Exploration of the shared pathways and common biomarker LY96 in Endometriosis and Systemic lupus erythematosus using integrated bioinformatics analysis

Exploration of the shared pathways and common biomarker LY96 in Endometriosis and Systemic lupus erythematosus using integrated bioinformatics analysis | 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 Article Exploration of the shared pathways and common biomarker LY96 in Endometriosis and Systemic lupus erythematosus using integrated bioinformatics analysis Jin Huang, Xuelian Ruan, Yongling Chen, Ziqing Zhong, Jiaqi Nie, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4150400/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 Endometriosis (EMS) is a chronic gynecological disorder that affects 5–10% of women of reproductive age, and Systemic lupus erythematosus (SLE) is one of the most prevalent systemic autoimmune diseases. Despite clinical evidence suggesting potential associations between EMS and SLE, the underlying pathogenesis is yet unclear. This article aimed to explore the shared gene signatures and potential molecular mechanisms in EMS and SLE. Microarray data were downloaded from the Gene Expression Omnibus (GEO) database and used to screen for differentially expressed genes (DEGs) in the SLE datasets. A weighted gene co-expression network analysis (WGCNA) was used to identify the co-expression modules of EMS. cytoscape software and three machine learning algorithms were used to determine critical biomarkers, and a diagnostic model was built using the XG-Boost machine learning algorithms. Immune cell infiltration analysis was used to investigate the correlation between immune cell infiltration and common biomarkers of EMS and SLE. Results revealed that shared genes enriched in immune-related pathways and inflammatory responses. The area under the receiver operating characteristic (AUROC) curve and the Precision-Recall (PR) curves showed satisfactory performance of the model. immune cell infiltration analysis showed that the expression of hub genes was closely associated with immune cells. RT-qPCR results indicated that LY96 might be the best biomarker for EMS and SLE. Biological sciences/Immunology/Autoimmunity Biological sciences/Immunology/Inflammation Endometriosis Systemic lupus erythematosus diagnostic biomarker bioinformatics machine learning immune response Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1 Introduction Endometriosis(EMS) affects 5–10% of women of reproductive age in the United States and is characterized by the presence of endometrioid tissue outside the uterine cavity, mainly in the pelvic peritoneum and ovaries ( 1 ). The cause of EMS is not yet fully understood, but there is evidence to suggest that it may be related to dysfunction in the immune system( 2 ). Systemic lupus erythematosus (SLE) is an autoimmune disease that affects multiple organs. Its pathogenesis involves abnormalities in the innate and adaptive immune system, resulting in inadequate responses to self-antigens( 3 ). Women of childbearing age are often susceptible to the disease, in a similar way to women with EMS. In 1987,Gleicher et al. proposed the theory that EMS may be an autoimmune disease, and since then there has been increasing evidence of a link between EMS and autoimmune diseases( 4 ) ( 5 ). However, the relationship between EMS and SLE has not been thoroughly researched, but there is evidence that women with EMS are at increased risk of developing SLE ( 6 ) ( 7 ) ( 8 ). Importantly, we have observed similarities between immune dysfunction in these two diseases. Women with EMS show deficits in immune surveillance( 9 ),such as abnormal activation of T-cell, B-cell( 10 ), high natural killer cell and macrophage cell counts with reduced activity( 11 , 12 ), decreased rate of neutrophils apoptosis( 13 ), increased incidence of serum autoantibodies( 14 ), and elevated inflammatory cytokines, including interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-alpha), and interleukin-1 (IL-1)).( 15 ) ( 16 ). As in EMS, abnormal activation of T-cell, B-cell, the production of autoantibodies and inflammatory cytokines, reduced natural killer cell activity( 17 ), and dysregulated neutrophil phagocytosis( 18 ) are important features of the pathophysiology of SLE. Notably, these studies tended to adopt a clinical approach and could not uncover molecular mechanisms at the gene level. With the rapid development of bioinformatics methods, common transcriptional features based on the molecular level of genes may provide new insights into the common pathogenesis of EMS and SLE. Therefore, in this study, we performed a comprehensive bioinformatics analysis combined with machine learning algorithms to identify the hub genes and pathways of EMS and SLE from the GEO database. In addition, we investigated the correlation between hub genes and immune cells in SLE and EMS. Finally, we validated the expression of the hub genes in an external dataset. In conclusion, this may be the first study to establish common biomarkers and associated therapeutic targets for SLE and EMS, which may provide clues for exploring the genetic aetiology and combined therapeutic strategies for SLE and EMS. 2 Results 2.1 Weighted gene co-expression network analysis of EMS WGCNA was used to explore the relationship between clinical traits and genes. First, in the GSE23339 datasets, all samples were clustered, and none were eliminated (Fig. 2 A). Then, β = 20 (scale-free R2 = 0.9) was selected as the optimal soft power value (Fig. 2 B). Next, a total of 6 gene co-expression modules (GCMs) were identified (Fig. 2 C). Furthermore, the correlations between the GCMs and the clinical features were calculated. The blue and yellow modules had high positive correlations with EMS (r = 0.59, p = 0.001 and r = 0.56, p = 0.01) and included 1530 genes (Fig. 2 D). Finally, a significant correlation between blue (r = 0.54) and yellow (r = 0.2) module membership and gene significance for EMS was observed (Fig. 2 E-F). Thus, the 1530 genes in the blue and yellow modules that are most associated with EMS were identified as crucial for further experiments. 2.2 Identification and Enrichment analysis of shared genes between SLE and EMS The R package limma was used to identify differentially expressed genes (DEGs) in the GSE72326 dataset. A total of 387 DEGs were identified, of which 116 were upregulated and 271 of these genes were downregulated showed in volcano plot (Fig. 3 A). Of these DEGs, the top 50 differentially expressed genes are displayed in heatmaps (Fig. 3 B). The DEGs and module genes were then intersected, and 93 shared genes were obtained (Fig. 3 C). These genes may be related to the pathogenesis of SLE and EMS. Therefore, we performed GO and KEGG enrichment analysis. The results of GO analysis showed that these genes were mainly enriched in immune system process (3.35E-18), immune response (2.39E-23), defense response (1.17E-15) and cytoplasmic vesicle (1.62E-12) (Fig. 3 D). The results of KEGG pathway enrichment showed that the most involved pathways were Cell adhesion molecules (CAMs) (1.57E-06), NOD-like receptor signaling pathway (1.82E-06), Staphylococcus aureus infection (6.06E-07), and Inflammatory bowel disease (IBD) (5.58E-07) (Fig. 3 E). These results are closely related to immune response and inflammation. 2.3 PPI network construction and analysis The PPI network of shared genes was constructed from the STRING database to reveal the interactions of each protein. There are 79 nodes and 724 edges in Fig. 4 A, which were then analyzed using Cytoscape software. The top 5 significant genes with the highest ranking were protein tyrosine phosphatase receptor type C ( PTPRC ), C-X-C motif chemokine ligand 10 ( CXCL10 ), signal transducer and activator of transcription 1 ( STAT1 ), C-C motif chemokine ligand 2 ( CCL2 ), and CD163 molecule. To obtain significant gene modules, we used the MCODE plug-in of Cytoscape, which revealed that there are 33 genes tightly connected within the four clusters, as shown in Fig. 4 B. GO analysis showed that these genes are related to immune system process (5.32E-18), immune response (1.29E-20), regulation of immune system process (1.45E-22),and positive regulation of immune system process (1.36E-21) (Fig. 4 C). KEGG analysis showed that these genes are related to Cytokine-cytokine receptor interaction (2.87E-06), Leishmaniasis (1.27E-07), Th1 and Th2 cell differentiation (4.00E-07), and Staphylococcus aureus infection (5.15E-07) (Fig. 4 D). Similarly, these findings indicate that the immune response and inflammatory processes are significant factors in both diseases. 2.4 Selection and analysis of hub genes To screen for genes with the highest diagnostic value, we used three machine learning algorithms. Figure 5 A-B shows that, following the 10-fold cross-validation procedure, the LASSO regression identified 20 diagnostic genes for SLE in the model. Similarly, Fig. 5 C-D shows that the LASSO regression identified 9 diagnostic genes for EMS in the model. Recursive Feature Elimination (RFE) is a feature selection method that is commonly used in many classification algorithms, such as Support Vector Machines and Random Forests, where a model is fitted and the weakest features are removed until an optimal subset of features is reached. The expression values of 93 shared genes were used as feature values for model training. A total of 68 and 12 genes of EMS were screened by SVM-RFE and RF-RFE, respectively (Fig. 5 E-F). 2 and 22 genes of SLE were screened by SVM-RFE and RF-RFE, respectively (Fig. 5 G-H). We then combined the overlapping genes of the two diseases screened by three machine learning algorithms and obtained 23 genes (Fig. 6 A). We also performed GO and KEGG enrichment analysis (Fig. 6 B-C). The results showed that these genes were predominantly enriched in pathways that are associated with the immune response. Finally, the Venn diagram illustrating the intersection of 7 hub genes: CD79A , CD163 , TNFSF13B , HERC5 , LY96 , STAT1 and IL7R (Fig. 6 D). 2.5 Validation of expression levels of hub genes and assessment of diagnostic value In the SLE dataset, differential expression analysis revealed significant upregulation of STAT1, LY96, CD163, TNFSF13B and HERC5 in GSE50772 (validation dataset) compared to GSE72326 (Fig. 7 A-B). In addition, in the EMS dataset, the results showed that STAT1, IL7R, LY96, CD163, TNFSF13B and HERC5 were significantly upregulated in GSE7305 (validation dataset) compared to GSE23339 (Fig. 7 C-D). The diagnostic value of the hub genes was determined by plotting the receiver operating characteristic (ROC) curve in RStudio. Most of the hub genes had significant diagnostic value in disease classification, as shown in Fig. 8 . After careful consideration, we considered STAT1, LY96, CD163, TNFSF13 and HERC5 to be the hub genes for the next set of analyses. For these hub genes, we collected fresh whole blood samples from 18 patients, extracted PBMC, and performed qRT-PCR analysis to further validate the differential expression of the hub genes in the patient samples. As shown in Fig. 7 E, LY96 was significantly highly expressed in both diseases compared with normal subjects (P < 0.05), suggesting that LY96 has the potential to be the best shared diagnostic marker for both diseases.(PBMC extraction and qRT-PCR analysis are described in Supplementary Material 1. The other four hub genes were not statistically different in the two diseases, see Supplementary Material 2). 2.6 Construction of diagnostic model based on XGBoost Our aim is to develop a model that enhances the diagnostic and predictive efficiency of the disease. Although each of the hub genes has a good diagnostic efficiency (AUC > 0.75), we constructed a model based on these genes using a machine learning algorithm – XGBoost. The feature values for model training were obtained from the expression values of hub genes in GSE72326. Validation sets, including GSE50772, GSE23339, and GSE7305, were used to assess the model using AUROC curves and Precision-Recall curve as indicators. The AUC and PR of the training set were both 1.0 (Fig. 9 A). In the SLE validation set GSE50772, the AUC was 0.818 and the PR was 0.989 (Fig. 9 B). Additionally, in the EMS validation set GSE 23339 and GSE7305, the AUC was 0.85 and the PR was 0.99 (Fig. 9 C), and 0.96 and 1.0, respectively (Fig. 9 D). The results show that the model performs satisfactorily in identifying and predicting SLE and EMS. 2.7 Immune cell infiltration and correlation with hub genes Enrichment analyses indicate that the immune response plays a crucial role in both diseases. Therefore, we chose the ssGSEA method, which is based on 28 types of immune cells, to identify different immune infiltration patterns. In GSE72326 and GSE23339, 28 immune cells were identified and displayed in heat and box plots (Fig. 10 A-D). The result found that in SLE patients, all 28 types of immune cells differed significantly from normal samples. A higher proportion was accounted for by activated CD8 T cell, central memory CD4 T cell, myeloid-derived suppressor cell, and monocyte cell. In patients with EMS, most immune cell types were significantly different from normal samples. Four cell types were highly represented: central memory CD4 T cell, plasmacytoid dendritic cell, immature dendritic cell, and monocyte. Subsequently, we also explored the correlation between hub genes and immune cell. As shown in Fig. 10 E, in the SLE dataset, all five hub genes were positively correlated with central memory CD8 T cell, natural killer cell, activated dendritic cell, and immature dendritic cell. In the EMS dataset, three hub genes showed strong positive associations with most types of immune cells, including central memory CD8 T cell, follicular helper T cell, type 1 helper T cell, regulatory T cell, activated B cell, immature B cell, natural killer cell, myeloid derived suppressor cell, macrophage, and mast cell, except for HERC5 and STAT1 (Fig. 10 F). 3 Discussion The health risks to women of childbearing age, who are at high risk of EMS and SLE, should not be ignored, and a recent Mendelian randomization study has shown that patients with EMS are more likely to develop SLE( 6 )and previous studies have shown similarities between immune dysfunction in EMS and SLE. Therefore, we consider that patients with EMS and SLE may share a common pathogenesis; however, there is little research on this mechanism. Therefore, a systems biology approach was used to investigate potential common pathways of action in these two diseases. The findings suggest that dysregulated immune responses and inflammation may be the drivers of both diseases. Additionally, persistently elevated LY96 in both diseases may be the best diagnostic marker. In this study, we obtained 93 shared genes by identifying DEGs and important modular genes using WGCNA. Enrichment analysis of these genes showed that immune and inflammatory processes are closely related to these two diseases. It has been suggested that abnormal immune responses and inflammatory processes are involved in the pathogenesis of EMS and SLE ( 1 ) ( 27 ).In EMS, due to immune system does not recognize and target the endometrial tissue that develops outside the uterine cavity, leading to dysregulation of autoimmune responses, which in turn produces high levels of the inflammatory cytokines IL-1, IL-6, IL-8, and TNF, leading to the development of SLE ( 27 ). Furthermore, the inflammatory microenvironment of ectopic lesions activates sensory nerve endings, leading to increased secretion of inflammatory mediators. This results in the transmission of painful stimuli to the spinal cord, causing the development and persistence of chronic pain. This chronic pain, mediated by inflammation, may contribute to stress, and promote the onset and progression of SLE ( 28 ) ( 29 ).Meanwhile,TNF-α can promote the adherence of ectopic endometrial cells to the peritoneum ( 30 ).IL-6 and IL-1β in the peritoneal cavity favor the establishment, proliferation and migration of endometriotic cells( 31 ). Importantly, these cytokines in turn regulate the stimulation of other cytokines and chemokines, such as vascular endothelial growth factor (VEGF), which is involved in the development of blood vessels, essential for the growth and maintenance of ectopic endometrial implants( 32 ). Likewise, these pro-inflammatory cytokines have been demonstrated to have a significant correlation with SLE and could be used as biomarkers to evaluate disease activity and severity ( 33 ). We then performed GO and KEGG analysis on the genes screened by the MCODE module and machine learning, and the results strengthened our hypothesis. In addition, seven hub genes obtained by cytoscape software and machine learning showed good diagnostic value in both EMS and SLE, suggesting that they may be important targets for the development of EMS in SLE patients. Although each of the shared hub genes can be used as predictive biomarkers, however, changes in the expression levels of individual genes may not be specific to EMS or SLE, meaning that relying on a single gene alone for disease diagnosis or prognosis may introduce greater uncertainty. Thus, we prefer to develop a model that improves the diagnostic and predictive efficiency of the disease. The AUROC and PR curves results for the training and validation sets indicate that the model built with the XGBoost algorithm performs satisfactorily. Given the crucial role of the immune response in the development of EMS and SLE, we examined the variations in immune cell infiltration across different samples. These findings indicate a strong correlation between immune cell infiltration and both EMS and SLE, which is in line with prior research( 9 ). Finally, we also collected blood samples from patients for qRT-PCR validation of these hub genes, which showed that only the LY96 gene was consistently elevated in patients with EMS and SLE, and we concluded that LY96 may be the best co-diagnostic marker for EMS and SLE. LY96 , also known as myeloid differentiation 2 (MD2), is a co-receptor for Toll-like receptor 4 (TLR4) and is required for its activation. ( 34 ). Research has demonstrated that the up-regulation of TLR4 is pathogenic at both the protein and gene levels, leading to the onset of spontaneous SLE-like autoimmunity and lupus nephritis ( 35 ).Tai-Ping Lee et al. demonstrated in a transgenic mouse model that pathogenic anti-dsDNA and TLR4 activation induce severe SLE syndromes through the overproduction of IL-10 and IFN-γ ( 36 ). Interestingly, it has been shown that LY96 has also been suggested as a diagnostic marker for SLE ( 37 ).Likewise, the study concluded that LY96 is believed to have a significant impact on inflammatory and immune-related diseases, including rheumatoid arthritis, Crohn's disease, and inflammatory diabetic cardiomyopathy( 38 – 40 ). Bae et al. further verified the high expression of the LY96 gene in EMS by immunohistochemistry and western blot analysis, concluded that TLR4/NF-κB and Wnt/frizzled signaling pathways, as well as estrogen receptors, regulate the progression of EMS( 41 ). Bao Guo et al. suggested that activation of the TLR4 signaling pathway promotes the recruitment and activation of immune cells, triggering a local inflammatory response, leading to the secretion of inflammatory factors (e.g., TNF-α) and growth factors (e.g., VEGF), which stimulate the proliferation of endometrial cells, invasion, and cause further tissue damage ( 42 ).Therefore, we hypothesize that the shared pathogenesis of EMS and SLE is mediated by the TLR4 signaling pathway. Overexpression of LY96 may enhance the activation of the TLR4 signaling pathway, which triggers the production of downstream signals from cytokines and chemokines, leading to the initiation of inflammatory and immune responses. In addition, it has been reported that rs9514828 of the TNFSF13B gene and rs7574865 of STAT4 are associated with EMS and SLE, and 10 microRNAs, including miR-20a, miR-99a, miR-125a, miR-126, miR-141, miR-142-3p, miR-143, miR-145, miR-199, and miR-200b, are also associated with these two diseases( 43 ). Our study has limitations. Firstly, EMS is a disease with three different phenotypes (OMA, SUP and DIE) and we do not know which phenotype is associated with a higher risk of SLE presentation ( 44 ) Second, the small number of patients with EMS may increase the bias of the results. Finally, we only studied peripheral blood lymphocytes from patients. We therefore plan to identify phenotype-specific EMS and expand the collection of blood tissue samples in future studies to validate these results. 4 Conclusion This study is the first to use bioinformatics analysis to explore common pathways and genetic diagnostic markers involved in EMS and SLE. The LY96 gene, confirmed in clinical PBMC samples, may be a potential diagnostic marker for patients with EMS and SLE, and the TLR4 signaling pathway may be involved in the pathogenesis of both diseases. EMS may accelerate the development of SLE through immune and inflammatory pathways. The study may provide diagnostic markers and therapeutic targets for patients with EMS combined with SLE. 5 Materials and methods 5.1 Data selection Gene expression profiles in the GEO database ( http://www.ncbi.nlm.nih.gov/geo ) ( 19 ) were searched with the keywords "systemic lupus erythematosus" and "Endometriosis “. Datasets GSE72326 and GSE50772 were selected for SLE; GSE23339 and GSE7305 were selected for EMS. For SLE, dataset GSE72326 contains 131 (26 male patients excluded) SLE patients and 17 (3 male controls excluded) normal controls (Platform: GPL10558 Illumina HumanHT-12 V4.0 expression bead chip) and dataset GSE50772 includes 61 SLE samples and 20 healthy control samples (Platform: GPL570 Affymetrix Human Genome U133 Plus 2.0 Array). For EMS, dataset GSE23339 includes 10 EMS samples and 9 healthy control samples (Platform: GPL6102 Illumina human-6 v2.0 expression bead chip) and dataset GSE7305 includes 10 EMS samples and 10 healthy control samples (Platform: GPL570 Affymetrix Human Genome U133 Plus 2.0 Array).We used the GEOquery package to read the data and match the probes to their gene symbol based on the annotation documents of the respective platforms. Finally, the gene matrix was obtained with the row names as the gene symbols and the column names as the sample names for the subsequent analyses. The flowchart of this investigation is provided in Fig. 1 . 5.2 Weighted gene co-expression network analysis To identify gene co-expression modules associated with EMS, we conducted weighted gene co-expression network analysis (WGCNA) on the GSE23339 datasets. First, the 4768 genes exhibiting the most significant expression changes (the top 25% of rank genes with the largest variance) were selected for further analysis. Second, hierarchical cluster analysis was performed using the Hclust function to exclude outlier samples. Third, we selected an optimal soft threshold ranging from 1 to 20 using the ‘pickSoftThreshold’ function to build an adjacency matrix, which was then transformed into a topological overlap matrix (TOM). Four, detection of gene modules based on TOM using dynamic tree-cutting technique. Finally, the correlations between gene significance (GS) and module membership (MM) were calculated, and the corresponding module gene information was extracted for further analysis. 5.3 Identification of DEGs The Limma package was used to analyze differentially expressed genes( 20 ). Probe sets without a corresponding gene symbol were eliminated. Adjusted P < 0.05 and |log2FC| ≥ 0.5 were used as our standard screening criteria for DEGs. 5.4 Enrichment analyses of shared genes Total 93 shared genes were obtained by intersection of DEGs in GSE72326 and WGCNA module genes in GSE23339.The Venn diagram obtained by Sanger-box ( http://www.sangerbox.com/tool ) ( 21 ). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were used to ascertain the distinctive biological and functional attributes ( 22 ) ( 23 ). GO and KEGG analyses were executed using the Sanger-box online platform. Enriched GO categories and KEGG pathways were identified at P value < 0.05, and the top 10 most enriched were visualized. 5.5 PPI network construction and module analysis STRING is an online search tool for the exploring interactions among genes( 24 ). A total of 93 shared genes were integrated into the STRING database to construct the protein- protein interaction (PPI) network with a PPI score threshold (medium confidence ≥ 0.4). Cytoscape software was used for analysis and visualization of the PPI network( 25 ). A Cytoscape plug-in, the molecular complex detection technology (MCODE), was used to identify core functional modules. The following parameters were used: degree cutoff = 2, max depth = 100, node score cutoff = 0.2 and K-core = 2. 5.6 Hub genes selection First, four significant gene clustering modules − 33 genes - were obtained using the MCODE plug-in. Second, three machine learning methods were used to predict genes associated with EMS combined with SLE disease: LASSO, RFE-RF and RFE-SVM. Using the three machine learning algorithms on both EMS and SLE datasets, we derived sets of characteristic genes. After merging the overlap from the two datasets, we finally identified a total of 23 common characteristic genes, and the hub genes were finally identified by the intersection of the Cytoscape software and machine learning results. 5.7 Validation of expression levels of hub genes and assessment of diagnostic value The expression levels of hub genes (p < 0.05) were detected using box plots (GSE72326, GSE50772, GSE23339 and GSE7305) using the R package ggplot2. The AUC of ROC was used to determine the diagnostic value of potential biomarkers in the dataset (GSE72326 and GSE23339) using the R package pROC. qRT-PCR analysis of hub gene mRNA expression in peripheral blood mononuclear cells (PBMC) from EMS(n = 5), SLE(n = 5) and normal controls(n = 8). The study was approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University (Approval Number: 2024-E171-01), and all subjects gave written informed consent to participate in the study. The study was conducted in accordance with relevant guidelines and regulations. 5.8 Construction of diagnostic model based on hub genes The Extreme Gradient Boosting (XGBoost) model, based on the Python package xgboost, is a machine learning algorithm with excellent features and functions. It has become a rising star in the field of machine learning and is widely favored ( 26 ). We use the SLE dataset GSE72326 as the training set, GSE50772, GSE23339 and GSE7305 as the validation set. AUROC and PR curve are used as evaluation indicators of the model. 5.9 Immune cell infiltration and correlation with hub gene The single sample Gene Set Enrichment Analysis (ssGSEA)was used to explore the relationship between different degrees of infiltration of immune cell types in the GSE72326 and GSE23339 datasets, followed by non-parametric correlation (spearman) to analyze the correlation between the expression of hub genes and immune infiltrating cells. 5.10 Statistical analysis All statistical analyses of bioinformatics studies in this study were conducted using R software. The differences between the groups were tested using a nonparametric Wilcoxon signed-rank test. Correlation analysis was performed using Spearman’s correlation. In comparison, p < 0.05 was considered statistically significant. Declarations Data availability statement The public datasets were downloaded and analyzed in this study, which can be found in GEO data repository and included the accession numbers as follows: GSE72326, GSE23339, GSE50772, GSE7305. Ethics approval and consent to participate. Human samples protocols obtained approval from the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University. Author contributions X.Q. and P.H.C designed research; J.H., L.X.R., and L.C.L performed research (data acquisition, analysis, and interpretation); Q.Z.Z., Q.J.N., Y.Q.M.Z., and X.T. drafted and revised the paper; and all authors read and approved the final manuscript. Funding This study was supported by grants from the Key Laboratory of Early Prevention and Treatment for Regional High-Incidence-Tumor, Guangxi Medical University, Ministry of Education (GKE-ZZ202132). Acknowledgements We sincerely thank the researchers who uploaded their data to the GEO database. Competing interests The authors declare no competing interests. References Zhang T, De Carolis C, Man GCW, Wang CC. The link between immunity, autoimmunity and endometriosis: a literature update. Autoimmun Rev. 2018 Oct;17(10):945–55. Symons LK, Miller JE, Kay VR, Marks RM, Liblik K, Koti M, et al. The Immunopathophysiology of Endometriosis. Trends Mol Med. 2018 Sep;24(9):748–62. Kaul A, Gordon C, Crow MK, Touma Z, Urowitz MB, van Vollenhoven R, et al. Systemic lupus erythematosus. Nat Rev Dis Primer. 2016 Jun 16;2:16039. Gleicher N, el-Roeiy A, Confino E, Friberg J. Is endometriosis an autoimmune disease? Obstet Gynecol. 1987 Jul;70(1):115–22. Shigesi N, Kvaskoff M, Kirtley S, Feng Q, Fang H, Knight JC, et al. The association between endometriosis and autoimmune diseases: a systematic review and meta-analysis. Hum Reprod Update. 2019 Jul;25(4):486–503. Tang T, Zhong Y, Xu S, Yu H. Causal effects of endometriosis on SLE, RA and SS risk: evidence from meta-analysis and Mendelian randomization. BMC Pregnancy Childbirth. 2024 Feb 23;24:162. Fan YH, Leong PY, Chiou JY, Wang YH, Ku MH, Wei JCC. Association between endometriosis and risk of systemic lupus erythematosus. Sci Rep. 2021 Jan 12;11:532. Harris HR, Costenbader KH, Mu F, Kvaskoff M, Malspeis S, Karlson EW, et al. Endometriosis and the risks of systemic lupus erythematosus and rheumatoid arthritis in the Nurses’ Health Study II. Ann Rheum Dis. 2016 Jul;75(7):1279–84. Izumi G, Koga K, Takamura M, Makabe T, Satake E, Takeuchi A, et al. Involvement of immune cells in the pathogenesis of endometriosis. J Obstet Gynaecol Res. 2018;44(2):191–8. Badawy SZ, Cuenca V, Stitzel A, Tice D. Immune rosettes of T and B lymphocytes in infertile women with endometriosis. J Reprod Med. 1987 Mar;32(3):194–7. Chuang PC, Wu MH, Shoji Y, Tsai SJ. Downregulation of CD36 results in reduced phagocytic ability of peritoneal macrophages of women with endometriosis. J Pathol. 2009 Oct;219(2):232–41. Kang YJ, Jeung IC, Park A, Park YJ, Jung H, Kim TD, et al. An increased level of IL-6 suppresses NK cell activity in peritoneal fluid of patients with endometriosis via regulation of SHP-2 expression. Hum Reprod Oxf Engl. 2014 Oct 10;29(10):2176–89. Kwak JY, Park SW, Kim KH, Na YJ, Lee KS. Modulation of neutrophil apoptosis by plasma and peritoneal fluid from patients with advanced endometriosis. Hum Reprod Oxf Engl. 2002 Mar;17(3):595–600. Eisenberg VH, Zolti M, Soriano D. Is there an association between autoimmunity and endometriosis? Autoimmun Rev. 2012 Sep;11(11):806–14. Burney RO, Giudice LC. Pathogenesis and Pathophysiology of Endometriosis. Fertil Steril. 2012 Sep;98(3):10.1016/j.fertnstert.2012.06.029. Agic A, Xu H, Finas D, Banz C, Diedrich K, Hornung D. Is endometriosis associated with systemic subclinical inflammation? Gynecol Obstet Invest. 2006;62(3):139–47. Ytterberg SR, Schnitzer TJ. Inhibition of natural killer cell activity by serum from patients with systemic lupus erythematosus: roles of disease activity and serum interferon. Ann Rheum Dis. 1984 Jun;43(3):457–61. Kaplan MJ. Neutrophils in the pathogenesis and manifestations of SLE. Nat Rev Rheumatol. 2011 Sep 27;7(12):691–9. Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 2013 Jan;41(Database issue):D991-995. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015 Apr 20;43(7):e47. Shen W, Song Z, Zhong X, Huang M, Shen D, Gao P, et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. iMeta. 2022;1(3):e36. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000 May;25(1):25–9. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000 Jan 1;28(1):27–30. Szklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023 Jan 6;51(D1):D638–46. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003 Nov;13(11):2498–504. Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining [Internet]. New York, NY, USA: Association for Computing Machinery; 2016 [cited 2024 Feb 16]. p. 785–94. (KDD ’16). Available from: https://dl.acm.org/doi/10.1145/2939672.2939785 Bianco B, André GM, Vilarino FL, Peluso C, Mafra FA, Christofolini DM, et al. The possible role of genetic variants in autoimmune-related genes in the development of endometriosis. Hum Immunol. 2012 Mar 1;73(3):306–15. Maddern J, Grundy L, Castro J, Brierley SM. Pain in Endometriosis. Front Cell Neurosci. 2020;14:590823. Sharif K, Watad A, Coplan L, Lichtbroun B, Krosser A, Lichtbroun M, et al. The role of stress in the mosaic of autoimmunity: An overlooked association. Autoimmun Rev. 2018 Oct;17(10):967–83. Zhang RJ, Wild RA, Ojago JM. Effect of tumor necrosis factor-alpha on adhesion of human endometrial stromal cells to peritoneal mesothelial cells: an in vitro system. Fertil Steril. 1993 Jun;59(6):1196–201. Berbic M, Schulke L, Markham R, Tokushige N, Russell P, Fraser IS. Macrophage expression in endometrium of women with and without endometriosis. Hum Reprod Oxf Engl. 2009 Feb;24(2):325–32. Kyama CM, Debrock S, Mwenda JM, D’Hooghe TM. Potential involvement of the immune system in the development of endometriosis. Reprod Biol Endocrinol RBE. 2003 Dec 2;1:123. Richter P, Macovei LA, Mihai IR, Cardoneanu A, Burlui MA, Rezus E. Cytokines in Systemic Lupus Erythematosus—Focus on TNF-α and IL-17. Int J Mol Sci. 2023 Sep 22;24(19):14413. Chen L, Fu W, Zheng L, Wang Y, Liang G. Recent progress in the discovery of myeloid differentiation 2 (MD2) modulators for inflammatory diseases. Drug Discov Today. 2018 Jun 1;23(6):1187–202. Liu B, Yang Y, Dai J, Medzhitov R, Freudenberg MA, Zhang PL, et al. TLR4 up-regulation at protein or gene level is pathogenic for lupus-like autoimmune disease. J Immunol Baltim Md 1950. 2006 Nov 15;177(10):6880–8. Lee TP, Tang SJ, Wu MF, Song YC, Yu CL, Sun KH. Transgenic overexpression of anti-double-stranded DNA autoantibody and activation of Toll-like receptor 4 in mice induce severe systemic lupus erythematosus syndromes. J Autoimmun. 2010 Dec;35(4):358–67. Chen J, Zhao X, Huang C, Lin J. Novel insights into molecular signatures and pathogenic cell populations shared by systemic lupus erythematosus and vascular dementia. Funct Integr Genomics. 2023 Nov 16;23(4):337. Wang Y, Hwang J, Yadav D, Oda T, Lee PCW, Jin JO. Inhibitory effect of porphyran on lipopolysaccharide-induced activation of human immune cells. Carbohydr Polym. 2020 Mar 15;232:115811. Yang Y. Cancer immunotherapy: harnessing the immune system to battle cancer. J Clin Invest. 2015 Sep 1;125(9):3335–7. Rychkov D, Neely J, Oskotsky T, Yu S, Perlmutter N, Nititham J, et al. Cross-Tissue Transcriptomic Analysis Leveraging Machine Learning Approaches Identifies New Biomarkers for Rheumatoid Arthritis. Front Immunol. 2021 Jun 8;12:638066. Bae SJ, Jo Y, Cho MK, Jin JS, Kim JY, Shim J, et al. Identification and analysis of novel endometriosis biomarkers via integrative bioinformatics. Front Endocrinol. 2022 Oct 20;13:942368. Guo B, Chen J hua, Zhang J hui, Fang Y, Liu X jing, Zhang J, et al. Pattern-recognition receptors in endometriosis: A narrative review. Front Immunol. 2023 Mar 23;14:1161606. Zervou MI, Matalliotakis M, Goulielmos GN. Comment on “Risk of systemic lupus erythematosus in patients with endometriosis: a nationwide population‑based cohort study.” Arch Gynecol Obstet. 2022 Feb 1;305(2):543–4. Nisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril. 1997 Oct;68(4):585–96. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial1.docx Supplementary material Supplementary Materials S1 | Extraction of peripheral blood mononuclear cells (PBMC) and RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR). SupplementaryMaterial2.docx Supplementary Materials S2 | Expression of the other four hub genes. 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-4150400","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":286778394,"identity":"b8f02211-a10c-4970-8df5-4e328ed0c5c3","order_by":0,"name":"Jin Huang","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jin","middleName":"","lastName":"Huang","suffix":""},{"id":286778395,"identity":"dd94cca6-b61f-49c8-8395-369c02b8b364","order_by":1,"name":"Xuelian Ruan","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xuelian","middleName":"","lastName":"Ruan","suffix":""},{"id":286778396,"identity":"513d8e27-e36a-47c1-9d4a-c5c0b00d2d83","order_by":2,"name":"Yongling Chen","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yongling","middleName":"","lastName":"Chen","suffix":""},{"id":286778397,"identity":"e8252198-85ea-46b0-a158-c1c99a12b256","order_by":3,"name":"Ziqing Zhong","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ziqing","middleName":"","lastName":"Zhong","suffix":""},{"id":286778398,"identity":"26224a5b-37dd-4cd6-9a05-3e800b2bc831","order_by":4,"name":"Jiaqi Nie","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jiaqi","middleName":"","lastName":"Nie","suffix":""},{"id":286778399,"identity":"bd729215-55b3-4530-8d09-85b61a13cd75","order_by":5,"name":"Moqiyi Zeng","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Moqiyi","middleName":"","lastName":"Zeng","suffix":""},{"id":286778400,"identity":"221e6fef-4a82-4f96-ba9d-63fb98dd3bed","order_by":6,"name":"Xiang Tao","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Tao","suffix":""},{"id":286778401,"identity":"75bd638a-2a87-4cbd-a697-3531fcb1e80a","order_by":7,"name":"Xue Qin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYDACZjBpw8PP3tj44AMJWtJkJHsONxvOIMGuwzYGN9LbpDmIUct3nPeYNE/FYR6Dmw8bpBkY7OR0GwhokTzMl2zMcyadR/J2YoNxAUOysdkBAloMDvMYPuZts+bhA2pJnsFwIHEbEVoMDvP+Y+ZhuHmw4TAPkVqAtjQ48wjcYGxsJkqL5GEeY8M5x9J4JHsSmxlnGBDhF77zZ8wk3tTY2POzH3/+40OFnRxBLQyoCgwIKcfUMgpGwSgYBaMACwAAvA1BSRx6R3EAAAAASUVORK5CYII=","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":true,"prefix":"","firstName":"Xue","middleName":"","lastName":"Qin","suffix":""},{"id":286778402,"identity":"53d6b8bb-6358-48c1-bd28-32a84a293d40","order_by":8,"name":"Hua Ping Chen","email":"","orcid":"","institution":"First Affiliated Hospital of Guangxi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Hua","middleName":"Ping","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2024-03-22 14:22:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4150400/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4150400/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":54006358,"identity":"5794fc89-4c7b-4e1d-b8a3-fb6395a92dc0","added_by":"auto","created_at":"2024-04-03 09:30:51","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":408082,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of Investigation.\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/0ffe8828c7933ded53696f54.jpg"},{"id":54005136,"identity":"353ef211-a8df-4513-84cb-30cdaa486a4c","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":678384,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of module genes \u003cem\u003evia\u003c/em\u003e WGCNA in date set GSE23339.\u003c/p\u003e\n\u003cp\u003e(A) Clustering according to the expression level of EMS patients in GSE23339. Each branch represents a sample in the data sets, and there is no outlier sample in each data set.\u003c/p\u003e\n\u003cp\u003e(B) The cluster dendrogram of co-expression in GSE23339.\u003c/p\u003e\n\u003cp\u003e(C) Determination of Soft Threshold power for GSE23339.\u003c/p\u003e\n\u003cp\u003e(D) Heatmap of the correlation between module eigengenes and the occurrence of EMS.\u003c/p\u003e\n\u003cp\u003e(E-F) The correlation between module membership and gene significance in EMS regarding the most positively correlated modules.\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/ac894bf32265012522f6c9b1.jpg"},{"id":54005844,"identity":"89b15e10-2b5f-4b31-ae93-44fc45a50fcf","added_by":"auto","created_at":"2024-04-03 09:22:51","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":726585,"visible":true,"origin":"","legend":"\u003cp\u003eScreening for differentially expressed genes and obtaining enrichment analyses of 93 shared genes.\u003c/p\u003e\n\u003cp\u003e(A) The volcano plot of DEGs in date set GSE72326.\u003c/p\u003e\n\u003cp\u003e(B) The heatmap plots of DEGs in GSE72326.\u003c/p\u003e\n\u003cp\u003e(C) Total 93 genes were obtained by intersection of EMS Positive module genes obtained by WGCNA in date set GSE23339 and DEGs from SLE in date set GSE72326.\u003c/p\u003e\n\u003cp\u003e(D-E) GO and KEGG enrichment analyses of the 93 genes. The size of the circle represents the number of genes involved, and the abscissa represents the frequency of the genes involved in the term total genes. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/c68004d771731876295229ed.jpg"},{"id":54005134,"identity":"12713a43-fe59-49ef-8441-a7fe7a8817ce","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":684696,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of PPI network and enrichment analysis of modular genes.\u003c/p\u003e\n\u003cp\u003e(A) PPI network of shared genes. The network has 79 nodes and 724 edges. PPI, protein–protein interaction.\u003c/p\u003e\n\u003cp\u003e(B) Four significant gene clustering modules were obtained by MCODE plug-in.\u003c/p\u003e\n\u003cp\u003e(C-D) GO and KEGG enrichment analyses of the four cluster of 33 genes.\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/cf87788a36f0a7b9ad7fe422.jpg"},{"id":54005842,"identity":"7b5081c8-3e88-4bc5-83ce-6e2b74eea030","added_by":"auto","created_at":"2024-04-03 09:22:51","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":696799,"visible":true,"origin":"","legend":"\u003cp\u003eScreening genes in EMS and SLE with three machine learning methods, respectively.\u003c/p\u003e\n\u003cp\u003e(A) Coefficient profiles of variables in the LASSO regression model in SLE.\u003c/p\u003e\n\u003cp\u003e(B) Ten-fold cross-validation for turning parameter, (\u003cem\u003eλ\u003c/em\u003e) selection in the LASSO regression model in SLE.\u003c/p\u003e\n\u003cp\u003e(C) Coefficient profiles of variables in the LASSO regression model in EMS.\u003c/p\u003e\n\u003cp\u003e(D) Ten-fold cross-validation for turning parameter, (\u003cem\u003eλ\u003c/em\u003e) selection in the LASSO regression model in EMS.\u003c/p\u003e\n\u003cp\u003e(E) \u0026nbsp;The Support Vector Machine - Recursive Feature Elimination (SVM-RFE) to screen best feature subset in EMS.\u003c/p\u003e\n\u003cp\u003e(F) The Random Forest-Recursive Feature Elimination (RF-RFE) to screen best feature subset in EMS.\u003c/p\u003e\n\u003cp\u003e(G) The Support Vector Machine - Recursive Feature Elimination (SVM-RFE) to screen best feature subset in SLE.\u003c/p\u003e\n\u003cp\u003e(H) The Random Forest-Recursive Feature Elimination (RF-RFE) to screen best feature subset in SLE. RMSE, the optimum root mean squared error.\u003c/p\u003e","description":"","filename":"Picture5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/77b160ddf340c18e0421665a.jpg"},{"id":54005841,"identity":"23ee6209-2a9b-4701-81ba-2a2848afc9d2","added_by":"auto","created_at":"2024-04-03 09:22:51","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":525306,"visible":true,"origin":"","legend":"\u003cp\u003eEnrichment analysis was conducted on the shared genes of machine learning and hub genes that were obtained.\u003c/p\u003e\n\u003cp\u003e(A) After merging the overlapping genes of the two diseases screened by three machine learning algorithms, 23 genes were obtained.\u003c/p\u003e\n\u003cp\u003e(B-C) GO and KEGG enrichment analyses of the 23 genes.\u003c/p\u003e\n\u003cp\u003e(D)Venn diagram demonstrates 7 hub genes obtained by the intersection of cytoscape software and machine learning algorithms.\u003c/p\u003e","description":"","filename":"Picture6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/c7fff816800a8fa33a7c7ad4.jpg"},{"id":54005142,"identity":"76625659-ace4-4b7d-bc23-4ac1a6befeb6","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":518944,"visible":true,"origin":"","legend":"\u003cp\u003eScreening genes in EMS and SLE with three machine learning methods, respectively.\u003c/p\u003e\n\u003cp\u003e(A) Expression of hub genes verified in GSE72326.\u003c/p\u003e\n\u003cp\u003e(B) Expression of hub genes verified in GSE50772.\u003c/p\u003e\n\u003cp\u003e(C) Expression of hub genes verified in GSE23339.\u003c/p\u003e\n\u003cp\u003e(D) Expression of hub genes verified in GSE7305.\u003c/p\u003e\n\u003cp\u003e(E) qRT-PCR analysis of mRNA expression levels of LY96 in PBMC from patients and healthy controls. The comparison in the two sets of data used the Mann-Whitney test, separately; \u003cem\u003ep\u003c/em\u003e -value \u0026lt; 0.05 was considered statistically significant. ns, not significant, *p \u0026lt; 0.05, **p \u0026lt; 0.01, ***p \u0026lt; 0.001, ****p \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"Picture7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/d1ddb3de936b3b870c1dd3bd.jpg"},{"id":54005140,"identity":"dbe27661-9858-4aec-8729-9a12b5660240","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":607466,"visible":true,"origin":"","legend":"\u003cp\u003eDiagnostic value of Hub genes in EMS and SLE.\u003c/p\u003e\n\u003cp\u003e(A, B, C, D, E, F,G) ROC curves of STAT1,CD79A，IL7R，LY96,CD163,TNFSF13B,HERC5 in the SLE dataset GSE72326, respectively.\u003c/p\u003e\n\u003cp\u003e(H, I, J, K, L, M,N) ROC curves of STAT1,CD79A，IL7R，LY96,CD163,TNFSF13B,HERC5 in the EMS dataset GSE23339, respectively.\u003c/p\u003e","description":"","filename":"Picture8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/6ad681bc32e6efb55f3975b5.jpg"},{"id":54005143,"identity":"c508bddb-d8ba-4feb-b343-1e08ea0a8429","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":570231,"visible":true,"origin":"","legend":"\u003cp\u003ePerformance of the XGBoost diagnostic model.\u003c/p\u003e\n\u003cp\u003e(A) Performance in the training set (GSE72326) using XGBoost.\u003c/p\u003e\n\u003cp\u003e(B) Performance in the validation set (GSE50772) using XGBoost.\u003c/p\u003e\n\u003cp\u003e(C) Performance in the validation set (GSE23339) using XGBoost.\u003c/p\u003e\n\u003cp\u003e(D) Performance in the validation set (GSE7305) using XGBoost.\u003c/p\u003e","description":"","filename":"Picture9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/cec2c91c8fa43e7ae1d286d5.jpg"},{"id":54005845,"identity":"d9d23860-5e4f-4ca6-9588-990500d2e77f","added_by":"auto","created_at":"2024-04-03 09:22:51","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":884359,"visible":true,"origin":"","legend":"\u003cp\u003eThe immune landscape of EMS and SLE and correlation analysis.\u003c/p\u003e\n\u003cp\u003e(A) Heatmap of 28 immune cell expression scores in SLE.\u003c/p\u003e\n\u003cp\u003e(B) Heatmap of 28 immune cell expression scores in EMS.\u003c/p\u003e\n\u003cp\u003e(C) Composition of infiltration of 28 immune cells in SLE.\u003c/p\u003e\n\u003cp\u003e(D) Composition of infiltration of 28 immune cells in EMS.\u003c/p\u003e\n\u003cp\u003e(E) The correlation between 5 hub genes and immune cells in SLE.\u003c/p\u003e\n\u003cp\u003e(F) The correlation between 5 hub genes and immune cells in EMS. Red: positive correlation; blue: negative correlation. \u0026nbsp;ns, not significant, *p \u0026lt; 0.05, **p \u0026lt; 0.01, ***p \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"Picture10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/0aa56e2620009818c009dda2.jpg"},{"id":58909831,"identity":"8972beba-c5da-4440-bf06-1e62e5425bac","added_by":"auto","created_at":"2024-06-24 04:14:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6936993,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/86222aa6-d688-4ba3-8459-59807e1b5502.pdf"},{"id":54005132,"identity":"6a86c160-edcb-4221-80e7-db18243e1052","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":14934,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary Materials S1 | \u003c/strong\u003eExtraction of peripheral blood mononuclear cells (PBMC) and RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR).\u003c/p\u003e","description":"","filename":"SupplementaryMaterial1.docx","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/cd81829075c8351533c29f4a.docx"},{"id":54005139,"identity":"094575f2-f86c-4ca6-8b66-5eddfd11935e","added_by":"auto","created_at":"2024-04-03 09:14:51","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":139167,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Materials S2 | \u003c/strong\u003eExpression of the other four hub genes\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"SupplementaryMaterial2.docx","url":"https://assets-eu.researchsquare.com/files/rs-4150400/v1/ca137c798c79010c3b795e76.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploration of the shared pathways and common biomarker LY96 in Endometriosis and Systemic lupus erythematosus using integrated bioinformatics analysis","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eEndometriosis(EMS) affects 5\u0026ndash;10% of women of reproductive age in the United States and is characterized by the presence of endometrioid tissue outside the uterine cavity, mainly in the pelvic peritoneum and ovaries (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). The cause of EMS is not yet fully understood, but there is evidence to suggest that it may be related to dysfunction in the immune system(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Systemic lupus erythematosus (SLE) is an autoimmune disease that affects multiple organs. Its pathogenesis involves abnormalities in the innate and adaptive immune system, resulting in inadequate responses to self-antigens(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Women of childbearing age are often susceptible to the disease, in a similar way to women with EMS. In 1987,Gleicher et al. proposed the theory that EMS may be an autoimmune disease, and since then there has been increasing evidence of a link between EMS and autoimmune diseases(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). However, the relationship between EMS and SLE has not been thoroughly researched, but there is evidence that women with EMS are at increased risk of developing SLE (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Importantly, we have observed similarities between immune dysfunction in these two diseases. Women with EMS show deficits in immune surveillance(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e),such as abnormal activation of T-cell, B-cell(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), high natural killer cell and macrophage cell counts with reduced activity(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), decreased rate of neutrophils apoptosis(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e), increased incidence of serum autoantibodies(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), and elevated inflammatory cytokines, including interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-alpha), and interleukin-1 (IL-1)).(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). As in EMS, abnormal activation of T-cell, B-cell, the production of autoantibodies and inflammatory cytokines, reduced natural killer cell activity(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), and dysregulated neutrophil phagocytosis(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e) are important features of the pathophysiology of SLE. Notably, these studies tended to adopt a clinical approach and could not uncover molecular mechanisms at the gene level. With the rapid development of bioinformatics methods, common transcriptional features based on the molecular level of genes may provide new insights into the common pathogenesis of EMS and SLE. Therefore, in this study, we performed a comprehensive bioinformatics analysis combined with machine learning algorithms to identify the hub genes and pathways of EMS and SLE from the GEO database. In addition, we investigated the correlation between hub genes and immune cells in SLE and EMS. Finally, we validated the expression of the hub genes in an external dataset. In conclusion, this may be the first study to establish common biomarkers and associated therapeutic targets for SLE and EMS, which may provide clues for exploring the genetic aetiology and combined therapeutic strategies for SLE and EMS.\u003c/p\u003e"},{"header":"2 Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Weighted gene co-expression network analysis of EMS\u003c/h2\u003e \u003cp\u003eWGCNA was used to explore the relationship between clinical traits and genes. First, in the GSE23339 datasets, all samples were clustered, and none were eliminated (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Then, β\u0026thinsp;=\u0026thinsp;20 (scale-free R2\u0026thinsp;=\u0026thinsp;0.9) was selected as the optimal soft power value (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Next, a total of 6 gene co-expression modules (GCMs) were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Furthermore, the correlations between the GCMs and the clinical features were calculated. The blue and yellow modules had high positive correlations with EMS (r\u0026thinsp;=\u0026thinsp;0.59, p\u0026thinsp;=\u0026thinsp;0.001 and r\u0026thinsp;=\u0026thinsp;0.56, p\u0026thinsp;=\u0026thinsp;0.01) and included 1530 genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). Finally, a significant correlation between blue (r\u0026thinsp;=\u0026thinsp;0.54) and yellow (r\u0026thinsp;=\u0026thinsp;0.2) module membership and gene significance for EMS was observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE-F). Thus, the 1530 genes in the blue and yellow modules that are most associated with EMS were identified as crucial for further experiments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Identification and Enrichment analysis of shared genes between SLE and EMS\u003c/h2\u003e \u003cp\u003eThe R package limma was used to identify differentially expressed genes (DEGs) in the GSE72326 dataset. A total of 387 DEGs were identified, of which 116 were upregulated and 271 of these genes were downregulated showed in volcano plot (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Of these DEGs, the top 50 differentially expressed genes are displayed in heatmaps (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The DEGs and module genes were then intersected, and 93 shared genes were obtained (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). These genes may be related to the pathogenesis of SLE and EMS. Therefore, we performed GO and KEGG enrichment analysis. The results of GO analysis showed that these genes were mainly enriched in immune system process (3.35E-18), immune response (2.39E-23), defense response (1.17E-15) and cytoplasmic vesicle (1.62E-12) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). The results of KEGG pathway enrichment showed that the most involved pathways were Cell adhesion molecules (CAMs) (1.57E-06), NOD-like receptor signaling pathway (1.82E-06), Staphylococcus aureus infection (6.06E-07), and Inflammatory bowel disease (IBD) (5.58E-07) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). These results are closely related to immune response and inflammation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 PPI network construction and analysis\u003c/h2\u003e \u003cp\u003eThe PPI network of shared genes was constructed from the STRING database to reveal the interactions of each protein. There are 79 nodes and 724 edges in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, which were then analyzed using Cytoscape software. The top 5 significant genes with the highest ranking were protein tyrosine phosphatase receptor type C (\u003cem\u003ePTPRC\u003c/em\u003e), C-X-C motif chemokine ligand 10 (\u003cem\u003eCXCL10\u003c/em\u003e), signal transducer and activator of transcription 1 (\u003cem\u003eSTAT1\u003c/em\u003e), C-C motif chemokine ligand 2 (\u003cem\u003eCCL2\u003c/em\u003e), and \u003cem\u003eCD163\u003c/em\u003e molecule. To obtain significant gene modules, we used the MCODE plug-in of Cytoscape, which revealed that there are 33 genes tightly connected within the four clusters, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB. GO analysis showed that these genes are related to immune system process (5.32E-18), immune response (1.29E-20), regulation of immune system process (1.45E-22),and positive regulation of immune system process (1.36E-21) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). KEGG analysis showed that these genes are related to Cytokine-cytokine receptor interaction (2.87E-06), Leishmaniasis (1.27E-07), Th1 and Th2 cell differentiation (4.00E-07), and Staphylococcus aureus infection (5.15E-07) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Similarly, these findings indicate that the immune response and inflammatory processes are significant factors in both diseases.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Selection and analysis of hub genes\u003c/h2\u003e \u003cp\u003eTo screen for genes with the highest diagnostic value, we used three machine learning algorithms. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-B shows that, following the 10-fold cross-validation procedure, the LASSO regression identified 20 diagnostic genes for SLE in the model. Similarly, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC-D shows that the LASSO regression identified 9 diagnostic genes for EMS in the model. Recursive Feature Elimination (RFE) is a feature selection method that is commonly used in many classification algorithms, such as Support Vector Machines and Random Forests, where a model is fitted and the weakest features are removed until an optimal subset of features is reached. The expression values of 93 shared genes were used as feature values for model training. A total of 68 and 12 genes of EMS were screened by SVM-RFE and RF-RFE, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE-F). 2 and 22 genes of SLE were screened by SVM-RFE and RF-RFE, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG-H). We then combined the overlapping genes of the two diseases screened by three machine learning algorithms and obtained 23 genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). We also performed GO and KEGG enrichment analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB-C). The results showed that these genes were predominantly enriched in pathways that are associated with the immune response. Finally, the Venn diagram illustrating the intersection of 7 hub genes: \u003cem\u003eCD79A\u003c/em\u003e, \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eTNFSF13B\u003c/em\u003e, \u003cem\u003eHERC5\u003c/em\u003e, \u003cem\u003eLY96\u003c/em\u003e, \u003cem\u003eSTAT1\u003c/em\u003e and \u003cem\u003eIL7R\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Validation of expression levels of hub genes and assessment of diagnostic value\u003c/h2\u003e \u003cp\u003eIn the SLE dataset, differential expression analysis revealed significant upregulation of STAT1, LY96, CD163, TNFSF13B and HERC5 in GSE50772 (validation dataset) compared to GSE72326 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA-B). In addition, in the EMS dataset, the results showed that \u003cem\u003eSTAT1, IL7R, LY96, CD163, TNFSF13B\u003c/em\u003e and \u003cem\u003eHERC5\u003c/em\u003e were significantly upregulated in GSE7305 (validation dataset) compared to GSE23339 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC-D). The diagnostic value of the hub genes was determined by plotting the receiver operating characteristic (ROC) curve in RStudio. Most of the hub genes had significant diagnostic value in disease classification, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e. After careful consideration, we considered \u003cem\u003eSTAT1, LY96, CD163, TNFSF13\u003c/em\u003e and \u003cem\u003eHERC5\u003c/em\u003e to be the hub genes for the next set of analyses. For these hub genes, we collected fresh whole blood samples from 18 patients, extracted PBMC, and performed qRT-PCR analysis to further validate the differential expression of the hub genes in the patient samples. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE, LY96 was significantly highly expressed in both diseases compared with normal subjects (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), suggesting that \u003cem\u003eLY96\u003c/em\u003e has the potential to be the best shared diagnostic marker for both diseases.(PBMC extraction and qRT-PCR analysis are described in Supplementary Material 1. The other four hub genes were not statistically different in the two diseases, see Supplementary Material 2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Construction of diagnostic model based on XGBoost\u003c/h2\u003e \u003cp\u003eOur aim is to develop a model that enhances the diagnostic and predictive efficiency of the disease. Although each of the hub genes has a good diagnostic efficiency (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.75), we constructed a model based on these genes using a machine learning algorithm \u0026ndash; XGBoost. The feature values for model training were obtained from the expression values of hub genes in GSE72326. Validation sets, including GSE50772, GSE23339, and GSE7305, were used to assess the model using AUROC curves and Precision-Recall curve as indicators. The AUC and PR of the training set were both 1.0 (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eA). In the SLE validation set GSE50772, the AUC was 0.818 and the PR was 0.989 (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eB). Additionally, in the EMS validation set GSE 23339 and GSE7305, the AUC was 0.85 and the PR was 0.99 (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eC), and 0.96 and 1.0, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eD). The results show that the model performs satisfactorily in identifying and predicting SLE and EMS.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Immune cell infiltration and correlation with hub genes\u003c/h2\u003e \u003cp\u003eEnrichment analyses indicate that the immune response plays a crucial role in both diseases. Therefore, we chose the ssGSEA method, which is based on 28 types of immune cells, to identify different immune infiltration patterns. In GSE72326 and GSE23339, 28 immune cells were identified and displayed in heat and box plots (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eA-D). The result found that in SLE patients, all 28 types of immune cells differed significantly from normal samples. A higher proportion was accounted for by activated CD8 T cell, central memory CD4 T cell, myeloid-derived suppressor cell, and monocyte cell. In patients with EMS, most immune cell types were significantly different from normal samples. Four cell types were highly represented: central memory CD4 T cell, plasmacytoid dendritic cell, immature dendritic cell, and monocyte. Subsequently, we also explored the correlation between hub genes and immune cell. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eE, in the SLE dataset, all five hub genes were positively correlated with central memory CD8 T cell, natural killer cell, activated dendritic cell, and immature dendritic cell. In the EMS dataset, three hub genes showed strong positive associations with most types of immune cells, including central memory CD8 T cell, follicular helper T cell, type 1 helper T cell, regulatory T cell, activated B cell, immature B cell, natural killer cell, myeloid derived suppressor cell, macrophage, and mast cell, except for \u003cem\u003eHERC5\u003c/em\u003e and \u003cem\u003eSTAT1\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eF).\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Discussion","content":"\u003cp\u003eThe health risks to women of childbearing age, who are at high risk of EMS and SLE, should not be ignored, and a recent Mendelian randomization study has shown that patients with EMS are more likely to develop SLE(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e)and previous studies have shown similarities between immune dysfunction in EMS and SLE. Therefore,\u003c/p\u003e \u003cp\u003ewe consider that patients with EMS and SLE may share a common pathogenesis; however, there is little research on this mechanism. Therefore, a systems biology approach was used to investigate potential common pathways of action in these two diseases. The findings suggest that dysregulated immune responses and inflammation may be the drivers of both diseases. Additionally, persistently elevated LY96 in both diseases may be the best diagnostic marker. In this study, we obtained 93 shared genes by identifying DEGs and important modular genes using WGCNA. Enrichment analysis of these genes showed that immune and inflammatory processes are closely related to these two diseases. It has been suggested that abnormal immune responses and inflammatory processes are involved in the pathogenesis of EMS and SLE (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e).In EMS, due to immune system does not recognize and target the endometrial tissue that develops outside the uterine cavity, leading to dysregulation of autoimmune responses, which in turn produces high levels of the inflammatory cytokines IL-1, IL-6, IL-8, and TNF, leading to the development of SLE (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Furthermore, the inflammatory microenvironment of ectopic lesions activates sensory nerve endings, leading to increased secretion of inflammatory mediators. This results in the transmission of painful stimuli to the spinal cord, causing the development and persistence of chronic pain. This chronic pain, mediated by inflammation, may contribute to stress, and promote the onset and progression of SLE (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e).Meanwhile,TNF-α can promote the adherence of ectopic endometrial cells to the peritoneum (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e).IL-6 and IL-1β in the peritoneal cavity favor the establishment, proliferation and migration of endometriotic cells(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Importantly, these cytokines in turn regulate the stimulation of other cytokines and chemokines, such as vascular endothelial growth factor (VEGF), which is involved in the development of blood vessels, essential for the growth and maintenance of ectopic endometrial implants(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). Likewise, these pro-inflammatory cytokines have been demonstrated to have a significant correlation with SLE and could be used as biomarkers to evaluate disease activity and severity (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe then performed GO and KEGG analysis on the genes screened by the MCODE module and machine learning, and the results strengthened our hypothesis. In addition, seven hub genes obtained by cytoscape software and machine learning showed good diagnostic value in both EMS and SLE, suggesting that they may be important targets for the development of EMS in SLE patients. Although each of the shared hub genes can be used as predictive biomarkers, however, changes in the expression levels of individual genes may not be specific to EMS or SLE, meaning that relying on a single gene alone for disease diagnosis or prognosis may introduce greater uncertainty. Thus, we prefer to develop a model that improves the diagnostic and predictive efficiency of the disease. The AUROC and PR curves results for the training and validation sets indicate that the model built with the XGBoost algorithm performs satisfactorily. Given the crucial role of the immune response in the development of EMS and SLE, we examined the variations in immune cell infiltration across different samples. These findings indicate a strong correlation between immune cell infiltration and both EMS and SLE, which is in line with prior research(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Finally, we also collected blood samples from patients for qRT-PCR validation of these hub genes, which showed that only the \u003cem\u003eLY96\u003c/em\u003e gene was consistently elevated in patients with EMS and SLE, and we concluded that \u003cem\u003eLY96\u003c/em\u003e may be the best co-diagnostic marker for EMS and SLE. \u003cem\u003eLY96\u003c/em\u003e, also known as myeloid differentiation 2 (MD2), is a co-receptor for Toll-like receptor 4 (TLR4) and is required for its activation. (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). Research has demonstrated that the up-regulation of TLR4 is pathogenic at both the protein and gene levels, leading to the onset of spontaneous SLE-like autoimmunity and lupus nephritis (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e).Tai-Ping Lee et al. demonstrated in a transgenic mouse model that pathogenic anti-dsDNA and TLR4 activation induce severe SLE syndromes through the overproduction of IL-10 and IFN-γ (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Interestingly, it has been shown that \u003cem\u003eLY96\u003c/em\u003e has also been suggested as a diagnostic marker for SLE (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e).Likewise, the study concluded that \u003cem\u003eLY96\u003c/em\u003e is believed to have a significant impact on inflammatory and immune-related diseases, including rheumatoid arthritis, Crohn's disease, and inflammatory diabetic cardiomyopathy(\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Bae et al. further verified the high expression of the \u003cem\u003eLY96\u003c/em\u003e gene in EMS by immunohistochemistry and western blot analysis, concluded that TLR4/NF-κB and Wnt/frizzled signaling pathways, as well as estrogen receptors, regulate the progression of EMS(\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). Bao Guo et al. suggested that activation of the TLR4 signaling pathway promotes the recruitment and activation of immune cells, triggering a local inflammatory response, leading to the secretion of inflammatory factors (e.g., TNF-α) and growth factors (e.g., VEGF), which stimulate the proliferation of endometrial cells, invasion, and cause further tissue damage (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e).Therefore, we hypothesize that the shared pathogenesis of EMS and SLE is mediated by the TLR4 signaling pathway. Overexpression of \u003cem\u003eLY96\u003c/em\u003e may enhance the activation of the TLR4 signaling pathway, which triggers the production of downstream signals from cytokines and chemokines, leading to the initiation of inflammatory and immune responses. In addition, it has been reported that rs9514828 of the \u003cem\u003eTNFSF13B\u003c/em\u003e gene and rs7574865 of STAT4 are associated with EMS and SLE, and 10 microRNAs, including miR-20a, miR-99a, miR-125a, miR-126, miR-141, miR-142-3p, miR-143, miR-145, miR-199, and miR-200b, are also associated with these two diseases(\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). Our study has limitations. Firstly, EMS is a disease with three different phenotypes (OMA, SUP and DIE) and we do not know which phenotype is associated with a higher risk of SLE presentation (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e) Second, the small number of patients with EMS may increase the bias of the results. Finally, we only studied peripheral blood lymphocytes from patients. We therefore plan to identify phenotype-specific EMS and expand the collection of blood tissue samples in future studies to validate these results.\u003c/p\u003e"},{"header":"4 Conclusion","content":"\u003cp\u003eThis study is the first to use bioinformatics analysis to explore common pathways and genetic diagnostic markers involved in EMS and SLE. The LY96 gene, confirmed in clinical PBMC samples, may be a potential diagnostic marker for patients with EMS and SLE, and the TLR4 signaling pathway may be involved in the pathogenesis of both diseases. EMS may accelerate the development of SLE through immune and inflammatory pathways. The study may provide diagnostic markers and therapeutic targets for patients with EMS combined with SLE.\u003c/p\u003e"},{"header":"5 Materials and methods","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Data selection\u003c/h2\u003e \u003cp\u003eGene expression profiles in the GEO database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.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) (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) were searched with the keywords \"systemic lupus erythematosus\" and \"Endometriosis \u0026ldquo;. Datasets GSE72326 and GSE50772 were selected for SLE; GSE23339 and GSE7305 were selected for EMS. For SLE, dataset GSE72326 contains 131 (26 male patients excluded) SLE patients and 17 (3 male controls excluded) normal controls (Platform: GPL10558 Illumina HumanHT-12 V4.0 expression bead chip) and dataset GSE50772 includes 61 SLE samples and 20 healthy control samples (Platform: GPL570 Affymetrix Human Genome U133 Plus 2.0 Array). For EMS, dataset GSE23339 includes 10 EMS samples and 9 healthy control samples (Platform: GPL6102 Illumina human-6 v2.0 expression bead chip) and dataset GSE7305 includes 10 EMS samples and 10 healthy control samples (Platform: GPL570 Affymetrix Human Genome U133 Plus 2.0 Array).We used the GEOquery package to read the data and match the probes to their gene symbol based on the annotation documents of the respective platforms. Finally, the gene matrix was obtained with the row names as the gene symbols and the column names as the sample names for the subsequent analyses. The flowchart of this investigation is provided in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Weighted gene co-expression network analysis\u003c/h2\u003e \u003cp\u003eTo identify gene co-expression modules associated with EMS, we conducted weighted gene co-expression network analysis (WGCNA) on the GSE23339 datasets. First, the 4768 genes exhibiting the most significant expression changes (the top 25% of rank genes with the largest variance) were selected for further analysis. Second, hierarchical cluster analysis was performed using the Hclust function to exclude outlier samples. Third, we selected an optimal soft threshold ranging from 1 to 20 using the \u0026lsquo;pickSoftThreshold\u0026rsquo; function to build an adjacency matrix, which was then transformed into a topological overlap matrix (TOM). Four, detection of gene modules based on TOM using dynamic tree-cutting technique. Finally, the correlations between gene significance (GS) and module membership (MM) were calculated, and the corresponding module gene information was extracted for further analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Identification of DEGs\u003c/h2\u003e \u003cp\u003eThe Limma package was used to analyze differentially expressed genes(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Probe sets without a corresponding gene symbol were eliminated. Adjusted P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log2FC| \u0026ge; 0.5 were used as our standard screening criteria for DEGs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e5.4 Enrichment analyses of shared genes\u003c/h2\u003e \u003cp\u003eTotal 93 shared genes were obtained by intersection of DEGs in GSE72326 and WGCNA module genes in GSE23339.The Venn diagram obtained by Sanger-box (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.sangerbox.com/tool\u003c/span\u003e\u003cspan address=\"http://www.sangerbox.com/tool\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were used to ascertain the distinctive biological and functional attributes (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e) (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). GO and KEGG analyses were executed using the Sanger-box online platform. Enriched GO categories and KEGG pathways were identified at P value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, and the top 10 most enriched were visualized.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e5.5 PPI network construction and module analysis\u003c/h2\u003e \u003cp\u003eSTRING is an online search tool for the exploring interactions among genes(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). A total of 93 shared genes were integrated into the STRING database to construct the protein- protein interaction (PPI) network with a PPI score threshold (medium confidence\u0026thinsp;\u0026ge;\u0026thinsp;0.4). Cytoscape software was used for analysis and visualization of the PPI network(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). A Cytoscape plug-in, the molecular complex detection technology (MCODE), was used to identify core functional modules. The following parameters were used: degree cutoff\u0026thinsp;=\u0026thinsp;2, max depth\u0026thinsp;=\u0026thinsp;100, node score cutoff\u0026thinsp;=\u0026thinsp;0.2 and K-core\u0026thinsp;=\u0026thinsp;2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e5.6 Hub genes selection\u003c/h2\u003e \u003cp\u003eFirst, four significant gene clustering modules \u0026minus;\u0026thinsp;33 genes - were obtained using the MCODE plug-in. Second, three machine learning methods were used to predict genes associated with EMS combined with SLE disease: LASSO, RFE-RF and RFE-SVM. Using the three machine learning algorithms on both EMS and SLE datasets, we derived sets of characteristic genes. After merging the overlap from the two datasets, we finally identified a total of 23 common characteristic genes, and the hub genes were finally identified by the intersection of the Cytoscape software and machine learning results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e5.7 Validation of expression levels of hub genes and assessment of diagnostic value\u003c/h2\u003e \u003cp\u003eThe expression levels of hub genes (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were detected using box plots (GSE72326, GSE50772, GSE23339 and GSE7305) using the R package ggplot2. The AUC of ROC was used to determine the diagnostic value of potential biomarkers in the dataset (GSE72326 and GSE23339) using the R package pROC. qRT-PCR analysis of hub gene mRNA expression in peripheral blood mononuclear cells (PBMC) from EMS(n\u0026thinsp;=\u0026thinsp;5), SLE(n\u0026thinsp;=\u0026thinsp;5) and normal controls(n\u0026thinsp;=\u0026thinsp;8). The study was approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University (Approval Number: 2024-E171-01), and all subjects gave written informed consent to participate in the study. The study was conducted in accordance with relevant guidelines and regulations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e5.8 Construction of diagnostic model based on hub genes\u003c/h2\u003e \u003cp\u003eThe Extreme Gradient Boosting (XGBoost) model, based on the Python package xgboost, is a machine learning algorithm with excellent features and functions. It has become a rising star in the field of machine learning and is widely favored (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). We use the SLE dataset GSE72326 as the training set, GSE50772, GSE23339 and GSE7305 as the validation set. AUROC and PR curve are used as evaluation indicators of the model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e5.9 Immune cell infiltration and correlation with hub gene\u003c/h2\u003e \u003cp\u003eThe single sample Gene Set Enrichment Analysis (ssGSEA)was used to explore the relationship between different degrees of infiltration of immune cell types in the GSE72326 and GSE23339 datasets, followed by non-parametric correlation (spearman) to analyze the correlation between the expression of hub genes and immune infiltrating cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e5.10 Statistical analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses of bioinformatics studies in this study were conducted using R software. The differences between the groups were tested using a nonparametric Wilcoxon signed-rank test. Correlation analysis was performed using Spearman\u0026rsquo;s correlation. In comparison, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe public datasets were downloaded and analyzed in this study, which can be found in GEO data repository and included the accession numbers as follows: GSE72326, GSE23339, GSE50772, GSE7305.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman samples protocols obtained approval from the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eX.Q. and P.H.C designed research; J.H., L.X.R., and L.C.L performed research (data acquisition, analysis, and interpretation); Q.Z.Z., Q.J.N., Y.Q.M.Z., and X.T. drafted and revised the paper; and all authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by grants from the Key Laboratory of Early Prevention and Treatment for Regional High-Incidence-Tumor, Guangxi Medical University, Ministry of Education (GKE-ZZ202132).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe sincerely thank the researchers who uploaded their data to the GEO database.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eZhang T, De Carolis C, Man GCW, Wang CC. The link between immunity, autoimmunity and endometriosis: a literature update. Autoimmun Rev. 2018 Oct;17(10):945\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eSymons LK, Miller JE, Kay VR, Marks RM, Liblik K, Koti M, et al. The Immunopathophysiology of Endometriosis. Trends Mol Med. 2018 Sep;24(9):748\u0026ndash;62. \u003c/li\u003e\n\u003cli\u003eKaul A, Gordon C, Crow MK, Touma Z, Urowitz MB, van Vollenhoven R, et al. Systemic lupus erythematosus. Nat Rev Dis Primer. 2016 Jun 16;2:16039. \u003c/li\u003e\n\u003cli\u003eGleicher N, el-Roeiy A, Confino E, Friberg J. Is endometriosis an autoimmune disease? Obstet Gynecol. 1987 Jul;70(1):115\u0026ndash;22. \u003c/li\u003e\n\u003cli\u003eShigesi N, Kvaskoff M, Kirtley S, Feng Q, Fang H, Knight JC, et al. The association between endometriosis and autoimmune diseases: a systematic review and meta-analysis. Hum Reprod Update. 2019 Jul;25(4):486\u0026ndash;503. \u003c/li\u003e\n\u003cli\u003eTang T, Zhong Y, Xu S, Yu H. Causal effects of endometriosis on SLE, RA and SS risk: evidence from meta-analysis and Mendelian randomization. BMC Pregnancy Childbirth. 2024 Feb 23;24:162. \u003c/li\u003e\n\u003cli\u003eFan YH, Leong PY, Chiou JY, Wang YH, Ku MH, Wei JCC. Association between endometriosis and risk of systemic lupus erythematosus. Sci Rep. 2021 Jan 12;11:532. \u003c/li\u003e\n\u003cli\u003eHarris HR, Costenbader KH, Mu F, Kvaskoff M, Malspeis S, Karlson EW, et al. Endometriosis and the risks of systemic lupus erythematosus and rheumatoid arthritis in the Nurses\u0026rsquo; Health Study II. Ann Rheum Dis. 2016 Jul;75(7):1279\u0026ndash;84. \u003c/li\u003e\n\u003cli\u003eIzumi G, Koga K, Takamura M, Makabe T, Satake E, Takeuchi A, et al. Involvement of immune cells in the pathogenesis of endometriosis. J Obstet Gynaecol Res. 2018;44(2):191\u0026ndash;8. \u003c/li\u003e\n\u003cli\u003eBadawy SZ, Cuenca V, Stitzel A, Tice D. Immune rosettes of T and B lymphocytes in infertile women with endometriosis. J Reprod Med. 1987 Mar;32(3):194\u0026ndash;7. \u003c/li\u003e\n\u003cli\u003eChuang PC, Wu MH, Shoji Y, Tsai SJ. Downregulation of CD36 results in reduced phagocytic ability of peritoneal macrophages of women with endometriosis. J Pathol. 2009 Oct;219(2):232\u0026ndash;41. \u003c/li\u003e\n\u003cli\u003eKang YJ, Jeung IC, Park A, Park YJ, Jung H, Kim TD, et al. An increased level of IL-6 suppresses NK cell activity in peritoneal fluid of patients with endometriosis via regulation of SHP-2 expression. Hum Reprod Oxf Engl. 2014 Oct 10;29(10):2176\u0026ndash;89. \u003c/li\u003e\n\u003cli\u003eKwak JY, Park SW, Kim KH, Na YJ, Lee KS. Modulation of neutrophil apoptosis by plasma and peritoneal fluid from patients with advanced endometriosis. Hum Reprod Oxf Engl. 2002 Mar;17(3):595\u0026ndash;600. \u003c/li\u003e\n\u003cli\u003eEisenberg VH, Zolti M, Soriano D. Is there an association between autoimmunity and endometriosis? Autoimmun Rev. 2012 Sep;11(11):806\u0026ndash;14. \u003c/li\u003e\n\u003cli\u003eBurney RO, Giudice LC. Pathogenesis and Pathophysiology of Endometriosis. Fertil Steril. 2012 Sep;98(3):10.1016/j.fertnstert.2012.06.029. \u003c/li\u003e\n\u003cli\u003eAgic A, Xu H, Finas D, Banz C, Diedrich K, Hornung D. Is endometriosis associated with systemic subclinical inflammation? Gynecol Obstet Invest. 2006;62(3):139\u0026ndash;47. \u003c/li\u003e\n\u003cli\u003eYtterberg SR, Schnitzer TJ. Inhibition of natural killer cell activity by serum from patients with systemic lupus erythematosus: roles of disease activity and serum interferon. Ann Rheum Dis. 1984 Jun;43(3):457\u0026ndash;61. \u003c/li\u003e\n\u003cli\u003eKaplan MJ. Neutrophils in the pathogenesis and manifestations of SLE. Nat Rev Rheumatol. 2011 Sep 27;7(12):691\u0026ndash;9. \u003c/li\u003e\n\u003cli\u003eBarrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, et al. NCBI GEO: archive for functional genomics data sets--update. Nucleic Acids Res. 2013 Jan;41(Database issue):D991-995. \u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015 Apr 20;43(7):e47. \u003c/li\u003e\n\u003cli\u003eShen W, Song Z, Zhong X, Huang M, Shen D, Gao P, et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. iMeta. 2022;1(3):e36. \u003c/li\u003e\n\u003cli\u003eAshburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000 May;25(1):25\u0026ndash;9. \u003c/li\u003e\n\u003cli\u003eKanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000 Jan 1;28(1):27\u0026ndash;30. \u003c/li\u003e\n\u003cli\u003eSzklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023 Jan 6;51(D1):D638\u0026ndash;46. \u003c/li\u003e\n\u003cli\u003eShannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003 Nov;13(11):2498\u0026ndash;504. \u003c/li\u003e\n\u003cli\u003eChen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining [Internet]. New York, NY, USA: Association for Computing Machinery; 2016 [cited 2024 Feb 16]. p. 785\u0026ndash;94. (KDD \u0026rsquo;16). Available from: https://dl.acm.org/doi/10.1145/2939672.2939785\u003c/li\u003e\n\u003cli\u003eBianco B, Andr\u0026eacute; GM, Vilarino FL, Peluso C, Mafra FA, Christofolini DM, et al. The possible role of genetic variants in autoimmune-related genes in the development of endometriosis. Hum Immunol. 2012 Mar 1;73(3):306\u0026ndash;15. \u003c/li\u003e\n\u003cli\u003eMaddern J, Grundy L, Castro J, Brierley SM. Pain in Endometriosis. Front Cell Neurosci. 2020;14:590823. \u003c/li\u003e\n\u003cli\u003eSharif K, Watad A, Coplan L, Lichtbroun B, Krosser A, Lichtbroun M, et al. The role of stress in the mosaic of autoimmunity: An overlooked association. Autoimmun Rev. 2018 Oct;17(10):967\u0026ndash;83. \u003c/li\u003e\n\u003cli\u003eZhang RJ, Wild RA, Ojago JM. Effect of tumor necrosis factor-alpha on adhesion of human endometrial stromal cells to peritoneal mesothelial cells: an in vitro system. Fertil Steril. 1993 Jun;59(6):1196\u0026ndash;201. \u003c/li\u003e\n\u003cli\u003eBerbic M, Schulke L, Markham R, Tokushige N, Russell P, Fraser IS. Macrophage expression in endometrium of women with and without endometriosis. Hum Reprod Oxf Engl. 2009 Feb;24(2):325\u0026ndash;32. \u003c/li\u003e\n\u003cli\u003eKyama CM, Debrock S, Mwenda JM, D\u0026rsquo;Hooghe TM. Potential involvement of the immune system in the development of endometriosis. Reprod Biol Endocrinol RBE. 2003 Dec 2;1:123. \u003c/li\u003e\n\u003cli\u003eRichter P, Macovei LA, Mihai IR, Cardoneanu A, Burlui MA, Rezus E. Cytokines in Systemic Lupus Erythematosus\u0026mdash;Focus on TNF-\u0026alpha; and IL-17. Int J Mol Sci. 2023 Sep 22;24(19):14413. \u003c/li\u003e\n\u003cli\u003eChen L, Fu W, Zheng L, Wang Y, Liang G. Recent progress in the discovery of myeloid differentiation 2 (MD2) modulators for inflammatory diseases. Drug Discov Today. 2018 Jun 1;23(6):1187\u0026ndash;202. \u003c/li\u003e\n\u003cli\u003eLiu B, Yang Y, Dai J, Medzhitov R, Freudenberg MA, Zhang PL, et al. TLR4 up-regulation at protein or gene level is pathogenic for lupus-like autoimmune disease. J Immunol Baltim Md 1950. 2006 Nov 15;177(10):6880\u0026ndash;8. \u003c/li\u003e\n\u003cli\u003eLee TP, Tang SJ, Wu MF, Song YC, Yu CL, Sun KH. Transgenic overexpression of anti-double-stranded DNA autoantibody and activation of Toll-like receptor 4 in mice induce severe systemic lupus erythematosus syndromes. J Autoimmun. 2010 Dec;35(4):358\u0026ndash;67. \u003c/li\u003e\n\u003cli\u003eChen J, Zhao X, Huang C, Lin J. Novel insights into molecular signatures and pathogenic cell populations shared by systemic lupus erythematosus and vascular dementia. Funct Integr Genomics. 2023 Nov 16;23(4):337. \u003c/li\u003e\n\u003cli\u003eWang Y, Hwang J, Yadav D, Oda T, Lee PCW, Jin JO. Inhibitory effect of porphyran on lipopolysaccharide-induced activation of human immune cells. Carbohydr Polym. 2020 Mar 15;232:115811. \u003c/li\u003e\n\u003cli\u003eYang Y. Cancer immunotherapy: harnessing the immune system to battle cancer. J Clin Invest. 2015 Sep 1;125(9):3335\u0026ndash;7. \u003c/li\u003e\n\u003cli\u003eRychkov D, Neely J, Oskotsky T, Yu S, Perlmutter N, Nititham J, et al. Cross-Tissue Transcriptomic Analysis Leveraging Machine Learning Approaches Identifies New Biomarkers for Rheumatoid Arthritis. Front Immunol. 2021 Jun 8;12:638066. \u003c/li\u003e\n\u003cli\u003eBae SJ, Jo Y, Cho MK, Jin JS, Kim JY, Shim J, et al. Identification and analysis of novel endometriosis biomarkers via integrative bioinformatics. Front Endocrinol. 2022 Oct 20;13:942368. \u003c/li\u003e\n\u003cli\u003eGuo B, Chen J hua, Zhang J hui, Fang Y, Liu X jing, Zhang J, et al. Pattern-recognition receptors in endometriosis: A narrative review. Front Immunol. 2023 Mar 23;14:1161606. \u003c/li\u003e\n\u003cli\u003eZervou MI, Matalliotakis M, Goulielmos GN. Comment on \u0026ldquo;Risk of systemic lupus erythematosus in patients with endometriosis: a nationwide population‑based cohort study.\u0026rdquo; Arch Gynecol Obstet. 2022 Feb 1;305(2):543\u0026ndash;4. \u003c/li\u003e\n\u003cli\u003eNisolle M, Donnez J. Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril. 1997 Oct;68(4):585\u0026ndash;96. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"[email protected]","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, Systemic lupus erythematosus, diagnostic biomarker, bioinformatics, machine learning, immune response","lastPublishedDoi":"10.21203/rs.3.rs-4150400/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4150400/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEndometriosis (EMS) is a chronic gynecological disorder that affects 5\u0026ndash;10% of women of reproductive age, and Systemic lupus erythematosus (SLE) is one of the most prevalent systemic autoimmune diseases. Despite clinical evidence suggesting potential associations between EMS and SLE, the underlying pathogenesis is yet unclear. This article aimed to explore the shared gene signatures and potential molecular mechanisms in EMS and SLE. Microarray data were downloaded from the Gene Expression Omnibus (GEO) database and used to screen for differentially expressed genes (DEGs) in the SLE datasets. A weighted gene co-expression network analysis (WGCNA) was used to identify the co-expression modules of EMS. cytoscape software and three machine learning algorithms were used to determine critical biomarkers, and a diagnostic model was built using the XG-Boost machine learning algorithms. Immune cell infiltration analysis was used to investigate the correlation between immune cell infiltration and common biomarkers of EMS and SLE. Results revealed that shared genes enriched in immune-related pathways and inflammatory responses. The area under the receiver operating characteristic (AUROC) curve and the Precision-Recall (PR) curves showed satisfactory performance of the model. immune cell infiltration analysis showed that the expression of hub genes was closely associated with immune cells. RT-qPCR results indicated that \u003cem\u003eLY96\u003c/em\u003e might be the best biomarker for EMS and SLE.\u003c/p\u003e","manuscriptTitle":"Exploration of the shared pathways and common biomarker LY96 in Endometriosis and Systemic lupus erythematosus using integrated bioinformatics analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-03 09:14:46","doi":"10.21203/rs.3.rs-4150400/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","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":"3a0e3fb4-fad0-4a9b-932b-12b8401ca5a5","owner":[],"postedDate":"April 3rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":30187428,"name":"Biological sciences/Immunology/Autoimmunity"},{"id":30187429,"name":"Biological sciences/Immunology/Inflammation"}],"tags":[],"updatedAt":"2024-06-24T04:06:24+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-03 09:14:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4150400","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4150400","identity":"rs-4150400","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}