{"paper_id":"a06128a4-eb44-4b21-a09f-5423fe0270e3","body_text":"RESEARCH Open Access\nMiddle East Fertility\nSociety Journal\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal           (2026) 31:37 \nhttps://doi.org/10.1186/s43043-026-00324-2\nA genome-wide association study identifies \nGNPNAT1 as a candidate risk locus \nfor endometriosis in the Tehran Cardio-\nmetabolic genetic study\nLeila Najd-Hassan-Bonab1† , Farzaneh Motafeghi2† , Marzieh Saei Ghare Naz2 , Samaneh Chegeni3 , \nFreidoun Azizi4, Maryam S. Daneshpour1*  and Fahimeh Ramezani Tehrani2,5*\n   †     L e i l a Najd-Hassan-Bonab and Farzaneh Motafeghi contributed \nequally to this work as the first authors.\n*Correspondence:\nMaryam S. Daneshpour\ndaneshpour@sbmu.ac.ir\nFahimeh Ramezani Tehrani\nfah.tehrani@gmail.com; framezan@post.harvard.edu\nFull list of author information is available at the end of the article\n© The Author(s) 2026. Open Access  This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, \nsharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and \nthe source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this \narticle are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included \nin the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will \nneed to obtain permission directly from the copyright holder. To view a copy of this licence, visit  h t t  p : / /  c r e  a t  i v e c o m m o n s . o r g / l i c e n s e s / b y / 4 . 0 /     .    \nAbstract\nBackground Population studies elucidating the genetic architecture of endometriosis have predominantly focused \non individuals of European ancestry, thereby leaving a gap in understanding the genetic influences within non-\nEuropean populations. This present study aims to identify potential genetic variants associated with endometriosis in \nIranian women.\nResearch design and methods We conducted a genome-wide association study on endometriosis involving a \ndiscovery group of 1978 women, comprising 305 women with endometriosis and 1673 unaffected individuals. An \nindependent confirmation cohort comprising 829 women (101 cases and 728 controls) was selected from the TCGS \ncohort. Over 9 million genetic variants were analyzed using the Genome-wide Complex Trait Analysis framework, \nfollowed by integrative bioinformatics and functional annotation analyses.\nResults In the discovery phase, we identified rs12886544 at the GNPNAT1 locus as genome-wide significant for \nendometriosis susceptibility (OR = 1.66, p-value = 5.89 × 10⁻⁸). In the confirmation cohort, the direction of effect was \nconsistent with the discovery phase, although the association did not remain statistically significant after correction \nfor multiple testing (OR = 1.4, 95% [CI 0.89–2.11], p = 0.14). A pooled analysis of the discovery and confirmation \ncohorts demonstrated a statistically significant association between rs12886544 and endometriosis risk (combined \np-value = 0.001, OR = 1.59, 95% CI 1.19–1.89). Epigenomics results suggest that rs12886544 may affect a regulatory \nmotif associated with the Bobby Sox homolog (BBX) transcription factor in women, implicating a potential role in \nmodulating gene expression.\n\nPage 2 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \n  Introduction\nEndometriosis is a chronic, estrogen-dependent inflam -\nmatory disorder characterized by the ectopic presence of \nendometrial-like tissue outside the uterine cavity, the pel-\nvic peritoneum, and the ovaries. Affecting approximately \n5–10% of women of reproductive age and up to 20–50% \nof those with infertility, it presents with significant clini -\ncal heterogeneity. It imposes a substantial burden on \nreproductive health, quality of life, and healthcare systems \nworldwide [ 1, 2]. Although primarily benign, endome -\ntriosis can, in rare cases, progress to become particularly \nsevere within reproductive organs, such as the ovaries.\nDespite its high prevalence, diagnosis remains chal -\nlenging due to nonspecific symptoms and the need for \nsurgical confirmation, often leading to delays of several \nyears [3]. The etiology of endometriosis is multifactorial, \ninvolving complex interactions between genetic predis -\npositions and environmental influences such as age, body \nmass index (BMI), and reproductive history. The genetic \nfactors are estimated to account for 50% of disease sus -\nceptibility. Family and twin studies provide strong evi -\ndence for heritability: first-degree relatives of affected \nwomen have up to a 7-fold increased risk, and monozy -\ngotic twins show significantly higher concordance than \ndizygotic twins, with heritability quantified at approxi -\nmately 51% [ 4– 7]. These findings highlight the central \nrole of gene architecture in disease development.\nEarly genetic studies of endometriosis focused on \ncandidate genes, particularly those involved in estrogen \nsignaling (ESR1, PGR) and inflammation (TNF, IL-1), \nreflecting the hypothesis that hormonal and immune \ndysregulation drive the disease. Later, genetic associa -\ntion shifted to genome-wide association studies (GWAS). \nThe first robust GWAS signal emerged from a Japanese \ncohort (Uno et al., 2010), identifying rs10965235 in \nCDKN2BAS on chromosome 9p21, a genomic region \nimplicated in cell cycle regulation [ 8– 11]. Soon after, \nPainter et al. [ 12] reported a significant association with \nConclusions Our findings provide supportive evidence that rs12886544 at the GNPNAT1 locus is associated with \nendometriosis susceptibility in Iranian women. The association was supported by a consistent direction of effect in an \nindependent cohort and a significant pooled analysis, although further validation in larger independent populations \nis required. Additionally, observed differences in allele frequencies between Iranian and European populations \nhighlight the importance of trans-ethnic studies for understanding population-specific genetic architecture in \nendometriosis.\nKeywords Tehran cardiometabolic genetic study (TCGS), GWAS, Endometriosis, GNPNAT1\nGraphical abstract\n\n\nPage 3 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nthe intergenic SNP rs12700667 on chromosome 7p15.2 \nin women of European ancestry, a locus harboring genes \ninvolved in uterine development. Subsequent studies in \nEuropean populations identified additional loci, includ -\ning HOXA10, which is a key transcriptional regulator of \nuterine development and endometrial receptivity, and \nNFE2L3, a transcription factor linked to cellular dif -\nferentiation and stress-responsive regulation that may \ninfluence endometrial remodeling [12– 15]. Variants near \nVEZT further implicate cell-cell adhesion and epithelial \nintegrity, processes relevant to endometrial implantation-\nlike behavior and lesion establishment, while GREB1, an \nestrogen-responsive gene, supports the central role of \nhormone-driven signaling in endometriosis pathogen -\nesis, with primary effects expected in endometrium and \nectopic pelvic lesions [ 16– 18]. Moreover, associations \ninvolving IL1A and STAT3 genes underscore the con -\ntribution of pro-inflammatory cytokine signaling and \ndownstream transcriptional programs, consistent with an \ninflammatory microenvironment in endometrium, peri -\ntoneal cavity, and lesion-associated immune cells [ 12]. \nAn Icelandic study identified signals near KDR (encod -\ning VEGFR2), TTC39B, and RTN4RL1 genes, pointing \ntoward angiogenesis (KDR/VEGFR2) that may support \nlesion vascularization in ovarian/peritoneal implants, \nalongside loci potentially related to metabolic regulation \n(TTC39B) and neurobiological processes (RTN4RL1), \nwhich may contribute to symptom heterogeneity, includ-\ning pain [ 19]. In contrast, a Polish cohort was unable to \nconfirm 22 previously reported SNPs [20]. These discrep-\nancies underscore the importance of ethnic diversity in \ngenetic studies, indicating that some risk variants may be \npopulation-specific.\nDespite recent advances, the genetic landscape of \nendometriosis in Middle Eastern and Iranian populations \nremains largely unexplored. While large-scale GWAS in \nEuropean and East Asian cohorts have identified numer -\nous risk loci, the generalizability of these findings and the \npotential for population-specific genetic risk factors in \nunderrepresented groups are unclear. This gap hinders \nthe development of precision medicine approaches in \nunderrepresented populations.\nTo address this gap in genetic data from Middle East -\nern populations and to investigate the potential for \nancestry-specific risk factors, we conducted a GWAS in \na large Iranian cohort with the following objectives: (1) \nto discover candiadate and population-specific genetic \nvariants associated with endometriosis; (2) evaluate the \ntransferability of established risk loci from prior large-\nscale studies in our cohort ancestries; and (3) to perform \nfunctional annotation of identified variants to elucidate \ntheir potential roles in disease pathogenesis. Our study \naims to expand the global understanding of endome -\ntriosis genetics and contribute to the development of \ntargeted diagnostics and therapies tailored to the Iranian \npopulation.\nPatients and methods\nStudy population\nWe utilised data from the Tehran Lipid and Glucose \nStudy (TLGS), an ongoing cohort study initiated in 1998 \nto investigate risk factors for non-communicable diseases \namong urban residents [ 21, 22]. Within this framework, \nthe Tehran Cardio-metabolic Genetic Study (TCGS) was \nestablished to identify genetic determinants of key car -\ndiometabolic traits using high-throughput genotyping \nand sequencing technologies [23].\nThe TLGS initially recruited 15,005 participants, all of \nwhom were aged three years or older. Trained personnel \ncollected detailed information regarding demographics, \nreproductive health, lifestyle factors, and medical history \nthrough standardized interviews. Skilled general practi -\ntioners conducted physical and anthropometric measure-\nments at baseline, with follow-ups occurring every three \nyears across seven phases.\nFor this analysis, we included non-menopausal women \naged 18–50 from the 6th follow-up visit, which included \nan expanded reproductive questionnaire [24]. Participants \nwere asked about a prior diagnosis of endometriosis, and \nall self-reported cases were verified through a review of \ntheir medical records. Medical records were also exam -\nined for women reporting moderate to severe chronic pel-\nvic pain, dysmenorrhea, or dyspareunia. An experienced \nsonographer performed a Transabdominal (3.5 MHz) or \ntransvaginal (5 MHz) ultrasound. For women without a \nprior diagnosis, suspected endometriosis was assessed \nusing the standardized protocol of the International Deep \nEndometriosis Analysis (IDEA) group [25].\nA total of 3,450 eligible female participants were ran -\ndomly assigned into two groups: 70% ( n = 2,415) for the \ndiscovery phase and 30% ( n = 1,035) for the confirmation \nphase. The study’s flowchart is presented in Fig. 1.\nDefinition of endometriosis\nIn the current study, participants were designated as \nendometriosis cases if they self-reported a diagnosis that \nwas subsequently confirmed through medical records. \nAdditionally, women with a positive ultrasound diag -\nnosis of endometriosis, regardless of the presence of \nsymptoms such as chronic pelvic pain, dysmenorrhea, or \ndyspareunia, were included in the endometriosis group. \nParticipants exhibiting symptoms like chronic pelvic pain \nor moderate to severe dysmenorrhea or dyspareunia but \nwith negative ultrasound findings were excluded, as a \nnegative ultrasound does not definitively exclude endo -\nmetriosis, particularly superficial peritoneal disease.\nThis classification follows the 2022 ESHRE guide -\nlines, which recommend reserving laparoscopy for \n\nPage 4 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nwith an Identity by Descent (IBD) value of ≥ 18.5%, and \n(4) extreme heterozygosity levels > ± 3SD.\nStatistical analysis\nFor our power calculations, we used the Genetic Asso -\nciation Study (GAS) Power Calculator [ 29]. The follow -\ning parameters were specified based on our available data \nand published estimates: Risk allele frequency: approxi -\nmately 0.49; Genotype relative risk: 1.5; Endometriosis \nprevalence: 20% and Genome-wide significance thresh -\nold of 5 × 10 − 8  for the discovery stage and 0.05 for the \nconfirmation stage.Under these assumptions and our \navailable cases (305 cases in discovery and 101 cases in \nconfirmation), the estimated power was 96% and 92% for \nthe discovery and confirmation phases, respectively.\nThe normality of continuous variables was assessed \nusing the Kolmogorov–Smirnov test. Variables follow -\ning a normal distribution are expressed as mean ± stan-\ndard deviation (SD) and compared between groups using \nStudent’s t-test. Categorical variables are presented as \npercentages and compared with the chi-squared (χ²) \ntest. The statistical method was conducted separately \nfor the discovery and confirmation cohorts. In the dis -\ncovery phase, genome-wide association testing was \nperformed for autosomal variants using the FastGWA \nmethod implemented in Genome-Wide Complex Trait \nAnalysis (GCTA; version 1.93 beta). FastGWA applies a \nmixed linear model under an additive genetic framework \nsymptomatic patients with negative imaging or ineffec -\ntive treatment and advice against routine laparoscopic \nscreening in asymptomatic individuals with negative \nimaging [26]. As a result, the final analysis included 406 \nwomen diagnosed with endometriosis and 2401 control \nparticipants.\nGenotyping quality control\nGenotyping for TCGS participants was performed using \nthe Illumina Human OmniExpress-24-v1-0 bead chip, \nwhich comprises 652,919 SNP loci, at deCODE genet -\nics/Amgen in Iceland, following the manufacturer’s \nprotocols (Illumina Inc., San Diego, CA, USA) [ 23]. A \nsubset of 1,462 samples was selected for whole-genome \nsequencing. Variant imputation was performed using the \nIMPUTE hidden Markov model (HMM), where chip-\ngenotyped individuals were imputed based on shared \nhaplotypes with the training set. Before association anal -\nysis, quality control procedures were applied to ensure \nthe reliability of genotype data at both the individual and \nSNP levels [27, 28]. Genetic variants were excluded based \non the following criteria: (1) genotyping rate below 95%, \n(2) minor allele frequency (MAF) less than 0.05, and (3) \ndeviation from Hardy-Weinberg Equilibrium ( p-value \n< 1 × 10 − 6 ). Participants were excluded for the follow -\ning reasons: (1) sex discrepancies identified via PLINK’s \n“sex check, ” (2) missing data exceeding 5%, (3) relatedness \nFig. 1 Overview of study design and analysis workflow used in the current GWAS on endometriosis. N indicates the sample size\n \n\nPage 5 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nand accounts for relatedness and population structure \nthrough a genetic relationship matrix (GRM). Case–con -\ntrol status for endometriosis was modeled as the out -\ncome variable, and all tests were two-sided.\nAnalyses were adjusted for age at menarche, body mass \nindex (BMI), smoking, with and without dysmenorrhea, \nand 10 principal components (PCs) capturing popula -\ntion structure. Although the Iranian population is rela -\ntively homogeneous, subtle genetic substructure arising \nfrom regional or ethnic diversity (Persian, Turkic, Kurd -\nish, and Baloch) may exist. To address this, PCs were \nderived from an LD-pruned subset of autosomal SNPs \nusing PLINK, and the genetic relationship matrix (GRM) \nwas estimated with GCTA. Genomic inflation was \nevaluated using quantile–quantile (Q–Q) plots and the \ngenomic control factor (λ). Manhattan and Q–Q plots \nwere generated using the qqman package in R (version \n4.0.2). Genome-wide statistical significance was defined \nas P-values < 5 × 10⁻⁸ to account for multiple testing. \nRegional association plots for significant loci were gener -\nated using LocusZoom.\nIn the confirmation phase, association testing was per -\nformed using logistic regression under an additive genetic \nmodel implemented in PLINK. Resampling permutation \ntesting with 100,000 iterations was applied to evaluate \nthe robustness of association signals. Odds ratios (ORs) \nand corresponding two-sided P-values were reported, \nwith statistical significance defined as P-values < 0.05.\nFunctional annotation of GWAS results\nFunctional interpretation of GWAS results was carried \nout using HaploReg, which helps identify tagging SNPs \nin high linkage disequilibrium and provides insights into \nchromatin states, conserved regions, and transcription \nfactor motif alterations [30]. To further contextualize the \nfindings, we integrated epigenomic mapping data from \nENCODE and the Roadmap Epigenomics Mapping Con -\nsortium, enabling interpretation at both the variant and \ngene levels.\nWe also leveraged additional resources, including \nGWAS Atlas, FinnGen, PheWeb, and Open Targets, to \nexplore phenotypic annotations and previously reported \ntrait associations [ 31– 34]. These open-access databases \nfacilitate the identification and prioritization of poten -\ntially causal variants and genes, providing a comprehen -\nsive functional understanding of GWAS results [35].\nResults\nStudy population characteristics\nA total of 2,807 women were included in the analysis, \ncomprising 406 women with endometriosis and 2,401 \ncontrols. They were randomly assigned to two phases: \nthe discovery phase (305 cases and 1,673 controls) and \nthe confirmation phase (101 cases and 728 controls). \nBaseline characteristics stratified by endometriosis status \nand study phase are presented in Table  1. As expected, \nwomen with endometriosis reported higher rates of dys -\nmenorrhea and chronic pelvic pain compared to controls.\nDiscovery phase\nAfter quality control, approximately 9  million imputed \nvariants were analyzed using FastGWA, adjusting for \nage at menarche, BMI, smoking status, and 10 principal \ncomponents in the primary GWAS model. We then rein-\ntroduced dysmenorrhea as an additional covariate in a \nsecondary analysis to assess the stability of our findings. \nThe genomic inflation factor (λ = 1.08) indicated minimal \ntest-statistic inflation, confirming adequate control for \npopulation stratification, suggesting the absence of false-\npositive association signals (Fig. 2A).\nTable 1 Baseline characteristics of the study population stratified by study phase and endometriosis status\nCharacteristic Discovery Phase Confirmation Phase\nEndometriosis\n(n = 305)\nNon-Endometriosis\n(n = 1673)\nP-value Endometriosis\n(n = 101)\nNon-Endometriosis\n(n = 728)\nP-value\nAge (years) 34.0 ± 12 34.8 ± 12 0.21 32.1 ± 12.5 32.3 ± 11.7 0.11\nAge at menarche (years) 13 ± 1.5 13 ± 1.50 0.046 13 ± 1.2 13 ± 1.3 0.35\nEver Smoke (n%) 11(3.7%) 56 (4.4%) 0.0001 4 (4.0%) 26 (3.6%) 0.0001\nNumber of parities 2.2 ± 1.0 2.3 ± 1.2 0.11 2.3 ± 1.1 2.2 ± 1.1 0.11\nNumber of gravidities 2.6 ± 1.2 2.7 ± 1.4 0.13 2.7 ± 1.3 2.6 ± 1.3 0.12\nNumber of abortions 1.4 ± 0.6 1.4 ± 0.7 0.29 1.3 ± 0.5 1.3 ± 0.7 0.23\nSBP (mm Hg), mean SD 126 ± 20.2 120 ± 20.5 0.04 109 ± 12.3 111 ± 13.5 0.02\nDBP (mm Hg), mean SD 80 ± 10.8 78 ± 11.02 0.07 75 ± 12.3 75 ± 9.2 0.1\nBMI (kg/m2) 27 ± 5.1 27 ± 5.03 0.53 27 ± 4.6 27 ± 5.2 0.51\nWaist Circumferences(cm) 93 ± 12.3 89 ± 12.3 0.02 83 ± 10.9 85 ± 11.08 0.2\nHip Circumferences (cm) 105 ± 9.9 104 ± 9.5 0.2 104 ± 8.9 103 ± 8.8 0.15\nWrist Circumferences(cm) 16.45 ± 1.09 16.1 ± 1.01 0.32 15.21 ± 0.9 15.86 ± 0.9 0.23\nData are presented as mean ± standard deviation, except for smoking status, which is presented as number (%)\nBMI body mass index, SBP Systolic Blood Pressure, DBP Diastolic Blood Pressure\nP-values were calculated using the independent samples t-test for continuous variables and the chi-square test for categorical variables\n\nPage 6 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nWe identified one genome-wide significant locus at \nchromosome 14q22.1 (rs12886544; OR = 1.66, 95% CI: \n1.35–2.04; p-value = 5.89 × 10⁻⁸) as shown in Manhattan \nplot in Fig.  2(B). This intronic variant is located within \nGNPNAT1 (Glucosamine-Phosphate N-Acetyltrans -\nferase 1), a gene encoding a key enzyme in the hexos -\namine biosynthesis pathway. Notably, the risk allele (C) \nshows marked population differentiation; minor allele \nfrequency (MAF) = 0.495 in the TCGS cohort versus \nMAF = 0.001 in European populations, suggesting this \nlocus may contribute to population-specific endometrio -\nsis susceptibility. Regional association plotting confirmed \na distinct peak centered on GNPNAT1 without evidence \nof multiple independent signals (Fig. 3). The results of \nthis sensitivity analysis demonstrated high consistency \nwith our original findings (primary GWAS model). The \nkey genome-wide significant association, represented by \nrs12886544 (OR = 1.66; p = 4.99 × 10⁻⁸), remained robust \nFig. 3 Regional association (LocusZoom) plot of the GNPNAT1 locus on chromosome 14. SNPs are plotted according to genomic position (x-axis) and \n−log10 ( P) value (y-axis), reflecting association results from the genome-wide association analysis. The lead SNP is indicated, and surrounding variants are \nshown based on their local linkage disequilibrium structure\n \nFig. 2 A Quantile–quantile (Q–Q) plot of observed versus expected −log10 ( P) values from the genome-wide association analysis of endometriosis. \nThe genomic inflation factor (λ) was 1.08. B Manhattan plot showing genome-wide association results for endometriosis. Each point represents a single \nSNP plotted according to its chromosomal position (x-axis) and −log10 ( P) value (y-axis). The horizontal red line denotes the genome-wide significance \nthreshold (P ≤ 5 × 10⁻⁸)\n \n\nPage 7 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nafter additional adjustment for dysmenorrhea and other \ncovariates. Furthermore, we observed a modest increase \nin the number of suggestive loci, from 13 SNPs in the dys-\nmenorrhea unadjusted model (Supplementary Table 1 A) \nto 17 SNPs in the dysmenorrhea-adjusted model (Sup -\nplementary Table 1B). These results collectively indicate \nthat the inclusion or exclusion of dysmenorrhea does \nnot materially alter the detected genetic associations, \nthereby reinforcing the robustness and interpretabil -\nity of our findings. To our knowledge, this locus has not \nbeen reported in prior endometriosis GWAS analyses. \nThree independent signals mapped to the MICAL3 locus \n(22q12.1), suggesting a potential regulatory hotspot.\nTo assess whether established genetic risk factors for \nendometriosis are shared in the Iranian population, we \nexamined previously reported loci and found that six \nvariants (near WNT4, FN1, and VEZT genes) reached \nnominal significance, each demonstrating effect direc -\ntions consistent with those reported in large prior \nmeta-analyses(Supplementary Table S2).\nConfirmation phase\nThe top SNP (rs12886544) identified in the discovery \nphase was further evaluated in an independent confirma -\ntion cohort consisting of 101 women with endometriosis \nand 728 controls. Logistic regression analyses were per -\nformed in PLINK, adjusting for the covariates used in \nthe discovery phase. Following correction for multiple \ntesting using the false discovery rate (FDR), none of the \ntested variants reached statistical significance in the con -\nfirmation phase.\nFor the lead variant rs12886544, the direction of effect \nsize in the confirmation phase was consistent with that \nobserved in the discovery phase. However, the asso -\nciation did not remain statistically significant after mul -\ntiple testing correction (OR = 1.4, 95% [CI 0.89–2.11], \np-value = 0.14). A pooled analysis combining the dis -\ncovery and confirmation phase was subsequently con -\nducted using logistic regression in PLINK, applying the \nsame covariates and analytical framework as used in \nthe confirmation phase. This combined analysis dem -\nonstrated a statistically significant association between \nrs12886544 and endometriosis susceptibility (combined \np-value = 0.001, OR = 1.59, 95% CI 1.19–1.89; Table  2). \nAlthough the association of rs12886544 did not reach \nstatistical significance after multiple testing correction in \nthe confirmation cohort, the consistent direction of effect \nand the statistically significant pooled analysis provide \nsupportive evidence for its role in endometriosis suscep -\ntibility, warranting further validation in larger indepen -\ndent populations.\nFunctional annotation analysis results of endometriosis-\nassociated SNPs\nTo characterize the potential biological function of \nrs12886544, we performed integrative functional annota -\ntion of rs12886544 using HaploReg, PheWeb, Open Tar -\ngets, ENCODE, Roadmap Epigenomics, GWAS Atlas, \nand FinnGen (Supplementary Table S3-S6). No prior \nGWAS has linked this variant to endometriosis. It is not \nin significant linkage disequilibrium (r² > 0.1) with known \ncoding variants in the 1000 Genomes Project, suggesting \nan independent signal.\nVariant-level annotation\nAt the variant level, rs12886544 is predicted to reside \nwithin genomic regions annotated with histone modifica-\ntions characteristic of enhancer and promoter elements, \naccording to HaploReg. The variant also overlaps DNase \nI hypersensitivity sites, indicative of open chromatin \nacross multiple cell types and tissues, notably in adipose \nand ovarian tissues. These tissues are critically involved \nin the pathophysiology of endometriosis, highlighting the \npotential functional relevance of rs12886544 in disease \nmechanisms through epigenetic modulation [36].\nAnnotations from the Roadmap Epigenomics and \nENCODE projects further indicate that rs12886544 is a \nfunctionally significant variant within the reproductive \nsystem. This SNP is predicted to modify a regulatory \nmotif and to interact with the transcription factor Bobby \nSox homolog (BBX). Subsequent analysis has revealed \nepigenetic signals associated with rs12886544, spe -\ncifically the trimethylation of histone H3 at lysine 4 \n(H3K4me3) in reproductive tissues, including the ovary, \nplacenta, and cervix. These findings underscore the pre -\ndicted regulatory role of the reproductive system, provid-\ning a basis for hypothesis generation regarding potential \nmechanisms in endometriosis.\nGene-level annotation\nTo further investigate at the gene level, associations by \nusing the PheWeb identified the GNPNAT1 gene as \nbeing correlated with non-inflammatory disorders of \nTable 2 Result for the top SNP in the pooled (combined) GWAS and confirmation study\nChr: Position SNP Risk allele Nearest Gene Consequence Phase Control/case P-value OR\nchr14:52837402 rs12886544 T GNPNAT1 intron variant GWAS 1673/305 5.89E-08 1.66\nConfirmation 728/101 0.143 1.39\nCombined 2401/406 0.001 1.59\nChr chromosome, OR odds ratio\n\nPage 8 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nthe cervix, conditions related to the amniotic cavity and \nmembranes, and Leukemia in UK Biobank data.\nMoreover, experimental data from the Open Target \nplatform revealed that mouse knockout studies impli -\ncate the GNPNAT1 gene as essential for multiple devel -\nopmental phenotypes, including the absence of amnion, \nallantois, and chorion, reduced cellular proliferation, \nabnormal morphology of extraembryonic tissue, and \nembryonic lethality occurring between implantation and \nsomite formation, with complete penetrance.\nResults derived from the GWAS Atlas further impli -\ncated GNPNAT1 in reproductive traits, such as age \nat menarche, menstrual quality of life, and gestational \nweight gain, as well as metabolic traits including waist-\nhip ratio, BMI, and fat mass distribution in the legs and \narms.\nConsistent with these findings, data from the Finn -\nGen cohort corroborated associations between GNP -\nNAT1 and diverse gynecological conditions, including \nadenomyosis (endometriosis of the uterus), deep endo -\nmetriosis, endometriosis involving the fallopian tube, \nrectovaginal septum, and vagina, in addition to female \ninfertility, pain syndromes, and other disorders related to \nthe female genital organs and menstrual cycle (Supple -\nmentary Table S6). So, gene-level evidence indicates that \nGNPNAT1 is associated with reproductive and metabolic \ntraits, highlighting the potential involvement of GNP -\nNAT1 in reproductive and related traits.\nPathway for the GNPNAT1 Gene\nWe utilized the Kyoto Encyclopedia of Genes and \nGenomes (KEGG) and Reactome databases to identify \nkey biochemical pathways associated with endometriosis. \nThe result revealed that the GNPNAT1 gene is involved \nin the biosynthesis of UDP-N-acetylglucosamine, a vital \ndonor molecule in the initial two steps of the N-glycan \nprecursor biosynthesis pathway. Subsequently, UDP-\nN-acetylglucosamine serves as a substrate for further \nenzymatic modifications following the attachment of the \nglycan precursor to the target protein. In addition, at the \nepigenetic level, it influences transcription factor binding \nactivity, notably that of the Bobby Sox homolog (BBX), \nwhich is critically involved in developmental processes \nand the regulation of Wnt signaling pathways. The signif-\nicantly enriched pathways identified in this analysis were \nsubsequently compared with those previously inves -\ntigated in the literature through candidate gene asso -\nciation studies to evaluate consistency [ 37]. These gene \nregulatory pathway links are predicted from functional \ngenomic and association databases. Further mechanistic \nstudies can clarify the functional roles of rs12886544 and \nGNPNAT1 in endometriosis pathogenesis.\nDiscussion\nThis study presents the first genome-wide association \nscan for endometriosis in an Iranian population, provid -\ning critical insights into the disease’s genetic architec -\nture within this underrepresented ancestry. Our analysis \nidentified a candidate susceptibility locus rs12886544, \nan intronic variant within the GNPNAT1 gene on \nchromosome 14, that demonstrates a significant asso -\nciation with increased endometriosis risk(OR = 1.66, \np-value = 5.89 × 10⁻⁸). Beyond being the first GWAS of \nendometriosis in an Iranian population, our study con -\ntributes by evaluating genetic risk in an understudied \nancestry, where differences in allele frequencies and LD \nstructure may reveal signals not captured in predomi -\nnantly European and East Asian cohorts. This finding is \nfurther supported by the consistency observed across our \nindependent discovery and confirmation cohorts (com -\nbined p-value = 0.001, OR = 1.59).\nTo contextualize our findings, we assessed previously \nreported loci and observed that six variants (near WNT4, \nFN1, and VEZT genes) showed nominal significance with \neffect directions consistent with prior large meta-analy -\nses, providing supportive internal validation and indicat -\ning that aspects of endometriosis genetic architecture \nare shared across.We also highlight the emerging role \nof GNPNAT1 in reproductive biology, while noting that \nmolecular mechanisms remain to be clarified. Impor -\ntantly, observed differences in allele frequencies between \nIranian and European populations, along with variability \nin sample sizes across existing GWASs, likely contribute \nto the heterogeneous results reported in the literature, \nhighlighting the value of studying diverse ancestries [38].\nAlthough the functional impact of rs12886544 on \nendometriosis pathogenesis has not been experimentally \nvalidated, genomic annotation reinforces the potential \ninvolvement of this non-coding variant near the GNP -\nNAT1 gene in reproductive disorders. UK Biobank data \ndemonstrate significant associations between the GNP -\nNAT1 gene and a spectrum of genital tract conditions, \nsupporting the gene’s involvement in relevant patho -\nphysiology. GNPNAT1 encodes a critical enzyme in the \nhexosamine biosynthesis pathway (HBP), responsible \nfor producing UDP-GlcNAc, a central metabolite inte -\ngrating glucose, amino acid, nucleotide, and fatty acid \nmetabolism. UDP-GlcNAc serves as an essential sub -\nstrate for both N-linked and O-linked glycosylation pro -\ncesses, which are pivotal for protein folding, stability, and \nfunction. Proper N-glycosylation in the Golgi apparatus \nensures the correct maturation of membrane and secre -\ntory proteins, while defects in this pathway lead to pro -\ntein misfolding, endoplasmic reticulum stress, altered \ncell surface proteomes, and subsequent disruptions in \ncellular signaling and fate determination [ 39– 41]. Based \non genomic and regulatory annotations of rs12886544, \n\nPage 9 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nsuggesting a potential role in gene regulation, we hypoth-\nesize that this variant may influence endometrial cell \nbehavior [ 42] by modulating GNPNAT1-dependent \nprocesses such as cell adhesion, migration, and signal \ntransduction. We emphasize that although GNPNAT1’s \nbiological role in glycosylation pathways is well estab -\nlished, direct experimental evidence linking rs12886544 \nto altered GNPNAT1 expression or function in endome -\ntrial tissue has yet to be demonstrated.\nFurthermore, O-GlcNAcylation, a dynamic post-\ntranslational modification mediated by HBP metabo -\nlites, is frequently upregulated in tumor cells, promoting \nmechanisms like proliferation. Inhibiting the HBP has \ninduced growth arrest and apoptosis in tumor models. \nConsistently, in vitro studies demonstrate that GNP -\nNAT1 knockdown reduces proliferation and invasive -\nness in breast cancer cell lines, supporting a functional \nrole for this gene in cell growth and migration [ 43, 44]. \nCollectively, these data highlight GNPNAT1 as a biologi -\ncally plausible candidate locus influencing endometriosis \nsusceptibility through its involvement in critical glycosyl-\nation-mediated cellular processes [39, 43].\nRecent studies have linked UDP-GlcNAc to the regu -\nlation of Wnt signaling. Neitzel et al. demonstrated that \nUDP-GlcNAc modulates Wnt pathway activity via protein \nglycosylation, a modification critical for proper receptor \nfunction and the downstream signaling cascade [ 45]. In \naddition to its role in Wnt signaling, UDP-GlcNAc influ -\nences hyaluronan-CD44 interactions and immune modu-\nlation, processes regulated in part by Leukemia inhibitory \nfactor (LIF), a cytokine essential for endometrial receptiv-\nity and embryo implantation. Disruption of LIF-mediated \nsignaling, as observed in infertile women, impairs down -\nstream effectors, including STAT3, MAPK, and Protein \nkinase C (PKC)pathways, which are integral to blastocyst \nadhesion [ 46– 48]. Notably, elevated STAT3 expression \nhas been reported in both endometriosis and endometrial \ncancer, implicating STAT3 as a potential risk factor for \nthese disorders [49, 50].\nGlycosylation plays a pivotal role in endometrial func -\ntion, with the FERM Domain Containing Kindlin 2 \n(FERMT2) gene, located adjacent to GNPNAT1, regu -\nlating the glycosylation of highly glycosylated proteins, \nsuch as MUC1, which are essential for integrin-mediated \nadhesion and implantation. In parallel, the O-GlcNAc \nmodification supports endometrial cell proliferation and \ninvasion during the secretory phase, underscoring the \ncontribution of glycosylation-dependent pathways to \nendometrial receptivity and fertility [ 51]. The FERMT2, \na member of the FERMT adaptor protein family, is \nhighly expressed in reproductive tissues, particularly the \nendometrium and uterus, and has been shown to facili -\ntate trophoblast adhesion and invasion throughout ges -\ntation [ 52]. Notably, gene-based association analyses \nreveal that FERMT2 and GNPNAT1 exhibit comparable \nsignificance levels and a strong correlation, suggesting \na potential co-regulatory relationship that influences \nglycosylation-mediated adhesion and implantation \nmechanisms processes that may also contribute to endo -\nmetriosis susceptibility [53].\nOur findings are consistent with prior studies. Based on \nPheWeb data, several of our suggestive variants showed \nassociations with reproductive and metabolic traits. The \nvariants rs8143037 and rs401910 (intron variants of the \nMICAL3 gene) were associated with irregular menstrual \nbleeding, anatomical abnormalities of the ureters, and \ndysmenorrhea. The variants rs706042 (intron variant of \nthe CHIC1 gene) and rs6593654 (nearest gene: RWDD3) \nwere linked to disorders of the cervix, prolapse of vagi -\nnal walls and vaginal vault after hysterectomy, premeno -\npausal menorrhagia, and ovarian cysts. The rs12981001 \nvariant (nearest gene: IL4I1) was associated with endo -\nmetrial hyperplasia, disorders of menstruation, abnormal \nbleeding from the female genital tract, irregular men -\nstrual cycle, and absent or infrequent menstruation. The \nrs396717 variant (intron variant of the MICAL3 gene) \nwas also associated with endometrial hyperplasia. The \nrs17058669 variant (nearest gene: ZNF516) was associ -\nated with menopause and placenta previa. The variants \nrs111741344 (intron variant of the B3GALT1 gene) and \nrs6593654 (nearest gene: RWDD3) were associated with \novarian cysts. Additionally, rs35913552 (nearest gene: \nAPOB) and rs13011615 (nearest gene: NCKAP1L) were \nassociated with ectopic pregnancy. Notably, variants \nrs706042 (CHIC1), rs35913552 (APOB), and rs13011615 \n(NCKAP1L) were also associated with type 2 diabetes, \ndisturbances in lipid metabolism, and hypercholester -\nolemia, conditions that may be more prevalent among \nwomen with endometriosis. The biological functions of \nthese genes, including roles in actin cytoskeleton regula -\ntion (MICAL3), immune modulation (IL4I1, NCKAP1L), \nlipid metabolism (APOB), and protein glycosylation \n(B3GALT1), may provide mechanistic insights into the \npathophysiology of endometriosis [54].\nAlthough evidence of epigenetic contributors to endo -\nmetriosis has historically been limited, recent in vitro \nand in vivo studies have increasingly elucidated a signifi -\ncant epigenetic role in the pathogenesis of this condition. \nComplementary data from the HaploReg database pro -\nvide evidence supporting the involvement of rs12886544 \nin epigenetic regulation, specifically through histone \nmodification marked by H3K4me1 in ovarian tissue. The \nrisk allele of this variant may influence transcription fac -\ntor binding activity, notably that of BBX. BBX harbors a \nhigh-mobility group (HMG) box domain, which is criti -\ncally involved in developmental processes and the regu -\nlation of Wnt signaling pathways [ 55]. Interestingly, the \nfindings indicate that Wnt signaling is involved in all \n\nPage 10 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nforms of endometriosis, as demonstrated by evidence \nfrom both single-SNP analyses and pathway-based asso -\nciation studies of Wnt genes [12– 57].\nA recent survey by Marquardt et al. demonstrated \nthat epigenetic dysregulation, particularly of histone \nmodifications such as H3K4me1, plays a crucial role in \naltering gene expression, contributing to ectopic endo -\nmetrial growth and infertility. H3K4me1 is differen -\ntially enriched in endometriotic stromal cells, regulating \ngenes involved in proliferation, invasion, and immune \nresponse [ 58]. Dysregulated H3K4me1 at this enhancer \nmay impact pathways such as the Wnt/β-catenin or TGF-\nβ pathways, which are known to drive the formation of \nendometriotic lesions and promote inflammation, as \npreviously explained. Additionally, experimental data \nfrom Timofeeva and colleagues suggest that the absence \nof specific proteins and their corresponding RNAs in the \noocyte affects the development of the preimplantation \nembryo. These key proteins include transcription fac -\ntors such as BBX and zinc finger protein 646, along with \nhistone modifier Ankyrin repeat domain 12 (ANKRD12) \n[59]. Collectively, these findings underscore the critical \nimportance of the epigenetic signature associated with \nthis variant, suggesting it may have a substantial influ -\nence on the molecular mechanisms underpinning the \npathophysiology of endometriosis. This warrants further \ninvestigation into its functional impact on the regulation \nof endometrial genes. The observed epigenetic modifica -\ntions suggest potential effects on chromatin accessibility, \ntranscription factor recruitment, and gene expression \nregulation within endometrial tissue [45].\nImportantly, we observed notable differences in allele \nfrequencies between our Iranian cohort and European-\nancestry reference populations (Ensembl), suggesting \npotential heterogeneity in genetic architecture across \nancestries. In addition to variability in sample sizes across \nprior GWAS (ranging from 171 to 17,045 participants) \n[60], differences in ethnic and genetic backgrounds likely \ncontribute to the inconsistent results reported in the lit -\nerature [61]. Therefore, our identification of rs12886544 \nas a candidate genetic locus associated with endome -\ntriosis may be partly attributable to these underlying \npopulation-specific genetic variations and the diverse \ndemographic characteristics of the study cohorts. For \nexample, SNPs such as rs706042, rs12981001, rs251151, \nrs13011615, rs111741344, and rs35913552 exhibit mark -\nedly divergent effect allele frequencies, with higher fre -\nquencies observed in European populations (0.816, 0.792, \n0.75, 0.83, 0.82, and 0.835) compared to Iranian popula -\ntions (0.14, 0.21, 0.15, 0.19, 0.23, and 0.19), respectively.\nIn contrast, the allele frequencies of most endometrio -\nsis-associated variants were broadly comparable between \nIranian and European populations, with the exception of \nrs868941587 and rs17058669 (Supplementary Table 1). \nThe markedly low frequency of these two variants in our \nIranian cohort may contribute to the significant associa -\ntion observed exclusively among Iranian women, poten -\ntially explaining why these signals remained undetected \nin previous Euro-centric studies where their effects may \nhave been masked.This pattern aligns with prior research \ndemonstrating that ancestry-specific analyses can reveal \ndistinct genetic contributors to endometriosis. For exam-\nple, studies focused solely on individuals of East Asian \ndescent have identified several variants associated with \nendometriosis that were not detected in predominantly \nEuropean cohorts, including five SNPs reported by two \nindependent studies [ 9, 62] and 14 SNPs identified by \nWang et al. [ 63]. In contrast, studies incorporating more \nethnically diverse populations, such as those by Sapkota \net al., [ 64] and Sobalska-Kwapis et al., have reported \nbroader but non-overlapping sets of associated SNPs, \nfurther highlighting substantial cross-population het -\nerogeneity. Collectively, these observations underscore \nthe importance of considering population genetic struc -\nture and ancestral background when interpreting genetic \nassociation results.\nThe principal strengths of the present study are embed-\nded within its robust methodological framework, which \nis distinguished by a large, population-based design. \nThis approach enhances the generalizability and statis -\ntical power of the findings by minimizing selection bias \nand ensuring a representative sample. Furthermore, \nthe comprehensive nature of the study design enables a \nmore accurate assessment of genetic associations within \ndiverse demographic groups, thereby enhancing the reli -\nability and validity of the results.\nIt is essential to acknowledge the limitations of our \nstudy. The primary limitation is the absence of a com -\nprehensive evaluation of all endometriosis cases via \nlaparoscopy, which remains the definitive diagnostic \nstandard for this condition. Additionally, our dataset \nlacks sufficient reliability for the precise identification \nof endometriosis severity. Moreover, individuals with \nmoderate to severe dysmenorrhea but without a con -\nfirmed diagnosis of endometriosis were excluded if their \nultrasound imaging did not indicate the presence of \nendometriosis. However, because a negative ultrasound \nresult cannot definitively exclude endometriosis, espe -\ncially in cases of superficial peritoneal involvement, our \nresults may not entirely reflect the population with less \nsevere manifestations of the condition. Furthermore, \nour study was conducted with a relatively modest sam -\nple size compared with contemporary GWAS, which \nmay limit precision in effect estimation. Power calcula -\ntions performed based on assumptions of a risk allele \nfrequency of 0.49, genotype relative risk of 1.5, disease \nprevalence of 20%, and standard genome-wide signifi -\ncance thresholds, indicated adequate statistical power. \n\nPage 11 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \nHowever, these estimates reflect optimistic assumptions \nand should be interpreted accordingly. We also recog -\nnize the potential impact of the winner’s curse, whereby \neffect sizes observed in initial discovery analyses may be \ninflated due to statistical selection bias [ 65]. Although \nthis bias affects studies across populations, the identi -\nfication of this variant specifically in our cohort raises \nthe possibility of ancestry-related heterogeneity or a \ngenuinely larger effect in this population. Nonetheless, \nindependent replication in a larger cohort of the same \nancestry will be essential to confirm this association and \nobtain an unbiased estimate of effect size.\nAnother key limitation of this study is the absence of \nan independent external cohort for replication. Because \nlarge-scale genomic datasets from Middle Eastern or Ira -\nnian populations are currently scarce, we were unable to \nvalidate our lead association in an ancestrally matched \ncohort. Establishing this locus will require replication \nin independent datasets, ideally including samples from \nancestrally relevant populations or large multi-ancestry \nbiobanks such as All of Us or similar resources. Such \nexternal validation is essential to confirm the robustness, \ngeneralizability, and true biological significance of the \ncandidate locus.\nOur research concentrated exclusively on genetic vari -\nants situated on the autosomal chromosomes, thereby \nomitting analysis of the X chromosome. Subsequent \nstudies incorporating sex chromosome data could yield \nfurther valuable insights. Furthermore, our analysis was \nlimited to single-variant association testing and did not \nincorporate haplotype-based analyses or cross-popula -\ntion comparisons of linkage disequilibrium (LD) struc -\nture surrounding the GNPNAT1 locus. Additionally, LD \nScore Regression (LDSC) could not be performed due \nto the absence of publicly available Iranian LD reference \npanels; applying European LD reference panels to our \ncohort would introduce bias, given documented popula -\ntion-specific differences in LD architecture.\nConclusion\nIn conclusion, our GWAS identification of rs12886544 is \npresented as a candidate population-specific locus that \nmay arise from these ancestry differences. In sum, our \nstudy adds (i) the largest Iranian GWAS dataset to date, \n(ii) evidence of concordance with established loci (mod -\nest internal validation), and (iii) population-specific \nallele-frequency that can inform future trans-ancestry \nmeta-analyses and functional follow-up. Independent \nreplication cohorts and inclusion in meta-analyses will \nbe required validate this association and elucidate its \nclinical relevance for risk stratification and therapeutic \ndevelopment.\nSupplementary Information\nThe online version contains supplementary material available at  h t t  p s : /  / d o  i .  o r \ng / 1 0 . 1 1 8 6 / s 4 3 0 4 3 - 0 2 6 - 0 0 3 2 4 - 2     .  \nSupplementary Material 1: Supplementary Tables.\nAuthors’ contributions\nLeila Najd-Hassan-Bonab contributed to conceptualization, methodology, \nsoftware, and writing the original draft. Farzaneh Motafeghi contributed \nto data curation and investigation. Samaneh Chegeni contributed to data \ncuration, review, and editing. Fereidoon Azizi contributed to the review and \nediting. Maryam S. Daneshpour contributed to methodology, validation, \nreview, and editing. Fahimeh Ramezani Tehrani contributed to methodology, \nvalidation, and review editing. All authors saw and approved the final version, \nand no other person made a substantial contribution to the paper.\nFunding\nResearch reported in this publication was supported by Elite Researcher Grant \nCommittee under award number [IR.NIMAD.REC.4040334] from the National \nInstitute for Medical Research Development (NIMAD), Tehran, Iran\nData availability\nData Availability On Request.\nDeclarations\nEthics approval and consent to participate\nThe study was conducted in accordance with the Declaration of Helsinki and \ninstitutional guidelines.\nConsent for publication\nNot applicable.\nCompeting interests\nThe authors declare no competing interests.\nAuthor details\n1Cellular and Molecular Endocrine Research Center, Research Institute for \nEndocrine Molecular Biology, Research Institute for Endocrine Sciences, \nShahid Beheshti University of Medical Sciences, Tehran, Iran\n2Reproductive Endocrinology Research Center, Research Institute for \nEndocrine Molecular Biology, Research Institute for Endocrine Sciences, \nShahid Beheshti University of Medical Sciences, Tehran, Iran\n3Department of Medical Genetics, Faculty of Medical Sciences, Tarbiat \nModares University, Tehran, Iran\n4Endocrine Research Center, Research Institute for Endocrine Disorders, \nResearch Institute for Endocrine Sciences, Shahid Beheshti University of \nMedical Sciences, Tehran, Iran\n5Foundation for Research & Education Excellence, Vestavia Hills, AL, USA\nReceived: 1 December 2025 / Accepted: 23 April 2026\nReferences\n1. Smolarz B, Szyłło K, Romanowicz H, Endometriosis (2021) Epidemiology, Clas-\nsification, Pathogenesis, Treatment and Genetics (Review of Literature). Int J \nMol Sci 22:19\n2. Macer ML, Taylor HS (2012) Endometriosis and infertility: a review of the \npathogenesis and treatment of endometriosis-associated infertility. Obstet \nGynecol Clin North Am 39(4):535–549\n3. De Corte P , Klinghardt M, von Stockum S, Heinemann K (2025) Time to Diag-\nnose Endometriosis: Current Status, Challenges and Regional Characteristics-\nA Systematic Literature Review. BJOG 132(2):118–130\n4. Bischoff F, Simpson JL (2004) Genetics of endometriosis: heritability and \ncandidate genes. Best Pract Res Clin Obstet Gynecol 18(2):219–232\n5. Saha R, Pettersson HJ, Svedberg P , Olovsson M, Bergqvist A, Marions L et al \n(2015) Heritability of endometriosis. Fertil Steril 104(4):947–952\n\nPage 12 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \n6. Treloar SA, O’Connor DT, O’Connor VM, Martin NG (1999) Genetic influences \non endometriosis in an Australian twin sample. sueT@qimr.edu.au. Fertil Steril \n71(4):701–710\n7. Asghari G, Yuzbashian E, Najd-Hassan-Bonab L, Mirmiran P , Daneshpour MS, \nAzizi F (2023) Association of rs2282679 polymorphism in vitamin D binding \nprotein gene (GC) with the risk of vitamin D deficiency in an iranian popula-\ntion: season-specific vitamin D status. BMC Endocr Disorders 23(1):217\n8. Sapkota Y, Steinthorsdottir V, Morris AP , Fassbender A, Rahmioglu N, De \nVivo I et al (2017) Meta-analysis identifies five novel loci associated with \nendometriosis highlighting key genes involved in hormone metabolism. Nat \nCommun 8(1):15539\n9. Uno S, Zembutsu H, Hirasawa A, Takahashi A, Kubo M, Akahane T et al \n(2010) A genome-wide association study identifies genetic variants in the \nCDKN2BAS locus associated with endometriosis in Japanese. Nat Genet \n42(8):707–710\n10. Dun EC, Taylor RN, Wieser F (2010) Advances in the genetics of endometriosis. \nGenome Med 2(10):75\n11. Vainio S, Heikkilä M, Kispert A, Chin N, McMahon AP (1999) Female \ndevelopment in mammals is regulated by Wnt-4 signalling. Nature \n397(6718):405–409\n12. Painter JN, Anderson CA, Nyholt DR, Macgregor S, Lin J, Lee SH et al (2011) \nGenome-wide association study identifies a locus at 7p15.2 associated with \nendometriosis. Nat Genet 43(1):51–54\n13. Nyholt DR, Low S-K, Anderson CA, Painter JN, Uno S, Morris AP et al (2012) \nGenome-wide association meta-analysis identifies new endometriosis risk \nloci. Nat Genet 44(12):1355–1359\n14. Mishra A, Modi D (2025) Role of HOXA10 in pathologies of the endometrium. \nReviews Endocr Metabolic Disorders 26(1):81–96\n15. Xiong G, Li J, Yao F, Yang F, Xiang Y (2024) New insight into the CNC-bZIP \nmember, NFE2L3, in human diseases. Front Cell Dev Biol 12:1430486.  h t t  p s : /  / \nd o  i .  o r g / 1 0 . 3 3 8 9 / f c e l l . 2 0 2 4 . 1 4 3 0 4 8 6     ,  h t t  p s : /  / w w  w .  f r o  n t i e  r s i  n .  o r g  / j o u  r n a  l s  / c e l \nl - a n d - d e v e l o p m e n t a l - b i o l o g y / a r t i c l e s / 1 0 . 3 3 8 9 / f c e l l . 2 0 2 4 . 1 4 3 0 4 8 6 / f u l l       \n16. Rahmioglu N, Mortlock S, Ghiasi M, Møller PL, Stefansdottir L, Galarneau G et \nal (2023) The genetic basis of endometriosis and comorbidity with other pain \nand inflammatory conditions. Nat Genet 55(3):423–436\n17. Holdsworth-Carson SJ, Fung JN, Luong HTT, Sapkota Y, Bowdler LM, Wallace \nL et al (2016) Endometrial vezatin and its association with endometriosis risk. \nHum Reprod 31(5):999–1013\n18. Chadchan SB, Popli P , Liao Z, Andreas E, Dias M, Wang T et al (2024) A GREB1-\nsteroid receptor feedforward mechanism governs differential GREB1 action in \nendometrial function and endometriosis. Nat Commun 15(1):1947\n19. Steinthorsdottir V, Thorleifsson G, Aradottir K, Feenstra B, Sigurdsson A, Ste-\nfansdottir L et al (2016) Common variants upstream of KDR encoding VEGFR2 \nand in TTC39B associate with endometriosis. Nat Commun 7(1):12350\n20. Sobalska-Kwapis M, Smolarz B, Słomka M, Szaflik T, Kępka E, Kulig B et al \n(2017) New variants near RHOJ and C2, HLA-DRA region and susceptibility \nto endometriosis in the Polish population—the genome-wide association \nstudy. Eur J Obstet Gynecol Reproductive Biology 217:106–112\n21. Azizi F, Zadeh-Vakili A, Takyar M (2018) Review of rationale, design, and \ninitial findings: Tehran lipid and glucose study. Int J Endocrinol Metab 16(4 \nSuppl):e84777\n22. Azizi F, Ghanbarian A, Momenan AA, Hadaegh F, Mirmiran P , Hedayati M et al \n(2009) Prevention of non-communicable disease in a population in nutrition \ntransition: Tehran Lipid and Glucose Study phase II. Trials 10(1):5\n23. Daneshpour MS, Akbarzadeh M, Lanjanian H, Sedaghati-Khayat B, Guity \nK, Masjoudi S et al (2023) Cohort profile update: Tehran cardiometabolic \ngenetic study. Eur J Epidemiol 38(6):699–711\n24. Tehrani FR, Behboudi-Gandevani S, Dovom MR, Farahmand M, Minooee \nS, Noroozzadeh M et al (2018) Reproductive assessment: findings from 20 \nyears of the Tehran lipid and glucose study. Int J Endocrinol Metab 16(4 \nSuppl):e84786\n25. Indrielle-Kelly T, Frühauf F, Fanta M, Burgetova A, Lavu D, Dundr P et al (2020) \nApplication of International Deep Endometriosis Analysis (IDEA) group con-\nsensus in preoperative ultrasound and magnetic resonance imaging of deep \npelvic endometriosis. Ultrasound Obstet Gynecol 56(1):115–116\n26. Becker CM, Bokor A, Heikinheimo O, Horne A, Jansen F, Kiesel L et al (2022) \nESHRE guideline: endometriosis. Hum Reprod open 2022(2):hoac009\n27. Najd-Hassan-Bonab L, Daneshpour MS, Jafarinia M, Akbarzadeh M, Moazzam-\nJazi M, Asgarian S et al (2024) Exploring sex-specific genetic architecture of \ncoronary artery disease in Tehran: a cardiometabolic genetic study. Expert \nRev Mol Diagn 24(12):1171–1180\n28. Najd-Hassan-Bonab L, Motafeghi F, Moazzam-jazi M, Farahmand M, Azizi \nF, Daneshpour MS et al (2025) Elucidating the genetic architecture of \nearly menopause in the Tehran cardiometabolic genetic study. Maturitas \n201:108692\n29. Johnson JL, Abecasis GR. GAS Power Calculator: web-based power calculator \nfor genetic association studies. bioRxiv; 2017. https:/ /doi.or g/10.11 01/1 64343, \nhttps:/ /europe pmc.org /art icle/ppr/ppr27422\n30. Ward LD, Kellis M (2016) HaploReg v4: systematic mining of putative causal \nvariants, cell types, regulators and target genes for human complex traits and \ndisease. Nucleic Acids Res 44(D1):D877–D81\n31. Tian D, Wang P , Tang B, Teng X, Li C, Liu X et al (2020) GWAS Atlas: a curated \nresource of genome-wide variant-trait associations in plants and animals. \nNucleic Acids Res 48(D1):D927–D32\n32. Kurki MI, Karjalainen J, Palta P , Sipilä TP , Kristiansson K, Donner KM et al (2023) \nFinnGen provides genetic insights from a well-phenotyped isolated popula-\ntion. Nature 613(7944):508–518\n33. Gagliano Taliun SA, VandeHaar P , Boughton AP , Welch RP , Taliun D, Schmidt \nEM et al (2020) Exploring and visualizing large-scale genetic associations by \nusing PheWeb. Nat Genet 52(6):550–552\n34. Carvalho-Silva D, Pierleoni A, Pignatelli M, Ong C, Fumis L, Karamanis N et al \n(2019) Open Targets Platform: new developments and updates two years on. \nNucleic Acids Res 47(D1):D1056–D65\n35. Strosahl J, Ye K, Pazdro R (2024) Novel insights into the pleiotropic health \neffects of growth differentiation factor 11 gained from genome-wide associa-\ntion studies in population biobanks. BMC Genomics 25(1):837\n36. Ernst J, Kellis M (2015) Large-scale imputation of epigenomic datasets for sys-\ntematic annotation of diverse human tissues. Nat Biotechnol 33(4):364–376\n37. Uimari O, Rahmioglu N, Nyholt DR, Vincent K, Missmer SA, Becker C et al \n(2017) Genome-wide genetic analyses highlight mitogen-activated protein \nkinase (MAPK) signaling in the pathogenesis of endometriosis. Hum Reprod \n32(4):780–793\n38. Lanjanian H, Najd Hassan Bonab L, Akbarzadeh M, Moazzam-Jazi M, Zahedi \nAS, Masjoudi S et al (2022) Sex, age, and ethnic dependency of lipoprotein \nvariants as the risk factors of ischemic heart disease: a detailed study on the \ndifferent age-classes and genders in Tehran Cardiometabolic Genetic Study \n(TCGS). Biology sex Differences 13(1):4\n39. Paneque A, Fortus H, Zheng J, Werlen G, Jacinto E (2023) The hexosamine \nbiosynthesis pathway: regulation and function. Genes 14(4):933\n40. He J, Li F, Jing Z, Ren X, Jia D, Zeng Y et al (2024) GNPNAT1 Serves as a Prog-\nnostic Biomarker Correlated with Immune Infiltration and Promotes Cancer \nCell Metastasis through Stabilization of Snai2 in Lung Adenocarcinoma. \nBiomedicines 12(7):1477\n41. Liu W, Jiang K, Wang J, Mei T, Zhao M, Huang D. Upregulation of GNPNAT1 \npredicts poor prognosis and correlates with immune infiltration in lung \nadenocarcinoma. Frontiers in Molecular Biosciences. 2021 Mar 25;8:605754.  h \nt t  p s : /  / w w  w .  f r o  n t i e  r s i  n .  o r g  / j o u  r n a  l s  / m o l e c u l a r - b i o s c i e n c e s / a r t i c l e s / 1 0 . 3 3 8 9 / f \nm o l b . 2 0 2 1 . 6 0 5 7 5 4 / f u l l       \n42. Ding P , Peng B, Li G, Sun X, Wang G (2022) Glucosamine-phosphate N-acet-\nyltransferase 1 and its DNA methylation can be biomarkers for the diagnosis \nand prognosis of lung cancer. J Clin Lab Anal 36(9):e24628\n43. Yuan R, Zhang Y, Wang Y, Chen H, Zhang R, Hu Z et al (2023) GNPNAT1 is a \npotential biomarker correlated with immune infiltration and immunotherapy \noutcome in breast cancer. Front Immunol 14:1152678\n44. Chokchaitaweesuk C, Kobayashi T, Izumikawa T, Itano N (2019) Enhanced \nhexosamine metabolism drives metabolic and signaling networks involving \nhyaluronan production and O-GlcNAcylation to exacerbate breast cancer. \nCell Death Dis 10(11):803\n45. Neitzel LR, Spencer ZT, Nayak A, Cselenyi CS, Benchabane H, Youngblood \nCQ et al (2019) Developmental regulation of Wnt signaling by Nagk and the \nUDP-GlcNAc salvage pathway. Mech Dev 156:20–31\n46. Zhu P , Gu S, Huang H, Zhong C, Liu Z, Zhang X et al (2021) Upregulation of \nglucosamine-phosphate N-acetyltransferase 1 is a promising diagnostic and \npredictive indicator for poor survival in patients with lung adenocarcinoma. \nOncol Lett 21(6):488\n47. Zutautas KB, Sisnett DJ, Miller JE, Lingegowda H, Childs T, Bougie O, Lessey \nBA, Tayade C. The dysregulation of leukemia inhibitory factor and its implica-\ntions for endometriosis pathophysiology. Frontiers in immunology. 2023 Mar \n23;14:1089098.  h t t  p s : /  / w w  w .  f r o  n t i e  r s i  n .  o r g  / j o u  r n a  l s  / i m m u n o l o g y / a r t i c l e s / 1 0 . \n3 3 8 9 / fi   m m u . 2 0 2 3 . 1 0 8 9 0 9 8 / f u l l       \n48. Canan S, İnan MA, Erdem A, Demirdağ E, Gündüz Mİ, Erdem Ö et al (2024) \nEvaluation of endometrial receptivity in recurrent pregnancy loss and recur-\nrent implantation failure. Turkish J Obstet Gynecol 21(1):22\n\nPage 13 of 13\nNajd-Hassan-Bonab et al. Middle East Fertility Society Journal            (2026) 31:37 \n49. van der Zee M, Sacchetti A, Cansoy M, Joosten R, Teeuwssen M, Heijmans-\nAntonissen C et al (2015) IL6/JAK1/STAT3 signaling blockade in endometrial \ncancer affects the ALDHhi/CD126 + stem-like component and reduces tumor \nburden. Cancer Res 75(17):3608–3622\n50. Kim BG, Yoo J-Y, Kim TH, Shin J-H, Langenheim JF, Ferguson SD et al (2015) \nAberrant activation of signal transducer and activator of transcription-3 \n(STAT3) signaling in endometriosis. Hum Reprod 30(5):1069–1078\n51. Zhang L, Feng Y, Zhang Y, Sun X, Ma Q, Ma F (2024) The sweet relation-\nship between the endometrium and protein glycosylation. Biomolecules \n14(7):770. https:/ /doi.or g/10.33 90/b iom14070770,  h t t    p  s :  / /  w w   w .  m   d p i . c o m / 2 2 \n1 8 - 2 7 3 X / 1 4 / 7 / 7 7 0       \n52. Kawamura E, Hamilton GB, Miskiewicz EI, MacPhee DJ (2018) Fermitin family \nhomolog-2 (FERMT2) is highly expressed in human placental villi and modu-\nlates trophoblast invasion. BMC Dev Biol 18(1):19\n53. Beecham GW, Vardarajan B, Blue E, Bush W, Jaworski J, Barral S et al (2018) \nRare genetic variation implicated in non-Hispanic white families with \nAlzheimer disease. Neurology: Genet 4(6):e286\n54. Saei Ghare Naz M, Noroozzadeh M, Ardebili SN, Mousavi M, Azizi F, Ramezani \nTehrani F (2024) Cardio-Metabolic Risk Profile of Women With Endometriosis: \nA Population‐Based Study. Endocrinol Diabetes Metabolism 7(6):e70008\n55. Lee SJ, Kim JA, Ihn HJ, Choi J-Y, Kwon T-Y, Shin H-I et al (2024) The transcrip-\ntion factor BBX regulates phosphate homeostasis through the modulation of \nFGF23. Exp Mol Med 56(11):2436–2448\n56. Rahmioglu N, Macgregor S, Drong AW, Hedman ÅK, Harris HR, Randall JC \net al (2015) Genome-wide enrichment analysis between endometriosis \nand obesity-related traits reveals novel susceptibility loci. Hum Mol Genet \n24(4):1185–1199\n57. Rahmioglu N, Montgomery GW, Zondervan KT (2015) Genetics of endome-\ntriosis. Women’s Health 11(5):577–586\n58. Marquardt RM, Tran DN, Lessey BA, Rahman MS, Jeong J-W (2023) Epigenetic \ndysregulation in endometriosis: implications for pathophysiology and thera-\npeutics. Endocr Rev 44(6):1074–1095\n59. Timofeeva A, Drapkina Y, Fedorov I, Chagovets V, Makarova N, Shamina M et \nal (2020) Small noncoding RNA signatures for determining the developmen-\ntal potential of an embryo at the morula stage. Int J Mol Sci 21(24):9399\n60. Cardoso JV, Perini JA, Machado DE, Pinto R, Medeiros R (2020) Systematic \nreview of genome-wide association studies on susceptibility to endometrio-\nsis. Eur J Obstet Gynecol Reproductive Biology 255:74–82\n61. Bonab LNH, Moazzam-Jazi M, Moosavi R-SM, Fallah M-S, Lanjanian H, \nMasjoudi S et al (2021) Low HDL concentration in rs2048327-G carriers can \npredispose men to develop coronary heart disease: Tehran Cardiometabolic \ngenetic study (TCGS). Gene 778:145485\n62. Adachi S, Tajima A, Quan J, Haino K, Yoshihara K, Masuzaki H et al (2010) \nMeta-analysis of genome-wide association scans for genetic susceptibility to \nendometriosis in Japanese population. J Hum Genet 55(12):816–821\n63. Wang W, Li Y, Li S, Wu Z, Yuan M, Wang T et al (2017) Pooling-based genome-\nwide association study identifies risk loci in the pathogenesis of ovarian \nendometrioma in Chinese Han women. Reprod Sci 24(3):400–406\n64. Sapkota Y, Vivo ID, Steinthorsdottir V, Fassbender A, Bowdler L, Buring JE et al \n(2017) Analysis of potential protein-modifying variants in 9000 endometriosis \npatients and 150000 controls of European ancestry. Sci Rep 7(1):11380\n65. Wojcik GL, Graff M, Nishimura KK, Tao R, Haessler J, Gignoux CR et al (2019) \nGenetic analyses of diverse populations improves discovery for complex \ntraits. Nature 570(7762):514–518\nPublisher’s note\nSpringer Nature remains neutral with regard to jurisdictional claims in \npublished maps and institutional affiliations.","source_license":"CC0","license_restricted":false}