Fig. 2: Human-specific chromatin contact ___domain (CCD) boundaries contribute divergent transcriptional concordance.

a Examples of CCD boundaries across species, showing insulation scores (IS), CTCF sites, peaks, and loops. Dashed lines indicate human CCD boundaries and their corresponding positions in non-human primates. b Heatmaps comparing the insulating function of variable boundaries (top) between humans and non-human primates, with conserved boundaries (bottom) as control. c Box plots of CTCF loop intensities across human-specific (HS), ape-specific (APE), miscellaneous (Misc), and conserved (Consv) boundaries. The box plot shows the median (central line), interquartile range (box), and whiskers extending to 1.5× the interquartile range. ****P < 0.0001; ns, not significant as determined by one-tailed Wilcoxon test. d Heritability enrichment for autism spectrum disorder susceptibility across variable and human-specific (HS) boundaries in 50 kb bins. The boundary bin is highlighted, with a red arrowhead indicating CTCF site orientation. Error bars show standard errors, and asterisks indicate significant enrichment after Bonferroni correction using LDSC (See Methods). e Heatmap showing unsupervised hierarchical clustering of normalized H3K27ac signals from 32 human primary tissues at variable boundaries in 50 kb bins. f H3K27ac intensity profiles at variable and conserved (Consv) boundaries with flanking regions, using data from three human brain regions. CTCF loop anchors (Random) serve as controls. Dashed lines denote the positions of boundaries or anchors. g Schematic outlining criteria for gene pairing to assess transcriptional concordance between human boundaries and their macaque counterparts (See Methods). h Smooth curve plots depicting transcriptional correlations between gene pairs as a function of genomic distance across expressions from human brain regions and their macaque counterparts. The gray shading indicates the 90% confidence intervals. P values were calculated using the one-tailed Wilcoxon test. i Scatter plots of transcriptional correlations for gene pairs separated by human-specific boundaries, using RNA-seq data from 34 human and 23 macaque medial frontal cortical tissues. Gray lines show linear regression of transcript levels for gene pairs, with shading indicating the 90% confidence intervals. Pearson correlation coefficients (r) and P values were calculated by two-sided t-test. The source data, n numbers, and exact P values for (c, d, i) are provided in the Source Data file.