Figure 1.
6q16.3 (Lead Variant rs57989773) Is an Erectile Dysfunction-Associated Locus and Exhibits Pleiotropic Phenotypic Effects
(A) Genome-wide meta-analysis revealed a single genome-wide significant locus for ED at 6q16.3. Only variants with a p value for association of <0.005 are shown. The red line indicates the genome-wide association significance threshold (set at 5 × 10−8).
(B) Six genome-wide significant variants at 6q16.3 are in high LD.
(C) The association of rs57989773 with ED shows a consistent direction of effect across the three cohorts and across clinically and therapy defined ED in UKBB. Estimates are per C-allele. Boxes represent point estimates of effects. Box sizes are drawn proportional to the precision of the estimates. Lines represent 95% confidence intervals.
(D) PheWAS reveals sex-specific associations of rs57989773 with waist-hip ratio and blood pressure. A PheWAS of 105 predefined traits using the lead ED SNP rs57989773 found associations with 12 phenotypes at p < 4.8 × 10−4 (surpassing the Bonferroni-corrected threshold of 0.05/105; Table S4). All allelic estimates are aligned to the ED risk allele (i.e., C-allele of rs57989773). Due to the nature of the ED phenotype and previously reported sex-specific effects in the MCHR2-SIM1 locus,5 sex-specific analyses were performed in significant traits. Diastolic blood pressure (DBP) and systolic blood pressure (SBP) are included here (despite not meeting the Bonferroni-corrected threshold in the original analysis) due to previous reports of effects on blood pressure in individuals with rare, coding variants in SIM1. Sexual heterogeneity was found to be present (surpassing a Bonferroni-corrected threshold of 0.05/7 for the number of traits where sex-specific analyses were conducted) for DBP (p valueheterogeneity = 6.52 × 10−3), SBP (p valueheterogeneity = 3.73 × 10−3), waist to hip ratio (WHR; p valueheterogeneity = 2.39 × 10−6), and WHR adjusted for BMI (p valueheterogeneity = 1.77 × 10−5). This plot shows sex-specific estimates only for traits showing presence of sexual heterogeneity. Continuous traits were standardized prior to analysis to facilitate comparison. Boxes represent point estimates of effects. Box sizes are drawn proportional to the precision of the estimates. Lines represent 95% confidence intervals.