Fig. 4: In the absence of HIRA, only non-transcribed early IZs switch from compartment A to B.

a Comparison of EV1 distribution genome-wide (50 kb bins, n = 40887, grey) or at blurred (n = 2382, green) and buried (n = 439, red) sites between WT and HIRA KO cells. Slope, intercept and R from linear regression are noted for each set. b Mean value of EV1 (1st eigenvector, indicating compartment) at blurred (n = 2382) and buried (n = 439) sites from WT (teal) and HIRA KO (light teal) cells. The boxplot centre is the median, the bounds of the box are the first and third quartiles and the whiskers extend to 1.5x IQR. Source data are provided as a Source data file. c Proportions of blurred and buried sites which remain in the same compartment (unchanged, light grey) or undergo a switch (changed, dark grey) from WT to HIRA KO cells. A set of randomised size-matched sites was quantified as control. d Scatterplot of mean EV1 value at buried sites (n = 439) from WT and HIRA KO cells. Colour represents transcriptional activity from HIRA WT cells. Proportion of sites which change from compartment A-to-B (lower right quadrant) or B-to-A (upper left quadrant) are quantified. EV1 was calculated from 50 kb-binned Hi-C matrices and then re-binned to 10 kb to compute mean EV1 value for blurred and buried sites. Transcriptional activity is measured as the mean log₂(cpm+1) RNA-seq signal binned at 10 kb for each site. Two-tailed Mann-Whitney U test corrected for multiple testing by FDR (5% cut-off) was used to determine significance of differences between WT and HIRA KO. Significance was noted as: * (p < =0.05), ** (p < =0.01), *** (p < =0.001) for all comparisons. P-values HIRA WT vs KO at blurred sites p = 0.45, at buried sites p = 5.14e^-14.