Figure 6

PREs are disproportionately disrupted by SNP risk alleles at enhancer-associated non-coding psoriasis susceptibility loci. We analyzed 572 psoriasis-associated SNPs, including (A, B) 324 non-coding SNPs and (C, D) 53 non-coding SNPs within an NHEK enhancer. Risk alleles for these SNPs were evaluated to assess whether they strengthened (effect > 0) or weakened (effect < 0) matches to the 2935 motifs included in our dictionary (see Methods, Equation 3). Parts (A) – (D) compare the median SNP effect between PRE motifs enriched in sequences upstream of psoriasis DEGs (FDR < 0.10) and all other non-enriched motifs (FDR > 0.10). For each motif group, boxes outline the middle 50% of risk allele effects and whiskers outline the middle 80% of effects (324 SNPs in parts A and B; 53 SNPs in parts C and D). P-values assess whether the median risk allele effect differs between motif groups (Wilcoxon rank sum test). Part (E) lists PREs associated with PP-increased DEGs (FDR < 0.10) with the lowest and highest effects (on average among the 324 non-coding SNPs). Part (F) lists PREs associated with PP-decreased DEGs (FDR < 0.10) with the lowest and highest effects (on average among the 324 non-coding SNPs).