Gene expression and demographic factors associated with endometriosis incidence: a landscape genetic approach

Background Endometriosis is a chronic inflammatory disease that results in female infertility. It is considered a complex disorder that plays a role in the impacts of endometriosis on infertility. The present study was performed in the Iranian women population to provide data on the genetic basis of endometriosis and the role played by differ- ent demographic variables like lifestyle factors, locality, ethnicity, etc.

The individuals were divided into two groups: 50 samples, including 25 women with endometriosis and 25 controls. Endometrial tissue and whole blood samples were used for gene expression of MFN2, PINK1, PRKN, and their nine SNPs genotyping, respectively. The multivariate computational methods used for analyzing data on the above- mentioned tasks included factor multiple logistic regression, factor analysis of mixed data (FAMD), and redundancy analysis (RDA). STRING was used for protein–protein interaction and K-means clustering.

The findings revealed a significant difference (P < 0.05) in the magnitude of gene expression in the target genes studied. PPI interaction (P < 0.0001) with FDR < 0.001 showed the interaction between three genes and clus- tered together. The FAMD analysis showed that the target genes’ SNP variability is the most contributing vari- able in differentiating the cases and controls studied. A significant association between the genes and the SNPs studied, as well as with demographic variables, was observed. The RDA analysis revealed a significant association between geographical variables, the gene’ expression magnitude, and the SNPs’ genotypes. In addition, sPCA analy- ses showed a significant positive and negative eigenvalue (global and local structuring, respectively) of the genetic content of the studied samples by geographical variables.

The present study, based on gene expressions and their related SNPs, showed the contribution of these data to geographical and demographical variables.

Endometriosis, Demography, Geographical variants, Redundancy analysis *Correspondence: Zahra Noormohammadi [email protected]; [email protected]; [email protected] Full list of author information is available at the end of the article Page 2 of 11Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43

Infertility is a disease of the reproductive system that affects the capacity of an individual to reproduce [1]. Successful reproduction results from complex pro - cesses required for developing functional gonads and other reproductive organs, along with sex determination, gametogenesis, and the ability to carry a pregnancy. Any defect or malfunction in these processes results in repro - ductive disorders and infertility, impacting approximately 10–15% of couples worldwide [1]. In general, about 10% of women experience infertility, with only about 35% are due to female factors that affect ovarian development, oocyte maturation, fertilization competence, etc. The rest is due to genetic disorders like chromosome abnormalities, DNA sequence mutations, and non-coding RNAs. In addition, epigenetic modifica - tions may also be associated with female infertility [2]. Endometriosis is a chronic inflammatory disease that

in female infertility in about 30%–50% of infertile women and causes pelvic adhesions and distorted pelvic anatomy. It is considered a complex disorder as different causes, like immunological, endocrine, biochemical, and genetic disorders, poor quality of the oocyte, embryo, and endometrial environment, play a role in the impacts of endometriosis on infertility [3]. Bougie et al. (2019) [4] reported that the risk of endome- triosis increased 3–15 times among first-degree relatives. Additionally, racial and ethnic differences could affect the prevalence of diagnosed endometriosis. For instance, Asian women had a higher risk, and Black women had a lower risk of endometriosis than White women [4]. The complex and varying nature of endometriosis is also evident from the results of genome-wide associa - tion studies (GWAS) performed in different groups. For example, the study of European and East Asian descent identified 42 genome-wide significant loci comprising 49 distinct association signals in endometriosis [5]. A similar study [6] reported single nucleotide polymor - phisms (SNPs) that appear over-represented in patients with endometriosis, particularly those with more exten - sive disease (stage III/IV) [7], and several groups have reported variants that are associated with endometriosis in individuals of European and Japanese origin [8]. How - ever, the Angioni et  al. (2020) [9] study concerned with the genotypes and allele frequency of target SNPs in the Sardinian population and their association with endome - triosis did not show a significant association between the studied variants of the genes and a greater risk of devel - oping endometriosis. Therefore, it has been suggested that specific risk alleles could act differently in the patho- genesis of the disease in different ethnic populations, and it is important to study the genetic basis of endometriosis in different populations [9]. Various demographic factors are considered to affect the prevalence of endometriosis in different parts of the world. These factors include body mass index (BMI) [10], ethnicity [4–11], age [12], chronic stress [13], and envi - ronmental factors [14]. Different studies based on several factors associated with endometriosis from various populations and geo - graphical areas greatly differ, making it very difficult to draw any definite conclusions, and therefore, these vari - ables should be studied locally and with a particular eth - nic group. Therefore, we performed the present study on Iranian women, both normal and diagnosed with endometriosis. This study considered the gene expression of three genes, namely, MFN2 (Mitofusin-2), PINK1 (PTEN-induced putative kinase 1), and PRKN (Parkin RBR E3 Ubiquitin Protein Ligase), and eleven SNPs related to these genes, namely, rs68121389, rs117341007, and rs1393563943 (from the PRKN gene), rs513414, rs3077908, rs512550, rs2078073, and rs1043502 (from the PINK1 gene), rs3088064, rs1042842, and rs41278636 (from the MFN2 gene). In addition, we used several demographic variables like age, ethnicity, BMI, and lifestyle factors related to eating, smoking, etc. The genes selected in the present study, including MFN2, PINK1, and PRKN, are known to be associated with female infertility or cervical cancer progression [see, for exam - ple, 15–18]. These genes are involved in the female repro- ductive gland and mitophagy. Moreover, the proteins of these three genes are interrelated in a network of protein interactions (https:// string- db. org/ cgi/ netwo rk). We used different computation methods to study sig - nificant differences between healthy women and those diagnosed with endometriosis, to reveal an association between environmental factors and gene expression, and the SNPs and the association between demographic vari - ables and gene expression. To our knowledge, this is the first report of its kind from Iran.

Sampling In total 50 individuals were studied comprised of normal unaffected persons and those diagnosed with endometriosis. Inclusion criteria for women with endometriosis included patients who were diagnosed with moderate to severe endometriosis based on sono - graphic or laparoscopic findings, characterized by the presence of at least one endometrioma exceeding 3 cm in diameter. Additional inclusion criteria required par - ticipants to be between 20 and 40 years old and eligible for oocyte or embryo cryopreservation. Endometrial tissue and whole blood of patients were collected and stored at—80C for further studies. The Page 3 of 11 Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 study was approved by the ethics committee (No. IR.IAU.SRB.REC. 1401.315). All patients signed a writ - ten informed consent. The questionnaire based on demographic data like age, ethnicity, BMI, smoking, education, job, lifestyle, and geographical variant was filled out by each participant. Reverse transcription and quantitative PCR (RT‑qPCR) We performed the gene expression analysis of MFN2, PINK1, and PRKN as the target genes, and the 18srRNA gene was utilized as a reference gene [19] for the nor - malization of each sample. Total RNA was extracted from endometrial tissue by Favor prep kit (CAT. NO: FABRK001) according to the manufacturer’s protocol. cDNA synthesis was performed using the Parstous kit (CAT. NO: A101161). Real-time PCR for three genes was conducted in a thermocycler Rotorgene (QIAGEN, Germany) in strip tubes including 12.5  µl AMPLI - CON SYBR Green master mix (Cat. No: A324402), 2 µl cDNA, and 10 pM forward and reverse primers of each gene (Table  1). A thermal program for each gene was performed based on the annealing temperature of each primer pair (Table  1). All reactions were performed in duplicate. We used the Pffafl formula for normalization and fold change calculation [20]. SNP genotyping For genotypes of nine SNPs related to three genes, namely, rs68121389, and rs117341007 (from the PRKN gene), rs513414, rs512550, rs2078073, and rs1043502 (from the PINK1 gene), rs3088064, rs1042842, and rs41278636 (from the gene MFN2), genomic DNA was extracted from blood samples by the Salting-out protocol and stored at – 20 °C. The quantity and quality of DNA were checked by a Nanodrop spectrometer and 0.8% agarose gel electrophoresis, respectively. PCR sequenc - ing was performed for the 3´UTR region of each gene as follows: 200 ng/µl DNA, 10 pg/µl of each forward and reverse primer (Table  1) along with the master mix PCR reaction (ParsTous Co., Iran). Amplification was per - formed using a 98-well thermal cycle (Applied Biosys - tems, USA) at annealing temperatures between 60 and 65ºC based on each primer pair’s melting temperature (Table 1). The PCR product was visualized by 1.5% aga - rose gel electrophoresis. PCR products were sequenced based on the Sanger sequencing protocol by Pishgam Co., Iran. Data analyses All data analyses were performed on 999 times permutations. Gene expression analysis The expression delta-CT data were log-transformed for all the following analyses. To study differences in the magnitude of gene expression between normal and affected individuals we performed a t-test (Independent and normal distribution data) and Mann–Whitney U test (nonparametric statistical test), followed by a box plot construction as performed in PAST ver.4. Table 1 Primer names and their sequences used in gene expression and SNP sequencing Primers for Gene Expression test Gene Accession number Primer Primer sequence Primer length TM Prod- uct size PINK1 NM_032409 Forward GAG TAT GGA GCA GTC ACT TACAG 23 58/32 144 Reverse CAG CAC ATC AGG GTA GTC G 19 57/65 PRKN NM_004562.3 Forward TGG GAG AAG AGC AGT ACA ACCG 22 98/61 207 Reverse CCC CTT CAT GGT ACG CTT CTT TAC 24 45/61 MFN2 NM_014874 Forward CTA CAC TGG CTC CAA CTG C 19 58/15 132 Reverse TCA ATT TTC TTG TTC ATG GCGG 22 58/09 18 s NC_002753.1 Forward GTA ACC CGT TGA ACC CCA TT 20 57/93 151 Reverse CCA TCC AAT CGG TAG TAG CG 20 58/09 Primers for SNP Sequencing Gene Primer Primer sequence Primer length TM Prod- uct size PINK1 Forward TAC TAA AAG AAC ATG GCA TCC TCT GT 26 60/30 936 Reverse TTT AAC TGT GAA ATG ATG GTT CTC CC 26 59/79 PRKN Forward GTC CCT CTT TTC CTA ACT GGC TAA GA 26 61/99 1351 Reverse GCT TGG AGT TGA TAT GAG AAT GGC TA 26 60/97 MFN2 Forward GAT TGT TGG AGG ATG ATG TAA GGG TGT 27 62/91 1214 Reverse GGA AAC ATG TCT CTT AAA GGG CAC AAC 27 62/80 Page 4 of 11Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 We used regression and correlation analyses to study the relationship between the genes studied as performed in R 4.3. Similarly, the collinearity between genes was checked by factor analysis of mixed data (FAMD) as per - formed in packages factoextra and FactoMineR in R 4.3. To check protein interactions of three genes studied in KEGG servers, we used STRING ver. 11.5. K-means clus - tering was conducted, and FDR and co-expression scores were calculated. Contribution of variables in differentiating cases and controls All data studied were coded as multinomial data and used in FAMD (Factor analysis of mixed data) analyses. The codings used are as follows: S = Status (case versus control samples), G1-G3 = Genes MFN2, PINK1, and PRKN , respec - tively. R1-R9 = SNPs 1–9, respectively, Personal_data = P1-P6 (Age, education, career, height, weight, and BMI, respectively. Locality = L1-L2 (ethnic and city, respectively). lifestyle factors = H1-H4 (Air pollution, smoking, fast food, and plastic_container, respectively. Association between variables studied We performed a chi-square test for ordinal data to study the relationships between the studied variables as per - formed in the package ordinal in R 4.3. For this, we used the same coded multinomial data as before, after 999 times permutations. Association studies between geographical variables and genetic data We performed the RDA (Redundancy Analysis) to study the association between geographical variables and gene expression, as well as SNPs’ genotypes. In addition, we used the spatial principal components analysis (sPCA), based on geographical variables, the longitude, lati - tude, and altitude of the studied samples to show their role in the genetic structuring of the samples studied. These analyses were performed in PAST ver. 4 and R 4.3, respectively.

Gene expression analyses T-test and Mann–Whitney test revealed a significant difference in the magnitude of gene expression between normal and affected individuals for the three genes stud - ied (P value < 0.01). The paired-sample regression and correlation analyses of the genes studied indicated a significant association between the studied genes (P value < 0.01, Fig.  1 A-C). In addition, the FAMD plot (Fig.  1, D) showed collinearity between the genes, and based on variance analysis, the PINK1 and MFN2 genes contributed the most to differ - entiating the normal and affected individuals. Based on K-means clustering of protein–protein inter - action (Fig.  2), 3 clusters were formed with PPI enrich - ment P-value = 6.09e-06 and average local clustering coefficient 0.816. Proteins in red color constructed a cluster including INK1, MFN2, UBC, UBA52, PRKN, PARK7, FBXO7, HTRA2. They are involved in mitophagy with a strength of 2.29 and FDR = 6.37E-13. Therefore, our experimental results of gene expression indicate that these genes are correlated in the same pathway and accord with their protein–protein interaction network. Genotyping Contribution of variables in differentiating cases and controls The FAMD plot showing the contribution of variables in differentiating the case and normal individuals is pre - sented in Fig. 3. It shows that the studied SNPs (R1–R9) are the most contributing variables in differentiating the cases and controls studied. However, these SNPs differ in the role they play as they are placed at different angles of the FAMD plot. For example, the SNPs coded R1 and R4 are placed differently from the rest of the studied SNPs. The expression level of the studied genes MFN2 and PRKN (coded G1 and G3 in Fig. 3), personal data, namely education (coded P2), lifestyle data, air pollution (coded H1), and the city where the sample lives (L2), grouped as the second most contributing variable that differentiates the case versus the control individuals. The rest of the variables studied seem to play a comparatively lesser role in differentiating the studied samples. Association between case/control samples and the variables studied The chi-square test for ordinal data showed a significant association between the status (Normal versus affected individuals) and the genes studied, as well as the SNPs 1 and 3–8 (P value < 0.01). A significant association was obtained for Status and fast food, age, and weight (P value < 0.01). Similarly, the analysis showed a significant association between the status and education and the locality (city) of the studied samples. These results indicate that the variables that are significantly associated with case versus control samples play a potential role in causing endometriosis. Page 5 of 11 Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 Association between gene expression and SNPs with other variables A significant association was obtained between the studied genes’ expression magnitude and air pollution, fast food, and plastic containers (P value < 0.01). The same holds for gene 1 and gene 2 expression magnitude with age, and gene 2 with all other personal data (P1– P6) studied. The chi-square test for ordinal data showed a signifi - cant association between the studied SNPs and ethnic - ity and locality (P -value < 0.01), and the gene expression data ( P -value < 0.001). Additionally, a significant asso - ciation was obtained between SNPs and age, education, weight, and BMI (P -value < 0.01), as well as with air pollution, smoking, fast food, and plastic containers (p-value < 0.01). A significant association was also obtained between personal data (P1–P6) and lifestyle factors (H1–H4, P value < 0.01). It was interesting to see that the ethnicity and the local area of the individuals may play a role in endometriosis and the genes’ expression level and their related studied SNPs’ genotype differences. Therefore, we performed the RDA (Redundancy analysis) to study the association between geographical variables and gene expression as well as SNPs’ genotypes. In addition, we used the spatial Fig. 1 Regression and FAMD plots of the gene studied show their significant association and collinearity. A PINK1 and MNF2, B PRKN and MNF2, C PINK1 and PRKN regressions, D FAMD plot based on gene expressions Page 6 of 11Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 Fig. 2 K-means clustering based on PPI with FDR < 0.001. Three clusters were constructed. Proteins studied grouped in one cluster (red color) Fig. 3 Representative plots of sPCA analysis based on the combination of longitude and latitude in the studied individuals. A The connection network. B, C Genetic clines formed due to spatial variables. D The positive and negative Eigenvalues show the significant role of the global and local structuring of the genetic data Page 7 of 11 Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 principal components analysis (sPCA), based on geo - graphical variables, the longitude, latitude, and altitude of the studied samples to show their role in the genetic structuring of the samples studied. RDA and sPCA results The RDA analysis after 999 times permutation revealed a significant association between geographical variables the genes’ expression magnitude, and the SNPs’ geno - types (R-squared = 0.40, P = 0.05). The results of sPCA are presented in Figs. 2, 3, and 4. In the first sPCA analysis, we used the combination of lon - gitude + latitude, while in the second sPCA, we used the combination of latitude + altitude. The connection network, eigenvalues, and genetic clines are provided in Figs.  3 and 4. These analyses showed significant positive eigenvalue (global structur - ing), and negative eigenvalue (local structuring) of the genetic content of the studied samples by geographical variables. These results indicate the role played by geographical variables and their association with the genes’ expression magnitude and the allele as well as genotype frequencies observed. The contribution of the studied genetic variables to the spatial structuring of the sample is presented in Fig.  5. In sPCA with the combination of longitude + latitude, the gene MFN2, and PINK1, as well as the SNPs 1, 3, and 7, are the most contributing variables. In the sPCA based on the combination of longi - tude + altitude, the most contributing variables are the status and the SNPs 2, 4, 5, 7, and 8. Similarly, in sPCA based on the combination of latitude + altitude, the most contributing variables are the gene PRKN, and the SNPs 1, 4, 5, 7, and 8. Therefore, the spatial variables operat - ing in the locality in which the studied individuals live play a significant role in structuring and affecting the gene expression as well as the genotypes of the samples studied.

In the present study, the FAMD plot showed collinear - ity between the genes. Also, PINK1 and MFN2 genes contributed the most to differentiating the normal and affected individuals. k-means clustering based on PPI also showed high interaction with the lowest FDR. The co- expression score of these proteins is 0.100. These proteins are ubiquitin protein ligase binding in the mitophagy pathway. PINK1 encoded mitochondrially targeted Ser- Thr kinase with cellular functions like autophagic deg - radation of dysfunctional mitochondria. PRKN, as a Ser-Thr kinase dysfunction, depolarized mitochondria through the phosphorylation of MFN2 with mitochon - drial outer membrane GTPase [21, 22]. Fig. 4 Representative plots of sPCA analysis based on the combination of latitude and altitude in the studied individuals. A The connection network. B The positive and negative Eigenvalues show the significant role of the global and local structuring of the genetic data. C, D Genetic clines formed due to spatial variables Page 8 of 11Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 Gene ontology of genes studied indicated that these genes are involved in the female reproductive gland and mitophagy. Also, several studies reported that genes selected in the present study are known to be associ - ated with female infertility or cervical cancer progression [15–18]. Genotyping and endometriosis In FAMD analysis of variables in the present work indi - cated that the studied SNPs (R1-R9) are the most contrib- uting variables in differentiating the cases and controls studied. However, these SNPs differ in the role they play as they are placed at different angles of the FAMD plot. In addition, the chi-square test for ordinal data showed a significant association between the status (Normal ver - sus affected individuals), and the gene studies, as well as the SNPs 1, 3–8. These SNPs were also significantly associated with the ethnicity and locality of the studied individuals. These results were supported by the RDA and spatial principal components analysis after 999 times permuta - tion revealed a significant association between geograph - ical variables and the genes’ expression magnitude and the SNPs’ genotypes and that some of the genetic vari - ants are associated with different combinations of geo - graphical variables. There have been controversial reports on the associa - tion of different SNPs and endometriosis. For example, [23], by using GWAS and functional network analyses reported an association between the risk genes and their variant SNPs with endometriosis in the Taiwanese-Han population, while Angioni et  al. (2020) [9], reported no association between the investigated variants of the genes and a greater risk of developing endometriosis in the Sar- dinian population. It has been suggested that the racial/ ethnic disparities in the process of hormone regulation and nutrition metabolism may be the reason for contro - versial results on the genetics of endometriosis across different populations [23]. Demographic variables and endometriosis The present study reports an association between some of the demographic variables and endometriosis. A sig - nificant association was obtained for the status and fast food, age, weight, and education of the studied samples. These variables play a potential role in causing endo - metriosis. Additionally, a significant association was obtained between SNP variation and age, education, weight, and BMI, as well as with air pollution, smoking, fast food, and plastic containers. Smoking Various studies report the controversial role of demo - graphic variables on the incidence of endometriosis, and there is no reasonable consistency in the role of environ - mental factors in endometriosis etiopathogenesis. For example, Polak et al. [14] report that about 1.7% of preg - nant women worldwide smoke. Smoking may reduce the Fig. 5 The contributing variables in spatial structuring of the studied individuals based on the first two eigenvalues of sPCA. LAT: latitude, Long: longitude, and Alt: altitude Page 9 of 11 Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 risk of endometriosis in later life among female fetuses. Though its mechanism is unknown, it has been suggested that “nicotine, along with its metabolite cotinine, may suppress the aromatase-dependent conversion of andro - gens to estrogen, stimulate apoptosis, and inhibit angio - genesis and therefore, may inhibit the development of endometriosis” [14]. Diet Several studies show an association between diet and the development of endometriosis [14]. Dietary factors may affect sex hormones and strive for pro- or antioxidant and pro-inflammatory effects which have considerable roles in endometriosis. Additionally, the contaminants accompanying food production may also affect the inci - dence of this disorder. A diet rich in fruits that supply large amounts of provi - tamin A can decrease the incidence of endometriosis, and vitamin A suppresses the formation of the pro-inflamma- tory interleukin-6, a cytokine whose high levels are found in the amniotic fluid in women with endometriosis. Simi- larly, citrus fruits with high amount of vitamin C, which inhibits inflammation and exerts antioxidant effects. However, there are reports on the opposite and negative effects of fruits which can increase the risk of disorder incidence in American women which may be associated with the large quantity of pesticides used during cultiva - tion in the US. A diet rich in red meat is also known to increase the risk of endometriosis [14]. Education The studies performed on the association between the socioeconomic status of populations and the incidence risk of endometriosis have shown that a high socioeco - nomic status (SES) or education level has been associated with a higher frequency of endometriosis. It may reflect better detection and patient care of women with high SES [24, 25]. Body mass index (BMI) Various studies differ in their reports on the role of BMI in the risk of endometriosis. In general, a lower body mass index (BMI) is thought to be associated with endo - metriosis [26, 27], but women with normal BMI also develop endometriosis [10]. Yunhui et  al. [10] studied the association between BMI and surgically diagnosed endometriosis in Chinese women. They concluded that there is no association between BMI and the incidence of endometriosis, but there was a significant increase in the incidence in obese women, compared with women of normal weight. In a similar study in Australia, women with a nor - mal BMI were more likely to have endometriosis, in comparison with women with underweight [28]. The dif - ferences in the inverse association of BMI and endome - triosis between Australia’s study and other studies could be explained by differences in the study populations. Spatial patterns of endometriosis incidence and the role of geographical variables The present study reported that the role played by geo - graphical variables, longitude, latitude, and altitude of the localities in which diagnosed individuals live can shape the genetic variants of the target risk genes and may affect the magnitude of gene expression leading to the incidence of endometriosis in the Iranian population. Le Moal et al. [29] studied the risk factors for the inci - dence of endometriosis in different geographical regions of France and reported geographical heterogeneity in its incidence and concluded that geography may influence this risk and state that this finding is the first step in the quest of clarifying environmental or other factors that may be contributing to the development of the disease. A similar study was performed by Cataby et al. [30] on the spatial pattern of endometriosis incidence by apply - ing Bayesian approaches to Disease Mapping. They used data on the incident cases of endometriosis in women aged 15–50  years in the Friuli Venezia Giulia region in the calendar period 2004–2017. They reported a very strong north–south spatial gradient related to endome - triosis incidence and identified a group of five neighbor - ing municipalities at higher risk in the industrialized and polluted southeast part of the region. Feng et al. [31] investigated the burden implication of endometriosis in 204 countries and territories from 1990 to 2019. They used estimated annual percentage changes (EAPCs) and disability-adjusted life-years (DALYs) of endometriosis with the classified data by region, country, age, and socio-demographic index (SDI). They reported “an increase in global incidence and DALYs of endome - triosis, but a decrease in the age-standardized incidence rate (ASIR) and age-standardized DALY rate of endome - triosis. The largest decreases in the ASIR and age-stand - ardized DALY rate of endometriosis were observed in Qatar and Oman, respectively” .

The present study reports a detailed analysis of an asso - ciation study concerned with endometriosis disorder in the Iranian population and its genetic basis, including gene expression magnitude, and SNP variability, as well as several demographic and geographical variables. Dif - ferent multivariate statistical and bioinformatic analyses revealed a significant difference in gene expression mag - nitude of the target genes, namely, MFN2, PINK1, and PRKN. A significant association was observed between Page 10 of 11Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 the SNP variability of the target genes and gene expres - sion magnitude and the incidence of the disorder. A sig - nificant association also occurred between the status (Normal versus affected individuals) and demographic variables like diet, age, weight, education, and the local - ity (city) of the studied samples. Similarly, an association was observed between the SNP variability and ethnic - ity, locality, and demographic data. The spatial principal components and redundancy analyses revealed a sig - nificant association between geographical variables, the genes’ expression magnitude, and the SNP variability. In addition, sPCA analyses showed a significant posi - tive eigenvalue (global structuring) and negative eigen - value (local structuring) of the genetic content of the studied samples by geographical variables. The present study, based on gene expressions and their related SNPs, showed the contribution of these data to geographical and demographic variables. There was some limitation in the present study such as sampling and collecting accurate samples with all included criteria. Furthermore, there was reliance on cross-sectional and bioinformatic data for selecting suit - able pathways for studies. Abbreviations FAMD Factor analysis of mixed data RDA Redundancy analysis sPCA Spatial principal components analysis PPI Protein-protein interaction FDR False discovery rate Supplementary Information The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s43043- 025- 00256-3. Supplementary Material 1: Table S1. The SNPs’ genotypes in the case and normal individuals studied

We thank the patients for their help. We also acknowledge the Science and Research Branch, Islamic Azad University for providing a laboratory. Authors’ contributions Z.N. and A.M. had conventionalization of the project, Z.N., H.M., K.J., and P .P . wrote the main manuscript, Z.N. did data analyses, H.M., K.J., and P .P . collected samples and performed laboratory work. Funding There is no funding to declare. Data availability The current study is not publicly available due to personal document confi- dentiality. Data are available from the corresponding author on request. Declarations Ethics approval and consent to participate The project proposal was reviewed by the Ethics Committee of the Islamic Azad University Science and Research Branch, and was approved with the ID number, No. IR.IAU.SRB.REC.1401.315. Informed consent was obtained from individuals. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Author details 1 Department of Biology, SR.C, Islamic Azad University, Tehran, Iran. 2 Depart- ment of Endocrinology and Female Infertility, Reproductive Biomedicine Research Center, Royan Institute for Reproductive Biomedicine, ACECR, Tehran, Iran. 3 Breast Disease Research Center (BDRC), Tehran University of Medical Sci- ence, Tehran, Iran. 4 Department of Obstetrics and Gynecology, Arash Women’s Hospital, Tehran University of Medical Sciences, Tehran, Iran. Received: 17 May 2025 Accepted: 16 September 2025

1. Yatsenko SA, Rajkovic A (2019) Genetics of human female infertility. Biol Reprod 101:549–566. https:// doi. org/ 10. 1093/ biolre/ ioz084 2. Bazrgar M, Gourabi H (2023) Editorial: genetics of female infertility. Front Genet 14:1297173. https:// doi. org/ 10. 3389/ fgene. 2023. 12971 73 3. Lee D, Kim SK, Lee JR, Jee BC (2020) Management of endometriosis- related infertility: Considerations and treatment options. Clin Exp Reprod Med 47:1–11. https:// doi. org/ 10. 5653/ cerm. 2019. 02971 4. Bougie O, Yap MI, Sikora L, Flaxman T, Singh S (2019) Influence of race/ ethnicity on prevalence and presentation of endometriosis: a systematic review and meta-analysis. BJOG 126:1104–1115. https:// doi. org/ 10. 1111/ 1471- 0528. 15692 5. Rahmioglu N, Mortlock S, Ghiasi M, Møller PL, Stefansdottir L, Galarneau G et al (2023) The genetic basis of endometriosis and comorbidity with other pain and inflammatory conditions. Nat Genet 55:423–436. https:// doi. org/ 10. 1038/ s41588- 023- 01323-z 6. Saunders PTK (2022) Insights from genomic studies on the role of sex steroids in the aetiology of endometriosis. Reprod Fertil 3:R51-65. https:// doi. org/ 10. 1530/ RAF- 21- 0078 7. Fung JN, Rogers PAW, Montgomery GW (2025) Identifying the biological basis of GWAS hits for endometriosis. Biol Reprod 92. https:// doi. org/ 10. 1095/ biolr eprod. 114. 126458. 8. Steinthorsdottir V, Thorleifsson G, Aradottir K, Feenstra B, Sigurdsson A, Stefansdottir L et al (2016) Common variants upstream of KDR encoding VEGFR2 and in TTC39B associate with endometriosis. Nat Commun 7:12350. https:// doi. org/ 10. 1038/ ncomm s12350 9. Angioni S, D’Alterio MN, Coiana A, Anni F, Gessa S, Deiana D (2020) Genetic characterization of endometriosis patients: review of the litera- ture and a prospective cohort study on a Mediterranean population. Int J Mol Sci 21:1765. https:// doi. org/ 10. 3390/ ijms2 10517 65 10. Tang Y, Zhao M, Lin L, Gao Y, Chen GQ, Chen S et al (2020) Is body mass index associated with the incidence of endometriosis and the severity of dysmenorrhoea: a case-control study in China? BMJ Open 10:e037095. https:// doi. org/ 10. 1136/ bmjop en- 2020- 037095 11. Arruda MS, Petta CA, Abrão MS, Benetti-Pinto CL (2003) Time elapsed from onset of symptoms to diagnosis of endometriosis in a cohort study of Brazilian women. Hum Reprod 18:756–759. https:// doi. org/ 10. 1093/ humrep/ deg136 12. Nnoaham KE, Webster P , Kumbang J, Kennedy SH, Zondervan KT (2012) Is early age at menarche a risk factor for endometriosis? A systematic review and meta-analysis of case-control studies. Fertil Steril 98:702-712. e6. https:// doi. org/ 10. 1016/j. fertn stert. 2012. 05. 035 13. Reis FM, Coutinho LM, Vannuccini S, Luisi S, Petraglia F (2020) Is stress a cause or a consequence of endometriosis? Reprod Sci 27:39–45. https:// doi. org/ 10. 1007/ s43032- 019- 00053-0 14. Polak G, Banaszewska B, Filip M, Radwan M, Wdowiak A (2021) Envi- ronmental factors and endometriosis. Int J Environ Res Public Health 18:11025. https:// doi. org/ 10. 3390/ ijerp h1821 11025 Page 11 of 11 Mahmoudi et al. Middle East Fertility Society Journal (2025) 30:43 15. Ahn SY, Li C, Zhang X, Hyun Y-M (2018) Mitofusin-2 expression is impli- cated in cervical cancer pathogenesis. Anticancer Res 38:3419–3426. https:// doi. org/ 10. 21873/ antic anres. 12610 16. Ahn SY, Song J, Kim YC, Kim MH, Hyun YM (2021) Mitofusin-2 promotes the epithelial-mesenchymal transition-induced cervical cancer progres- sion. Immune Netw 21:e30. https:// doi. org/ 10. 4110/ in. 2021. 21. e30 17. Holzer I, Machado Weber A, Marshall A, Freis A, Jauckus J, Strowitzki T et al (2020) GRN, NOTCH3, FN1, and PINK1 expression in eutopic endome- trium – potential biomarkers in the detection of endometriosis – a pilot study. J Assist Reprod Genet 37:2723–2732. https:// doi. org/ 10. 1007/ s10815- 020- 01905-4 18. Skodvin SN, Gjessing HK, Jugessur A, Romanowska J, Page CM, Corfield EC et al (2023) Statistical methods to detect mother-father genetic interaction effects on risk of infertility: a genome-wide approach. Genet Epidemiol 47:503–519. https:// doi. org/ 10. 1002/ gepi. 22534 19. Bahrami N, Nazari A, Afshari Z, Aftabsavad S, Moini A, Noormohammadi Z (2023) Gene expression and demographic analyses in women with the poor ovarian response: a computational approach. J Assist Reprod Genet 40:2627–2638. https:// doi. org/ 10. 1007/ s10815- 023- 02919-4 20. Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29:e45. https:// doi. org/ 10. 1093/ nar/ 29.9. e45 21. Vives-Bauza C, Zhou C, Huang Y, Cui M, de Vries RLA, Kim J et al (2010) PINK1-dependent recruitment of Parkin to mitochondria in mitophagy. Proc Natl Acad Sci U S A 107:378–383. https:// doi. org/ 10. 1073/ pnas. 09111 87107 22. Chen Y, Dorn GW (2013) PINK1-phosphorylated mitofusin 2 is a Parkin receptor for culling damaged mitochondria. Science 340:471–475. https:// doi. org/ 10. 1126/ scien ce. 12310 31 23. Sheu JJ-C, Lin W-Y, Liu T-Y, Chang CY-Y, Cheng J, Li Y-H et al (2024) Ethnic- specific genetic susceptibility loci for endometriosis in Taiwanese-Han population: a genome-wide association study. J Hum Genet. https:// doi. org/ 10. 1038/ s10038- 024- 01270-5 24. Hemmings R, Rivard M, Olive DL, Poliquin-Fleury J, Gagné D, Hugo P et al (2004) Evaluation of risk factors associated with endometriosis. Fertil Steril 81:1513–1521. https:// doi. org/ 10. 1016/j. fertn stert. 2003. 10. 038 25. Peterson CM, Johnstone EB, Hammoud AO, Stanford JB, Varner MW, Kennedy A et al (2013) Risk factors associated with endometriosis: impor- tance of study population for characterizing disease in the ENDO Study. Am J Obstet Gynecol 208:451.e1–11. https:// doi. org/ 10. 1016/j. ajog. 2013. 02. 040 26. Ferrero S, Anserini P , Remorgida V, Ragni N (2005) Body mass index in endometriosis. Eur J Obstet Gynecol Reprod Biol 121:94–98. https:// doi. org/ 10. 1016/j. ejogrb. 2004. 11. 019 27. Vitonis AF, Baer HJ, Hankinson SE, Laufer MR, Missmer SA (2010) A pro- spective study of body size during childhood and early adulthood and the incidence of endometriosis. Hum Reprod 25:1325–1334. https:// doi. org/ 10. 1093/ humrep/ deq039 28. Holdsworth-Carson SJ, Dior UP , Colgrave EM, Healey M, Montgomery GW, Rogers PAW et al (2018) The association of body mass index with endo- metriosis and disease severity in women with pain. J Endometr Pelvic Pain Disord 10:79–87. https:// doi. org/ 10. 1177/ 22840 26518 773939 29. Le Moal J, Goria S, Chesneau J, Fauconnier A, Kvaskoff M, De Crouy- Chanel P et al (2022) Increasing incidence and spatial hotspots of hospitalized endometriosis in France from 2011 to 2017. Sci Rep 12:6966. https:// doi. org/ 10. 1038/ s41598- 022- 11017-x 30. Catelan D, Giangreco M, Biggeri A, Barbone F, Monasta L, Ricci G et al (2021) Spatial patterns of endometriosis incidence. A study in Friuli Ven- ezia Giulia (Italy) in the period 2004–2017. Int J Environ Res Public Health 18:7175. https:// doi. org/ 10. 3390/ ijerp h1813 7175 31. Feng J, Zhang S, Chen J, Zhu J, Yang J (2022) Global burden of endome- triosis in 204 countries and territories from 1990 to 2019. Clin Exp Obstet Gynecol 49:235. https:// doi. org/ 10. 31083/j. ceog4 910235 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations.