Fig. 4
mQTL mapping results in the wolf RRBS data. Methods for analysis include IMAGE (red), IMAGE-I (orange), IMAGE-A (green), MACAU (pink), GEMMA (brown), and BB (blue). a IMAGE identified more associated SNP-CpG pairs than the other five methods across a range of empirical FDRs constructed by permutation. b IMAGE identifies more consistent associations than the other methods in the subset analysis. Here, we randomly split individuals into two approximately equal-sized subsets and applied methods to analyze the two subsets separately. We count the number of overlapping associations between the top SNP-CpG pairs in the two subsets. The overlap ratio (y-axis) is plotted against the percentage of top SNP-CpG pairs (x-axis). c Upper panel: log2 odds ratio of detecting associated SNP-CpG pairs, together with the 95% CI, is computed for CpG sites residing in different annotated genomic regions. CpG sites associated with SNPs identified by IMAGE are enriched in open sea regions (p value < 2.2 × 10−16) and depleted in CpG island regions (p value < 2.2 × 10−16). Shores are defined as the 2000-bp regions flanking CpG islands; shelves are defined as the 2000-bp regions flanking the island shores (2000–4000 bp from CpG islands). Bottom panel: all analyzed CpG sites were annotated to genomic regions based on their relation to the nearest CpG island. CpG islands were annotated based on the UCSC Genome Browser (average length = 830 bp in the data; min = 201 bp; max = 322,257 bp). Shore is the flanking region of CpG islands covering 0–2000 bp distant from the CpG island. Shelf is the region flanking island shores covering 2000–4000 bp distant from the CpG island. d A higher percentage of CpG sites are directly disrupted by the SNP in the mQTL pairs compared to by chance alone (horizontal dashed line), and more so than in non-mQTL pairs (p value < 2.2 × 10−16). Such enrichment decays with increased FDR thresholds. *p < 0.05, **p < 0.01