD values of this co-localization control. b–d, f–h, Distribution of average/95% percentile distances across non overlapping time windows in individual nuclei (with both foci detected for at least half of the timepoints in the window). Boxplots within the violin plots, show median, edges are 25th and 75th percentiles, whiskers extend to non-outlier data points (for the comparison of any two distributions within the same panel Mann Whitney or KS tests p value << 1*\({10}^{-4}\)). b, c, Distributions of time averaged distance measurements between fluorescent foci marking transcribing genes (corresponding to the instantaneous data plotted in Fig. 2). From left to right: for a co-localization control reporter gene with interlaced MS2 and PP7 stem loops driven by the Hunchback (hb) p2promoter/enhancer, for the scyl/chrb tagged genes and for the chrb/CG11652 tagged genes. Across non-overlapping 5min time windows (b), or 25 min windows (c). d, Distribution of the 95% percentile distance across 25min windows from individual nuclei. e, Same as a but for co-localization control measured in the same imaging conditions (see methods) as the knrl/kni data (mean ± STD, N = 3, n > 1.6x10⁴). f–h, Same as b–d except that knrl/kni is compared to the corresponding co-localization control. i, Instantaneous distances for a control reporter gene with interlaced MS2 and PP7, and for knrl/kni tagged genes measured in the anterior stripe domain (as in Fig. 2a), and posterior domain (regulated by enhancers proximal to kni). Boxplots within the violin plots, show median, edges are 25th, 75th percentiles, whiskers extend to non-outlier data points. j, Micro-C map encompassing the scyl/chrb/CG11652 region. Arrows mark the focal contact between scyl and chrb and the lack of such focal contact between chrb and CG11652. k, Computed frequency of co-initiation events (within 1.5 min) out of knrl initiation events, across all measured nuclei in individual embryos, for embryos where the genes are tagged in cis (purple) or in trans (blue) is shown. The pooled data from these embryos are presented in Fig. 2b. A boxplot showing the distribution of such frequencies computed by 100 random shuffling of the single-nucleus associations between green and red traces in the cis tagged embryos (see methods), is shown in grey (centre is median, edges are 25th, 75th percentiles, whiskers extend to non-outlier data points). As an additional control frequency of co-initiation events (within 1.5 min) is also computed for embryos where the kni gene is tagged with an interlaced MS2-PP7 cassette. To serve as an appropriate control, imaging was done with the same condition as the knrl/kni tagged embryos imaging (consequently green signal is slightly stronger). l, Mean transcriptional activity in the anterior stripe domain (arbitrary units) ± SEM over time in nc14 for the cis-tagged (purple, N = 7) and trans-tagged (blue, N = 6) embryos shown in k.