Supplementary Figure 3: DNA methylation concordance between spatially proximal regions.

a. Genome-wide DNA methylation correlation calculated in loop anchors by overlapping loops (n=3000) from in situ Hi-C and WGBS datasets from the same cell line (Hi-C+WGBS) or by overlapping loops (n=3110) and DNA methylation tracks from the same experiment (Methyl-HiC). Significance were assessed by Pearson Correlation Coefficient. b. Illustration of the calculation for methylation concordance on DNA loop anchors. Only reads have CpGs on both ends are considered. Read ends locate in anchor regions but are not linked are used as controls. c. An illustration shows the methylation status on read pairs supporting a chromatin loop. Methylation status of each end is determined by the average methylation level of CpG sites on the reads. d. Pearson correlation coefficient distribution of individual loops (n=3110). Boxes are median with quartiles and whiskers extend to 5th and 95th percentile. e. DNA methylation concordance comparison between loops with both anchors overlapping with CTCF peaks (n=2,837) and non-CTCF loops (n=3,145). Significance was assessed using Fisher’s (1925) z in cocor package. f. DNA methylation concordance comparison between reads (n=167,464) with both ends originated from the same TADs and reads (n=52,675) from two different TADs. Only reads containing 2 or more CpGs on each end were included. Significance was assessed using Fisher’s (1925) z in cocor package. g. DNA methylation concordance comparison between reads with both ends originated from compartment A (n=115,286) and B (n=218,488). Only reads containing 2 or more CpGs on each end were included. Significance was assessed using Fisher’s (1925) z in cocor package. h. Pearson correlation coefficient of DNA methylation concordance between 1Mb bins from different chromosomes. Inter-chromosomal reads connecting every 1Mb bins in two chromosomes (n=773,026) were extracted and methylation concordance was calculated. Paired reads located within the same bins were shuffled and used to calculate the expected values. Only reads containing 2 or more CpGs on each end were included. Significance was assessed using Fisher’s (1925) z in cocor package.