Machine Learning Algorithms for a Novel Cuproptosis-related Gene Signature of Diagnostic and Immune Infiltration in Endometriosis

Machine Learning Algorithms for a Novel Cuproptosis-related Gene Signature of Diagnostic and Immune Infiltration in Endometriosis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Machine Learning Algorithms for a Novel Cuproptosis-related Gene Signature of Diagnostic and Immune Infiltration in Endometriosis Jiajia Wang, Yiming Lu, Yongchang Ling, Guangyu Sun, Zhihao Fang, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2742573/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 (EMT) is a chronic hormone-dependent disease where in viable endometrial tissue is transplanted outside the uterus. Interestingly, immune infiltration is significantly involved in EMT pathogenesis. Currently, no studies have shown the involvement of cuproptosis-related genes (CRGs) in regulating immune infiltration in EMT. This study identified three CRGs such as GLS, NFE2L2, and PDHA1, associated with EMT using machine learning algorithms. These three CRGs were upregulated in the endometrium of patients with moderate/severe EMT and downregulated in patients with infertility. Single sample genomic enrichment analysis (ssGSEA) revealed that these CRGs were closely correlated with autoimmune diseases such as systemic lupus erythematosus. Furthermore, these CRGs were correlated with immune cells such as eosinophils, natural killer cells, and macrophages. Therefore, profiling patients based on these genes aid in a more accurate diagnosis of EMT progression. These findings provide a new idea for the pathology and treatment of endometriosis, suggesting that CRGs such as GLS, NFE2L2, and PDHA1 may play a key role in the occurrence and development of endometriosis. cuproptosis endometriosis immune infiltration machine learning algorithms Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Introduction Endometriosis (EMT) is a common estrogen-dependent disease wherein the endometrial tissue implants and grows on the peritoneal surface, ovaries, rectovaginal septum, perineum, and surgical scars [ 1 ]. The primary symptoms of EMT include lower abdominal pain, dysmenorrhea, menstrual abnormalities, and discomfort during intercourse. Globally, nearly 5–10% of women of reproductive age are affected by EMT, and 30–50% of women with EMT suffer from infertility [ 2 , 3 ]. However, the pathogenesis of EMT has not yet been elucidated. Several studies have identified markers associated with EMT [ 4 ]. however, the sample size of these studies was small, which prevents accurate diagnosis. Therefore, there is an urgent need to identify new biomarkers to understand the pathophysiological mechanisms of EMT. This would aid in designing better and more accurate diagnostic as well as therapeutic approaches for EMT. Copper is an essential cofactor regulating several enzymes involved in various cellular functions, including mitochondrial respiration, antioxidation, and synthesis of hormones, neurotransmitters, and pigment. However, abnormal copper levels can induce oxidative stress and cytotoxicity [ 5 ]. Recent studies have demonstrated that copper induces cell death by altering the tricarboxylic acid cycle (TCA), which causes the accumulation of lipid-acylated proteins and a decrease in iron-sulfur cluster protein levels. This triggers a proteotoxic stress response called cuproptosis [ 6 ]. Moreover, studies have demonstrated that copper homeostasis is critically involved in immune cell infiltration. Tan et al. showed that copper-induced H2O2 production mediated by LOXL4 activates the type I interferon signaling pathway, which promotes PD-L1 presentation by macrophages [ 7 ]. A study has demonstrated a close correlation between the copper-related protein STEAP2 and the prognosis, as well as the infiltration of immune cells in gliomas [ 8 ]. Ji et al. indicated that cuproptosis could influence the development and prognosis of patients with renal clear cell carcinoma by regulating immune cell infiltration [ 9 ]. Therefore, it is necessary to study the involvement of cuproptosis in the diagnosis and immune cell infiltration in diseases. However, no studies have determined the influence of cuproptosis-mediated immune cell infiltration on EMT pathogenesis. Therefore, in this study, we combined data on EMTs and cuproptosis-related genes (CRGs) to determine the underlying mechanism by which cuproptosis mediates immune cell infiltration in EMT. First, we obtained data on 19 CRGs from previous studies [ 10 ] and the GSE7307 datasets from the "Gene Expression Omnibus (GEO)” database to perform differential and functional correlation analysis. Next, we identified three key CRGs using machine learning algorithms. Furthermore, we analyzed the correlation between key CRGs, immune cells, and pathways. Finally, we constructed a nomogram and screened for drug targets to provide insights into cuproptosis in EMT. The outcomes of this study are summarized in the following table. These results would aid in enhancing our understanding of the involvement of cuproptosis in regulating EMT. Material and Methods Data download and pre-processing We retrieved data on 19 CRGs from previously published studies and the GSE7307 dataset consisting of microarray data on EMT of 677 samples from GEO. To analyze the data, different probes corresponding to the same gene were identified, and the null probes were removed. (3) Next, a probe corresponding to a gene was identified, and (4) log2-transformed quantile normalization was performed. Finally, we identified 41 samples, consisting of 18 patients with EMT and 23 healthy individuals. The GSE7305 dataset consisting of ten patients with EMT and ten healthy individuals was used as a validation set. Screening for differentially expressed CRGs We employed the “linear models for microarray data” package for screening differentially expressed genes (DEG) in tissues of patients with EMT and healthy individuals based on the following criteria: “P < 0.05.” Functional enrichment Analysis of DEGs We used the “Disease Ontology (DO),” “Gene Ontology (GO),” and “Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses” to determine the biological roles of CRGs. P < 0.05 was considered statistically significant. Further, the visual enrichment maps were constructed using the "ggplot2" and "GOplot" R packages. Machine learning algorithm to identify key CRGs We used three machine learning algorithms such as "LASSO regression analysis,” “SVM-RFE,” and “RF” R package, to identify DEGs. Finally, the results were visualized by constructing a Venn diagram using the "venneuler" R package. Diagnostic value of key CRGs in EMT and external validation We determined the diagnostic value of these key CRGs in GSE7307 and validated these genes in GSE7305. For GSE51981 and GSE120103, we plotted the receiver operating characteristic (ROC) curves to determine the diagnostic value of these genes using “P < 0.05” as the threshold. We calculated the area under the ROC curve (AUC), and the AUC value between 0 and 1 was used for analysis. The greater the AUC value, the better the performance of these genes in predicting diagnosis. Screening key regulatory pathways using Single sample genomic enrichment analysis (ssGSEA) The median expression of key genes was set as the criteria to classify tissues of patients with EMT into the high- and low-expression groups. Furthermore, analysis of the variance was performed using the “ssGSEA” R package on gene sets (P < 0.05) to identify pathways enriched by key CRGs. Immunological analysis The relative proportions of the 22 types of infiltrating immune cells in all EMT samples were determined using “Cell-type Identification by Estimating Relative Subsets of RNA Transcripts” (CIBERSORT, http://cibersortx.stanford.edu ). Next, we calculated immune scores using the “Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data" algorithm. Finally, Spearman rank correlation analysis was performed to determine the correlation between CRGs and the number of infiltrating immune cells. The correlation was visualized using the “ggplot2” package. Construction of nomogram prediction model and efficacy assessment for clinical diagnosis of EMT We constructed a nomogram prediction model and validated it internally using the "rms" R package. The performance of the model was assessed using calibration curves, clinical decision curve analysis (DCA), and ROC curves. Drug screening for targeting key CRGs The “DrugBank” database ( https://go.drugbank.com/drugs ) was used to predict potential therapeutic targets of key CRGs associated with EMT. The “DrugBank database is a unique bioinformatics and cheminformatics resource that contains more than 13,791 drug entries, including small molecules, biotechnology, nutraceuticals, and investigational drugs. Additionally, there are > 5236 nonredundant protein sequences associated with these drug entries. Statistical analysis R 4.0.5 statistical software was used to complete the analysis. We used the t-test and Wilcoxon rank-sum test for quantifying variables. Furthermore, Spearman correlation analysis was used to investigate the correlation between CRGs and immune cell infiltration. P < 0.05 was considered statistically significant. Results Screening for DEGs in patients with EMT We combined 19 CRGs from previously published studies and DEGs screened from patients with EMT. A total of 10 DEGs were screened from GSE7307, of which six were upregulated, and four were downregulated (Figs. 1 A, B). Functional enrichment Analysis of DEGs We performed the DO, GO, and KEGG pathway enrichment analyses on DEGs. The DO enrichment analysis demonstrated that these key CRGs were mainly enriched in the nutrient deficiency disease pathway associated with carbohydrate metabolism disorder (Fig. 2 A). The GO enrichment analysis revealed that these key CRGs were enriched in various processes such as acetyl-CoA biosynthesis from pyruvate, acetyl-CoA metabolism, and thioester, acyl-a CoA, a sulfur compound, nucleoside bisphosphate, ribonucleoside, and acyl-CoA biosynthesis processes. The KEGG enrichment analysis showed pathways, including TCA, Glycolysis/Gluconeogenesis, pyruvate, and carbon metabolism, as well as central carbon metabolism in cancer, resistance to platinum drugs, and the glucagon and HIF-1 signaling pathways were enriched by these CRGs. However, no significant difference was observed between these CRGs (Fig. 2 C). Screening of EMT markers using machine learning algorithms A total of ten DEGs were identified using machine learning as signature genes. Five signature genes, including NFE2L2, NLRP3, DLD, PDHA1, and GLS, were identified using the “LASSO regression analysis” (Figs. 3 A, B). Next, six signature genes, including GLS, PDHB, DLD, NFE2L2, PDHA1, and SLC31A1, were identified using the “SVM-RFE” algorithm (Fig. 3 C, D). Finally, five signature genes, including PDHA1, GLS, NFE2L2, DLAT, and MTF1, were identified using the “RF” algorithm (Figs. 3 E, F). Further, genes identified using these three machine learning algorithms were intersected, and three signature genes, including GLS, NFE2L2, and PDHA1, were identified. The three genes were intersected to obtain three signature genes, GLS, NFE2L2, and PDHA1, and plotted on a Wayne diagram (Figs. 3 G, H). Expression and validation of the diagnostic significance of the signature genes First, we determined the differential expression and diagnostic value of GLS, NFE2L2, and PDHA1 in GSE7307. Further, these genes were externally validated in GSE7305. The results revealed a significant increase in GLS and NFE2L2 expression and a significant decrease in PDHA1 expression in both datasets (Figs. 4 A, B). The AUC values of PDHA1 in the two datasets were 0.879 and 0.920, respectively (Figs. 5 A, B). The AUC values of all three diagnostic genes were > 0.800, thereby indicating these CRGs had high diagnostic values. Expression of key CRGs in EMT with different severity and combined infertility To validate the confidence and applicability of the key CRGs, external datasets such as GSE51981 and GSE120103 were selected to validate the correlation between key CRGs, EMT severity, and combined infertility. Since PDA1 was not expressed in GSE120103, PDHA1 was not analyzed in EMT samples. The patients with micro/light EMT were categorized into one group, and patients with moderate/severe EMT were categorized into another group. The results revealed high GLS and NFE2L2 expression in patients with moderate/severe EMT (Fig. 6 A). ROC curves were performed on these upregulated key CRGs to analyze the severity of EMT (P 0.7 (Fig. 6 B). Next, we used GSE120103, consisting of patients with EMT who may or may not be suffering from infertility, for subsequent validation. The results revealed an increase in GLS and NFE2L2 expression in the endometrium of patients with EMT; however, a decrease in GLS and NFE2L2 expression was observed in the endometrium of patients with infertility (Fig. 7 A). On the contrary, we performed ROC analysis to determine the significance of key CRGs in diagnosing patients with EMT. The AUC values validated the sensitivity and specificity of CRGs (Fig. 7 B). Analyzing Regulatory pathway enriched by key CRGs The ssGSEA results showed that the complement and coagulation pathways, Legionellosis, Malaria, Pertussis, Systemic lupus erythematosus (SLE), etc., were enriched in the high GLS and high NFE2L2 expression groups (Figs. 8 A, B). In addition, the cell cycle, the Fanconi anemia and protein export pathways, DNA replication, and Homologous recombination are enriched in the PDHA1 expression group (Fig. 8 C). Correlation between three key CRGs and proportion of immune cell infiltration in EMT We utilized the “CIBERSORT” method to determine the proportion of 22 immune cell subpopulations infiltrating in patients with EMT. GLS, NFE2L2, and PDHA1 expression were evaluated as a function of the immune microenvironment. A significant positive correlation was observed between GLS and immune cells like M2 macrophages, Plasma cells, naïve CD4 T cells, and activated Mast cells. GLS was negatively correlated with activated NK cells (Fig. 9 A). NFE2L2 was significantly correlated with immune cells such as Plasma cells, memory B cells, resting memory CD4 T cells, Eosinophils, M1 and M2 macrophages, resting mast cells, and CD8 T cells (Fig. 9 B). Furthermore, a significant positive correlation was observed between NFE2L2 and immune cells such as activated NK cells, CD8T cells, M1 and M2 Macrophages, resting mast cells, naïve B cells, resting memory CD4 T cells, eosinophils, memory B cells, and Plasma cells (Fig. 9 C). Construction and validation of the Nomogram prediction model We constructed a nomogram prediction model using GLS, NFE2L2, and PDHA1 for predicting the risk of EMT development (Fig. 10 A). Nomogram was externally validated using GSE7305. The rate of incidences was derived from each risk factor score and summed to obtain a total score. Next, we plotted the calibration curve (Fig. 10 B). The results revealed that the three curves were consistent and converged, thereby indicating a good agreement between the actual and predicted outcomes of the model. In addition, DCA showed the clinical benefit of the nomogram prediction model (Fig. 10 C). Finally, we plotted the ROC curve (the AUC value = 0.942) to determine the predictive ability of the columnar graph model (Fig. 10 D). The AUC value of the nomogram prediction model was 0.942, thereby indicating that the model had a good predictive ability. Drug prediction We identified five drugs targeting three key CRGs using data from the “DrugBank database (Fig. 11 ), and all of these drugs have been approved. First, L-glutamine (DB00130) was identified as a drug targeting CRGs. It is an important energy source in cells and is involved in metabolic processes. It is used to treat patients with sickle cell disease gastric and duodenal ulcers, improving brain function. Next, we identified Glutamate (DB00142), a common amino acid and an excitatory neurotransmitter. It is a precursor molecule required for GABA synthesis in GABAergic neurons. It is mainly used as total parenteral nutrition and as an adjunct for treating psycho-neurological disorders. Next, Dimethyl fumarate (DMF, DB08908) binds to NFE2L2. It is used for treating patients with relapsing multiple sclerosis. Finally, NADH (DB00157) targets PDHA1 and is used as supplementary nutrition. Discussion Currently, invasive techniques, such as laparoscopic histopathological examination, are used for EMT diagnosis. The most common treatment modalities include hormonal suppression or surgical excision/ablation of visible lesions. However, the therapeutic outcomes are unsatisfactory and often accompanied by adverse effects. Studies have identified several key molecules, such as adhesion molecules, matrix metalloproteinases, lncRNA HOTAIR, oxidative stress, and regulatory T cells, that are involved in EMT development [ 11 , 12 ]. However, the exact mechanism is still unclear. Therefore, there is an urgent need to identify new biomarkers associated with EMT pathogenesis. This would aid in developing non-invasive diagnostic and screening tools for patients with EMT. Copper is an essential cofactor of enzymes. It is critically involved in various physiological processes and activities. Turgut et al. showed that copper was associated with the pathogenesis of EMT and oxidative stress [ 13 ]. Cuproptosis is a newly discovered form of cell death involving TCA regulation. Lipid-acylated proteins induce cytotoxic stress, which leads to cell death 6. However, no studies have shown the involvement of cuproptosis in EMT. Therefore, in this study, we combined data on EMT and CRGs from GEO. We identified 10 DEGs, including NFE2L2, NLRP3, SLC31A1, DLAT, DLD, PDHB, PDHA1, MTF1, CDKN2A, and GLS. The DO, GO, and KEGG pathway enrichment analysis showed that key CRGs were primarily enriched in the acetyl-CoA and pyruvate metabolism, TCA, and Glycolysis/Gluconeogenesis. Bradbeer et al. showed a correlation between TCA enzyme succinate dehydrogenase and EMT [ 14 ]. Interestingly, alteration in TCA occurs in cuproptosis, which induces cell death. Therefore, these results indicate that these CRGs could be a promising target for modulating cuproptosis in EMT, which could be a new therapeutic strategy for treating EMT. Next, we used three machine learning algorithms, such as “LASSO regression,” “SVM-RFE,” and “RF,” to screen for EMT and cuproptosis-related biomarkers. We identified three CRGs associated with immune cell infiltration in EMT. Of these three CRGs, an increase in GLS and NFE2L2 expression and a decrease in PDHA1 expression were observed. Furthermore, the ROC curve analysis validated the diagnostic value of these key CRGs in distinguishing the endometrium of patients with EMT and healthy controls. Finally, we validated our results using the external validation dataset GSE7305, and the results obtained were consistent. Glutaminase (GLS) is located on chromosome 2. It is a key enzyme mediating glutamine-to-glutamate conversion. GLS is converted to alpha-ketoglutarate, which enters TCA to produce energy. GLS is an oncogene, and an increase in GLS expression was observed in pancreatic ductal adenocarcinoma. Under oxidative stress, SUCLA2 increases glutamine catabolism and the production of nicotinamide adenine dinucleotide phosphate as well as glutathione by activating GLS, thereby ameliorating oxidative stress and promoting tumor growth [ 15 ]. Our results revealed a significant correlation between CRGs like GLS and EMT progression. Nuclear transcription factor E2-related factor 2 (NFE2 like bZIP transcription factor 2, NFE2L2, also known as Nrf2) is primarily localized in the cytoplasm. It is an antioxidant transcription factor that regulates inflammatory responses and oxidative stress in cells [ 16 ]. NFE2L2 variants are closely linked to the risk of human disease. Interestingly, oxidative damage promotes EMT progression. Previous studies have shown that NFE2L2 increases EMT risk by inhibiting processes such as oxidative stress against the growth and development of EMT [ 17 , 18 ]. The pyruvate dehydrogenase complex E1 alpha subunit (PDHA1) is the E1 subunit of the pyruvate dehydrogenase complex. It is primarily localized in the mitochondria and contains three serine residues, which could be activated by four inhibitory PDHK1-4 and two reactivated phosphatases (PDP1-2). These PDP1-2 are reversibly phosphorylated and regulate glycolysis and TCA [ 19 , 20 ]. Spakova et al. demonstrated that silencing HIF-1α/miR210 expression increases PDHA1 and decreases MITF-M expression. This promotes mitochondrial respiratory activity, which aids in eliminating reactive oxygen species in melanoma cells [ 21 ]. Our results indicate a significant role of GLS, NFE2L2, and PDHA1 in EMT pathogenesis. ROC curve analysis revealed that these key CRGs are significantly involved in the onset and progression of EMT, thereby indicating the potential diagnostic value of these CRGs in clinical settings. Moreover, risk modeling results suggested that GLS, NFE2L2, and PDHA1 could be risk factors for EMT. Based on previous studies and our results, CRGs may play an important role in EMT development, and the underlying mechanisms should be further investigated. We also determined the expression of key CRGs in patients with varying severity of EMT and those suffering from infertility. The results revealed an increase in GLS and NFE2L2 expression in patients with moderate/severe EMT. Furthermore, an increase in the expression of key CRGs could lead to infertility in patients with EMT. Qiao et al. showed that isoniazid activates the Keap1/Nrf2 signaling pathway by inducing oxidative stress and apoptosis, thereby impairing the reproductive system and reducing fertility in mammals [ 22 ]. Ivanov et al. showed that melatonin could protect the developing embryo from oxidative stress by modulating NFE2L2, SOD1, and GPX1 expression [ 23 ]. These results indicate a close correlation between NFE2L2 and infertility, consistent with our results. Next, we performed ssGSEA to verify the involvement of key CRGs in EMT. Our results revealed significant enrichment of the complement and coagulation pathways, Legionellosis, Malaria, Pertussis, and SLE in the high GLS and high NFE2L2 expression groups. On the contrary, the cell cycle, the Fanconi anemia and protein export pathways, DNA replication, and homologous recombination were enriched in the high PDHA1 expression group. SLE is an autoimmune disease characterized by the presence of specific autoantibodies, including antinuclear antibodies. Studies have shown a correlation between EMT and autoimmune diseases such as SLE and Sjögren's syndrome [ 24 , 25 ]. However, the underlying mechanism of the pathogenesis of EMT is still unclear. Implantation of reflux menstrual blood, endocrine, genetic factors, angiogenesis, and stem cell differentiation are important factors associated with EMT pathogenesis. Of these factors, abnormal immune responses could be the underlying mechanism of EMT pathogenesis. In fact, the immune component of EMT has received widespread attention from researchers. Some studies have demonstrated that an increase in the secretion of proinflammatory cytokines due to immune dysfunction, impaired immune surveillance, and altered immune cell profiles contribute to EMT progression. Prolonged persistent immune dysregulation could lead to a state of chronic inflammation, thereby creating a conducive environment for promoting adhesion and angiogenesis. This could lead to a vicious cycle of EMT development and progression [ 26 ]. The “CIBERSORT” method is widely used to study EMT in humans. It is based on the principles of linear support vector regression to deconvolute the expression matrix of immune cell subtypes. Therefore, we used CIBERSORT to assess infiltrating immune cells for determining the role of immune cell infiltration in EMT. Our results revealed that these three CRGs were associated with various immune cell types, including eosinophils, activated NK cells, and macrophages. Eidukaite et al. showed that epithelial cells in endometriotic foci secrete high levels of eosinophil-specific chemokines compared to in situ endometrial cells. These eosinophils and other myeloid cells enter the peritoneum, which triggers inflammatory and allergic responses [ 27 ]. Ścieżyńska et al. showed that endometriotic cells' survival and growth in the peritoneal cavity might also be due to their recognition and clearance by local immune cells such as macrophages and NK cells [ 28 ]. Ramírez-Pavez et al. demonstrated that macrophages accumulate in the peritoneal cavity of patients with EMT; however, their ability to clear migrating endometrial debris is reduced [ 29 ]. Furthermore, combining the results of differential immune cell infiltration and the correlation between cuproptosis and immune cell infiltration revealed that immune cells such as eosinophils, activated NK cells, and M2 Macrophages could be critically involved in the modulation of CRG-regulated immune function in EMT. EMT is an aggressive disease. Koninckx et al. have suggested that oxidative stress and peritoneal microbiota due to retrograde menstruation could cause EMT. Therefore, pharmacological treatment could prevent new lesions and be prescribed post-surgery. Additionally, Sirohi et al. showed that inhibiting copper-induced toxicity increases the survival of endometriotic cells [ 30 ]. Therefore, copper chelators could be used as therapeutic agents for treating patients with EMT [ 31 ]. We identified five drugs from the DrugBank database targeting these three CRGs. Qinpi methicin (DB13155) has anti-inflammatory and antioxidant activities. Hence, qinpi methicin could be used for treating diseases caused by inflammation and oxidative stress, such as EMT. As-Sanie et al. demonstrated high forebrain insula glutamine (DB00130)-glutamate (DB00142) binding in EMT patients with chronic pelvic pain. High islet activity could cause chronic pelvic pain in patients with EMT; therefore, reducing islet glutamate levels could alleviate chronic pelvic pain [ 32 ]. DMF is an immunomodulatory and antioxidant molecule commonly used to treat patients with relapsing multiple sclerosis and psoriasis. It binds with NFE2L2. Chen et al. demonstrated that DMF, an Nrf2 agonist, could significantly influence immune responses and oxidative stress. Additionally, Yan et al. suggested that DMF could attenuate inflammation, oxidative stress, and iron death via the NRF2/ARE/NF-kB signaling pathway. Moreover, DMF improves cognitive dysfunction in rats with chronic hypoperfusion [ 33 ]. Also, EMT is closely associated with oxidative stress [ 34 ]. Therefore, it is necessary to investigate if DMF could be used for treating patients with EMT. NADH is a nutritional supplement. It has antioxidant properties, reduces anxiety, promotes neurotransmitter synthesis, and prevents dementia. Govatati et al. showed that altering the mitochondrial membrane complex I could be a risk factor for EMT. Mitochondrial membrane complex I catalyze the transfer of electrons from NADH to ubiquinone and is closely correlated with EMT, which could be a new therapeutic approach [ 35 ]. However, our study has a few limitations. First, the sample size of our study was relatively small since data were only obtained from GEO. Therefore, additional studies using a larger sample size should be performed to validate our results. Second, although we have successfully identified three CRGs as potential biomarkers for immunophenotyping of EMT, no in vivo or in vitro studies were conducted to validate these results. Thus, future studies should focus on conducting in vivo or in vitro studies to validate these biomarkers. Conclusions GLS, NFE2L2, and PDHA1 could be novel diagnostic markers for EMT, representing a significant breakthrough. Additionally, the correlation between these CRGs and immune cell infiltration could aid in developing effective immunotherapy for patients with EMT. Declarations Acknowledgements The authors express their gratitude for the invaluable assistance and insightful discussions provided by the members of the Department of Youjiang Medical College for Nationalities. Additionally, the authors would like to thank the GEO database for granting access to the data used in this study. Conflict of Interest The authors declare no competing interests. Ethics approval Since the GEO database information is public, ethical approval or informed consent is not required. Consent to participate All authors have agreed to take part in the study. Consent for Publication All authors have read and agreed to the published version of the manuscript. Availability of data and material All the data in this study are available. Code availability All codes for this study are available. Author contribution J.L.W. designed the study. J.J.W. produced the initial draft of the manuscript. M.Y.D. produced the methodology. Y.M.L. and Y.C.L. contributed to analyzing data. G.L.S. contributed to drafting of the manuscript. All authors have read and approved the final submitted manuscript. References Kapoor R, Stratopoulou CA, Dolmans M. Pathogenesis of Endometriosis: New Insights into Prospective Therapies. Int J Mol Sci. 2021;22(21):11700. Taylor HS, Kotlyar AM, Flores VA. Endometriosis is a chronic systemic disease: clinical challenges and novel innovations. Lancet. 2021;397(10276):839-852. Hodgson RM, Lee HL, Wang R, Mol BW, Johnson N. Interventions for endometriosis-related infertility: a systematic review and network meta-analysis. Fertil Steril. 2020;113(2):374-382.e2. Anastasiu CV, Moga MA, Elena Neculau A, et al. Biomarkers for the Noninvasive Diagnosis of Endometriosis: State of the Art and Future Perspectives. Int J Mol Sci. 2020;21(5):1750. Ge EJ, Bush AI, Casini A, et al. Connecting copper and cancer: from transition metal signalling to metalloplasia. Nat Rev Cancer. 2022;22(2):102-113. Tsvetkov P, Coy S, Petrova B, et al. Copper induces cell death by targeting lipoylated TCA cycle proteins. Science. 2022;375(6586):1254-1261. Tan HY, Wang N, Zhang C, Chan YT, Yuen MF, Feng Y. Lysyl Oxidase-Like 4 Fosters an Immunosuppressive Microenvironment During Hepatocarcinogenesis. Hepatology. 2021;73(6):2326-2341. Wang X, Han M, Chen S, Sun Y, Tan R, Huang B. The copper-associated protein STEAP2 correlated with glioma prognosis and immune infiltration. Front Cell Neurosci. 2022;16. Ji Z, Ren W, Wang H, Gao W, Yuan B. Molecular Subtyping Based on Cuproptosis-Related Genes and Characterization of Tumor Microenvironment Infiltration in Kidney Renal Clear Cell Carcinoma. Front Oncol. 2022;12. Jiang X, Ke J, Jia L, et al. A novel cuproptosis-related gene signature of prognosis and immune microenvironment in head and neck squamous cell carcinoma cancer. J Cancer Res Clin Oncol. 2023;149(1):203-218. Zhang L, Yu Z, Qu Q, Li X, Lu X, Zhang H. Exosomal lncRNA HOTAIR Promotes the Progression and Angiogenesis of Endometriosis via the miR-761/HDAC1 Axis and Activation of STAT3-Mediated Inflammation. Int J Nanomedicine. 2022;17:1155-1170. Koninckx PR, Fernandes R, Ussia A, et al. Pathogenesis Based Diagnosis and Treatment of Endometriosis. Front Endocrinol (Lausanne). 2021;12. Turgut A, Ozler A, Goruk NY, Tunc SY, Evliyaoglu O, Gul T. Copper, ceruloplasmin and oxidative stress in patients with advanced-stage endometriosis. Eur Rev Med Pharmacol Sci. 2013;17(11):1472-8. Bradbeer JN, Lindsay PC, Reeve J. Fluctuation of mineral apposition rate at individual bone-remodeling sites in human iliac cancellous bone: independent correlations with osteoid width and osteoblastic alkaline phosphatase activity. J Bone Miner Res. 1994;9(11):1679-86. Tong Y, Guo D, Lin S, et al. SUCLA2-coupled regulation of GLS succinylation and activity counteracts oxidative stress in tumor cells. Mol Cell. 2021;81(11):2303-2316.e8. Lee DH, Park JS, Lee YS, et al. SQSTM1/p62 activates NFE2L2/NRF2 via ULK1-mediated autophagic KEAP1 degradation and protects mouse liver from lipotoxicity. Autophagy. 2020;16(11):1949-1973. Siracusa R, D Amico R, Cordaro M, et al. The Methyl Ester of 2-Cyano-3,12-Dioxooleana-1,9-Dien-28-Oic Acid Reduces Endometrial Lesions Development by Modulating the NFkB and Nrf2 Pathways. Int J Mol Sci. 2021;22(8):3991. Osiński M, Mostowska A, Wirstlein P, Wender-Ożegowska E, Jagodziński PP, Szczepańska M. The assessment of GWAS — identified polymorphisms associated with infertility risk in Polish women with endometriosis. Ginekol Pol. 2018;89(6):304-310. Echeverri Ruiz NP, Mohan V, Wu J, et al. Dynamic regulation of mitochondrial pyruvate metabolism is necessary for orthotopic pancreatic tumor growth. Cancer Metab. 2021;9(1). Xu X, Tang Y, Zhu J, Luo J. Endoplasmic Reticulum Stress-Related Four-Biomarker Risk Classifier for Survival Evaluation in Esophageal Cancer. J Oncol. 2022;2022:1-14. Apaková I, Rabajdová M, Mičková H, Graier WF, Mareková M. Effect of hypoxia factors gene silencing on ROS production and metabolic status of A375 malignant melanoma cells. Sci Rep. 2021;11(1). Qiao P, Zhang Y, Yang Y, et al. Oral isoniazid causes oxidative stress, oocyte deterioration and infertility in mice. Toxicology. 2021;455:152749. Ivanov D, Mazzoccoli G, Anderson G, et al. Melatonin, Its Beneficial Effects on Embryogenesis from Mitigating Oxidative Stress to Regulating Gene Expression. Int J Mol Sci. 2021;22(11):5885. Geng R, Huang X, Li L, et al. Gene expression analysis in endometriosis: Immunopathology insights, transcription factors and therapeutic targets. Front Immunol. 2022;13. Shigesi N, Kvaskoff M, Kirtley S, et al. The association between endometriosis and autoimmune diseases: a systematic review and meta-analysis. Hum Reprod Update. 2019;25(4):486-503. Jiang I, Yong PJ, Allaire C, Bedaiwy MA. Intricate Connections between the Microbiota and Endometriosis. Int J Mol Sci. 2021;22(11):5644. Eidukaite A, Tamosiunas V. Activity of eosinophils and immunoglobulin E concentration in the peritoneal fluid of women with endometriosis. Clin Chem Lab Med. 2004;42(6):590-4. Ścieżyńska AA, Komorowski MM, Soszyńska MM, Malejczyk JJ. NK Cells as Potential Targets for Immunotherapy in Endometriosis. J Clin Med. 2019;8(9):1468. Ramírez-Pavez TN, Martínez-Esparza M, Ruiz-Alcaraz AJ, Marín-Sánchez P, Machado-Linde F, García-Peñarrubia P. The Role of Peritoneal Macrophages in Endometriosis. Int J Mol Sci. 2021;22(19):10792. Sirohi D, Al RR, Knibbs LD. Environmental exposures to endocrine disrupting chemicals (EDCs) and their role in endometriosis: a systematic literature review. Rev Environ Health. 2021;36(1):101-115. Delsouc MB, Conforti RA, Vitale DL, et al. Antiproliferative and antiangiogenic effects of ammonium tetrathiomolybdate in a model of endometriosis. Life Sci. 2021;287:120099. As-Sanie S, Kim J, Schmidt-Wilcke T, et al. Functional Connectivity is Associated with Altered Brain Chemistry in Women with Endometriosis-Associated Chronic Pelvic Pain. J Pain. 2016;17(1):1-13. Yan N, Xu Z, Qu C, Zhang J. Dimethyl fumarate improves cognitive deficits in chronic cerebral hypoperfusion rats by alleviating inflammation, oxidative stress, and ferroptosis via NRF2/ARE/NF-κB signal pathway. Int Immunopharmacol. 2021;98:107844. Samimi M, Pourhanifeh MH, Mehdizadehkashi A, Eftekhar T, Asemi Z. The role of inflammation, oxidative stress, angiogenesis, and apoptosis in the pathophysiology of endometriosis: Basic science and new insights based on gene expression. J Cell Physiol. 2019;234(11):19384-19392. Govatati S, Deenadayal M, Shivaji S, Bhanoori M. Mitochondrial NADH:ubiquinone oxidoreductase alterations are associated with endometriosis. Mitochondrion. 2013;13(6):782-790. 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-2742573","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":199988854,"identity":"67eb73e2-aa97-4a27-862f-e92c9363894e","order_by":0,"name":"Jiajia Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYFCCAwkHEgwk5IAsxgMMDAlEaUl88KDAwpiBgZmBWC0MzIYPPlQkNhCtxeDggWcSQIel90vkHzjwoSKNgb+9G78+gwMH0kBacmfOSGY4OONMDoPEmbMb8Goxg2nZcCOZ4TBvWwUDiE2UlnQDUrQkGwC1JEC15BDWYn8AGMhALYYzex4bAP2SxkPQL5IzziQc/PGnTp6fPfEhMLCT5fjbe/FrAQZQAgqfB79yEOBvP0BY0SgYBaNgFIxsAABsS1GQbZa/0wAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-4465-0622","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":true,"prefix":"","firstName":"Jiajia","middleName":"","lastName":"Wang","suffix":""},{"id":199988855,"identity":"a05f2858-070c-4583-83d0-160318019adb","order_by":1,"name":"Yiming Lu","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Yiming","middleName":"","lastName":"Lu","suffix":""},{"id":199988856,"identity":"4b591c9e-f310-4ca2-a54c-fa43d7832f64","order_by":2,"name":"Yongchang Ling","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Yongchang","middleName":"","lastName":"Ling","suffix":""},{"id":199988857,"identity":"fc2c88c0-c70b-4a7d-bf61-644cb2b7f797","order_by":3,"name":"Guangyu Sun","email":"","orcid":"","institution":"Shantou University Medical College","correspondingAuthor":false,"prefix":"","firstName":"Guangyu","middleName":"","lastName":"Sun","suffix":""},{"id":199988858,"identity":"1a0e5dce-e2e4-4364-a33a-2ea018aef2be","order_by":4,"name":"Zhihao Fang","email":"","orcid":"","institution":"Shantou University Medical College","correspondingAuthor":false,"prefix":"","firstName":"Zhihao","middleName":"","lastName":"Fang","suffix":""},{"id":199988859,"identity":"f0ce6462-7271-4e4e-b7e0-e9d514a217a7","order_by":5,"name":"Liqiao He","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Liqiao","middleName":"","lastName":"He","suffix":""},{"id":199988860,"identity":"373780c9-662a-41b8-80bc-7db86caf27d5","order_by":6,"name":"Zhiyong Xing","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Zhiyong","middleName":"","lastName":"Xing","suffix":""},{"id":199988861,"identity":"14950a7f-9106-406d-88f0-a29877b3373e","order_by":7,"name":"Weihua Nong","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Weihua","middleName":"","lastName":"Nong","suffix":""},{"id":199988862,"identity":"25c83664-ee4b-4369-a6a6-a06fc9cfc00b","order_by":8,"name":"Yunbao Wei","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Yunbao","middleName":"","lastName":"Wei","suffix":""},{"id":199988863,"identity":"b8f58e47-0a62-409c-a6cf-8158dbb38661","order_by":9,"name":"Shan Wang","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Shan","middleName":"","lastName":"Wang","suffix":""},{"id":199988864,"identity":"8c2ed6a1-2cd0-4b8d-b67e-512585b7a1c5","order_by":10,"name":"Guiling Shi","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Guiling","middleName":"","lastName":"Shi","suffix":""},{"id":199988865,"identity":"4c2a53df-680b-46b4-95d5-067ef34c4337","order_by":11,"name":"Mingyou Dong","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Mingyou","middleName":"","lastName":"Dong","suffix":""},{"id":199988866,"identity":"7f5e27a6-f332-4222-a468-4d9334a2395b","order_by":12,"name":"Junli Wang","email":"","orcid":"","institution":"Youjiang Medical University for Nationalities","correspondingAuthor":false,"prefix":"","firstName":"Junli","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2023-03-27 14:18:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2742573/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2742573/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":37095689,"identity":"7bf1dde7-abad-42c9-bcfc-8516ee1af108","added_by":"auto","created_at":"2023-05-16 15:16:12","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":130321,"visible":true,"origin":"","legend":"\u003cp\u003eThe dysregulated genes were shown in the \u003cstrong\u003eA\u003c/strong\u003e heat map via analyzing GSE7307 datasets. 10 CRGs were identified between EMT and normal samples. \u003cstrong\u003eB\u003c/strong\u003e CRGs expression in EMT and normal samples, blue: normal samples, red: tumor samples (*\u003cem\u003ep\u003c/em\u003e\u0026lt;0.05, **\u003cem\u003ep\u003c/em\u003e\u0026lt;0.01, ***\u003cem\u003ep\u003c/em\u003e\u0026lt;0.001).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/2f7b184365fcae07d8a9b604.png"},{"id":37094939,"identity":"e6963179-d2c7-4be8-9c58-34db0740d663","added_by":"auto","created_at":"2023-05-16 15:08:12","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":81438,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e DO enrichment analysis of 10 CRGs in EMT. \u003cstrong\u003eB\u003c/strong\u003e GO enrichment analysis of 10 CRGs in EMT. \u0026nbsp;\u003cstrong\u003eC\u003c/strong\u003e KEGG pathway analyses of 10 CRGs in EMT.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/f86c5a926f84997806acc25b.png"},{"id":37096656,"identity":"622356d3-00bb-4b86-b91e-10756ffdbd5b","added_by":"auto","created_at":"2023-05-16 15:24:12","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":137730,"visible":true,"origin":"","legend":"\u003cp\u003eSelection procedure of diagnostic markers for EMT diagnoses. \u003cstrong\u003eA, B\u003c/strong\u003e Tuning feature selection in the LASSO model. \u003cstrong\u003eC, D\u003c/strong\u003e Biomarker signature gene expression validation through the SVM-RFE arithmetic. \u003cstrong\u003eE\u003c/strong\u003e RandomForest error rate versus the number of classification trees. \u003cstrong\u003eF\u003c/strong\u003e The top 10 relatively important genes. \u003cstrong\u003eG, H\u003c/strong\u003e Venn graph presenting 3 diagnostic biomarkers shared by the LASSO, SVM-RFE and RandomForest arithmetic methods.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/8b54143179495701e174b69b.png"},{"id":37096655,"identity":"77966d3e-ef64-4ba1-a87b-c891d37cd3ee","added_by":"auto","created_at":"2023-05-16 15:24:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":61399,"visible":true,"origin":"","legend":"\u003cp\u003eThe expression pattern of the 3 critical genes in EMT. \u003cstrong\u003eA\u003c/strong\u003e GLS and NFE2L2 were highly expressed in EMT, and PDHA1 was lowly expressed in EMT. \u003cstrong\u003eB\u003c/strong\u003e The expression pattern of the 3 critical genes was further demonstrated in GSE7305 datasets.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/30c4bd05fa525d31abd5e8c3.png"},{"id":37094943,"identity":"e453c647-c726-45af-8b08-dc376b3364bc","added_by":"auto","created_at":"2023-05-16 15:08:12","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":83577,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e The diagnostic value of the 3 critical genes was studied using ROC assays in GSE7307. \u003cstrong\u003eB\u003c/strong\u003eThree genes (GLS, NFE2L2, and PDHA1) were further demonstrated to be diagnostic genes in GSE7305.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/5beb89cba5da838ed2730229.png"},{"id":37094936,"identity":"220eccaa-d7ca-4ee2-a40f-3209a57c50e1","added_by":"auto","created_at":"2023-05-16 15:08:12","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":93552,"visible":true,"origin":"","legend":"\u003cp\u003eThe expression pattern of the 3 critical genes in different severity groups of endometrioses. \u003cstrong\u003eA\u003c/strong\u003e Expression of CRGs in different severity groups of endometrioses. \u003cstrong\u003eB\u003c/strong\u003e ROC analysis of CRGs predicting the severity of endometriosis in GSE51981.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/5d7cbcb2fa4a195eeda2ca98.png"},{"id":37097504,"identity":"46bbce0e-5c1c-412c-9733-68d04068eca0","added_by":"auto","created_at":"2023-05-16 15:40:12","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":92590,"visible":true,"origin":"","legend":"\u003cp\u003eExpression of CRGs in the endometria of the infertile and fertile females with endometriosis. \u003cstrong\u003eA\u003c/strong\u003e Upregulated CRGs were decreased in the infertile group. \u003cstrong\u003eB\u003c/strong\u003e The ROC analysis of CRGs predicting infertility of endometriosis in GSE120103.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/d39b32b4c5927a17923c3753.png"},{"id":37095693,"identity":"64c557ef-d7a9-4a56-a527-294fccd3581c","added_by":"auto","created_at":"2023-05-16 15:16:12","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":150729,"visible":true,"origin":"","legend":"\u003cp\u003eGSEA revealed the enriched pathways of the hub genes. \u003cstrong\u003eA\u003c/strong\u003e GLS, \u003cstrong\u003eB\u003c/strong\u003e NFE2L2, and \u003cstrong\u003eC\u003c/strong\u003e PDHA1.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/8951a997124aac51bf1ef668.png"},{"id":37095690,"identity":"6ed5b17a-fd41-481f-b2e1-03f4be5862cf","added_by":"auto","created_at":"2023-05-16 15:16:12","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":185042,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e Correlation between GLS and infiltrating immune cells in EMT. \u003cstrong\u003eB\u003c/strong\u003e Correlation between\u003cstrong\u003e \u003c/strong\u003eNFE2L2 and infiltrating immune cells in EMT. \u003cstrong\u003eC\u003c/strong\u003e Correlation between PDHA1 and infiltrating immune cells in EMT.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/43e8362a716a236e716722f0.png"},{"id":37097236,"identity":"d920d67a-aa11-488f-b684-497a79ec956f","added_by":"auto","created_at":"2023-05-16 15:32:12","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":126182,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction and validation of the EMT diagnostic column line graph model. \u003cstrong\u003eA\u003c/strong\u003e Column line graphs are used to predict the occurrence of EMT. \u003cstrong\u003eB\u003c/strong\u003e Calibration curves to assess the predictive power of the column line graph model. \u003cstrong\u003eC\u003c/strong\u003eDCA curves to assess the clinical value of the column line graph model.\u003cstrong\u003e D\u003c/strong\u003eROC curves to assess the clinical value of the column line graph model.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/1e798cf0738239756434e5e6.png"},{"id":37094940,"identity":"466c313d-100e-4192-bb79-b166d50dd20c","added_by":"auto","created_at":"2023-05-16 15:08:12","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":115119,"visible":true,"origin":"","legend":"\u003cp\u003eDrug screening for targeting key CRGs\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/f862288259478eeebb4f7a13.png"},{"id":38362086,"identity":"f64b4b02-6f93-4e63-a35e-ffdbd556348a","added_by":"auto","created_at":"2023-06-11 22:53:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2569267,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2742573/v1/eadedf23-cb0c-440a-83ba-09df6e1d6717.pdf"}],"financialInterests":"","formattedTitle":"Machine Learning Algorithms for a Novel Cuproptosis-related Gene Signature of Diagnostic and Immune Infiltration in Endometriosis","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEndometriosis (EMT) is a common estrogen-dependent disease wherein the endometrial tissue implants and grows on the peritoneal surface, ovaries, rectovaginal septum, perineum, and surgical scars [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The primary symptoms of EMT include lower abdominal pain, dysmenorrhea, menstrual abnormalities, and discomfort during intercourse. Globally, nearly 5\u0026ndash;10% of women of reproductive age are affected by EMT, and 30\u0026ndash;50% of women with EMT suffer from infertility [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. However, the pathogenesis of EMT has not yet been elucidated. Several studies have identified markers associated with EMT [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. however, the sample size of these studies was small, which prevents accurate diagnosis. Therefore, there is an urgent need to identify new biomarkers to understand the pathophysiological mechanisms of EMT. This would aid in designing better and more accurate diagnostic as well as therapeutic approaches for EMT.\u003c/p\u003e \u003cp\u003eCopper is an essential cofactor regulating several enzymes involved in various cellular functions, including mitochondrial respiration, antioxidation, and synthesis of hormones, neurotransmitters, and pigment. However, abnormal copper levels can induce oxidative stress and cytotoxicity [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Recent studies have demonstrated that copper induces cell death by altering the tricarboxylic acid cycle (TCA), which causes the accumulation of lipid-acylated proteins and a decrease in iron-sulfur cluster protein levels. This triggers a proteotoxic stress response called cuproptosis [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Moreover, studies have demonstrated that copper homeostasis is critically involved in immune cell infiltration. Tan et al. showed that copper-induced H2O2 production mediated by LOXL4 activates the type I interferon signaling pathway, which promotes PD-L1 presentation by macrophages [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. A study has demonstrated a close correlation between the copper-related protein STEAP2 and the prognosis, as well as the infiltration of immune cells in gliomas [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Ji et al. indicated that cuproptosis could influence the development and prognosis of patients with renal clear cell carcinoma by regulating immune cell infiltration [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTherefore, it is necessary to study the involvement of cuproptosis in the diagnosis and immune cell infiltration in diseases. However, no studies have determined the influence of cuproptosis-mediated immune cell infiltration on EMT pathogenesis. Therefore, in this study, we combined data on EMTs and cuproptosis-related genes (CRGs) to determine the underlying mechanism by which cuproptosis mediates immune cell infiltration in EMT. First, we obtained data on 19 CRGs from previous studies [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] and the GSE7307 datasets from the \"Gene Expression Omnibus (GEO)\u0026rdquo; database to perform differential and functional correlation analysis. Next, we identified three key CRGs using machine learning algorithms. Furthermore, we analyzed the correlation between key CRGs, immune cells, and pathways. Finally, we constructed a nomogram and screened for drug targets to provide insights into cuproptosis in EMT. The outcomes of this study are summarized in the following table. These results would aid in enhancing our understanding of the involvement of cuproptosis in regulating EMT.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData download and pre-processing\u003c/h2\u003e \u003cp\u003eWe retrieved data on 19 CRGs from previously published studies and the GSE7307 dataset consisting of microarray data on EMT of 677 samples from GEO. To analyze the data, different probes corresponding to the same gene were identified, and the null probes were removed. (3) Next, a probe corresponding to a gene was identified, and (4) log2-transformed quantile normalization was performed. Finally, we identified 41 samples, consisting of 18 patients with EMT and 23 healthy individuals. The GSE7305 dataset consisting of ten patients with EMT and ten healthy individuals was used as a validation set.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eScreening for differentially expressed CRGs\u003c/h3\u003e\n\u003cp\u003eWe employed the \u0026ldquo;linear models for microarray data\u0026rdquo; package for screening differentially expressed genes (DEG) in tissues of patients with EMT and healthy individuals based on the following criteria: \u0026ldquo;P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u0026rdquo;\u003c/p\u003e\n\u003ch3\u003eFunctional enrichment Analysis of DEGs\u003c/h3\u003e\n\u003cp\u003eWe used the \u0026ldquo;Disease Ontology (DO),\u0026rdquo; \u0026ldquo;Gene Ontology (GO),\u0026rdquo; and \u0026ldquo;Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses\u0026rdquo; to determine the biological roles of CRGs. P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant. Further, the visual enrichment maps were constructed using the \"ggplot2\" and \"GOplot\" R packages.\u003c/p\u003e\n\u003ch3\u003eMachine learning algorithm to identify key CRGs\u003c/h3\u003e\n\u003cp\u003eWe used three machine learning algorithms such as \"LASSO regression analysis,\u0026rdquo; \u0026ldquo;SVM-RFE,\u0026rdquo; and \u0026ldquo;RF\u0026rdquo; R package, to identify DEGs. Finally, the results were visualized by constructing a Venn diagram using the \"venneuler\" R package.\u003c/p\u003e\n\u003ch3\u003eDiagnostic value of key CRGs in EMT and external validation\u003c/h3\u003e\n\u003cp\u003eWe determined the diagnostic value of these key CRGs in GSE7307 and validated these genes in GSE7305. For GSE51981 and GSE120103, we plotted the receiver operating characteristic (ROC) curves to determine the diagnostic value of these genes using \u0026ldquo;P\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u0026rdquo; as the threshold. We calculated the area under the ROC curve (AUC), and the AUC value between 0 and 1 was used for analysis. The greater the AUC value, the better the performance of these genes in predicting diagnosis.\u003c/p\u003e\n\u003ch3\u003eScreening key regulatory pathways using Single sample genomic enrichment analysis (ssGSEA)\u003c/h3\u003e\n\u003cp\u003eThe median expression of key genes was set as the criteria to classify tissues of patients with EMT into the high- and low-expression groups. Furthermore, analysis of the variance was performed using the \u0026ldquo;ssGSEA\u0026rdquo; R package on gene sets (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) to identify pathways enriched by key CRGs.\u003c/p\u003e\n\u003ch3\u003eImmunological analysis\u003c/h3\u003e\n\u003cp\u003eThe relative proportions of the 22 types of infiltrating immune cells in all EMT samples were determined using \u0026ldquo;Cell-type Identification by Estimating Relative Subsets of RNA Transcripts\u0026rdquo; (CIBERSORT, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://cibersortx.stanford.edu\u003c/span\u003e\u003cspan address=\"http://cibersortx.stanford.edu\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Next, we calculated immune scores using the \u0026ldquo;Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data\" algorithm. Finally, Spearman rank correlation analysis was performed to determine the correlation between CRGs and the number of infiltrating immune cells. The correlation was visualized using the \u0026ldquo;ggplot2\u0026rdquo; package.\u003c/p\u003e\n\u003ch3\u003eConstruction of nomogram prediction model and efficacy assessment for clinical diagnosis of EMT\u003c/h3\u003e\n\u003cp\u003eWe constructed a nomogram prediction model and validated it internally using the \"rms\" R package. The performance of the model was assessed using calibration curves, clinical decision curve analysis (DCA), and ROC curves.\u003c/p\u003e\n\u003ch3\u003eDrug screening for targeting key CRGs\u003c/h3\u003e\n\u003cp\u003eThe \u0026ldquo;DrugBank\u0026rdquo; database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://go.drugbank.com/drugs\u003c/span\u003e\u003cspan address=\"https://go.drugbank.com/drugs\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used to predict potential therapeutic targets of key CRGs associated with EMT. The \u0026ldquo;DrugBank database is a unique bioinformatics and cheminformatics resource that contains more than 13,791 drug entries, including small molecules, biotechnology, nutraceuticals, and investigational drugs. Additionally, there are \u0026gt;\u0026thinsp;5236 nonredundant protein sequences associated with these drug entries.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eR 4.0.5 statistical software was used to complete the analysis. We used the t-test and Wilcoxon rank-sum test for quantifying variables. Furthermore, Spearman correlation analysis was used to investigate the correlation between CRGs and immune cell infiltration. P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003eScreening for DEGs in patients with EMT\u003c/h2\u003e\n \u003cp\u003eWe combined 19 CRGs from previously published studies and DEGs screened from patients with EMT. A total of 10 DEGs were screened from GSE7307, of which six were upregulated, and four were downregulated (Figs. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA, B).\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eFunctional enrichment Analysis of DEGs\u003c/h3\u003e\n\u003cp\u003eWe performed the DO, GO, and KEGG pathway enrichment analyses on DEGs. The DO enrichment analysis demonstrated that these key CRGs were mainly enriched in the nutrient deficiency disease pathway associated with carbohydrate metabolism disorder (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA). The GO enrichment analysis revealed that these key CRGs were enriched in various processes such as acetyl-CoA biosynthesis from pyruvate, acetyl-CoA metabolism, and thioester, acyl-a CoA, a sulfur compound, nucleoside bisphosphate, ribonucleoside, and acyl-CoA biosynthesis processes. The KEGG enrichment analysis showed pathways, including TCA, Glycolysis/Gluconeogenesis, pyruvate, and carbon metabolism, as well as central carbon metabolism in cancer, resistance to platinum drugs, and the glucagon and HIF-1 signaling pathways were enriched by these CRGs. However, no significant difference was observed between these CRGs (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC).\u003c/p\u003e\n\u003ch3\u003eScreening of EMT markers using machine learning algorithms\u003c/h3\u003e\n\u003cp\u003eA total of ten DEGs were identified using machine learning as signature genes. Five signature genes, including NFE2L2, NLRP3, DLD, PDHA1, and GLS, were identified using the \u0026ldquo;LASSO regression analysis\u0026rdquo; (Figs. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA, B). Next, six signature genes, including GLS, PDHB, DLD, NFE2L2, PDHA1, and SLC31A1, were identified using the \u0026ldquo;SVM-RFE\u0026rdquo; algorithm (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC, D). Finally, five signature genes, including PDHA1, GLS, NFE2L2, DLAT, and MTF1, were identified using the \u0026ldquo;RF\u0026rdquo; algorithm (Figs. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eE, F). Further, genes identified using these three machine learning algorithms were intersected, and three signature genes, including GLS, NFE2L2, and PDHA1, were identified. The three genes were intersected to obtain three signature genes, GLS, NFE2L2, and PDHA1, and plotted on a Wayne diagram (Figs. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eG, H).\u003c/p\u003e\n\u003ch3\u003eExpression and validation of the diagnostic significance of the signature genes\u003c/h3\u003e\n\u003cp\u003eFirst, we determined the differential expression and diagnostic value of GLS, NFE2L2, and PDHA1 in GSE7307. Further, these genes were externally validated in GSE7305. The results revealed a significant increase in GLS and NFE2L2 expression and a significant decrease in PDHA1 expression in both datasets (Figs. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003eA, B). The AUC values of PDHA1 in the two datasets were 0.879 and 0.920, respectively (Figs. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA, B). The AUC values of all three diagnostic genes were \u0026gt;\u0026thinsp;0.800, thereby indicating these CRGs had high diagnostic values.\u003c/p\u003e\n\u003ch3\u003eExpression of key CRGs in EMT with different severity and combined infertility\u003c/h3\u003e\n\u003cp\u003eTo validate the confidence and applicability of the key CRGs, external datasets such as GSE51981 and GSE120103 were selected to validate the correlation between key CRGs, EMT severity, and combined infertility. Since PDA1 was not expressed in GSE120103, PDHA1 was not analyzed in EMT samples. The patients with micro/light EMT were categorized into one group, and patients with moderate/severe EMT were categorized into another group. The results revealed high GLS and NFE2L2 expression in patients with moderate/severe EMT (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA). ROC curves were performed on these upregulated key CRGs to analyze the severity of EMT (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) and the AUC values\u0026thinsp;\u0026gt;\u0026thinsp;0.7 (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eB). Next, we used GSE120103, consisting of patients with EMT who may or may not be suffering from infertility, for subsequent validation. The results revealed an increase in GLS and NFE2L2 expression in the endometrium of patients with EMT; however, a decrease in GLS and NFE2L2 expression was observed in the endometrium of patients with infertility (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA). On the contrary, we performed ROC analysis to determine the significance of key CRGs in diagnosing patients with EMT. The AUC values validated the sensitivity and specificity of CRGs (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eB).\u003c/p\u003e\n\u003ch3\u003eAnalyzing Regulatory pathway enriched by key CRGs\u003c/h3\u003e\n\u003cp\u003eThe ssGSEA results showed that the complement and coagulation pathways, Legionellosis, Malaria, Pertussis, Systemic lupus erythematosus (SLE), etc., were enriched in the high GLS and high NFE2L2 expression groups (Figs. \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003eA, B). In addition, the cell cycle, the Fanconi anemia and protein export pathways, DNA replication, and Homologous recombination are enriched in the PDHA1 expression group (Fig. \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003eC).\u003c/p\u003e\n\u003ch3\u003eCorrelation between three key CRGs and proportion of immune cell infiltration in EMT\u003c/h3\u003e\n\u003cp\u003eWe utilized the \u0026ldquo;CIBERSORT\u0026rdquo; method to determine the proportion of 22 immune cell subpopulations infiltrating in patients with EMT. GLS, NFE2L2, and PDHA1 expression were evaluated as a function of the immune microenvironment. A significant positive correlation was observed between GLS and immune cells like M2 macrophages, Plasma cells, na\u0026iuml;ve CD4 T cells, and activated Mast cells. GLS was negatively correlated with activated NK cells (Fig. \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eA). NFE2L2 was significantly correlated with immune cells such as Plasma cells, memory B cells, resting memory CD4 T cells, Eosinophils, M1 and M2 macrophages, resting mast cells, and CD8 T cells (Fig. \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eB). Furthermore, a significant positive correlation was observed between NFE2L2 and immune cells such as activated NK cells, CD8T cells, M1 and M2 Macrophages, resting mast cells, na\u0026iuml;ve B cells, resting memory CD4 T cells, eosinophils, memory B cells, and Plasma cells (Fig. \u003cspan class=\"InternalRef\"\u003e9\u003c/span\u003eC).\u003c/p\u003e\n\u003ch3\u003eConstruction and validation of the Nomogram prediction model\u003c/h3\u003e\n\u003cp\u003eWe constructed a nomogram prediction model using GLS, NFE2L2, and PDHA1 for predicting the risk of EMT development (Fig. \u003cspan class=\"InternalRef\"\u003e10\u003c/span\u003eA). Nomogram was externally validated using GSE7305. The rate of incidences was derived from each risk factor score and summed to obtain a total score. Next, we plotted the calibration curve (Fig. \u003cspan class=\"InternalRef\"\u003e10\u003c/span\u003eB). The results revealed that the three curves were consistent and converged, thereby indicating a good agreement between the actual and predicted outcomes of the model. In addition, DCA showed the clinical benefit of the nomogram prediction model (Fig. \u003cspan class=\"InternalRef\"\u003e10\u003c/span\u003eC). Finally, we plotted the ROC curve (the AUC value\u0026thinsp;=\u0026thinsp;0.942) to determine the predictive ability of the columnar graph model (Fig. \u003cspan class=\"InternalRef\"\u003e10\u003c/span\u003eD). The AUC value of the nomogram prediction model was 0.942, thereby indicating that the model had a good predictive ability.\u003c/p\u003e\n\u003ch3\u003eDrug prediction\u003c/h3\u003e\n\u003cp\u003eWe identified five drugs targeting three key CRGs using data from the \u0026ldquo;DrugBank database (Fig. \u003cspan class=\"InternalRef\"\u003e11\u003c/span\u003e), and all of these drugs have been approved. First, L-glutamine (DB00130) was identified as a drug targeting CRGs. It is an important energy source in cells and is involved in metabolic processes. It is used to treat patients with sickle cell disease gastric and duodenal ulcers, improving brain function. Next, we identified Glutamate (DB00142), a common amino acid and an excitatory neurotransmitter. It is a precursor molecule required for GABA synthesis in GABAergic neurons. It is mainly used as total parenteral nutrition and as an adjunct for treating psycho-neurological disorders. Next, Dimethyl fumarate (DMF, DB08908) binds to NFE2L2. It is used for treating patients with relapsing multiple sclerosis. Finally, NADH (DB00157) targets PDHA1 and is used as supplementary nutrition.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eCurrently, invasive techniques, such as laparoscopic histopathological examination, are used for EMT diagnosis. The most common treatment modalities include hormonal suppression or surgical excision/ablation of visible lesions. However, the therapeutic outcomes are unsatisfactory and often accompanied by adverse effects. Studies have identified several key molecules, such as adhesion molecules, matrix metalloproteinases, lncRNA HOTAIR, oxidative stress, and regulatory T cells, that are involved in EMT development [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, the exact mechanism is still unclear. Therefore, there is an urgent need to identify new biomarkers associated with EMT pathogenesis. This would aid in developing non-invasive diagnostic and screening tools for patients with EMT.\u003c/p\u003e \u003cp\u003eCopper is an essential cofactor of enzymes. It is critically involved in various physiological processes and activities. Turgut et al. showed that copper was associated with the pathogenesis of EMT and oxidative stress [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Cuproptosis is a newly discovered form of cell death involving TCA regulation. Lipid-acylated proteins induce cytotoxic stress, which leads to cell death 6. However, no studies have shown the involvement of cuproptosis in EMT. Therefore, in this study, we combined data on EMT and CRGs from GEO. We identified 10 DEGs, including NFE2L2, NLRP3, SLC31A1, DLAT, DLD, PDHB, PDHA1, MTF1, CDKN2A, and GLS. The DO, GO, and KEGG pathway enrichment analysis showed that key CRGs were primarily enriched in the acetyl-CoA and pyruvate metabolism, TCA, and Glycolysis/Gluconeogenesis. Bradbeer et al. showed a correlation between TCA enzyme succinate dehydrogenase and EMT [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Interestingly, alteration in TCA occurs in cuproptosis, which induces cell death. Therefore, these results indicate that these CRGs could be a promising target for modulating cuproptosis in EMT, which could be a new therapeutic strategy for treating EMT.\u003c/p\u003e \u003cp\u003eNext, we used three machine learning algorithms, such as \u0026ldquo;LASSO regression,\u0026rdquo; \u0026ldquo;SVM-RFE,\u0026rdquo; and \u0026ldquo;RF,\u0026rdquo; to screen for EMT and cuproptosis-related biomarkers. We identified three CRGs associated with immune cell infiltration in EMT. Of these three CRGs, an increase in GLS and NFE2L2 expression and a decrease in PDHA1 expression were observed. Furthermore, the ROC curve analysis validated the diagnostic value of these key CRGs in distinguishing the endometrium of patients with EMT and healthy controls. Finally, we validated our results using the external validation dataset GSE7305, and the results obtained were consistent.\u003c/p\u003e \u003cp\u003eGlutaminase (GLS) is located on chromosome 2. It is a key enzyme mediating glutamine-to-glutamate conversion. GLS is converted to alpha-ketoglutarate, which enters TCA to produce energy. GLS is an oncogene, and an increase in GLS expression was observed in pancreatic ductal adenocarcinoma. Under oxidative stress, SUCLA2 increases glutamine catabolism and the production of nicotinamide adenine dinucleotide phosphate as well as glutathione by activating GLS, thereby ameliorating oxidative stress and promoting tumor growth [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Our results revealed a significant correlation between CRGs like GLS and EMT progression. Nuclear transcription factor E2-related factor 2 (NFE2 like bZIP transcription factor 2, NFE2L2, also known as Nrf2) is primarily localized in the cytoplasm. It is an antioxidant transcription factor that regulates inflammatory responses and oxidative stress in cells [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. NFE2L2 variants are closely linked to the risk of human disease. Interestingly, oxidative damage promotes EMT progression. Previous studies have shown that NFE2L2 increases EMT risk by inhibiting processes such as oxidative stress against the growth and development of EMT [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The pyruvate dehydrogenase complex E1 alpha subunit (PDHA1) is the E1 subunit of the pyruvate dehydrogenase complex. It is primarily localized in the mitochondria and contains three serine residues, which could be activated by four inhibitory PDHK1-4 and two reactivated phosphatases (PDP1-2). These PDP1-2 are reversibly phosphorylated and regulate glycolysis and TCA [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Spakova et al. demonstrated that silencing HIF-1α/miR210 expression increases PDHA1 and decreases MITF-M expression. This promotes mitochondrial respiratory activity, which aids in eliminating reactive oxygen species in melanoma cells [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Our results indicate a significant role of GLS, NFE2L2, and PDHA1 in EMT pathogenesis. ROC curve analysis revealed that these key CRGs are significantly involved in the onset and progression of EMT, thereby indicating the potential diagnostic value of these CRGs in clinical settings. Moreover, risk modeling results suggested that GLS, NFE2L2, and PDHA1 could be risk factors for EMT. Based on previous studies and our results, CRGs may play an important role in EMT development, and the underlying mechanisms should be further investigated.\u003c/p\u003e \u003cp\u003eWe also determined the expression of key CRGs in patients with varying severity of EMT and those suffering from infertility. The results revealed an increase in GLS and NFE2L2 expression in patients with moderate/severe EMT. Furthermore, an increase in the expression of key CRGs could lead to infertility in patients with EMT. Qiao et al. showed that isoniazid activates the Keap1/Nrf2 signaling pathway by inducing oxidative stress and apoptosis, thereby impairing the reproductive system and reducing fertility in mammals [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Ivanov et al. showed that melatonin could protect the developing embryo from oxidative stress by modulating NFE2L2, SOD1, and GPX1 expression [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. These results indicate a close correlation between NFE2L2 and infertility, consistent with our results.\u003c/p\u003e \u003cp\u003eNext, we performed ssGSEA to verify the involvement of key CRGs in EMT. Our results revealed significant enrichment of the complement and coagulation pathways, Legionellosis, Malaria, Pertussis, and SLE in the high GLS and high NFE2L2 expression groups. On the contrary, the cell cycle, the Fanconi anemia and protein export pathways, DNA replication, and homologous recombination were enriched in the high PDHA1 expression group. SLE is an autoimmune disease characterized by the presence of specific autoantibodies, including antinuclear antibodies. Studies have shown a correlation between EMT and autoimmune diseases such as SLE and Sj\u0026ouml;gren's syndrome [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHowever, the underlying mechanism of the pathogenesis of EMT is still unclear. Implantation of reflux menstrual blood, endocrine, genetic factors, angiogenesis, and stem cell differentiation are important factors associated with EMT pathogenesis. Of these factors, abnormal immune responses could be the underlying mechanism of EMT pathogenesis. In fact, the immune component of EMT has received widespread attention from researchers. Some studies have demonstrated that an increase in the secretion of proinflammatory cytokines due to immune dysfunction, impaired immune surveillance, and altered immune cell profiles contribute to EMT progression. Prolonged persistent immune dysregulation could lead to a state of chronic inflammation, thereby creating a conducive environment for promoting adhesion and angiogenesis. This could lead to a vicious cycle of EMT development and progression [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. The \u0026ldquo;CIBERSORT\u0026rdquo; method is widely used to study EMT in humans. It is based on the principles of linear support vector regression to deconvolute the expression matrix of immune cell subtypes.\u003c/p\u003e \u003cp\u003eTherefore, we used CIBERSORT to assess infiltrating immune cells for determining the role of immune cell infiltration in EMT. Our results revealed that these three CRGs were associated with various immune cell types, including eosinophils, activated NK cells, and macrophages. Eidukaite et al. showed that epithelial cells in endometriotic foci secrete high levels of eosinophil-specific chemokines compared to in situ endometrial cells. These eosinophils and other myeloid cells enter the peritoneum, which triggers inflammatory and allergic responses [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Ścieżyńska et al. showed that endometriotic cells' survival and growth in the peritoneal cavity might also be due to their recognition and clearance by local immune cells such as macrophages and NK cells [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Ram\u0026iacute;rez-Pavez et al. demonstrated that macrophages accumulate in the peritoneal cavity of patients with EMT; however, their ability to clear migrating endometrial debris is reduced [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Furthermore, combining the results of differential immune cell infiltration and the correlation between cuproptosis and immune cell infiltration revealed that immune cells such as eosinophils, activated NK cells, and M2 Macrophages could be critically involved in the modulation of CRG-regulated immune function in EMT.\u003c/p\u003e \u003cp\u003eEMT is an aggressive disease. Koninckx et al. have suggested that oxidative stress and peritoneal microbiota due to retrograde menstruation could cause EMT. Therefore, pharmacological treatment could prevent new lesions and be prescribed post-surgery. Additionally, Sirohi et al. showed that inhibiting copper-induced toxicity increases the survival of endometriotic cells [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Therefore, copper chelators could be used as therapeutic agents for treating patients with EMT [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. We identified five drugs from the DrugBank database targeting these three CRGs. Qinpi methicin (DB13155) has anti-inflammatory and antioxidant activities. Hence, qinpi methicin could be used for treating diseases caused by inflammation and oxidative stress, such as EMT. As-Sanie et al. demonstrated high forebrain insula glutamine (DB00130)-glutamate (DB00142) binding in EMT patients with chronic pelvic pain. High islet activity could cause chronic pelvic pain in patients with EMT; therefore, reducing islet glutamate levels could alleviate chronic pelvic pain [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. DMF is an immunomodulatory and antioxidant molecule commonly used to treat patients with relapsing multiple sclerosis and psoriasis. It binds with NFE2L2. Chen et al. demonstrated that DMF, an Nrf2 agonist, could significantly influence immune responses and oxidative stress. Additionally, Yan et al. suggested that DMF could attenuate inflammation, oxidative stress, and iron death via the NRF2/ARE/NF-kB signaling pathway. Moreover, DMF improves cognitive dysfunction in rats with chronic hypoperfusion [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Also, EMT is closely associated with oxidative stress [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Therefore, it is necessary to investigate if DMF could be used for treating patients with EMT. NADH is a nutritional supplement. It has antioxidant properties, reduces anxiety, promotes neurotransmitter synthesis, and prevents dementia. Govatati et al. showed that altering the mitochondrial membrane complex I could be a risk factor for EMT. Mitochondrial membrane complex I catalyze the transfer of electrons from NADH to ubiquinone and is closely correlated with EMT, which could be a new therapeutic approach [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHowever, our study has a few limitations. First, the sample size of our study was relatively small since data were only obtained from GEO. Therefore, additional studies using a larger sample size should be performed to validate our results. Second, although we have successfully identified three CRGs as potential biomarkers for immunophenotyping of EMT, no in vivo or in vitro studies were conducted to validate these results. Thus, future studies should focus on conducting in vivo or in vitro studies to validate these biomarkers.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eGLS, NFE2L2, and PDHA1 could be novel diagnostic markers for EMT, representing a significant breakthrough. Additionally, the correlation between these CRGs and immune cell infiltration could aid in developing effective immunotherapy for patients with EMT.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003eThe authors express their gratitude for the invaluable assistance and insightful discussions provided by the members of the Department of Youjiang Medical College for Nationalities. Additionally, the authors would like to thank the GEO database for granting access to the data used in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u0026nbsp;\u003c/strong\u003eSince the GEO database information is public, ethical approval or informed consent is not required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u0026nbsp;\u003c/strong\u003eAll authors have agreed to take part in the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e All the data in this study are available.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u0026nbsp;\u003c/strong\u003eAll codes for this study are available.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contribution\u0026nbsp;\u003c/strong\u003eJ.L.W. designed the study. J.J.W. produced the initial draft of the manuscript. M.Y.D. produced the methodology. Y.M.L. and Y.C.L. contributed to analyzing data. G.L.S. contributed to drafting of the manuscript. All authors have read and approved the final submitted manuscript.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eKapoor R, Stratopoulou CA, Dolmans M. Pathogenesis of Endometriosis: New Insights into Prospective Therapies. Int J Mol Sci. 2021;22(21):11700. \u003c/li\u003e\n\u003cli\u003eTaylor HS, Kotlyar AM, Flores VA. Endometriosis is a chronic systemic disease: clinical challenges and novel innovations. Lancet. 2021;397(10276):839-852. \u003c/li\u003e\n\u003cli\u003eHodgson RM, Lee HL, Wang R, Mol BW, Johnson N. Interventions for endometriosis-related infertility: a systematic review and network meta-analysis. Fertil Steril. 2020;113(2):374-382.e2. \u003c/li\u003e\n\u003cli\u003eAnastasiu CV, Moga MA, Elena Neculau A, et al. Biomarkers for the Noninvasive Diagnosis of Endometriosis: State of the Art and Future Perspectives. Int J Mol Sci. 2020;21(5):1750. \u003c/li\u003e\n\u003cli\u003eGe EJ, Bush AI, Casini A, et al. Connecting copper and cancer: from transition metal signalling to metalloplasia. Nat Rev Cancer. 2022;22(2):102-113. \u003c/li\u003e\n\u003cli\u003eTsvetkov P, Coy S, Petrova B, et al. Copper induces cell death by targeting lipoylated TCA cycle proteins. Science. 2022;375(6586):1254-1261.\u003c/li\u003e\n\u003cli\u003eTan HY, Wang N, Zhang C, Chan YT, Yuen MF, Feng Y. Lysyl Oxidase-Like 4 Fosters an Immunosuppressive Microenvironment During Hepatocarcinogenesis. Hepatology. 2021;73(6):2326-2341. \u003c/li\u003e\n\u003cli\u003eWang X, Han M, Chen S, Sun Y, Tan R, Huang B. The copper-associated protein STEAP2 correlated with glioma prognosis and immune infiltration. Front Cell Neurosci. 2022;16. \u003c/li\u003e\n\u003cli\u003eJi Z, Ren W, Wang H, Gao W, Yuan B. Molecular Subtyping Based on Cuproptosis-Related Genes and Characterization of Tumor Microenvironment Infiltration in Kidney Renal Clear Cell Carcinoma. Front Oncol. 2022;12.\u003c/li\u003e\n\u003cli\u003eJiang X, Ke J, Jia L, et al. A novel cuproptosis-related gene signature of prognosis and immune microenvironment in head and neck squamous cell carcinoma cancer. J Cancer Res Clin Oncol. 2023;149(1):203-218.\u003c/li\u003e\n\u003cli\u003eZhang L, Yu Z, Qu Q, Li X, Lu X, Zhang H. Exosomal lncRNA HOTAIR Promotes the Progression and Angiogenesis of Endometriosis via the miR-761/HDAC1 Axis and Activation of STAT3-Mediated Inflammation. Int J Nanomedicine. 2022;17:1155-1170. \u003c/li\u003e\n\u003cli\u003eKoninckx PR, Fernandes R, Ussia A, et al. Pathogenesis Based Diagnosis and Treatment of Endometriosis. Front Endocrinol (Lausanne). 2021;12. \u003c/li\u003e\n\u003cli\u003eTurgut A, Ozler A, Goruk NY, Tunc SY, Evliyaoglu O, Gul T. Copper, ceruloplasmin and oxidative stress in patients with advanced-stage endometriosis. Eur Rev Med Pharmacol Sci. 2013;17(11):1472-8.\u003c/li\u003e\n\u003cli\u003eBradbeer JN, Lindsay PC, Reeve J. Fluctuation of mineral apposition rate at individual bone-remodeling sites in human iliac cancellous bone: independent correlations with osteoid width and osteoblastic alkaline phosphatase activity. J Bone Miner Res. 1994;9(11):1679-86. \u003c/li\u003e\n\u003cli\u003eTong Y, Guo D, Lin S, et al. SUCLA2-coupled regulation of GLS succinylation and activity counteracts oxidative stress in tumor cells. Mol Cell. 2021;81(11):2303-2316.e8. \u003c/li\u003e\n\u003cli\u003eLee DH, Park JS, Lee YS, et al. SQSTM1/p62 activates NFE2L2/NRF2 via ULK1-mediated autophagic KEAP1 degradation and protects mouse liver from lipotoxicity. Autophagy. 2020;16(11):1949-1973.\u003c/li\u003e\n\u003cli\u003eSiracusa R, D Amico R, Cordaro M, et al. The Methyl Ester of 2-Cyano-3,12-Dioxooleana-1,9-Dien-28-Oic Acid Reduces Endometrial Lesions Development by Modulating the NFkB and Nrf2 Pathways. Int J Mol Sci. 2021;22(8):3991. \u003c/li\u003e\n\u003cli\u003eOsiński M, Mostowska A, Wirstlein P, Wender-Ożegowska E, Jagodziński PP, Szczepańska M. The assessment of GWAS \u0026mdash; identified polymorphisms associated with infertility risk in Polish women with endometriosis. Ginekol Pol. 2018;89(6):304-310. \u003c/li\u003e\n\u003cli\u003eEcheverri Ruiz NP, Mohan V, Wu J, et al. Dynamic regulation of mitochondrial pyruvate metabolism is necessary for orthotopic pancreatic tumor growth. Cancer Metab. 2021;9(1). \u003c/li\u003e\n\u003cli\u003eXu X, Tang Y, Zhu J, Luo J. Endoplasmic Reticulum Stress-Related Four-Biomarker Risk Classifier for Survival Evaluation in Esophageal Cancer. J Oncol. 2022;2022:1-14. \u003c/li\u003e\n\u003cli\u003eApakov\u0026aacute; I, Rabajdov\u0026aacute; M, Mičkov\u0026aacute; H, Graier WF, Marekov\u0026aacute; M. Effect of hypoxia factors gene silencing on ROS production and metabolic status of A375 malignant melanoma cells. Sci Rep. 2021;11(1). \u003c/li\u003e\n\u003cli\u003eQiao P, Zhang Y, Yang Y, et al. Oral isoniazid causes oxidative stress, oocyte deterioration and infertility in mice. Toxicology. 2021;455:152749. \u003c/li\u003e\n\u003cli\u003eIvanov D, Mazzoccoli G, Anderson G, et al. Melatonin, Its Beneficial Effects on Embryogenesis from Mitigating Oxidative Stress to Regulating Gene Expression. Int J Mol Sci. 2021;22(11):5885. \u003c/li\u003e\n\u003cli\u003eGeng R, Huang X, Li L, et al. Gene expression analysis in endometriosis: Immunopathology insights, transcription factors and therapeutic targets. Front Immunol. 2022;13. \u003c/li\u003e\n\u003cli\u003eShigesi N, Kvaskoff M, Kirtley S, et al. The association between endometriosis and autoimmune diseases: a systematic review and meta-analysis. Hum Reprod Update. 2019;25(4):486-503. \u003c/li\u003e\n\u003cli\u003eJiang I, Yong PJ, Allaire C, Bedaiwy MA. Intricate Connections between the Microbiota and Endometriosis. Int J Mol Sci. 2021;22(11):5644. \u003c/li\u003e\n\u003cli\u003eEidukaite A, Tamosiunas V. Activity of eosinophils and immunoglobulin E concentration in the peritoneal fluid of women with endometriosis. Clin Chem Lab Med. 2004;42(6):590-4. \u003c/li\u003e\n\u003cli\u003eŚcieżyńska AA, Komorowski MM, Soszyńska MM, Malejczyk JJ. NK Cells as Potential Targets for Immunotherapy in Endometriosis. J Clin Med. 2019;8(9):1468. \u003c/li\u003e\n\u003cli\u003eRam\u0026iacute;rez-Pavez TN, Mart\u0026iacute;nez-Esparza M, Ruiz-Alcaraz AJ, Mar\u0026iacute;n-S\u0026aacute;nchez P, Machado-Linde F, Garc\u0026iacute;a-Pe\u0026ntilde;arrubia P. The Role of Peritoneal Macrophages in Endometriosis. Int J Mol Sci. 2021;22(19):10792. \u003c/li\u003e\n\u003cli\u003eSirohi D, Al RR, Knibbs LD. Environmental exposures to endocrine disrupting chemicals (EDCs) and their role in endometriosis: a systematic literature review. Rev Environ Health. 2021;36(1):101-115. \u003c/li\u003e\n\u003cli\u003eDelsouc MB, Conforti RA, Vitale DL, et al. Antiproliferative and antiangiogenic effects of ammonium tetrathiomolybdate in a model of endometriosis. Life Sci. 2021;287:120099. \u003c/li\u003e\n\u003cli\u003eAs-Sanie S, Kim J, Schmidt-Wilcke T, et al. Functional Connectivity is Associated with Altered Brain Chemistry in Women with Endometriosis-Associated Chronic Pelvic Pain. J Pain. 2016;17(1):1-13.\u003c/li\u003e\n\u003cli\u003eYan N, Xu Z, Qu C, Zhang J. Dimethyl fumarate improves cognitive deficits in chronic cerebral hypoperfusion rats by alleviating inflammation, oxidative stress, and ferroptosis via NRF2/ARE/NF-\u0026kappa;B signal pathway. Int Immunopharmacol. 2021;98:107844. \u003c/li\u003e\n\u003cli\u003eSamimi M, Pourhanifeh MH, Mehdizadehkashi A, Eftekhar T, Asemi Z. The role of inflammation, oxidative stress, angiogenesis, and apoptosis in the pathophysiology of endometriosis: Basic science and new insights based on gene expression. J Cell Physiol. 2019;234(11):19384-19392. \u003c/li\u003e\n\u003cli\u003eGovatati S, Deenadayal M, Shivaji S, Bhanoori M. Mitochondrial NADH:ubiquinone oxidoreductase alterations are associated with endometriosis. Mitochondrion. 2013;13(6):782-790.\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":"cuproptosis, endometriosis, immune infiltration, machine learning algorithms","lastPublishedDoi":"10.21203/rs.3.rs-2742573/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2742573/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEndometriosis (EMT) is a chronic hormone-dependent disease where in viable endometrial tissue is transplanted outside the uterus. Interestingly, immune infiltration is significantly involved in EMT pathogenesis. Currently, no studies have shown the involvement of cuproptosis-related genes (CRGs) in regulating immune infiltration in EMT. This study identified three CRGs such as GLS, NFE2L2, and PDHA1, associated with EMT using machine learning algorithms. These three CRGs were upregulated in the endometrium of patients with moderate/severe EMT and downregulated in patients with infertility. Single sample genomic enrichment analysis (ssGSEA) revealed that these CRGs were closely correlated with autoimmune diseases such as systemic lupus erythematosus. Furthermore, these CRGs were correlated with immune cells such as eosinophils, natural killer cells, and macrophages. Therefore, profiling patients based on these genes aid in a more accurate diagnosis of EMT progression. These findings provide a new idea for the pathology and treatment of endometriosis, suggesting that CRGs such as GLS, NFE2L2, and PDHA1 may play a key role in the occurrence and development of endometriosis.\u003c/p\u003e","manuscriptTitle":"Machine Learning Algorithms for a Novel Cuproptosis-related Gene Signature of Diagnostic and Immune Infiltration in Endometriosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-05-16 15:08:07","doi":"10.21203/rs.3.rs-2742573/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":"54d9671f-e580-49b4-97e0-5ce1a370fe7a","owner":[],"postedDate":"May 16th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2023-06-11T22:53:07+00:00","versionOfRecord":[],"versionCreatedAt":"2023-05-16 15:08:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-2742573","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2742573","identity":"rs-2742573","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}