Oral Presentation GENEMAPPERS 2026

Genome-wide analysis in over 1.6 million participants uncovers 147 loci associated with obstructive sleep apnoea (#29)

Luis M. Garcia-Marin 1 , Zuriel Ceja 1 , Abishna Parasuraman 1 , Jia Wen Xu 2 , Santiago Díaz-Torres 1 , Victor Flores-Ocampo 1 , Asma M. Aman 3 , Mateo Maya-Martínez 4 , Xueyan Huang 1 , Camilla Pasquali 1 , Aura Aguilar-Roldán 1 , Bade Uckac 1 , Fangyuan Cao 1 , Natalia S. Ogonowski 1 , Nicholas G. Martin 1 , Stuart MacGregor 3 , Xianjun Dong 5 , Sarah Lewis 6 , Mathias Seviiri 3 , Jiao Wang 2 , Miguel E. Renteria 1
  1. Brain & Mental Health Program, QIMR Berghofer, Brisbane, QLD, Australia
  2. School of Public Health, Sun Yat-sen University, Guangzhou, Guangdong, China
  3. Population Health Program, QIMR Berghofer, Brisbane, QLD, Australia
  4. Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, MA, USA
  5. Stephen & Denise Adams Center for Parkinson's Disease Research, Yale School of Medicine, New Haven, CT, USA
  6. Department of Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK

We conducted the largest GWAS meta-analysis for obstructive sleep apnoea (OSA; Ncases= 230,657; Ncontrols= 1,377,442) using individual- and summary-level European ancestry genetic data for OSA from eight international cohorts, including the UK Biobank, All of Us, Million Veteran Program, Mass General Brigham Biobank, FinnGen, the Canadian Longitudinal Study on Aging, the Australian Genetics of Bipolar Disorder Study, and the Australian Genetics of Depression Study. We identified 147 independent loci associated with OSA, of which 141 have not been reported previously, explaining 16% of its phenotypic variance, with the top hit within the FTO gene. We report six independent loci in a separate African population meta-analysis (Ncases= 46,834; Ncontrols= 149,192). Post-GWAS analyses for the European ancestry meta-analysis included fine-mapping, gene-based tests, expression quantitative trait loci (eQTL) mapping, and the integration of spatial transcriptomics data to investigate the putative role of specific brain structures in OSA. We observed spatially resolved gene enrichment involving GABAergic and glutamate pathways, synaptic transmission, and cytoskeletal remodeling in cortical brain tissue. We also evaluated the predictive utility of OSA-derived polygenic risk scores (PRS), demonstrating their predictive ability for clinician‑ascertained OSA status, Fitbit-derived sleep features, and self-reported sleep traits in participants of diverse ancestral backgrounds. Finally, we explored the genetic overlap and investigated putative causal relationships between OSA and a range of physical and neuropsychiatric phenotypes, sleep disturbances, lifestyle factors, and structural brain neuroimaging measures. We identified putative causal genetic effects with ADHD, depression, multisite chronic pain, body mass index, and schizophrenia, among others. Crucially, our findings provide significant new insights into the genetic basis of OSA beyond the well-established effects of BMI. Altogether, our findings demonstrate a robust genetic component underlying OSA risk, independent of body mass index, implicating gene expression patterns in diverse neural cell types.