Additional file 8: of Putative bovine topological association domains and CTCF binding motifs can reduce the search space for causative regulatory variants of complex traits

Figure S1. Runs of genes with allele-specific expression within regulatory units. Only runs of genes within TAD and between CTCF binding motifs are shown. Our ANOVA models showed that both TAD and CTCF are significant factors (P-value ≤ 10−6 and false discovery rate < 0.01) explaining the observed ASE variation, while gene is a also a significant factor in Figure S1A but not a significant factor in Figure S1B. The ASE scores (y-axis) for heterozygous loci (x-axis) are plotted as black dot points in each Manhattan plot. Since ASE score was a division of paternal to maternal allelic read counts, the heterozygous locus whose ASE score is larger than 0 favours paternal expression, and the heterozygous locus whose ASE score is less than 0 favours maternal expression. The putative bovine TAD is represented as a rectangle centring at y = 0. The CTCF binding motifs (motif score ≥ 80 and motif P-value ≤ 10−8) are represented as the start and end position of each arrowed curve, where the direction of the arrow is the direction of transcription. Genes are represented as coloured bars starting from y = 0 towards either the top (gene on forward strand) or the bottom (gene on reverse strand) of the graph. Gene names or IDs from Ensembl UMD3.1 annotation (release 75) are listed in legend. (ZIP 260 kb)