Table 2.
Genome-wide association study single nucleotide polymorphisms that correlate with Δphotoperiod
SNP | Gene(s) a | Disease | Risk allele | Risk allele frequency decreases with Δphotoperiod | Kendall’s correlation P value b | τ rank c | F ST rank d | lnRsb rank e | Gene functional connection to circadian rhythm | ||
---|---|---|---|---|---|---|---|---|---|---|---|
Biaka-Orcadian | Biaka-Maya | CEPH | CHB + JPT | ||||||||
rs1930961 | LRP5L, ADRBK2 | BPD | C | Yes | 1.182 × 10−6 | 0.993 | 0.996 | 0.942 | 0.530 | 0.700 | ADRBK2 (aka GRK3) phosphorylates melanopsin [34] |
rs6746896 | LMAN2L, CNNM4 | BPD | A | Yes | 8.421 × 10−7 | 0.995 | 0.550 | 0.350 | 0.780 | 0.790 | CNNM4 is an RNAi hit (the gene does not appear in Table 1 because the SNP is non-genic) |
rs8099939 | GRIK5 | BPD | T | Yes | 6.538 × 10−5 | 0.956 | 0.968 | 0.615 | 0.960 | 0.980 | - |
rs2286492 | FAM126A | BPD | G | No | 8.521 × 10−5 | 0.966 | 0.811 | 0.000 | 0.160 | 0.260 | - |
rs7570682 | TMEM18, POU3F3 | BPD | A | Yes | 2.320 × 10−4 | 0.950 | 0.455 | 0.004 | 0.160 | 0.090 | - |
rs4075511 | KCNS3, RDH14 | BPD | A | No | 7.345 × 10−9 | 0.999 | 0.797 | 0.156 | 0.720 | 0.730 | - |
rs7319311 | COL4A2 | BPD and SCZ | A | No | 1.083 × 10−4 | 0.952 | 0.474 | 0.319 | 0.400 | 0.470 | Clock-regulated in mouse cardiomyocytes [35] |
rs1124376 | KAT2B | BPD and SCZ | G | Yes | 3.852 × 10−4 | 0.950 | 0.860 | 0.076 | 0.880 | 0.150 | Directly interacts with CLOCK and NPAS2 [36] |
rs1605834 | APOB, KLHL29 | BPD and SCZ | C | Yes | 2.857 × 10−4 | 0.950 | 0.966 | 0.877 | 0.350 | 0.240 | - |
rs17002034 | MKL1 | BPD and SCZ | G | Yes | 2.322 × 10−5 | 0.978 | 0.689 | 0.000 | 0.990 | 0.990 | RNAi hit (gene does not appear among selected hits because of heavier Bonferroni correction) |
rs589249 | CSF3R, GRIK3 | SCZ | G | Yes | 4.565 × 10−5 | 0.964 | 0.996 | 0.624 | 1.000 | 1.000 | Grik3 expression is regulated by Clock in the mouse ventral tegmental area [37] |
rs7004633 | MMP16, RIPK2 | SCZ | A | No | 1.433 × 10−6 | 0.992 | 0.925 | 0.538 | 0.260 | 0.330 | - |
rs1009080 | PTPRU, MATN1 | SCZ | G | Yes | 2.419 × 10−6 | 0.988 | 0.987 | 0.708 | 0.900 | 0.690 | - |
rs1572299 | TLR4, DBC1 | SCZ | A | Yes | 3.901 × 10−4 | 0.951 | 0.950 | 0.321 | 1.000 | 0.200 | DBC1 regulates NR1D1 stability [38] |
rs892055 | RASGRP4 | Asperger | nr | na | 4.268 × 10−5 | 0.964 | 0.603 | 0.585 | 0.610 | 0.420 | - |
rs9601248 | NDFIP2, SPRY2 | MDD | C | No | 7.198 × 10−5 | 0.958 | 0.896 | 0.085 | 0.390 | 0.330 | SPRY2 is a direct target of ARNTL in mouse liver [39] |
rs8020095 | GPHN | Depression | A | Yes | 1.273 × 10−4 | 0.950 | 0.944 | 0.504 | 0.620 | 0.980 | GPHN shows rhythmic expression in mouse SCN [40] |
rs12593813 | MAP2K5 | RLS | A | No | 2.066 × 10−5 | 0.971 | 0.498 | 0.175 | 1.000 | 0.830 | - |
aOne gene is reported if the variant is genic, for intergenic SNPs the two flanking protein-coding genes are listed; bKendall’s P values for the correlation between allele frequency and Δphotoperiod; cpercentile rank in the distribution of Kendall’s correlation coefficients (τ) calculated for minor allele frequency (MAF)-matched SNPs; dFST percentile rank in the distribution of SNPs showing a similar average MAF (calculated over all populations); elnRsb percentile rank; significant values are shown in bold. BPD, bipolar disorder; CEPH, 1000 Genomes Phase I data for Utah Residents with Northern and Western European ancestry; CHB + JPT, 1000 Genomes Phase I data for Han Chinese in Beijing plus Japanese in Tokyo; MDD, major depressive disorder; na, not available; nr, not reported; RLS, restless leg syndrome; SCZ, schizophrenia; SNP, single nucleotide polymorphism.