rs10947690 (MDGA1): Insomnia and Psychiatric GWAS Locus

Key takeaways

• rs10947690 is a variant near MDGA1 that has been linked to insomnia risk by three independent large-scale genetic studies covering a combined total of millions of participants
• The same region also appears in large-scale genetic analyses of depression and tobacco and alcohol use behaviors
• GTEx tissue data show this variant is associated with lower MDGA1 gene activity in blood, skin, esophagus, and muscle
• Insomnia associations held up after statistical adjustment for caffeine use, BMI, lifestyle factors, and psychiatric medication
• Depression genetic data suggest approximately 11,700 variants collectively account for 90% of SNP heritability for that condition, illustrating how small any single variant contribution is

What the research says The MDGA1 locus has been implicated in insomnia across three independent large-scale analyses: a UK Biobank GWAS of 453,379 individuals identifying 57 total risk loci explaining 1% of insomnia variance PMID 31427789, a meta-analysis of 1,331,010 individuals identifying 202 loci collectively explaining 2.6% of variance with enrichment in axonal neuronal gene sets and specific brain cell types PMID 30804565, and a larger meta-analysis of 593,724 cases and 1,771,286 controls identifying 554 loci using expression-QTL and functional genomic prioritization approaches PMID 35538174. The region also falls within the genetic landscapes described by a depression meta-analysis of over 1.3 million individuals (243 risk loci) PMID 36702997 and a multi-ancestry tobacco and alcohol GWAS of 3.4 million individuals (2,143 associated loci, 3,823 independent variants) PMID 37012456.

Reported associations

• Insomnia symptoms (UK Biobank): Among 57 genome-wide significant loci in 453,379 participants (129,270 frequent insomnia cases, 108,357 controls); replicated in the HUNT study (14,923 cases, 47,610 controls) and physician-diagnosed insomnia in Partners Biobank (2,217 cases, 14,240 controls); associations were independent of BMI, lifestyle, caffeine, and psychiatric medication use PMID 31427789
• Insomnia complaints (large GWAS): Among 202 risk loci in 1,331,010 individuals; the full locus set collectively explained 2.6% of insomnia variance; enriched cell types included striatal, hypothalamic, and claustrum neurons; Mendelian randomization supported possible causal effects of insomnia on depression, diabetes, and cardiovascular disease PMID 30804565
• Insomnia (meta-analysis, 554 loci): Among 554 risk loci (364 novel) in 593,724 cases and 1,771,286 controls; a novel gene prioritization strategy narrowed 3,898 candidate genes to 289, emphasizing brain-tissue expression specificity and synaptic signaling gene set enrichment PMID 35538174
• Depression: Implicated within a genetic landscape of 243 risk loci identified in over 1.3 million individuals (371,184 depression cases); 64 loci were novel, including genes encoding glutamate and GABA receptors; enriched cell types included prenatal GABAergic neurons, astrocytes, and oligodendrocyte lineages; the trait is highly polygenic, with approximately 11,700 variants explaining 90% of SNP heritability PMID 36702997
• Tobacco and alcohol use: Implicated within 2,143 associated loci (3,823 independent variants) identified across smoking initiation, cigarettes per day, smoking cessation, and drinks per week in 3.4 million individuals from four major ancestry groups (approximately 21% non-European) PMID 37012456

Evidence quality The insomnia associations are the most extensively replicated for this locus, appearing across three independent studies. Lane et al. (n = 453,379) confirmed signals in two independent replication cohorts and found a genetic correlation between sex-stratified GWASes of rg = 0.807, suggesting partially distinct genetic architecture for insomnia between men and women; Mendelian randomization also supported possible causal links between insomnia and coronary artery disease, depressive symptoms, and subjective well-being PMID 31427789. Jansen et al. found considerable genetic overlap between insomnia and psychiatric traits and modest overlap with sleep duration, and Mendelian randomization supported possible causal effects of insomnia on depression, diabetes, and cardiovascular disease PMID 30804565. Watanabe et al. noted that their novel gene prioritization approach reduces candidates but adds an inference layer beyond standard positional mapping, and the study specifically flags that many more loci likely remain undiscovered given the extreme polygenicity observed PMID 35538174. No variant-specific p-value or odds ratio for rs10947690 alone is provided in the study excerpts. For depression, the highly polygenic architecture and finding that over 95% of risk variants for schizophrenia, bipolar disorder, ADHD, and anxiety also influence depression risk place this locus within a broadly pleiotropic genetic landscape PMID 36702997. The tobacco and alcohol GWAS found that polygenic risk scores trained in European ancestry populations performed poorly when applied to other ancestry groups, underscoring limits to cross-population generalizability PMID 37012456.

Tissue-specific expression effects

• MDGA1: The alternative allele is associated with reduced expression in seven tissues: esophageal mucosa (strongest reduction), unexposed skin, sun-exposed skin, whole blood, esophageal muscularis, cultured fibroblasts, and skeletal muscle GTEx Portal
• CCDC167: The alternative allele is also associated with reduced expression in esophageal mucosa GTEx Portal

Lifestyle considerations No lifestyle considerations on file for this variant.