Fig. 4: Haplotyping and allele-specific analysis of RNA-seq data.

a Haplotyping: Based on SNPs from Illumina sequencing larger haplotype blocks are created using PacBio long-reads. Large haplotype blocks are connected with the help of Hi-C. The breakpoints of rearrangements are also phased and used to label WT allele and affected allele. The haplotype information is used to phase RNA-seq data at positions with informative SNPs and to derive allele-specific gene expression values. b Differential gene expression analysis around breakpoints: Fraction of allelic imbalance genes (AIGs) with respect to the distance between an expressed gene and the closest breakpoint (red line). As a control, simulations of random breakpoints were performed. The light gray area with the gray line indicates the 5th percentile, median and 95th percentile, respectively, expected by chance. P-values were computed by comparing each observed value against values obtained from an empirical background model. The tests were performed right-sided, adjustments for multiple testing were not performed. c Same as b, but the distance of a gene to the next breakpoint is measured in TAD units (Same TAD, 1 TAD, etc.). P-values were computed as described in b.