Extended Data Fig. 10: NIPBL-dependence is associated with unique genomic structures across lineages.
a, Cross-TSS connectivity at example genes sensitive to NIPBL depletion during mitotic exit. The top-right corner of each box corresponds to the TSS of each gene. b, c, d, Quantification of Fig. 4d, specifically the mean observed/expected contact frequency for the flanking regions (b), central regions (c), and the ratio between flanking and central regions (d) for each TSS. P-values were determined by independent, two-sided t-tests for (b) and (c), and using two-sided Wilcoxon rank-sum tests for (d). (b) non-sig. DMSO n = 12964, non-sig. dTAG n = 12966, non-sig. (decreased NN) n = 450, decreased n = 450, non-sig. (increased NN) n = 87, and increased n = 91. (c & d) non-sig. DMSO n = 12889, non-sig. dTAG n = 12893, non-sig. (decreased NN) n = 450, decreased n = 449, non-sig. (increased NN) n = 87, and increased n = 91. The inset shows the regions that were quantified, with the mean observed/expected contact frequency for a given TSS taken from the shown regions for plotting. The center lines mark the median, the boxes show the IQR, and the whiskers extend to 1.5xIQR; all box plots in this figure use the same metrics. e, Ratio of NIPBL to PDS5 (left) and WAPL (right) at a 20 kb region centered on the TSSs of non-significant, decreased and increased genes. Two-sided Wilcoxon rank-sum tests were performed to give the shown p-values (non-sig. (decreased NN) n = 436, decreased n = 449, non-sig. (increased NN) n = 91, increased n = 92). Only non-significant nearest neighbors (NN) are shown. f, Balanced counts between super-enhancers (left) or typical-enhancers (right) and enhancers on the opposite side of a TSS. For each gene, the mean of different, non-zero enhancer-enhancer pairs was taken before plotting. P-values were determined using a two-sided Wilcoxon rank-sum test (SE-E: non-sig. n = 8204, non-sig. (decreased NN) n = 307, decreased n = 359, non-sig. (increased NN) n = 66, and increased n = 71; TE-E: non-sig. n = 11995, non-sig. (decreased NN) n = 431, decreased n = 440, non-sig. (increased NN) n = 87, and increased n = 91). All non-significantly changed genes and nearest neighbor (NN) non-significantly changed genes are shown. g, Examples of TSS insulation at a non-significantly changed gene (NECTIN3) and reduced TSS insulation at a significantly decreased gene (JARID2). For each gene we show Virtual 4 C anchored at two different super-enhancers (SE-A: blue, SE-B: pink) that are upstream of the TSS. The arrow marks the transcription start site (TSS). h, Immunofluorescence of hiPSC-derived cardiomyocytes following treatment with DMSO or dTAGV-1 for 4 hours. cTnT is a marker for cardiomyocyte differentiation. i, For the significantly decreased genes in hiPSC-derived neurons and cardiomyocytes, we performed nearest neighbor analysis to identify the closest non-significant genes based on FPKM and the number of enhancers and TSSs within 1 Mb up- or down-stream. A grey line connects each significantly changed gene with its nearest neighbor. j, Comparison of FPKM, enhancer count, and gene count metrics between significantly changed genes and the non-significant nearest neighbors. P-values were determined using a two-sided Wilcoxon rank-sum test. k, Observed/expected pile-up analysis at a 10 kb resolution between neighborhood (±1 Mb of TSS) typical-enhancers (TE) and non-significant genes, nearest neighbors, or decreased genes in hiPSC-derived neurons and cardiomyocytes.
