Table 2. Top associations between single risk alleles and subphenotypes.
PHENOTYPE* | SNP | GENE | Unadjusted p† | FDR‡ p | OR† (95% CI) |
anti-dsDNA antibodies | rs7574865 | STAT4 | 4.40E–05 | 0.00097 | 1.40 (1.19–1.64) |
anti–dsDNA antibodies | rs2187668 | HLA-DR3 | 0.00090 | 0.0099 | 1.37 (1.14–1.65) |
renal disorder | rs2187668 | HLA-DR3 | 0.00060 | 0.013 | 1.37 (1.14–1.64) |
anti-dsDNA antibodies | rs5754217 | UBE2L3 | 0.0033 | 0.020 | 1.31 (1.09–1.56) |
anti-dsDNA antibodies | rs9888739 | ITGAM | 0.0037 | 0.020 | 1.32 (1.09–1.59) |
age at diagnosis<34 years | rs7574865 | STAT4 | 0.0013 | 0.029 | 1.27 (1.01–1.47) |
immunologic disorder | rs4963128 | KIAA1542 | 0.0019 | 0.036 | 0.79 (0.68–0.92) |
immunologic disorder | rs9462015 | UHRF1BP1 | 0.0033 | 0.036 | 1.25 (1.08–1.44) |
arthritis | rs9888739 | ITGAM | 0.0017 | 0.038 | 0.72 (0.59–0.88) |
*anti-dsDNA positivity defined by presence of any positive test versus presence of all negative tests; renal disorder, immunologic disorder, and arthritis defined as in the 1987 ACR criteria [13], [14].
†: Logistic regression adjusted for PC1-PC4, parent study, disease duration, and gender.
‡: 5% False Discovery Rate by Benjamin-Hochberg method [16].