Funct Integr Genomics. 2026 Jul 25;26(1):203. doi: 10.1007/s10142-026-01979-w.
ABSTRACT
We examined clinical, genetic, and molecular links between sleep disorders and endometriosis. Women aged 20-54 years in NHANES 2005-2006 were analyzed with survey-weighted logistic regression. Bidirectional two-sample Mendelian randomization (MR) was used to assess genetic directionality. Endometriosis-related genes were obtained from the Endometriosis eQTL analysis, whereas sleep disorder-related proteins were obtained from the sleep disorder pQTL analysis and mapped to their encoding genes. The two candidate lists were intersected by gene symbol, and shared candidates were evaluated by pathway enrichment, transcriptomic validation, machine-learning prioritization, single-cell localization, drug-gene annotation, and exploratory docking. Among 1,460 women, clinician-diagnosed sleep disorder was more common in those with endometriosis than in those without endometriosis (weighted prevalence, 13.4% vs 5.1%). In the core adjusted model, endometriosis was associated with higher odds of clinician-diagnosed sleep disorder (OR 2.64, 95% CI 1.03-6.76; P = 0.044). Forward MR suggested a modest association between genetic liability to sleep disorders and endometriosis risk (IVW OR 1.11, 95% CI 1.00-1.22; P = 0.040), whereas reverse MR provided no clear evidence for the opposite direction. Intersecting endometriosis-related eQTL genes with sleep disorder-related pQTL protein-coding genes yielded 37 shared xQTL-prioritized gene/protein candidates. Downstream analyses highlighted glucose metabolism, MAPK signaling, inflammatory pathways, and stromal-cell expression of SPARC and YEATS4. Sleep disorders and endometriosis showed convergent epidemiologic, genetic, and expression-based signals. The core clinical association remained positive after adjustment for major demographic and lifestyle covariates. The genetic and molecular findings should be interpreted as hypothesis-generating pending prospective and experimental validation.
PMID:42498900 | DOI:10.1007/s10142-026-01979-w
