{"paper_id":"a30b0111-24dc-41f2-8fa0-6956c50593a5","body_text":"Oxidative stress-related genes as diagnostic markers for endometriosis and their associated immunoassays | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Oxidative stress-related genes as diagnostic markers for endometriosis and their associated immunoassays yanlun song, hui wu, jian wang, qiumei huang, siyu liao, yi wei, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5483387/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose Endometriosis (EMT) affects millions of women worldwide and is closely associated with the body's response to oxidative stress. The role of oxidative stress markers in the diagnosis and treatment of endometriosis is a potentially fruitful avenue of research. Methods In this study, we employed a machine learning approach to model and screen key biomarkers, integrating five independent datasets from the Omnibus (GEO) database to construct a comprehensive training set. The identification of key genes was achieved through a process of cross-referencing with the aim of locating those that were differentially expressed and known to be involved in oxidative stress. Nine machine learning algorithms were employed for model selection, followed by the evaluation of immune infiltration and immune correlation through single sample gene set enrichment analysis (ssGSEA) and the CIBERSORT algorithm. Results After comparing the performance of different machine learning algorithms, the gradient boosting algorithm (GBM) was selected as the best model. Eventually it screened five featured genes (FOS, CFH, AOX1, FMO1, FCGR2B). The expression patterns of these genes showed diagnostic and predictive potential in the constructed nomograms and external validation. In addition, the association of these genes with pregnancy status and disease severity was explored. The results of immune infiltration analysis showed significant correlation between these key genes and the immune system. Conclusions This study identifies genes at the intersection of endometriosis and oxidative stress, thereby providing reliable molecular markers for clinical application. This offers a new avenue for subsequent diagnosis and treatment of endometriosis. Endometriosis oxidative stress machine learning algorithms immune infiltration subtypes Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Endometriosis (EMT) is a chronic gynecological disease characterized by the implantation of active endometrial tissue outside the uterine cavity. This leads to the development of chronic pelvic pain and infertility, which are the main clinical manifestations of the disease [ 1 ]. EMT is common in women of childbearing age, with a clinical morbidity rate of 10–15% [ 2 ]. It is estimated that 80% of patients experience varying degrees of chronic pelvic pain, with 35–50% of cases resulting in infertility. This has a significant impact on the quality of life and mental health of women. [ 3 ]. Reactive oxygen species (ROS) represent the intermediate products of normal oxygen metabolism, but their detrimental effects have been demonstrated [ 4 ]. Under normal conditions the body protects itself from oxidative damage and stress generates an antioxidant system to inactivate reactive oxygen species to maintain homeostasis. Oxidative stress (OS) is defined as a state of imbalance between the oxidative and antioxidative systems, whereby the former exerts a greater influence than the latter. The existence of excess oxygen free radicals may cause serious oxidative damage to DNA, lipid, protein and other cellular structures. This damage can lead to the destruction of cell structure and physiological function. Moreover, it can function as a secondary messenger, indirectly influencing the occurrence and progression of numerous diseases through the activation of associated factors and signaling pathways, of which EMT is one such pathway. [ 5 ]. In recent years, more and more evidence has demonstrated a correlation between the occurrence and progression of endometriosis and the body's oxidative stress response [ 6 ]. The pathogenesis of endometriosis remains poorly defined, which makes it difficult to diagnose and treat the disease at an early stage due to the lack of specificity of the condition. Despite endometriosis being a benign disease, it is a challenging condition to treat, and there are notable individual differences in clinical symptoms and a high recurrence rate among different patients, so it is important to strengthen the value of early diagnosis[ 7 ].Since laparoscopy and pathology are difficult to be widely used in the preoperative evaluation of endometriosis, clinical scholars are more interested in searching for hematological indexes that are closely related to the disease, and it is expected that stress indexes will become an important indicator for evaluating the severity of the disease [ 8 ]. The development of the disease is related to oxidative stress. Nevertheless, the expression of oxidative stress markers in endometriosis patients and their relationship have been less extensively investigated in both domestic and international research contexts. The objective of this paper is to examine the potential of bioinformatics in the study of oxidative stress as a diagnostic instrument for the identification of endometriosis. 2. Materials And Methods 2.1 Downloading and processing of data Microcolumn matrix datasets related to endometriosis were retrieved and downloaded from the public Gene Expression Omnibus (GEO) database ( www.ncbi.nlm.nih.gov/geo ), GSE6364 (16 cases normal group, 21 cases disease group), GSE7305 (10 cases normal group, 10 cases disease group), GSE23339 (9 cases normal group, 10 cases of disease group), GSE25628 (6 cases of normal group, 16 cases of disease group), GSE86534 (4 cases of normal group, 8 cases of disease group) were used as the training set, and the differences were analyzed using the \"limma\" R package (logFCfilter = 1 ,version3.58.1), and using the \"RAA\" package (version 1.2.1) and \"SVA\" package (version 3.50.0) were used to normalize the differential gene expression data of these five datasets.GSE51981 (71 cases of normal group, 26 cases of mild, 48 cases of severe), GSE120103 (18 cases of normal, 9 cases of fertile woman patients, 9 cases of infertile woman patients), GSE7307 (14 cases of normal group, 17 cases of disease group) were used as validation sets and their differential gene expression data were extracted using the same method. 2.2 Vene plotting and gene enrichment analysis The normalized differential genes were intersected with 1399 genes related to oxidative stress [ 9 ] by the \"VennDiagram\" R package (version 1.12) and plotted as a vene plot. The intersecting genes were subjected to analysis using the Kyoto Encyclopedia of Genes and Genomes (KEGG), the Gene Ontology (GO) and the Disease Ontology (DO) through the utilisation of R packages such as \"clusterProfiler \", \"org.Hs.eg.db\", \"ggplot2\" and other R packages for analysis. 2.3 Gene correlation analysis To investigate the expression correlation between the intersecting genes, the correlation analysis of the intersecting genes was carried out using the \"corrplot\" R package (version 0.92) to reveal the functional linkage and regulatory mechanism between the genes. 2.4 Machine learning algorithms to select feature models A variety of machine learning algorithms are constructed for intersecting genes, including nine machine learning algorithms of Decision Tree (DT), Gradient Booster (GBM), Generalized Linear Model (GLM), K Nearest Neighbor Algorithm (KNN), Least Absolute Shrinkage and Selection Operator (LASSO), Neural Networks (NNET), Random Forests (RF), Support Vector Machines (SVM), Extreme Gradient Boosting (XGB). Using these nine algorithms, models with different diagnostic capabilities were developed. The residuals of the machine learning algorithms are visualized using the \"DALEX\" R package (version 2.4.3), the receiver operating characteristic (ROC) curves were plotted using the \"PROC\" package to facilitate a comparison of the performance of the algorithms., and finally the optimal model is selected based on the residuals of the samples and the area under the receiver operating characteristic (ROC) curve. 2.5 Construction of nomogram and validation of diagnostic efficacy The five genes selected by the dominant algorithm were internally validated and externally validated in three datasets, and the ROC curves of the model and the genes were plotted separately. The \"rms\" package (version 6.0–8) was employed for the construction of disease prediction models for the genes selected by the dominant algorithm, and DCA curves and calibration curves were used to validate the prediction models. 2.6 Uncovering mRNA - miRNA - lncRNA interactions of model genes To explore the intrinsic relationship between the messenger RNA (mRNA) - microRNA (miRNA) - long stranded non-coding RNA (lncRNA) networks of five key genes, we used MiRanda ( http://www.microRNA.org ), miRDB ( http://www.miRDB.org/ ) and TargetScan ( http://www.TargetScan.org/vert_71/ ) to carefully characterize the relationship between the relevant miRNAs and endometriosis. Next, we identified the lncRNA corresponding to these miRNAs using the SpongeScan website site. The predictions were then visualized using Cystoscope 3.8.2. 2.7 Single-sample Gene Set Enrichment Analysis Single-sample Gene Set Enrichment Analysis (ssGSEA) was performed on five key genes using the \"GSVA\" (version 1.50.1) and \"GSEAbase\" (version 1.64.0) packages and the \"h.all.v7.5.1.symbols.gmt\" file. we used the \"ggplot\" package to draw box plots to visualize the differential pathways and carried out correlation analysis between the five key genes and these pathways. 2.8 Differential and clinical analysis of key genes Heatmaps and volcano maps of the key genes were plotted using the \"pheamap\" and \"ggplot2\" packages (version 3.5.0) with logFCfilter = 1 and adj.P.Val.Filter = 0.05. GSE51981 was used to study the difference between mild and severe endometriosis for these genes, GSE120103 was used to study the difference between fertile and infertile women for these genes, and GSE7307 was employed as an external validation set to provide additional validation of the differential expression. 2.9 Immune infiltration analysis and immune function studies Gene expression data was employed to utilize the CIBERSORT algorithm ( https://cibersort.stanford.edu/ ) to assess the relative expression levels of each sample in 22 immune cells. The CIBERSORT algorithm employs Monte Carlo sampling to generate back-convolution p-values for all samples. The aim of this study was to assess the correlations between key genes and immune cells, as well as their interactions, with different proportions of immune cells. Spearman correlation coefficients were employed to ascertain the relationship between the variables, and the results were subsequently visualized using the ggplot2 package in R, allowing for the generation of illustrative representations. 2.10 Unsupervised clustering of endometriosis samples An unsupervised clustering method (Consensus ClusterPlus R package version 2.60) and the K-means algorithm was employed for the purpose of categorizing endometriosis samples based on the expression levels of nine differential genes. After iteration, the samples were categorized into different clusters and different clusters were selected to ascertain the optimal number of clusters. After the subtyping of endometriosis by unsupervised clustering, the gene expression differences among the identified subtypes and the immune infiltration analysis of each subtype were subjected to further investigation. 2.11 Reverse transcription–quantitative polymerase chain reaction validation Samples were obtained from the Hospital 1 and the Hospital 2. The organization and experiment adopted in the experiment were approved by the Ethics Committee of Hospital. Total RNA was extracted using TRIzol reagent and reverse transcribed into cDNA. cDNA was then mixed with primers and fluorescent dyes, and PCR was performed using a real-time fluorescent quantitative PCR system (see Table 1 for the primer sequences). The GAPDH gene was employed as an internal reference gene, and the 2-ΔΔCt method was utilized to calculate reverse transcription-quantitative polymerase chain reaction (RT-qPCR) mRNA expression. Table 1 Primer sequences for reverse transcription–quantitative polymerase chain reaction Target gene Primer Sequence (5′ − 3′) GAPDH Forward CAGGAGGATTGCTGATGAT Reverse GAAGGCTGGGGCTCATTT FOS Forward TGTGAAGACCATGACAGGAGG Reverse TCTAGTTGGTCTGTCTCCGCT CFH Forward GGACGTTGTGAACAGAGTTAGC Reverse GATAGCCTGGGTGCCTTCTG AOX1 Forward GATGGCAGAATCTTGGCCCT Reverse CCCACGAAAAGCTGTGTTGG FMO1 Forward TCCATCAAGTGCTGTCTGGAAG Reverse TGGCTGACTCTTGCTTCTCTT FCGR2B Forward GATTGCTGTAGCGGCCATTG Reverse ATCCGGGTGCATGAGAAGTG 2.12 Western blotting Clinical specimens were lysed using RIPA lysate and a grinder, and proteins were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Isolated proteins were transferred to polyvinylidene fluoride membranes. The membranes were closed with 5% skimmed milk powder for 2 h. The membranes were incubated with GAPDH (1:10,000; Proteintech), CFH (1:000; Biodragon, BD-PB0853), AOX1(1:1000; Biodragon, BD-PB0243), FCGR2B(1: 1000; Biodragon, BD-PP0325), FOS (1:1000; Biodragon, BD-PE2556) and FMO1(1:750; ABclonal, A25236) via primary antibody overnight at 4°C, and then with secondary antibody at room temperature for 2h. Bands were detected with enhanced chemiluminescence chromogenic solution, and GAPDH was used as a quantitative control. 2.13 Statistical analyses Using the R statistical computing software(version 4.3.2)to complete all statistical analyses. A one-way analysis of variance was employed to ascertain the existence of differences between two or more groups, and the t-test was employed to ascertain the extent of the discrepancies between the two groups. The statistical significance level was P < 0.05. 3. Results 3.1 Processing of data and analysis of variance Data sets GSE23339, GSE25628, GSE6364, GSE7305, and GSE86534 were selected from the GEO database (Fig. 1 A). The batch effect was removed to obtain the integrated dataset (Fig. 1 B). Principal component analyses were then performed separately between the pre and post de-batch datasets (Fig. 1 C). Subsequently, in the obtained integrated dataset, the differences between endometriosis samples and control samples were investigated, and 66 DEGs were identified; a total of 34 genes exhibited increased expression, while 32 genes displayed decreased expression. (Fig. 1 D). 3.2 Screening of oxidative stress key genes and enrichment analysis The initial step involved the point of convergence of oxidative stress-related genes with differentially expressed genes (DEGs), resulting in the identification of nine overlapping DEG candidates. (Fig. 2 A). The overlapping DEGs were subjected to GO, KEGG and DO pathway enrichment analysis. Enriched in GO enrichment analysis, 'regulation of inflammatory response ', 'regulation of humoral immune response ,' and 'complement-dependent cytotoxicity' were predominantly in Biological Process ( BP ), 'plasma membrane valves', 'blood particles', and 'serine-type endopeptidase complex' were predominantly enriched in Cellular Component ( CC ), whereas 'flavin adenine dinucleotide conjugate', 'monooxygenated vivo', and ' ferric ion conjugate' were mainly enriched in Molecular Function ( MF ) (Fig. 2 B, 2 C, 2 E). In KEGG enrichment analysis, 'Staphylococcus aureus infection', 'arachidonic acid metabolism' and 'drug metabolism-cytochrome p450' were predominantly enriched (Fig. 2 D, 2 F). Enrichment in DO enrichment analysis, these key genes were predominantly enriched in pathways associated with glomerulonephritis as well as systemic lupus erythematosus (Fig. 2 G). Subsequently, a correlation analysis was conducted on the differentially expressed key genes (Fig. 2 H, 2 I) to elucidate their role in the pathogenesis of endometriosis. 3.3 Machine learning algorithms to identify key genes We used a machine learning approach to screen key genes for endometriosis associated with oxidative stress. Nine machine learning model algorithms, GBM, GLM, KNN, LASSO, NNET, RF, SVM, and XGB, were built to identify key genes by R language and residual distributions were plotted to determine the optimal model. GBM was found to be the best match for minimal sample residuals (Fig. 3 A). Subsequently, the principal feature variables of each model were ranked in accordance with the root mean square error (RMSE). (Fig. 3 B), and it was found that the residuals of the GBM model were lower than those of the other models for most of the samples (Fig. 3 C). AUC = 0.725; KNN, AUC = 0.628; NNET, AUC = 0.528; LASSO, AUC = 0.704; DT, AUC = 0.585) (Fig. 3 D). Together with these results, the GBM model was the best discriminator between different patient groups. Finally, the best five predictive genes from the GBM model, namely FOS, CFH, AOX-1, FMO-1 and FCGR-2B, were chosen for further analysis. 3.4 The key gene and endometriosis incidence modeling We validated our predictive model consisting of 5 genes on a training set including normal individuals and endometriosis patients, as well as on three external validation datasets. The ROC curves showed that the predictive model consisting of 5 genes performed satisfactorily, with AUC values of 0.849 in the training set, 0.849 in the GSE7307 dataset, 0.955 in the GSE51981 dataset and 0.948 in the GSE120103 dataset (Fig. 4 A). We further evaluated the diagnostic values of FOS, CFH, AOX1, FMO1, and FCGR2B in a training set as well as in three external validation datasets by ROC curves. We found that they all had high accuracy (Fig. 4 B). Further evaluation of the predictive ability of the GBM model, we demonstrated the risk relationship between the key genes and the incidence of endometriosis through a nomogram (Fig. 4 C). Calibration curves and decision curve analysis (DCA) were used to assess the predictive ability of the nomogram model. Based on the calibration curve (Fig. 4 D), the error between the actual risk of endometriosis aggregation and the predicted risk is small, and the DCA shows that our nomination map is very accurate and can inform clinical decision making (Fig. 4 E). 3.5Validation of key genes and their expression under different types of endometrioses For further analysis, the five most important key genes in the GBM model were selected (FOS, CFH, AOX1, FMO1, FCGR2B). Normal samples and samples from patients with endometriosis in the combined data (GSE23339, GSE25628, GSE6364, GSE7305, and GSE86534) cohort were analyzed for differential expression. Figure 5 A and Fig. 5 B depict heatmaps and volcano maps of five key genes between the normal and disease groups in the combined data cohort. It was found that all five key genes were up-regulated genes in the training set. To identify the expression of key genes in normal uterine tissue and in tissue samples from patients with moderate and severe endometriosis, the dataset GSE51981 was used. The findings of the study indicate that four of the five key genes, namely FOS, FMO1, FCGR2B and CFH, exhibited differential expression levels between the tissue samples from patients with moderate endometriosis and those from patients with severe endometriosis (Fig. 5 C). To identify the expression of key genes in normal uterine tissue and in uterine tissue from patients with endometriosis in fertile women and in uterine tissue from patients with endometriosis in infertile women, GSE120103 was used. It was found that four genes-FOS, FMO1, FCGR2B, and AOX1-were differentially expressed in normal uterine tissues and in uterine tissues of patients with endometriosis of fertile woman, as well as uterine tissues of patients with endometriosis of infertile woman (Fig. 5 D). In addition, external validation of these five genes was performed in GSE7307 (Fig. 5 E). The results demonstrated a notable elevation in the expression of FOS, FMO1, CFH, AOX1, and FCGR2B in the uterine tissues of patients with endometriosis, a finding that was corroborated by the results obtained using the Training Set. 3.6 Analysis of immune infiltration The functional and pathway analysis of oxidative stress-related pathogenic genes in endometriosis demonstrated a close association with inflammatory and immune processes. The CIBERSORT algorithm was employed to ascertain the characteristics of immune cells in patients with endometriosis, with the objective of elucidating the immunomodulatory mechanism of endometriosis and establishing a correlation between diagnostic markers and immune cell infiltration. Figure 6 A shows the proportion of 22 immune cells in each sample, demonstrating notable distinctions between endometriosis samples and normal samples across the three subpopulations of immune cells. There were lower proportions of naïve B cells, lower proportions of T-cell follicular helper cells, and higher proportions of M2 macrophages compared with controls (Fig. 6 B). Correlations between five key genes (FOS, CFH, AOX1, FMO1, FCGR2B) and three significantly different immune cells (naïve B cells, T-cell follicular helper cells, and M2 macrophages) were analyzed in endometriosis tissues, in which FMO1 and CFH were positively correlated with T-cell follicular helper cells and FMO1, CFH, FCGR2B were positively correlated with M2 macrophages (Fig. 6 C). Figure 6 D shows the correlation analysis of FMO1, FCGR2B, and CFH with M2 macrophages and T cell follicular helper cells. 3.7 Immune cell and immune function analysis To gain further insight into the immune profile of endometriosis, the ssGSEA algorithm was utilized to forecast discrepancies in immune response and immune cell infiltration between the disease and control groups (Fig. 7 A) and it was found that there was a significant enrichment of major histocompatibility complex class I (MHC Class I). The immune cells and immune responses were further correlated (Fig. 7 B and Fig. 7 C), and all the immune cells were coordinated with each other, except for the negative correlation between natural killer cells (NK cells) and follicular helper T cells (Tfh) and other immune cells. Furthermore, Major Histocompatibility Complex Class I (MHC Class I) demonstrated antagonistic effects on the functions of various T cells. Afterwards, we analyzed the disease and control groups for differences in immune cells, and the findings indicated that the disease group showed differences in aDCs, DCs, Macrophages, Mast cells, Neutrophils, NK cells, aDCs, T helper cells, Tfh, and Til (Fig. 7 D). The results in the immune function analysis in the disease and control groups showed differences in APC co inhibition, CCR, HLA, MHC class 1, Parainflammation, T cell co-inhibition, T cell co- stimulation, Type 1 IFN Reponse, type 2 IFN Reponse were differentially present in type 2 IFN Reponse (Fig. 7 E). Finally, we also correlated the five selected key genes with immune cell infiltration and immune response. The results demonstrated a positive correlation between FCGR2B and HLA, macrophages, parainflammation, and T cell co-inhibition (Fig. 7 F). Furthermore, a notable positive correlation was identified between AOX1 and APC co-stimulation and the type 2 IFN response. Furthermore, there was a relatively significant positive correlation between FMO1 and parainflammation. 3.8 The objective of this study was to validate five signature genes associated with endometriosis in vivo. To ascertain the potential clinical applications of the screened signature genes, we determined the differential expression levels of five signature genes between endometriosis patients and healthy controls. To this end, the differential expression levels of five signature genes were determined between endometriosis patients and healthy controls. Tissue samples were obtained from patients diagnosed with endometriosis and age- and sex-matched healthy individuals. The results of the Western blotting analysis demonstrated that the expression of the five signature genes was significantly up regulated compared to the control group. Furthermore, the expression levels were statistically significant between the endometriosis patient group and the healthy group (Fig. 8 A), with a p-value of less than 0.01. The qPCR results demonstrated that all the signature genes exhibited significantly elevated expression in endometriosis patients compared to healthy controls. Furthermore, the expression levels were statistically significant between the endometriosis patient group and the healthy group (Fig. 8 B to 8 F). These results were statistically significant (p < 0.01). In the present study, the expression of the five signature genes was verified at the human level, and the results were consistent. 3.9 mRNA-miRNA-lncRNA interaction network and functional analysis of ssGASE marker gene set To investigate the complex molecular interactions among FOS, CFH, AOX1, FMO1 and FCGR2B in more depth, we constructed an mRNA-miRNA-lncRNA network diagram (Fig. 8 G). This complex network consists of 151 nodes and 156 interactions, including 75 miRNA and 81 lncRNA. The network structure and interconnectivity provide valuable insights into the potential regulatory associations between these genes, miRNA and lncRNA. Visualizing these interactions through network diagrams provides a holistic view that enriches our overall understanding of potential molecular pathways. Considering the pivotal function of miRNA and lncRNA in modulating gene expression, the integration of these two regulatory elements underscores the existence of intricate regulatory crosstalk. This integrative approach considers the interactions between mRNA, miRNA and lncRNA and is expected to provide new clues to the pathogenesis of endometriosis. A differential pathway analysis of five key genes by ssGSEA in normal and disease groups indicated that the key genes were closely associated with the KRAS UP pathways, spermatogenesis, allograft rejection, glycolysis and oxidative phosphorylation, early response to estrogen, late phase, and apoptosis (Fig. 8 H). Notably, in the correlation analysis, FOS, CFH, AOX1, FMO1, and FCGR2B were all significantly negatively correlated with the late estrogen response, glycolysis, and the G2M checkpoint, and the four genes, CFH, AOX1, FMO1, and FCGR2B, were also significantly negatively correlated with the early estrogen response. A significant positive correlation was observed between FOS, CFH, AOX1, FMO1, and FCGR2B and the interferon gamma response, interferon alpha response, and inflammatory response (Fig. 8 I). 3.10 Consistency clustering analysis To clarify the expression of endometriosis associated with oxidative stress in the model, we have nine differentially expressed genes expression the consistency of clustering analysis. The concordance index demonstrates a high level of concordance, with a minimum range of 0.2 to 0.6. (Fig S1 A, S1B). For values of k between 2 and 9, the area under the cumulative distribution function (CDF) exhibits two distinct CDF curves (k and k-1), which display notable differences. Furthermore, when k = 2, all subtypes exhibited consistency scores exceeding 0.9 (Fig S1 C). The consistency matrix heat map was employed in conjunction with the categorization process, resulting in the formation of two clusters comprising 65 patients. These clusters were designated as Cluster 1 (n = 34) and Cluster 2 (n = 31) (Fig S1 D). Subsequently, the expression of nine genes was visualized using a heatmap (Fig S1 E). Subsequently, the patients were clustered using T-distributed random neighborhood embedding (t-SNE), and the results demonstrated a distinct separation between the two clusters. To gain insight into the molecular characteristics of different cohorts, a comprehensive evaluation of the expression differences of nine differentially expressed genes was conducted in cohorts 1 and 2. The two clusters exhibited distinct differential gene expression patterns. Cluster 1 displayed high expression levels for all genes, whereas Cluster 2 exhibited low expression levels for all genes (Fig S1 F). 3.11 Immune infiltration analysis as well as enrichment analysis of endometriosis subtypes Furthermore, the results of the immunoinfiltration analysis demonstrated that the immune microenvironment of Cluster 1 and Cluster 2 had undergone a transformation (Fig S2 B). The proportion of plasma cells and NK cells in a resting state was found to be lower in Cluster 1 (Fig S2 A). To further elucidate the functional differences associated with specific differentially expressed genes in the two clusters, genomic set variation analysis (GSVA) was employed. The results demonstrated that oxidative phosphorylation, RNA degradation, long duration enhancement, and metabolic signaling activities were enhanced in Cluster 1, whereas TCA cycling, immune response, cytokine receptor, TGF-β, and Notch signaling activities were up-regulated in Cluster 2 (Fig S2 C). Furthermore, the functional enrichment analysis revealed that Cluster 1 was markedly linked to the modulation of synaptic and axonal growth, dendritic spine development, and mitochondrial localization. The study revealed that immune-related pathways, including T cell activation, B cell differentiation, and interferon β-1α production, were more prominent in Cluster 2 (Fig S2 D). 4. Discussion Endometriosis is a common gynecologic disorder that is easily misdiagnosed in adolescent girls and young women due to the presence of retrograde menstruation [ 10 ]. In addition to the typical clinical symptoms, the disease is also likely to have a detrimental impact on the quality of sleep and perceived stress, which subsequently affects the quality of life of the patient. [ 7 , 11 ]. Although endometriosis has become a growing social health problem, the causative mechanism has not yet been clarified. Although endometriosis has become a growing social health problem, the pathogenic mechanisms have not been fully clarified to date. Therefore, new diagnostic strategies are needed to evaluate patients with endometriosis. In the last few years, a substantial body of evidence has emerged indicating that oxidative stress is a pivotal factor in the pathogenesis and progression of EMT, as well as in the clinical manifestations of EMT such as pain and infertility [ 12 ]. The products of reactions between oxygen and electrons are referred to as reactive oxygen species (ROS), and oxidative stress (OS) occurs when ROS are produced in excess or when the ability to eliminate ROS is diminished, which can lead to a variety of infertility-related diseases [ 13 ]. Oxidative stress plays a role in the development of a variety of diseases, including polycystic ovary syndrome, endometriosis and recurrent miscarriages. Furthermore, it is also involved in the regression of these diseases [ 14 ]. Therefore, the objective of this study was to examine the relationship between oxidative stress-related genes and endometriosis, and to use a series of bioinformatics analyses to better identify key genes and diagnostic modalities. The integration of the GEO dataset with oxidative stress-related genes yielded nine key genes, as determined by intersection. GSEA enrichment analysis demonstrated that the differentially expressed genes (DEGs) were significantly enriched in multiple oxidative stress-related pathways. To further analyze these differential genes, five signature genes (FOS, CFH, AOX1, FMO1, FCGR2B) were selected for further study using nine different machine learning algorithms based on validation of independent datasets. To further confirm the accuracy of the integrated bioinformatics analysis, the expression patterns of the five signature genes were examined in patients recruited from an external cohort. The results were consistent with the predicted outcomes, indicating that the five signature genes were elevated in the diseased tissues of patients with endometriosis. The proto-oncogene FOS is a member of the immediate early gene family [ 15 ]. Upon stimulation by external factors, the FOS gene is rapidly expressed to produce the FOS protein, which, in conjunction with the JUN protein, constitutes the transcription factor AP-1 [ 16 ]. This factor regulates the expression of effector genes, thereby contributing to the physiological and pathological processes of female reproduction. In the present study, we observed that the FOS gene was rapidly expressed to produce the FOS protein in the early stages of female reproduction. The function of the FOS gene has been investigated in the field of gynecology and obstetrics reproduction [ 18 ], with findings indicating that the FOS gene important in the physiological and pathological aspects of female reproductive function. These include follicular development, menstrual cycle formation, placenta formation, initiation of labor, uterine fibroids, and the development of EMT [ 19 ]. These findings have been further verified by our bioinformatics analyses. This indicates that further investigation into the role of FOS may assist in elucidating the etiology of endometriosis and potentially lead to the development of novel diagnostic techniques. The CFH gene belongs to the regulatory factor region of the complement-activated gene cluster [ 20 ]. The main functions of the CFH gene are to encode Factor H protein, Factor H-like protein 1 (FHL-1), and five Factor H-related proteins (FHR1-5) [ 21 – 22 ]. It plays a pivotal role in the negative regulation of the alternative complement pathway, thereby safeguarding its own cells from complement activation. However, it is not an effective defense against bacterial and viral attacks. In recent years, it has been found that mutations in the CFH gene are associated with a variety of serious diseases, including, but not limited to, rare kidney diseases and, more commonly, age-related macular degeneration (AMD) [ 23 ]. However, its mechanism in endometriosis is not yet known, and our study shows that it may prove to be an important diagnostic tool in the identification of endometriosis, offering a promising avenue for advancing the diagnosis of this condition. The AOX1 gene, also designated as aldehyde oxidase 1, represents a phase I isozyme belonging to the xanthine oxidase family of cytoplasmic molybdenum enzymes [ 24 – 25 ]. AOX1 is important in the metabolism of exogenous and endogenous compounds, including the oxidation of aldehydes and aromatic N-heterocycles [ 26 ]. In the extant literature, mutations or sequence abnormalities in AOX1 have been linked to an elevated risk of bladder cancer and other neoplastic diseases [ 27 ]. Nevertheless, our study demonstrated that AOX1 exerts disparate functions across the spectrum of endometriotic subtypes, which may indicate that it is a pivotal factor in the genesis of diverse forms of endometriosis., and an in-depth study will help us to have a new understanding of the diagnosis of endometriosis. The FMO1 gene encodes flavin-containing monooxygenase 1. FMO functions as a catalyst in the oxidative metabolism of a variety of foreign chemicals, which are dependent on NADPH. [ 28 – 29 ]. In the present study, the gene in question exhibited differential expression in samples from patients with oxidative stress-associated endometriosis. This finding offers a promising avenue for subsequent diagnostic studies in patients with endometriosis. The FCGR2B gene encodes the low-affinity immunoglobulin gamma Fc receptor for the Fc region of the IgG1 complex [ 30 ]. This receptor is a protein on the cell membrane [ 31 ] that recognizes and binds to IgG antibodies. Consequently, it is involved in the phagocytosis of immune complexes and in the regulation of B-cell antibody production. [ 32 ]. The data indicate that the FCGR2B gene is a key regulator of M2 macrophages, offering a potential avenue for advancing the diagnosis of immune infiltrating endometriosis. Furthermore, the data provide new ideas and data to support the diagnosis of endometriosis. The risk relationship between key genes and the incidence of endometriosis was evaluated using a nomogram, and the results of the model were found to be consistent with the differential gene analysis. The differential analysis data showed that FOS, FMO1, AOX1, FCGR2B, CFH were all highly expressed in endometriosis. This provides considerable diagnostic and therapeutic data support for the diagnosis and treatment of endometriosis. We confirmed the expression differences of these five key genes in different types of endometriosis patients by enrichment analysis, and the results demonstrated that FOS, FMO1, FCGR2B, and CFH exhibited notable differences in patients with endometriosis of varying severity and age groups. It suggests the importance of key genes in endometriosis and provides an important reference for disease prevention and diagnosis. The immune response is a crucial component of numerous defensive mechanisms in the human body as well as in the development of diseases, and pathway enrichment analysis of the samples likewise indicates that the immune response may be a contributing factor in the pathogenesis of endometriosis [ 33 ]. In order to elucidate the mechanism through which the immune response contributes to the pathogenesis of endometriosis, the samples were subjected to immunoinfiltration analysis, the results of which demonstrated disparities in the profiles of endometriosis M2 macrophages and T-cell runner replicators between patients and normal tissue, and the correlation between FMO1 and CFH indicated that these two genes may be involved in causing endometriosis by regulating these two types of immune cells. This opens up potential therapeutic targets for endometriosis for further research and treatment. In our analysis, several of these key genes have early or late estrogen changes associated with them, suggesting that disturbances in estrogen metabolism increase reactive oxygen species and that oxidative imbalance is common in patients with endometriosis, and that hormonal contraceptives with estradiol can lead to improved oxidative stress in patients suffering from endometriosis [ 34 ] It is evident that there is a clear link between the five key genes associated with oxidative stress. Interferon is one of the cytokines that has been demonstrated to be elevated in endometriotic tissues in comparison to normal endometrium. It has been demonstrated that elevated levels of interferon are a pivotal factor in the pathogenesis of endometriosis [ 35 ]. From a clinical perspective, endometriosis can be classified into four main categories: peritoneal endometriosis, ovarian endometriosis, deep infiltrating endometriosis, and other sites of endometriosis. This classification is based on the specific pathology associated with each type [ 36 ]. A growing body of evidence is already available which emphasizes the importance of subtyping classification in the diagnosis, treatment and prognosis of the disease. Thus, endometriosis subtyping can serve as a more accurate guide for diagnostic and therapeutic decisions. Here, unsupervised cluster analysis revealed the existence of two distinct subtypes. Accordingly, nine genes associated with oxidative stress were highly expressed in subtype 1. This analysis has some limitations. On the one hand, these results are subject to further experimental analysis and clinical trials to verify them. In addition, these results were derived from large public databases with no access to raw sequencing data, which may introduce a degree of selection bias. Clinical and experimental validation to confirm these findings. 5. Conclusion Therefore, in this paper, we screened five endometriosis diagnostic genes associated with oxidative stress using nine different machine learning methods and explored their expression in patients with different types of endometrioses, and we also utilized these five genes to classify endometriosis into two subtypes. Furthermore, the correlation between these diagnostic genes and immune cell infiltration may facilitate the development of efficacious immunotherapy for patients with EMT. Declarations Acknowledgements The authors express their gratitude for the invaluable assistance and insightful discussions provided by the members of the Department of Youjiang Medical College for Nationalities. Additionally, the authors would like to thank the GEO database for granting access to the data used in this study. Ethics approval and consent to participate The study was conducted according to the 2013 version of the Declaration of Helsinki. Consent for publication All authors have read and agreed to the published version of the manuscript. Availability of data and materials All the data in this study are available. Competing interests All the authors declare that they have no actual, potential, or perceived conflict of interest regarding the manuscript submitted for review. Funding This study was supported by the National Natural Science Foundation of China (no. 82060293) and Innovation Project of Guangxi Graduate Education (no. YCSW2024532). Authors’ contributions Yanlun Song: Data curation, Formal Analysis, Writing – original draft. Hui Wu: Formal analysis, Writing – review & editing. Jian Wang: Formal analysis, Writing – review & editing. Qiumei Huang: Supervision, Writing –review & editing. Siyu Liao: Supervision, Writing – review & editing. Yi Wei: Supervision, Writing – review & editing. Changxue Ceng: Visualization, Writing – review & editing. Yuehua Huang: Visualization, Writing – review & editing. Rong Wang: Visualization, Writing – review & editing Haimei Qin: Conceptualization, Writing – review & editing. Junli Wang: Conceptualization, Funding acquisition, Writing – review & editing. References Liakopoulou M-K et al (2022) Medical and Behavioral Aspects of Adolescent Endometriosis: A Review of the Literature. Children 9:384 Vitale SG et al (2018) The Role of Oxidative Stress and Membrane Transport Systems during Endometriosis: A Fresh Look at a Busy Corner. Oxidative Medicine and Cellular Longevity. 1–14 (2018) Vitale SG, La Rosa VL, Rapisarda AMC, Laganà AS (2017) Impact of endometriosis on quality of life and psychological well-being. J Psychosom Obstet Gynecol 38:317–319 Chang C-H, Yu F-Y, Wu T-S, Wang L-T, Liu B-H (2011) Mycotoxin Citrinin Induced Cell Cycle G2/M Arrest and Numerical Chromosomal Aberration Associated with Disruption of Microtubule Formation in Human Cells. Toxicol Sci 119:84–92 Santulli P et al (2015) Protein oxidative stress markers in peritoneal fluids of women with deep infiltrating endometriosis are increased. Hum Reprod 30:49–60 Hung SW et al (2021) Pharmaceuticals targeting signaling pathways of endometriosis as potential new medical treatment: A review. Med Res Rev 41:2489–2564 Habib N et al (2020) Bowel Endometriosis: Current Perspectives on Diagnosis and Treatment. IJWH Volume 12:35–47 Yang H et al (2018) Pleiotropic roles of melatonin in endometriosis, recurrent spontaneous abortion, and polycystic ovary syndrome. Am J Rep Immunol 80:e12839 Wang H et al (2023) A four oxidative stress gene prognostic model and integrated immunity-analysis in pancreatic adenocarcinoma. Front Oncol 12:1015042 Zullo F et al (2017) Endometriosis and obstetrics complications: a systematic review and meta-analysis. Fertil Steril 108:667–672e5 Ana C, Nuria P, Antonio P, Hortensia F (2021) Novel therapeutic targets to improve IVF outcomes in endometriosis patients: a review and future prospects. Hum Reprod Update 27 Clower L, Fleshman T, Geldenhuys WJ, Santanam N (2022) Targeting Oxidative Stress Involved in Endometriosis and Its Pain. Biomolecules 12:1055 Fan H (2019) In–vitro models of human endometriosis (Review). Exp Ther Med. 10.3892/etm.2019.8363 Cordaro M et al (2021) Hidrox® and Endometriosis: Biochemical Evaluation of Oxidative Stress and Pain. Antioxidants 10:720 Pettit NL, Yap E-L, Greenberg ME, Harvey C (2022) D. Fos ensembles encode and shape stable spatial maps in the hippocampus. Nature 609:327–334 Choi Y, Jeon H, Akin JW, Curry TE, Jo M (2021) The FOS/AP-1 Regulates Metabolic Changes and Cholesterol Synthesis in Human Periovulatory Granulosa Cells. Endocrinology 162:bqab127 Mechta-Grigoriou F, Gerald D, Yaniv M The mammalian Jun proteins: redundancy and speci®city Jiang Z et al (2021) The m6A mRNA demethylase FTO in granulosa cells retards FOS-dependent ovarian aging. Cell Death Dis 12:744 Choi Y et al (2018) FOS, a Critical Downstream Mediator of PGR and EGF Signaling Necessary for Ovulatory Prostaglandins in the Human Ovary. J Clin Endocrinol Metabolism 103:4241–4252 Kopp A, Hebecker M, Svobodová E, Józsi M, Factor H (2012) A Complement Regulator in Health and Disease, and a Mediator of Cellular Interactions. Biomolecules 2:46–75 Ding J-D et al (2015) Expression of Human Complement Factor H Prevents Age-Related Macular Degeneration–Like Retina Damage and Kidney Abnormalities in Aged Cfh Knockout Mice. Am J Pathol 185:29–42 Sándor N et al (2024) The human factor H protein family – an update. Front Immunol 15 Fakhouri F et al (2010) Treatment with human complement factor H rapidly reverses renal complement deposition in factor H-deficient mice. Kidney Int 78:279–286 Garattini E, Fratelli M, Terao M (2009) The mammalian aldehyde oxidase gene family. Hum Genomics 4:119 Garattini E, Fratelli M, Terao M (2008) Mammalian aldehyde oxidases: genetics, evolution and biochemistry. Cell Mol Life Sci 65:1019–1048 Kitamura S, Sugihara K, Ohta S (2006) Drug-Metabolizing Ability of Molybdenum Hydroxylases. Drug Metab Pharmacokinet 21:83–98 Vantaku V et al (2020) Epigenetic loss of AOX1 expression via EZH2 leads to metabolic deregulations and promotes bladder cancer progression. Oncogene 39:6265–6285 Veeravalli S et al (2014) The phenotype of a flavin-containing monooyxgenase knockout mouse implicates the drug-metabolizing enzyme FMO1 as a novel regulator of energy balance. Biochem Pharmacol 90:88–95 Krueger SK, Williams DE (2005) Mammalian flavin-containing monooxygenases: structure/function, genetic polymorphisms and role in drug metabolism. Pharmacol Ther 106:357–387 Sjef VJ, Sachiko H, Hiroyuki N (2019) The Complex Association of FcγRIIb With Autoimmune Susceptibility. Front Immunol 10 Arce Vargas F et al (2017) Fc-Optimized Anti-CD25 Depletes Tumor-Infiltrating Regulatory T Cells and Synergizes with PD-1 Blockade to Eradicate Established Tumors. Immunity 46:577–586 Sharp PEH et al (2012) Increased incidence of anti-GBM disease in Fcgamma receptor 2b deficient mice, but not mice with conditional deletion of Fcgr2b on either B cells or myeloid cells alone. Mol Immunol 50:49–56 Jiang I, Yong PJ, Allaire C, Bedaiwy MA (2021) Intricate Connections between the Microbiota and Endometriosis. IJMS 22:5644 Biasioli A et al (2022) Systemic Oxidative Stress in Women with Ovarian and Pelvic Endometriosis: Role of Hormonal Therapy. JCM 11:7460 Park Y, Han SJ (2022) Interferon Signaling in the Endometrium and in Endometriosis. Biomolecules 12:1554 Ashkenazi MS et al (2023) The Clinical Presentation of Endometriosis and Its Association to Current Surgical Staging. JCM 12, 2688 Additional Declarations No competing interests reported. Supplementary Files SupplementaryFigures.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-5483387\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":380537460,\"identity\":\"c7a58189-dc2c-4fa6-89d1-38a58c2199ab\",\"order_by\":0,\"name\":\"yanlun song\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"yanlun\",\"middleName\":\"\",\"lastName\":\"song\",\"suffix\":\"\"},{\"id\":380537461,\"identity\":\"a967d925-eb16-4d0b-8bbe-fee152feae56\",\"order_by\":1,\"name\":\"hui wu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"hui\",\"middleName\":\"\",\"lastName\":\"wu\",\"suffix\":\"\"},{\"id\":380537462,\"identity\":\"0be8ab02-17a3-42e8-9c47-3f06c70597fc\",\"order_by\":2,\"name\":\"jian wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"jian\",\"middleName\":\"\",\"lastName\":\"wang\",\"suffix\":\"\"},{\"id\":380537463,\"identity\":\"2d4e53d7-05bb-4b7d-90bf-b8cbe7488da6\",\"order_by\":3,\"name\":\"qiumei huang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"qiumei\",\"middleName\":\"\",\"lastName\":\"huang\",\"suffix\":\"\"},{\"id\":380537464,\"identity\":\"38612e8f-bfde-40f0-9997-77580fa0893d\",\"order_by\":4,\"name\":\"siyu liao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"siyu\",\"middleName\":\"\",\"lastName\":\"liao\",\"suffix\":\"\"},{\"id\":380537465,\"identity\":\"6bbb5110-1116-4e18-8184-885706bd8839\",\"order_by\":5,\"name\":\"yi wei\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"yi\",\"middleName\":\"\",\"lastName\":\"wei\",\"suffix\":\"\"},{\"id\":380537466,\"identity\":\"4fecda8d-f312-47b8-8eca-e9a33eb7814f\",\"order_by\":6,\"name\":\"changxue ceng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"changxue\",\"middleName\":\"\",\"lastName\":\"ceng\",\"suffix\":\"\"},{\"id\":380537467,\"identity\":\"6832dba8-a779-4ed9-8305-10ceb2bad44d\",\"order_by\":7,\"name\":\"yuehua huang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"yuehua\",\"middleName\":\"\",\"lastName\":\"huang\",\"suffix\":\"\"},{\"id\":380537468,\"identity\":\"2f7fab13-ed35-4422-b46c-c918dc1f71e4\",\"order_by\":8,\"name\":\"rong wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Affiliated Hospital of Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"rong\",\"middleName\":\"\",\"lastName\":\"wang\",\"suffix\":\"\"},{\"id\":380537469,\"identity\":\"bd6194da-8003-4f23-a6de-f535566313c8\",\"order_by\":9,\"name\":\"haimei qin\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzUlEQVRIiWNgGAWjYBACAwbGBhDJwyB/+MCBDz9I0iLBlnhwZg9RWmBAgsf4MAcbEVrM2Q+3bvhQYC1jLt3z4TDIffxiB/BrsexJbLs5wyCdx3LO2Q2HCywYDGfOTiDgsAOJbbd5DA7zGBzI3XB4Bg9DgsFtQlrOP2y7/QesJefBYR42YrTcANrCANJyI4eBWC0P2272AP1icOaYATCQJYjwy/n0Zzd+/LG2Nzje/PjDhx828vzSBLRAATOMIUGUchQto2AUjIJRMAowAQCF4kuZYwJBVAAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"Affiliated Hospital of Youjiang Medical University for Nationalities\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"haimei\",\"middleName\":\"\",\"lastName\":\"qin\",\"suffix\":\"\"},{\"id\":380537470,\"identity\":\"853cd7fd-c966-429e-a1e3-f79659752a5f\",\"order_by\":10,\"name\":\"junli wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Affiliated Hospital of Youjiang Medical University for Nationalities\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"junli\",\"middleName\":\"\",\"lastName\":\"wang\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-11-19 12:08:30\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-5483387/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-5483387/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":69526083,\"identity\":\"669ccc5c-854c-4cdf-af56-f14f8dabdfa6\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:40:50\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":287483,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eGEO data de-batching. (A) Gene expression level statistics of the dataset before de-batching. (B) Gene expression level statistics of the integrated dataset after de-batching. (C) Principal component analysis (PCA) between and after the prede-batch dataset. (D) Heatmap depicting the expression patterns of DEGs in the integrated GEO dataset.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/9223f9170b6f193bcdb74f5f.png\"},{\"id\":69526229,\"identity\":\"80044ef1-c06b-4f7f-835f-d5c0443e17e9\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:48:50\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":278698,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eVene and enrichment analysis. (A)Venn diagram showing intersecting genes shared by DEG and oxidative stress-related genes. (B, C, E) GO results are presented by circle and bar graphs; (D, F) KEGG results are shown by circle and bar graphs; (G) DO results are shown by bar graphs. (H, I) Correlation heatmap depicting the correlation between different differential gene compositions.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/97a7d85046720e81cf4c8789.png\"},{\"id\":69525088,\"identity\":\"63d52a58-0a1a-427d-a99f-1dfa65bcd45e\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:32:50\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":163673,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eConstruction and evaluation of GBM, GLM, KNN, LASSO, NNET, RF, SVM, and XGB machine models. (A) Cumulative residual distributions for each machine learning model. (B) Important features in the GBM, GLM, KNN, LASSO, NNET, RF, SVM, and XGB machine models. (C) Box-and-line plots showing the residuals for each machine learning model. The red dots indicate the root mean square of the residuals (RMSE). (D) ROC analysis of the nine machine learning models based on the test cohort.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/d59bcd7fae4328c68bd9f353.png\"},{\"id\":69526231,\"identity\":\"bb0462fe-80cd-45fc-bf35-bbc035909b91\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:48:50\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":172798,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eConstruction and validation of column line diagrams (A) ROC analysis of GBM models based on key genes in Train, GSE7307, GSE5981, GSE120103 datasets. (B) ROC curve evaluation of the diagnostic validity of candidate biomarkers using Train, GSE707, GSE5981, GSE120103 datasets. (C) Construction of a column-line graph for predicting the risk of EMT clustering based on the key gene GBM model. (D, E) Construction of DCA (D) and calibration curves (E) for evaluating the predictive efficiency of the column-line diagram model.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/453d7ba501edc88b3ff87382.png\"},{\"id\":69526086,\"identity\":\"4c6a8cff-7c32-4a75-a4c1-83b67f485b00\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:40:50\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":189338,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eValidation of gene expression.(A) Heatmap of the expression patterns of FOS, CFH, AOX1, FMO1, and FCGR2B. (B) Volcano plot of FOS, CFH, AOX1, FMO1, and FCGR2B; (C) Volcano plot of the GSE51981 dataset of FOS, CFH, AOX1, FMO1, and FCGR2B between normal, EMT mild, and EMT severe relative expression levels of FOS, CFH, AOX1, FCGR2B in the GSE51981 data set. (D) Relative expression levels of FOS, CFH, AOX1, FMO1, FCGR2B between normal, fw of EMT and Ifw of EMT in GSE120103. (E), FOS, CFH, AOX1, FMO1, FCGR2B in the dataset GSE7307 relative expression levels between EMT and non-EMT.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/000684edc871815291026a2e.png\"},{\"id\":69525090,\"identity\":\"4a4bf27d-2dc0-42bd-aa70-49d625467a25\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:32:50\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":246655,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eAnalysis of immune cell infiltration in endometriosis. (A) Stacked histogram showing the proportion of immune cells between endometriosis and control groups. (B)Violin plot showing the comparison of 22 immune cells between endometriosis and control groups. (C)Gene-immune cell correlation analysis. (D)Scatterplot analysis of FMO1, FCGR2B, and CFH in relation to M2-type macrophages and T-cell follicular helper cells correlation analysis.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/e8872ae1eb2dedf3022674b9.png\"},{\"id\":69527637,\"identity\":\"c90d0c03-4886-45ed-a76b-c34d34624611\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:56:50\",\"extension\":\"png\",\"order_by\":7,\"title\":\"Figure 7\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":332130,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eAnalysis of immune cells and immune functions. (A) Analysis of differences in immune cells and immune function between disease and control groups. (B) Correlation analysis between immune cells. (C) Correlation analysis between immune response reactions. (D) Difference analysis between disease and control groups in immune cells. (E) Difference analysis between disease and control groups in immune responses. (F) Correlation analysis between the five key genes and immune function.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/d3d191d9c3189cf17f386176.png\"},{\"id\":69525093,\"identity\":\"5522914b-4987-4515-8580-1bf1978624e5\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:32:50\",\"extension\":\"png\",\"order_by\":8,\"title\":\"Figure 8\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":321602,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eqRT‑PCR and Western blotting of the expression levels of five hub OS-related genes，and pathway analysis of genes. (A) Western blotting analysis of the expression levels of five hub OS-related genes in whole blood samples from EMT patients and healthy individuals. （B-F）qRT-PCR showing increased mRNA levels of the expression levels of five hub OS-related genes in EMT patients samples. (G)miRNA gene interaction network among the five genes. (H) Differential analysis of the key genes in the pathways between normal and disease groups. (I) Correlation analysis between the five key genes and the pathways.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"image8.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/b7d1ce205e29b805e98ec4d8.png\"},{\"id\":69587128,\"identity\":\"ac3c93ad-972a-40df-8f6c-f32b4c550806\",\"added_by\":\"auto\",\"created_at\":\"2024-11-22 02:32:01\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":2803314,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/d98ea791-120a-40ef-b165-0224f4266a9e.pdf\"},{\"id\":69525084,\"identity\":\"929ca790-c06a-4c63-b937-cb47c01efcab\",\"added_by\":\"auto\",\"created_at\":\"2024-11-21 09:32:50\",\"extension\":\"docx\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":500433,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryFigures.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5483387/v1/011a44eaad292fedd67f7780.docx\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Oxidative stress-related genes as diagnostic markers for endometriosis and their associated immunoassays\",\"fulltext\":[{\"header\":\"1. Introduction\",\"content\":\"\\u003cp\\u003eEndometriosis (EMT) is a chronic gynecological disease characterized by the implantation of active endometrial tissue outside the uterine cavity. This leads to the development of chronic pelvic pain and infertility, which are the main clinical manifestations of the disease [\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e]. EMT is common in women of childbearing age, with a clinical morbidity rate of 10\\u0026ndash;15% [\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e]. It is estimated that 80% of patients experience varying degrees of chronic pelvic pain, with 35\\u0026ndash;50% of cases resulting in infertility. This has a significant impact on the quality of life and mental health of women. [\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e].\\u003c/p\\u003e \\u003cp\\u003eReactive oxygen species (ROS) represent the intermediate products of normal oxygen metabolism, but their detrimental effects have been demonstrated [\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e]. Under normal conditions the body protects itself from oxidative damage and stress generates an antioxidant system to inactivate reactive oxygen species to maintain homeostasis. Oxidative stress (OS) is defined as a state of imbalance between the oxidative and antioxidative systems, whereby the former exerts a greater influence than the latter. The existence of excess oxygen free radicals may cause serious oxidative damage to DNA, lipid, protein and other cellular structures. This damage can lead to the destruction of cell structure and physiological function. Moreover, it can function as a secondary messenger, indirectly influencing the occurrence and progression of numerous diseases through the activation of associated factors and signaling pathways, of which EMT is one such pathway. [\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e].\\u003c/p\\u003e \\u003cp\\u003eIn recent years, more and more evidence has demonstrated a correlation between the occurrence and progression of endometriosis and the body's oxidative stress response [\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e]. The pathogenesis of endometriosis remains poorly defined, which makes it difficult to diagnose and treat the disease at an early stage due to the lack of specificity of the condition. Despite endometriosis being a benign disease, it is a challenging condition to treat, and there are notable individual differences in clinical symptoms and a high recurrence rate among different patients, so it is important to strengthen the value of early diagnosis[\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e].Since laparoscopy and pathology are difficult to be widely used in the preoperative evaluation of endometriosis, clinical scholars are more interested in searching for hematological indexes that are closely related to the disease, and it is expected that stress indexes will become an important indicator for evaluating the severity of the disease [\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e]. The development of the disease is related to oxidative stress. Nevertheless, the expression of oxidative stress markers in endometriosis patients and their relationship have been less extensively investigated in both domestic and international research contexts. The objective of this paper is to examine the potential of bioinformatics in the study of oxidative stress as a diagnostic instrument for the identification of endometriosis.\\u003c/p\\u003e\"},{\"header\":\"2. Materials And Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.1 Downloading and processing of data\\u003c/h2\\u003e \\u003cp\\u003eMicrocolumn matrix datasets related to endometriosis were retrieved and downloaded from the public Gene Expression Omnibus (GEO) database (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ewww.ncbi.nlm.nih.gov/geo\\u003c/span\\u003e\\u003cspan address=\\\"http://www.ncbi.nlm.nih.gov/geo\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), GSE6364 (16 cases normal group, 21 cases disease group), GSE7305 (10 cases normal group, 10 cases disease group), GSE23339 (9 cases normal group, 10 cases of disease group), GSE25628 (6 cases of normal group, 16 cases of disease group), GSE86534 (4 cases of normal group, 8 cases of disease group) were used as the training set, and the differences were analyzed using the \\\"limma\\\" R package (logFCfilter\\u0026thinsp;=\\u0026thinsp;1 ,version3.58.1), and using the \\\"RAA\\\" package (version 1.2.1) and \\\"SVA\\\" package (version 3.50.0) were used to normalize the differential gene expression data of these five datasets.GSE51981 (71 cases of normal group, 26 cases of mild, 48 cases of severe), GSE120103 (18 cases of normal, 9 cases of fertile woman patients, 9 cases of infertile woman patients), GSE7307 (14 cases of normal group, 17 cases of disease group) were used as validation sets and their differential gene expression data were extracted using the same method.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.2 Vene plotting and gene enrichment analysis\\u003c/h2\\u003e \\u003cp\\u003eThe normalized differential genes were intersected with 1399 genes related to oxidative stress [\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e] by the \\\"VennDiagram\\\" R package (version 1.12) and plotted as a vene plot. The intersecting genes were subjected to analysis using the Kyoto Encyclopedia of Genes and Genomes (KEGG), the Gene Ontology (GO) and the Disease Ontology (DO) through the utilisation of R packages such as \\\"clusterProfiler \\\", \\\"org.Hs.eg.db\\\", \\\"ggplot2\\\" and other R packages for analysis.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.3 Gene correlation analysis\\u003c/h2\\u003e \\u003cp\\u003eTo investigate the expression correlation between the intersecting genes, the correlation analysis of the intersecting genes was carried out using the \\\"corrplot\\\" R package (version 0.92) to reveal the functional linkage and regulatory mechanism between the genes.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec6\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.4 Machine learning algorithms to select feature models\\u003c/h2\\u003e \\u003cp\\u003eA variety of machine learning algorithms are constructed for intersecting genes, including nine machine learning algorithms of Decision Tree (DT), Gradient Booster (GBM), Generalized Linear Model (GLM), K Nearest Neighbor Algorithm (KNN), Least Absolute Shrinkage and Selection Operator (LASSO), Neural Networks (NNET), Random Forests (RF), Support Vector Machines (SVM), Extreme Gradient Boosting (XGB). Using these nine algorithms, models with different diagnostic capabilities were developed. The residuals of the machine learning algorithms are visualized using the \\\"DALEX\\\" R package (version 2.4.3), the receiver operating characteristic (ROC) curves were plotted using the \\\"PROC\\\" package to facilitate a comparison of the performance of the algorithms., and finally the optimal model is selected based on the residuals of the samples and the area under the receiver operating characteristic (ROC) curve.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.5 Construction of nomogram and validation of diagnostic efficacy\\u003c/h2\\u003e \\u003cp\\u003eThe five genes selected by the dominant algorithm were internally validated and externally validated in three datasets, and the ROC curves of the model and the genes were plotted separately. The \\\"rms\\\" package (version 6.0\\u0026ndash;8) was employed for the construction of disease prediction models for the genes selected by the dominant algorithm, and DCA curves and calibration curves were used to validate the prediction models.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.6 Uncovering mRNA - miRNA - lncRNA interactions of model genes\\u003c/h2\\u003e \\u003cp\\u003eTo explore the intrinsic relationship between the messenger RNA (mRNA) - microRNA (miRNA) - long stranded non-coding RNA (lncRNA) networks of five key genes, we used MiRanda (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://www.microRNA.org\\u003c/span\\u003e\\u003cspan address=\\\"http://www.microRNA.org\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), miRDB (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://www.miRDB.org/\\u003c/span\\u003e\\u003cspan address=\\\"http://www.miRDB.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) and TargetScan (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://www.TargetScan.org/vert_71/\\u003c/span\\u003e\\u003cspan address=\\\"http://www.TargetScan.org/vert_71/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) to carefully characterize the relationship between the relevant miRNAs and endometriosis. Next, we identified the lncRNA corresponding to these miRNAs using the SpongeScan website site. The predictions were then visualized using Cystoscope 3.8.2.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.7 Single-sample Gene Set Enrichment Analysis\\u003c/h2\\u003e \\u003cp\\u003eSingle-sample Gene Set Enrichment Analysis (ssGSEA) was performed on five key genes using the \\\"GSVA\\\" (version 1.50.1) and \\\"GSEAbase\\\" (version 1.64.0) packages and the \\\"h.all.v7.5.1.symbols.gmt\\\" file. we used the \\\"ggplot\\\" package to draw box plots to visualize the differential pathways and carried out correlation analysis between the five key genes and these pathways.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec10\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.8 Differential and clinical analysis of key genes\\u003c/h2\\u003e \\u003cp\\u003eHeatmaps and volcano maps of the key genes were plotted using the \\\"pheamap\\\" and \\\"ggplot2\\\" packages (version 3.5.0) with logFCfilter\\u0026thinsp;=\\u0026thinsp;1 and adj.P.Val.Filter\\u0026thinsp;=\\u0026thinsp;0.05. GSE51981 was used to study the difference between mild and severe endometriosis for these genes, GSE120103 was used to study the difference between fertile and infertile women for these genes, and GSE7307 was employed as an external validation set to provide additional validation of the differential expression.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.9 Immune infiltration analysis and immune function studies\\u003c/h2\\u003e \\u003cp\\u003eGene expression data was employed to utilize the CIBERSORT algorithm (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://cibersort.stanford.edu/\\u003c/span\\u003e\\u003cspan address=\\\"https://cibersort.stanford.edu/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) to assess the relative expression levels of each sample in 22 immune cells. The CIBERSORT algorithm employs Monte Carlo sampling to generate back-convolution p-values for all samples. The aim of this study was to assess the correlations between key genes and immune cells, as well as their interactions, with different proportions of immune cells. Spearman correlation coefficients were employed to ascertain the relationship between the variables, and the results were subsequently visualized using the ggplot2 package in R, allowing for the generation of illustrative representations.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.10 Unsupervised clustering of endometriosis samples\\u003c/h2\\u003e \\u003cp\\u003eAn unsupervised clustering method (Consensus ClusterPlus R package version 2.60) and the K-means algorithm was employed for the purpose of categorizing endometriosis samples based on the expression levels of nine differential genes. After iteration, the samples were categorized into different clusters and different clusters were selected to ascertain the optimal number of clusters. After the subtyping of endometriosis by unsupervised clustering, the gene expression differences among the identified subtypes and the immune infiltration analysis of each subtype were subjected to further investigation.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec13\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.11 Reverse transcription\\u0026ndash;quantitative polymerase chain reaction validation\\u003c/h2\\u003e \\u003cp\\u003eSamples were obtained from the Hospital 1 and the Hospital 2. The organization and experiment adopted in the experiment were approved by the Ethics Committee of Hospital. Total RNA was extracted using TRIzol reagent and reverse transcribed into cDNA. cDNA was then mixed with primers and fluorescent dyes, and PCR was performed using a real-time fluorescent quantitative PCR system (see Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e for the primer sequences). The GAPDH gene was employed as an internal reference gene, and the 2-ΔΔCt method was utilized to calculate reverse transcription-quantitative polymerase chain reaction (RT-qPCR) mRNA expression.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e \\u003ccaption language=\\\"En\\\"\\u003e \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003ePrimer sequences for reverse transcription\\u0026ndash;quantitative polymerase chain reaction\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"3\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTarget gene\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003ePrimer\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eSequence (5\\u0026prime;\\u0026thinsp;\\u0026minus;\\u0026thinsp;3\\u0026prime;)\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eGAPDH\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eCAGGAGGATTGCTGATGAT\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eGAAGGCTGGGGCTCATTT\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFOS\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eTGTGAAGACCATGACAGGAGG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eTCTAGTTGGTCTGTCTCCGCT\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCFH\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eGGACGTTGTGAACAGAGTTAGC\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eGATAGCCTGGGTGCCTTCTG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eAOX1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eGATGGCAGAATCTTGGCCCT\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eCCCACGAAAAGCTGTGTTGG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFMO1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eTCCATCAAGTGCTGTCTGGAAG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eTGGCTGACTCTTGCTTCTCTT\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eFCGR2B\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eForward\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eGATTGCTGTAGCGGCCATTG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eReverse\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eATCCGGGTGCATGAGAAGTG\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec14\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.12 Western blotting\\u003c/h2\\u003e \\u003cp\\u003eClinical specimens were lysed using RIPA lysate and a grinder, and proteins were subjected to sodium dodecyl sulfate\\u0026ndash;polyacrylamide gel electrophoresis. Isolated proteins were transferred to polyvinylidene fluoride membranes. The membranes were closed with 5% skimmed milk powder for 2 h. The membranes were incubated with GAPDH (1:10,000; Proteintech), CFH (1:000; Biodragon, BD-PB0853), AOX1(1:1000; Biodragon, BD-PB0243), FCGR2B(1: 1000; Biodragon, BD-PP0325), FOS (1:1000; Biodragon, BD-PE2556) and FMO1(1:750; ABclonal, A25236) via primary antibody overnight at 4\\u0026deg;C, and then with secondary antibody at room temperature for 2h. Bands were detected with enhanced chemiluminescence chromogenic solution, and GAPDH was used as a quantitative control.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec15\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.13 Statistical analyses\\u003c/h2\\u003e \\u003cp\\u003eUsing the R statistical computing software(version 4.3.2)to complete all statistical analyses. A one-way analysis of variance was employed to ascertain the existence of differences between two or more groups, and the t-test was employed to ascertain the extent of the discrepancies between the two groups. The statistical significance level was P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05.\\u003c/p\\u003e \\u003c/div\\u003e \"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec16\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.1 Processing of data and analysis of variance\\u003c/h2\\u003e \\u003cp\\u003eData sets GSE23339, GSE25628, GSE6364, GSE7305, and GSE86534 were selected from the GEO database (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eA). The batch effect was removed to obtain the integrated dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eB). Principal component analyses were then performed separately between the pre and post de-batch datasets (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eC). Subsequently, in the obtained integrated dataset, the differences between endometriosis samples and control samples were investigated, and 66 DEGs were identified; a total of 34 genes exhibited increased expression, while 32 genes displayed decreased expression. (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eD).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec17\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.2 Screening of oxidative stress key genes and enrichment analysis\\u003c/h2\\u003e \\u003cp\\u003eThe initial step involved the point of convergence of oxidative stress-related genes with differentially expressed genes (DEGs), resulting in the identification of nine overlapping DEG candidates. (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA). The overlapping DEGs were subjected to GO, KEGG and DO pathway enrichment analysis. Enriched in GO enrichment analysis, 'regulation of inflammatory response ', 'regulation of humoral immune response ,' and 'complement-dependent cytotoxicity' were predominantly in Biological Process ( BP ), 'plasma membrane valves', 'blood particles', and 'serine-type endopeptidase complex' were predominantly enriched in Cellular Component ( CC ), whereas 'flavin adenine dinucleotide conjugate', 'monooxygenated vivo', and ' ferric ion conjugate' were mainly enriched in Molecular Function ( MF ) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB, \\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eC, \\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eE). In KEGG enrichment analysis, 'Staphylococcus aureus infection', 'arachidonic acid metabolism' and 'drug metabolism-cytochrome p450' were predominantly enriched (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eD, \\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eF). Enrichment in DO enrichment analysis, these key genes were predominantly enriched in pathways associated with glomerulonephritis as well as systemic lupus erythematosus (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eG). Subsequently, a correlation analysis was conducted on the differentially expressed key genes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eH, \\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eI) to elucidate their role in the pathogenesis of endometriosis.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec18\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.3 Machine learning algorithms to identify key genes\\u003c/h2\\u003e \\u003cp\\u003eWe used a machine learning approach to screen key genes for endometriosis associated with oxidative stress. Nine machine learning model algorithms, GBM, GLM, KNN, LASSO, NNET, RF, SVM, and XGB, were built to identify key genes by R language and residual distributions were plotted to determine the optimal model. GBM was found to be the best match for minimal sample residuals (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eA). Subsequently, the principal feature variables of each model were ranked in accordance with the root mean square error (RMSE). (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eB), and it was found that the residuals of the GBM model were lower than those of the other models for most of the samples (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eC). AUC\\u0026thinsp;=\\u0026thinsp;0.725; KNN, AUC\\u0026thinsp;=\\u0026thinsp;0.628; NNET, AUC\\u0026thinsp;=\\u0026thinsp;0.528; LASSO, AUC\\u0026thinsp;=\\u0026thinsp;0.704; DT, AUC\\u0026thinsp;=\\u0026thinsp;0.585) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eD). Together with these results, the GBM model was the best discriminator between different patient groups. Finally, the best five predictive genes from the GBM model, namely FOS, CFH, AOX-1, FMO-1 and FCGR-2B, were chosen for further analysis.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec19\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.4 The key gene and endometriosis incidence modeling\\u003c/h2\\u003e \\u003cp\\u003eWe validated our predictive model consisting of 5 genes on a training set including normal individuals and endometriosis patients, as well as on three external validation datasets. The ROC curves showed that the predictive model consisting of 5 genes performed satisfactorily, with AUC values of 0.849 in the training set, 0.849 in the GSE7307 dataset, 0.955 in the GSE51981 dataset and 0.948 in the GSE120103 dataset (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eA). We further evaluated the diagnostic values of FOS, CFH, AOX1, FMO1, and FCGR2B in a training set as well as in three external validation datasets by ROC curves. We found that they all had high accuracy (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eB). Further evaluation of the predictive ability of the GBM model, we demonstrated the risk relationship between the key genes and the incidence of endometriosis through a nomogram (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eC). Calibration curves and decision curve analysis (DCA) were used to assess the predictive ability of the nomogram model. Based on the calibration curve (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eD), the error between the actual risk of endometriosis aggregation and the predicted risk is small, and the DCA shows that our nomination map is very accurate and can inform clinical decision making (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eE).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec20\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.5Validation of key genes and their expression under different types of endometrioses\\u003c/h2\\u003e \\u003cp\\u003eFor further analysis, the five most important key genes in the GBM model were selected (FOS, CFH, AOX1, FMO1, FCGR2B). Normal samples and samples from patients with endometriosis in the combined data (GSE23339, GSE25628, GSE6364, GSE7305, and GSE86534) cohort were analyzed for differential expression. Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eA and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eB depict heatmaps and volcano maps of five key genes between the normal and disease groups in the combined data cohort. It was found that all five key genes were up-regulated genes in the training set. To identify the expression of key genes in normal uterine tissue and in tissue samples from patients with moderate and severe endometriosis, the dataset GSE51981 was used. The findings of the study indicate that four of the five key genes, namely FOS, FMO1, FCGR2B and CFH, exhibited differential expression levels between the tissue samples from patients with moderate endometriosis and those from patients with severe endometriosis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eC). To identify the expression of key genes in normal uterine tissue and in uterine tissue from patients with endometriosis in fertile women and in uterine tissue from patients with endometriosis in infertile women, GSE120103 was used. It was found that four genes-FOS, FMO1, FCGR2B, and AOX1-were differentially expressed in normal uterine tissues and in uterine tissues of patients with endometriosis of fertile woman, as well as uterine tissues of patients with endometriosis of infertile woman (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eD). In addition, external validation of these five genes was performed in GSE7307 (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eE). The results demonstrated a notable elevation in the expression of FOS, FMO1, CFH, AOX1, and FCGR2B in the uterine tissues of patients with endometriosis, a finding that was corroborated by the results obtained using the Training Set.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec21\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.6 Analysis of immune infiltration\\u003c/h2\\u003e \\u003cp\\u003eThe functional and pathway analysis of oxidative stress-related pathogenic genes in endometriosis demonstrated a close association with inflammatory and immune processes. The CIBERSORT algorithm was employed to ascertain the characteristics of immune cells in patients with endometriosis, with the objective of elucidating the immunomodulatory mechanism of endometriosis and establishing a correlation between diagnostic markers and immune cell infiltration. Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eA shows the proportion of 22 immune cells in each sample, demonstrating notable distinctions between endometriosis samples and normal samples across the three subpopulations of immune cells. There were lower proportions of na\\u0026iuml;ve B cells, lower proportions of T-cell follicular helper cells, and higher proportions of M2 macrophages compared with controls (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eB). Correlations between five key genes (FOS, CFH, AOX1, FMO1, FCGR2B) and three significantly different immune cells (na\\u0026iuml;ve B cells, T-cell follicular helper cells, and M2 macrophages) were analyzed in endometriosis tissues, in which FMO1 and CFH were positively correlated with T-cell follicular helper cells and FMO1, CFH, FCGR2B were positively correlated with M2 macrophages (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eC). Figure\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eD shows the correlation analysis of FMO1, FCGR2B, and CFH with M2 macrophages and T cell follicular helper cells.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec22\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.7 Immune cell and immune function analysis\\u003c/h2\\u003e \\u003cp\\u003eTo gain further insight into the immune profile of endometriosis, the ssGSEA algorithm was utilized to forecast discrepancies in immune response and immune cell infiltration between the disease and control groups (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eA) and it was found that there was a significant enrichment of major histocompatibility complex class I (MHC Class I). The immune cells and immune responses were further correlated (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eB and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eC), and all the immune cells were coordinated with each other, except for the negative correlation between natural killer cells (NK cells) and follicular helper T cells (Tfh) and other immune cells. Furthermore, Major Histocompatibility Complex Class I (MHC Class I) demonstrated antagonistic effects on the functions of various T cells. Afterwards, we analyzed the disease and control groups for differences in immune cells, and the findings indicated that the disease group showed differences in aDCs, DCs, Macrophages, Mast cells, Neutrophils, NK cells, aDCs, T helper cells, Tfh, and Til (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eD). The results in the immune function analysis in the disease and control groups showed differences in APC co inhibition, CCR, HLA, MHC class 1, Parainflammation, T cell co-inhibition, T cell co- stimulation, Type 1 IFN Reponse, type 2 IFN Reponse were differentially present in type 2 IFN Reponse (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eE). Finally, we also correlated the five selected key genes with immune cell infiltration and immune response. The results demonstrated a positive correlation between FCGR2B and HLA, macrophages, parainflammation, and T cell co-inhibition (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003eF). Furthermore, a notable positive correlation was identified between AOX1 and APC co-stimulation and the type 2 IFN response. Furthermore, there was a relatively significant positive correlation between FMO1 and parainflammation.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003e3.8 The objective of this study was to validate five signature genes associated with endometriosis in vivo.\\u003c/b\\u003e \\u003c/p\\u003e \\u003cp\\u003eTo ascertain the potential clinical applications of the screened signature genes, we determined the differential expression levels of five signature genes between endometriosis patients and healthy controls. To this end, the differential expression levels of five signature genes were determined between endometriosis patients and healthy controls. Tissue samples were obtained from patients diagnosed with endometriosis and age- and sex-matched healthy individuals. The results of the Western blotting analysis demonstrated that the expression of the five signature genes was significantly up regulated compared to the control group. Furthermore, the expression levels were statistically significant between the endometriosis patient group and the healthy group (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eA), with a p-value of less than 0.01. The qPCR results demonstrated that all the signature genes exhibited significantly elevated expression in endometriosis patients compared to healthy controls. Furthermore, the expression levels were statistically significant between the endometriosis patient group and the healthy group (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eB to \\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eF). These results were statistically significant (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01). In the present study, the expression of the five signature genes was verified at the human level, and the results were consistent.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec23\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.9 mRNA-miRNA-lncRNA interaction network and functional analysis of ssGASE marker gene set\\u003c/h2\\u003e \\u003cp\\u003eTo investigate the complex molecular interactions among FOS, CFH, AOX1, FMO1 and FCGR2B in more depth, we constructed an mRNA-miRNA-lncRNA network diagram (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eG). This complex network consists of 151 nodes and 156 interactions, including 75 miRNA and 81 lncRNA. The network structure and interconnectivity provide valuable insights into the potential regulatory associations between these genes, miRNA and lncRNA. Visualizing these interactions through network diagrams provides a holistic view that enriches our overall understanding of potential molecular pathways. Considering the pivotal function of miRNA and lncRNA in modulating gene expression, the integration of these two regulatory elements underscores the existence of intricate regulatory crosstalk. This integrative approach considers the interactions between mRNA, miRNA and lncRNA and is expected to provide new clues to the pathogenesis of endometriosis.\\u003c/p\\u003e \\u003cp\\u003eA differential pathway analysis of five key genes by ssGSEA in normal and disease groups indicated that the key genes were closely associated with the KRAS UP pathways, spermatogenesis, allograft rejection, glycolysis and oxidative phosphorylation, early response to estrogen, late phase, and apoptosis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eH). Notably, in the correlation analysis, FOS, CFH, AOX1, FMO1, and FCGR2B were all significantly negatively correlated with the late estrogen response, glycolysis, and the G2M checkpoint, and the four genes, CFH, AOX1, FMO1, and FCGR2B, were also significantly negatively correlated with the early estrogen response. A significant positive correlation was observed between FOS, CFH, AOX1, FMO1, and FCGR2B and the interferon gamma response, interferon alpha response, and inflammatory response (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e8\\u003c/span\\u003eI).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec24\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.10 Consistency clustering analysis\\u003c/h2\\u003e \\u003cp\\u003eTo clarify the expression of endometriosis associated with oxidative stress in the model, we have nine differentially expressed genes expression the consistency of clustering analysis. The concordance index demonstrates a high level of concordance, with a minimum range of 0.2 to 0.6. (Fig \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eA, S1B). For values of k between 2 and 9, the area under the cumulative distribution function (CDF) exhibits two distinct CDF curves (k and k-1), which display notable differences. Furthermore, when k\\u0026thinsp;=\\u0026thinsp;2, all subtypes exhibited consistency scores exceeding 0.9 (Fig \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eC). The consistency matrix heat map was employed in conjunction with the categorization process, resulting in the formation of two clusters comprising 65 patients. These clusters were designated as Cluster 1 (n\\u0026thinsp;=\\u0026thinsp;34) and Cluster 2 (n\\u0026thinsp;=\\u0026thinsp;31) (Fig \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eD). Subsequently, the expression of nine genes was visualized using a heatmap (Fig \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eE). Subsequently, the patients were clustered using T-distributed random neighborhood embedding (t-SNE), and the results demonstrated a distinct separation between the two clusters. To gain insight into the molecular characteristics of different cohorts, a comprehensive evaluation of the expression differences of nine differentially expressed genes was conducted in cohorts 1 and 2. The two clusters exhibited distinct differential gene expression patterns. Cluster 1 displayed high expression levels for all genes, whereas Cluster 2 exhibited low expression levels for all genes (Fig \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eF).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec25\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.11 Immune infiltration analysis as well as enrichment analysis of endometriosis subtypes\\u003c/h2\\u003e \\u003cp\\u003eFurthermore, the results of the immunoinfiltration analysis demonstrated that the immune microenvironment of Cluster 1 and Cluster 2 had undergone a transformation (Fig \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eB). The proportion of plasma cells and NK cells in a resting state was found to be lower in Cluster 1 (Fig \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eA).\\u003c/p\\u003e \\u003cp\\u003eTo further elucidate the functional differences associated with specific differentially expressed genes in the two clusters, genomic set variation analysis (GSVA) was employed. The results demonstrated that oxidative phosphorylation, RNA degradation, long duration enhancement, and metabolic signaling activities were enhanced in Cluster 1, whereas TCA cycling, immune response, cytokine receptor, TGF-β, and Notch signaling activities were up-regulated in Cluster 2 (Fig \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eC). Furthermore, the functional enrichment analysis revealed that Cluster 1 was markedly linked to the modulation of synaptic and axonal growth, dendritic spine development, and mitochondrial localization. The study revealed that immune-related pathways, including T cell activation, B cell differentiation, and interferon β-1α production, were more prominent in Cluster 2 (Fig \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eD).\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"4. Discussion\",\"content\":\"\\u003cp\\u003eEndometriosis is a common gynecologic disorder that is easily misdiagnosed in adolescent girls and young women due to the presence of retrograde menstruation [\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e]. In addition to the typical clinical symptoms, the disease is also likely to have a detrimental impact on the quality of sleep and perceived stress, which subsequently affects the quality of life of the patient. [\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e]. Although endometriosis has become a growing social health problem, the causative mechanism has not yet been clarified. Although endometriosis has become a growing social health problem, the pathogenic mechanisms have not been fully clarified to date. Therefore, new diagnostic strategies are needed to evaluate patients with endometriosis.\\u003c/p\\u003e \\u003cp\\u003eIn the last few years, a substantial body of evidence has emerged indicating that oxidative stress is a pivotal factor in the pathogenesis and progression of EMT, as well as in the clinical manifestations of EMT such as pain and infertility [\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e]. The products of reactions between oxygen and electrons are referred to as reactive oxygen species (ROS), and oxidative stress (OS) occurs when ROS are produced in excess or when the ability to eliminate ROS is diminished, which can lead to a variety of infertility-related diseases [\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e]. Oxidative stress plays a role in the development of a variety of diseases, including polycystic ovary syndrome, endometriosis and recurrent miscarriages. Furthermore, it is also involved in the regression of these diseases [\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e]. Therefore, the objective of this study was to examine the relationship between oxidative stress-related genes and endometriosis, and to use a series of bioinformatics analyses to better identify key genes and diagnostic modalities.\\u003c/p\\u003e \\u003cp\\u003eThe integration of the GEO dataset with oxidative stress-related genes yielded nine key genes, as determined by intersection. GSEA enrichment analysis demonstrated that the differentially expressed genes (DEGs) were significantly enriched in multiple oxidative stress-related pathways. To further analyze these differential genes, five signature genes (FOS, CFH, AOX1, FMO1, FCGR2B) were selected for further study using nine different machine learning algorithms based on validation of independent datasets. To further confirm the accuracy of the integrated bioinformatics analysis, the expression patterns of the five signature genes were examined in patients recruited from an external cohort. The results were consistent with the predicted outcomes, indicating that the five signature genes were elevated in the diseased tissues of patients with endometriosis.\\u003c/p\\u003e \\u003cp\\u003eThe proto-oncogene FOS is a member of the immediate early gene family [\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e]. Upon stimulation by external factors, the FOS gene is rapidly expressed to produce the FOS protein, which, in conjunction with the JUN protein, constitutes the transcription factor AP-1 [\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e]. This factor regulates the expression of effector genes, thereby contributing to the physiological and pathological processes of female reproduction. In the present study, we observed that the FOS gene was rapidly expressed to produce the FOS protein in the early stages of female reproduction. The function of the FOS gene has been investigated in the field of gynecology and obstetrics reproduction [\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e], with findings indicating that the FOS gene important in the physiological and pathological aspects of female reproductive function. These include follicular development, menstrual cycle formation, placenta formation, initiation of labor, uterine fibroids, and the development of EMT [\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e]. These findings have been further verified by our bioinformatics analyses. This indicates that further investigation into the role of FOS may assist in elucidating the etiology of endometriosis and potentially lead to the development of novel diagnostic techniques. The CFH gene belongs to the regulatory factor region of the complement-activated gene cluster [\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. The main functions of the CFH gene are to encode Factor H protein, Factor H-like protein 1 (FHL-1), and five Factor H-related proteins (FHR1-5) [\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e]. It plays a pivotal role in the negative regulation of the alternative complement pathway, thereby safeguarding its own cells from complement activation. However, it is not an effective defense against bacterial and viral attacks. In recent years, it has been found that mutations in the CFH gene are associated with a variety of serious diseases, including, but not limited to, rare kidney diseases and, more commonly, age-related macular degeneration (AMD) [\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e]. However, its mechanism in endometriosis is not yet known, and our study shows that it may prove to be an important diagnostic tool in the identification of endometriosis, offering a promising avenue for advancing the diagnosis of this condition. The AOX1 gene, also designated as aldehyde oxidase 1, represents a phase I isozyme belonging to the xanthine oxidase family of cytoplasmic molybdenum enzymes [\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e]. AOX1 is important in the metabolism of exogenous and endogenous compounds, including the oxidation of aldehydes and aromatic N-heterocycles [\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e]. In the extant literature, mutations or sequence abnormalities in AOX1 have been linked to an elevated risk of bladder cancer and other neoplastic diseases [\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e]. Nevertheless, our study demonstrated that AOX1 exerts disparate functions across the spectrum of endometriotic subtypes, which may indicate that it is a pivotal factor in the genesis of diverse forms of endometriosis., and an in-depth study will help us to have a new understanding of the diagnosis of endometriosis. The FMO1 gene encodes flavin-containing monooxygenase 1. FMO functions as a catalyst in the oxidative metabolism of a variety of foreign chemicals, which are dependent on NADPH. [\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e]. In the present study, the gene in question exhibited differential expression in samples from patients with oxidative stress-associated endometriosis. This finding offers a promising avenue for subsequent diagnostic studies in patients with endometriosis. The FCGR2B gene encodes the low-affinity immunoglobulin gamma Fc receptor for the Fc region of the IgG1 complex [\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e]. This receptor is a protein on the cell membrane [\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e] that recognizes and binds to IgG antibodies. Consequently, it is involved in the phagocytosis of immune complexes and in the regulation of B-cell antibody production. [\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e]. The data indicate that the FCGR2B gene is a key regulator of M2 macrophages, offering a potential avenue for advancing the diagnosis of immune infiltrating endometriosis. Furthermore, the data provide new ideas and data to support the diagnosis of endometriosis.\\u003c/p\\u003e \\u003cp\\u003eThe risk relationship between key genes and the incidence of endometriosis was evaluated using a nomogram, and the results of the model were found to be consistent with the differential gene analysis. The differential analysis data showed that FOS, FMO1, AOX1, FCGR2B, CFH were all highly expressed in endometriosis. This provides considerable diagnostic and therapeutic data support for the diagnosis and treatment of endometriosis. We confirmed the expression differences of these five key genes in different types of endometriosis patients by enrichment analysis, and the results demonstrated that FOS, FMO1, FCGR2B, and CFH exhibited notable differences in patients with endometriosis of varying severity and age groups. It suggests the importance of key genes in endometriosis and provides an important reference for disease prevention and diagnosis. The immune response is a crucial component of numerous defensive mechanisms in the human body as well as in the development of diseases, and pathway enrichment analysis of the samples likewise indicates that the immune response may be a contributing factor in the pathogenesis of endometriosis [\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e]. In order to elucidate the mechanism through which the immune response contributes to the pathogenesis of endometriosis, the samples were subjected to immunoinfiltration analysis, the results of which demonstrated disparities in the profiles of endometriosis M2 macrophages and T-cell runner replicators between patients and normal tissue, and the correlation between FMO1 and CFH indicated that these two genes may be involved in causing endometriosis by regulating these two types of immune cells. This opens up potential therapeutic targets for endometriosis for further research and treatment. In our analysis, several of these key genes have early or late estrogen changes associated with them, suggesting that disturbances in estrogen metabolism increase reactive oxygen species and that oxidative imbalance is common in patients with endometriosis, and that hormonal contraceptives with estradiol can lead to improved oxidative stress in patients suffering from endometriosis [\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e] It is evident that there is a clear link between the five key genes associated with oxidative stress. Interferon is one of the cytokines that has been demonstrated to be elevated in endometriotic tissues in comparison to normal endometrium. It has been demonstrated that elevated levels of interferon are a pivotal factor in the pathogenesis of endometriosis [\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e].\\u003c/p\\u003e \\u003cp\\u003eFrom a clinical perspective, endometriosis can be classified into four main categories: peritoneal endometriosis, ovarian endometriosis, deep infiltrating endometriosis, and other sites of endometriosis. This classification is based on the specific pathology associated with each type [\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e]. A growing body of evidence is already available which emphasizes the importance of subtyping classification in the diagnosis, treatment and prognosis of the disease. Thus, endometriosis subtyping can serve as a more accurate guide for diagnostic and therapeutic decisions. Here, unsupervised cluster analysis revealed the existence of two distinct subtypes. Accordingly, nine genes associated with oxidative stress were highly expressed in subtype 1.\\u003c/p\\u003e \\u003cp\\u003eThis analysis has some limitations. On the one hand, these results are subject to further experimental analysis and clinical trials to verify them. In addition, these results were derived from large public databases with no access to raw sequencing data, which may introduce a degree of selection bias. Clinical and experimental validation to confirm these findings.\\u003c/p\\u003e\"},{\"header\":\"5. Conclusion\",\"content\":\"\\u003cp\\u003eTherefore, in this paper, we screened five endometriosis diagnostic genes associated with oxidative stress using nine different machine learning methods and explored their expression in patients with different types of endometrioses, and we also utilized these five genes to classify endometriosis into two subtypes. Furthermore, the correlation between these diagnostic genes and immune cell infiltration may facilitate the development of efficacious immunotherapy for patients with EMT.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors express their gratitude for the invaluable assistance and insightful discussions provided by the members of the Department of Youjiang Medical College for Nationalities. Additionally, the authors would like to thank the GEO database for granting access to the data used in this study.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthics approval and consent to participate\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe study was conducted according to the 2013 version of the Declaration of Helsinki.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eAll authors have read and agreed to the published version of the manuscript.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and materials\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eAll the data in this study are available.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eAll the authors declare that they have no actual, potential, or perceived conflict of interest regarding the manuscript submitted for review.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was supported by the National Natural Science Foundation of China (no. 82060293) and Innovation Project of Guangxi Graduate Education (no. YCSW2024532).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors\\u0026rsquo; contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eYanlun Song: Data curation, Formal Analysis, Writing \\u0026ndash; original draft. Hui Wu: Formal analysis, Writing \\u0026ndash; review \\u0026amp; editing. Jian Wang: Formal analysis, Writing \\u0026ndash; review \\u0026amp; editing. Qiumei Huang: Supervision, Writing \\u0026ndash;review \\u0026amp; editing. Siyu Liao: Supervision, Writing \\u0026ndash; review \\u0026amp; editing. Yi Wei: Supervision, Writing \\u0026ndash; review \\u0026amp; editing. Changxue Ceng: Visualization, Writing \\u0026ndash; review \\u0026amp; editing. Yuehua Huang: Visualization, Writing \\u0026ndash; review \\u0026amp; editing. Rong Wang: Visualization, Writing \\u0026ndash; review \\u0026amp; editing Haimei Qin: Conceptualization, Writing \\u0026ndash; review \\u0026amp; editing. Junli Wang: Conceptualization, Funding acquisition, Writing \\u0026ndash; review \\u0026amp; editing.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eLiakopoulou M-K et al (2022) Medical and Behavioral Aspects of Adolescent Endometriosis: A Review of the Literature. Children 9:384\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVitale SG et al (2018) The Role of Oxidative Stress and Membrane Transport Systems during Endometriosis: A Fresh Look at a Busy Corner. Oxidative Medicine and Cellular Longevity. 1\\u0026ndash;14 (2018)\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVitale SG, La Rosa VL, Rapisarda AMC, Lagan\\u0026agrave; AS (2017) Impact of endometriosis on quality of life and psychological well-being. J Psychosom Obstet Gynecol 38:317\\u0026ndash;319\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChang C-H, Yu F-Y, Wu T-S, Wang L-T, Liu B-H (2011) Mycotoxin Citrinin Induced Cell Cycle G2/M Arrest and Numerical Chromosomal Aberration Associated with Disruption of Microtubule Formation in Human Cells. Toxicol Sci 119:84\\u0026ndash;92\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSantulli P et al (2015) Protein oxidative stress markers in peritoneal fluids of women with deep infiltrating endometriosis are increased. Hum Reprod 30:49\\u0026ndash;60\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHung SW et al (2021) Pharmaceuticals targeting signaling pathways of endometriosis as potential new medical treatment: A review. Med Res Rev 41:2489\\u0026ndash;2564\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHabib N et al (2020) Bowel Endometriosis: Current Perspectives on Diagnosis and Treatment. IJWH Volume 12:35\\u0026ndash;47\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eYang H et al (2018) Pleiotropic roles of melatonin in endometriosis, recurrent spontaneous abortion, and polycystic ovary syndrome. Am J Rep Immunol 80:e12839\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang H et al (2023) A four oxidative stress gene prognostic model and integrated immunity-analysis in pancreatic adenocarcinoma. Front Oncol 12:1015042\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZullo F et al (2017) Endometriosis and obstetrics complications: a systematic review and meta-analysis. Fertil Steril 108:667\\u0026ndash;672e5\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAna C, Nuria P, Antonio P, Hortensia F (2021) Novel therapeutic targets to improve IVF outcomes in endometriosis patients: a review and future prospects. Hum Reprod Update 27\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eClower L, Fleshman T, Geldenhuys WJ, Santanam N (2022) Targeting Oxidative Stress Involved in Endometriosis and Its Pain. Biomolecules 12:1055\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFan H (2019) In\\u0026ndash;vitro models of human endometriosis (Review). Exp Ther Med. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3892/etm.2019.8363\\u003c/span\\u003e\\u003cspan address=\\\"10.3892/etm.2019.8363\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCordaro M et al (2021) Hidrox\\u0026reg; and Endometriosis: Biochemical Evaluation of Oxidative Stress and Pain. Antioxidants 10:720\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePettit NL, Yap E-L, Greenberg ME, Harvey C (2022) D. Fos ensembles encode and shape stable spatial maps in the hippocampus. Nature 609:327\\u0026ndash;334\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChoi Y, Jeon H, Akin JW, Curry TE, Jo M (2021) The FOS/AP-1 Regulates Metabolic Changes and Cholesterol Synthesis in Human Periovulatory Granulosa Cells. Endocrinology 162:bqab127\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMechta-Grigoriou F, Gerald D, Yaniv M The mammalian Jun proteins: redundancy and speci\\u0026reg;city\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJiang Z et al (2021) The m6A mRNA demethylase FTO in granulosa cells retards FOS-dependent ovarian aging. Cell Death Dis 12:744\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChoi Y et al (2018) FOS, a Critical Downstream Mediator of PGR and EGF Signaling Necessary for Ovulatory Prostaglandins in the Human Ovary. J Clin Endocrinol Metabolism 103:4241\\u0026ndash;4252\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKopp A, Hebecker M, Svobodov\\u0026aacute; E, J\\u0026oacute;zsi M, Factor H (2012) A Complement Regulator in Health and Disease, and a Mediator of Cellular Interactions. Biomolecules 2:46\\u0026ndash;75\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDing J-D et al (2015) Expression of Human Complement Factor H Prevents Age-Related Macular Degeneration\\u0026ndash;Like Retina Damage and Kidney Abnormalities in Aged Cfh Knockout Mice. Am J Pathol 185:29\\u0026ndash;42\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eS\\u0026aacute;ndor N et al (2024) The human factor H protein family \\u0026ndash; an update. Front Immunol 15\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFakhouri F et al (2010) Treatment with human complement factor H rapidly reverses renal complement deposition in factor H-deficient mice. Kidney Int 78:279\\u0026ndash;286\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGarattini E, Fratelli M, Terao M (2009) The mammalian aldehyde oxidase gene family. Hum Genomics 4:119\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGarattini E, Fratelli M, Terao M (2008) Mammalian aldehyde oxidases: genetics, evolution and biochemistry. Cell Mol Life Sci 65:1019\\u0026ndash;1048\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKitamura S, Sugihara K, Ohta S (2006) Drug-Metabolizing Ability of Molybdenum Hydroxylases. Drug Metab Pharmacokinet 21:83\\u0026ndash;98\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVantaku V et al (2020) Epigenetic loss of AOX1 expression via EZH2 leads to metabolic deregulations and promotes bladder cancer progression. Oncogene 39:6265\\u0026ndash;6285\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eVeeravalli S et al (2014) The phenotype of a flavin-containing monooyxgenase knockout mouse implicates the drug-metabolizing enzyme FMO1 as a novel regulator of energy balance. Biochem Pharmacol 90:88\\u0026ndash;95\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKrueger SK, Williams DE (2005) Mammalian flavin-containing monooxygenases: structure/function, genetic polymorphisms and role in drug metabolism. Pharmacol Ther 106:357\\u0026ndash;387\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSjef VJ, Sachiko H, Hiroyuki N (2019) The Complex Association of FcγRIIb With Autoimmune Susceptibility. Front Immunol 10\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eArce Vargas F et al (2017) Fc-Optimized Anti-CD25 Depletes Tumor-Infiltrating Regulatory T Cells and Synergizes with PD-1 Blockade to Eradicate Established Tumors. Immunity 46:577\\u0026ndash;586\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSharp PEH et al (2012) Increased incidence of anti-GBM disease in Fcgamma receptor 2b deficient mice, but not mice with conditional deletion of Fcgr2b on either B cells or myeloid cells alone. Mol Immunol 50:49\\u0026ndash;56\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJiang I, Yong PJ, Allaire C, Bedaiwy MA (2021) Intricate Connections between the Microbiota and Endometriosis. IJMS 22:5644\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBiasioli A et al (2022) Systemic Oxidative Stress in Women with Ovarian and Pelvic Endometriosis: Role of Hormonal Therapy. JCM 11:7460\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePark Y, Han SJ (2022) Interferon Signaling in the Endometrium and in Endometriosis. Biomolecules 12:1554\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAshkenazi MS et al (2023) The Clinical Presentation of Endometriosis and Its Association to Current Surgical Staging. JCM 12, 2688\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Endometriosis, oxidative stress, machine learning algorithms, immune infiltration, subtypes\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-5483387/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-5483387/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003ch2\\u003ePurpose\\u003c/h2\\u003e \\u003cp\\u003eEndometriosis (EMT) affects millions of women worldwide and is closely associated with the body's response to oxidative stress. The role of oxidative stress markers in the diagnosis and treatment of endometriosis is a potentially fruitful avenue of research.\\u003c/p\\u003e\\u003ch2\\u003eMethods\\u003c/h2\\u003e \\u003cp\\u003eIn this study, we employed a machine learning approach to model and screen key biomarkers, integrating five independent datasets from the Omnibus (GEO) database to construct a comprehensive training set. The identification of key genes was achieved through a process of cross-referencing with the aim of locating those that were differentially expressed and known to be involved in oxidative stress. Nine machine learning algorithms were employed for model selection, followed by the evaluation of immune infiltration and immune correlation through single sample gene set enrichment analysis (ssGSEA) and the CIBERSORT algorithm.\\u003c/p\\u003e\\u003ch2\\u003eResults\\u003c/h2\\u003e \\u003cp\\u003eAfter comparing the performance of different machine learning algorithms, the gradient boosting algorithm (GBM) was selected as the best model. Eventually it screened five featured genes (FOS, CFH, AOX1, FMO1, FCGR2B). The expression patterns of these genes showed diagnostic and predictive potential in the constructed nomograms and external validation. In addition, the association of these genes with pregnancy status and disease severity was explored. The results of immune infiltration analysis showed significant correlation between these key genes and the immune system.\\u003c/p\\u003e\\u003ch2\\u003eConclusions\\u003c/h2\\u003e \\u003cp\\u003eThis study identifies genes at the intersection of endometriosis and oxidative stress, thereby providing reliable molecular markers for clinical application. This offers a new avenue for subsequent diagnosis and treatment of endometriosis.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Oxidative stress-related genes as diagnostic markers for endometriosis and their associated immunoassays\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-11-21 09:32:45\",\"doi\":\"10.21203/rs.3.rs-5483387/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"08866780-884e-4854-a95e-c9b172386d5e\",\"owner\":[],\"postedDate\":\"November 21st, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2024-11-22T02:23:50+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-11-21 09:32:45\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-5483387\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-5483387\",\"identity\":\"rs-5483387\",\"version\":[\"v1\"]},\"buildId\":\"B-jG_2CBjPDmsCi4Wdhf-\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC0","license_restricted":false}