Fig. 3: Haplotype-specific NO analysis shows local effects of a copy-neutral driver SV in AML.

a, Balanced t(8;21) translocation in AML_1, discovered based on strand cosegregation (P value = 0.00003 for translocation discovery using strand cosegregation²⁴, FDR-adjusted Fisher’s exact test; Supplementary Fig. 16). The SV breakpoint was fine-mapped to the region highlighted in light blue. Composite reads shown were taken from all informative cells in which reads could be phased (WC or CW configuration; Methods). b, A violin plot demonstrates haplotype-specific NO at the RUNX1T1 gene body (10% FDR; two-sided Wilcoxon rank sum test followed by Benjamini–Hochberg multiple correction; n = 17 single cells; boxplot was defined by minima = 25th percentile – 1.5× IQR, maxima = 75th percentile + 1.5× IQR, center = median and bounds of box = 25th and 75th percentile), consistent with aberrant activity of the locus on der(8). c, Haplotype-specific NO around the SV breakpoint. FCs of haplotype-specific NO, measured between the RUNX1-RUNX1T1 containing derivative chromosome (der(8)) and corresponding regions on the unaffected homolog (Ref), are shown in black, and –log₁₀(P values) in light blue. Enhancer-target gene physical interactions based on chromatin conformation capture^56,93 are depicted in orange (interactions involving RUNX1 and RUNX1T1) and gray (involving other loci). d, Significant CREs located within the distal peak region, demonstrating haplotype-specific absence of NO on der(8) at 10% FDR, suggesting increased CRE accessibility on der(8). Within the segment around 0.8 to 1.1 Mb upstream of RUNX1, which showed pronounced haplotype-specific NO, we tested 69 CREs for haplotype-specific NO, which identified two significant CREs. e, Haplotype-specific NO measured between der(8) and corresponding regions of the unaffected homolog. Red, regions corresponding to the fused TAD. f, A beeswarm plot shows that the fused TAD (red) is an outlier in terms of haplotype-specific NO on der(8) (P values based on Kolmogorov–Smirnov tests; n = 83 TADs in der(8); boxplot was defined by minima = 25th percentile – 1.5× IQR, maxima = 75th percentile + 1.5× IQR, center = median and bounds of box = 25th and 75th percentile).