A Three-Gene Based Diagnos Tic Model for Endometriosis

A Three-Gene Based Diagnos Tic Model for 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 A Three-Gene Based Diagnos Tic Model for Endometriosis Fangfang Fan, Qipeng Wei, Yue Xie, Xiaofang Yin, Xuezhou Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-824061/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 Background: Endometriosis is a common gynecologic pathology among reproductive-aged women. The lack of a definitive clinical symptom and a minimally invasive diagnostic method lead to a diagnostic latency of endometriosis. In this study, we aim to establish a gene-based diagnostic model that can easily diagnose endometriosis. Results: Gene expression profiles of 226 tissue samples from the Gene Expression Omnibus and Genotype tissue expression databases were used to construct model for diagnosis of endometriosis. Using differential gene expression analysis, a total of 170 genes were identified. The weighted gene co-expression network analysis was utilized to identify hub genes. Lasso Cox regression was performed to identify diagnostic biomarkers and construct diagnosis model. Three genes including ELOVL6, UTP20 and ARHGAP18 were screened in the hub genes as the promising targets used for diagnosis of endometriosis. The three genes based diagnostic model was established by Least absolute shrinkage and selection operator and then validated in an independent validation cohort. Receiver operating characteristic curve analysis inferred the diagnostic model has a good performance in both training and validation cohort (The area under the curve of ROC achieved 0.75 and 0.73 in training and validation cohort, respectively). Conclusions: Therefore, our study established a novel three gene-based diagnostic model for endometriosis, which may assist clinicians in diagnosis of endometriosis. Biomedical Engineering Endometriosis Machine Learning weighted gene co-expression network analysis Least Absolute Shrinkage and Selection Operator Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Background Endometriosis is characterized by the aberrant presence of estrogen-sensitive endometrium-like tissue outside the uterine cavity. It is a chronic inflammatory disease that affects 6–10% of reproductive-aged women[ 1 – 3 ]. It usually has a 5–10 years of delay from the presentation of symptoms to the definitive diagnosis, which have a long-term negative effect on their daily life and work[ 4 , 5 ]. Endometriosis varies significantly in clinical presentations. For endometrial lesions frequently located on the surrounding organs situated in the pelvic cavity, patients with endometriosis present with a collection of symptoms, including chronic pelvic pain, urinary tract symptoms and gastrointestinal symptoms[ 4 , 5 ]. These symptoms resemble other pain-associated disorders, which makes great challenge for doctors to distinguish whether these symptoms are caused by endometriosis or other diseases associated with chronic pain. This in part lead to the delay in diagnosis. For now, exploratory laparoscopy under general anesthesia and subsequent histological confirmation of suspicious lesions remains the gold standard for definitive diagnosis of endometriosis. However, it has to be weighted against the anesthetic risks, the potential decreases in ovarian reserve and the cost. These disadvantages of laparoscopy also contribute to the aforementioned diagnosis delay. Thus, a non-invasive or minimal invasive, safe, and cost-effective approach, such as endometrial biopsy, is needed to shorten time to the diagnosis. It has been reported that the eutopic endometrium of endometriosis patients displays a number of molecular alterations. It has been reported that adrenomedullin (ADM) was elevated in the eutopic endometrium of women with endometriosis compared with healthy donors[ 8 ]. An integrated quantitative proteomic analysis of eutopic endometrium of women with endometriosis versus controls had identified 1214 differential expressed proteins[ 9 ]. Endometrial BCL6 over-expressed in eutopic endometrium of women with endometriosis[ 10 ]. These altered molecules could distinguish endometriosis from healthy women. Eutopic endometrium is easily obtainable by the semi-invasive sampling procedure, thus it is an excellent source for biomarker discovery and developing a less invasive diagnostic approach for endometriosis. In this work, we integrated 132 endometriosis cases and 94 normal endometrium controls with gene expression data from 3 independent cohorts, including GSE51981, GSE135485 and Genotype-Tissue Expression (GTEx), to develop a highly accurate diagnostic predictive model for classifying endometriosis samples from the control. We also validated this diagnostic predictive model on an independent cohort, GSE7305. Results Differentially expressed genes and pathway analyses Differentially expressed genes (DEGs) among GSE51981 and merged dataset of GSE135485 and GTEx were analyzed. In total, 97 and 73 DEGs (|log2FC|>1 and adj.Pvalue<0.05) were identified (Figure2). Identification of the Most Relevant Co-Expression Modules for Endometriosis A total of 166 DEGS from the GSE51981 dataset were included to construct co-expression network. According to the scale-free topology criterion, the soft threshold power of β= 6 was selected (Figure 3A). Modules with a minimum size of 30 were identified by dynamic tree cut method. After the highly similar modules had been merged, a total of 3 modules were obtained (Blue, Turquoise, and Grey). Correlation analysis between different co-expression modules and endometriosis sample traits indicated that the turquoise module had the most significant association with endometriosis (correlation coefficient=−0.43, p= 4e-08, respectively) (Figure 3B and 3C). Hence, genes in turquoise module were considered to be the most significant module gene for further analysis. Hub Genes Identification For the turquoise module gene interaction inspection, we constructed a PPI network of 64 nodes and 1336 edges by Cytoscape software (Figure 4). The intra-modular connectivity of each gene was analyzed by CytoHubba and the top 10 genes with the highest connectivity, including MORC4, ELOVL6, JUNB, UTP20, CCDC146, MSH2, ARHGAP18, ANKRD12, SP100 and SRGAP2B, were considered as hub genes (Figure 5). Establishment of the three-gene-based prognostic gene model 3 out of the 10 hub genes, including ELOVL6, UTP20 and ARHGAP18, were selected for endometriosis model construction in GSE5198 (training cohort) by the minimizing λ method of LASSO Cox regression analysis (Figure 6A and 6B). According to the weighted coefficients in the training cohort, the three-gene-based classifier for predicting endometriosis was built as follows: Risk score = EXP ELOVL6 * (-0.303243) + EXP UTP20 * (-0.163774) + EXP ARHGAP18 * (-0.296289) (Figure 6C and 6D). ROC analysis was applied to estimate the performance of the three-gene-based model in the endometriosis classification and the result showed that this classifier exhibited good classification accuracy (area under the curve (AUC) =0.75) (Figure 7A). Then, the classifier was validated in an independent validation cohort and achieved a similar predictive accuracy (AUC=0.73) (Figure 7B). Discussion The lack of easily accessible and non-invasive tool is responsible for the delayed diagnosis of endometriosis. Therefore, there is an urgent need to develop a novel strategy for the earlier diagnosis of endometriosis. Gene expression of eutopic endometrium of endometriosis patients are different from healthy women and eutopic endometrium is easily obtainable. Thus identifying altered genes from eutopic endometrium is a way to explore potential biomarkers for endometriosis. Endometriosis is a highly heterogeneous disease. The molecular heterogeneity will negatively affect the diagnostic performance of individual biomarkers. Compared with single biomarkers, a combination of multiple biomarkers has a more accurate and effective diagnostic capacity for endometriosis. Consequently, we tried to establish a multi-gene diagnostic model for endometriosis via integrative bioinformatics analysis. In this study, we identified DEGs between endometriosis cases and controls from a combined cohort of GSE135485 (an endometriosis cohort including 55 endometriosis samples and 3 control samples) and 20 normal endometrial tissue samples from GTEx. Meanwhile, we also identified DEGs by analyzing the gene expression profiles of GSE51981, another endometriosis cohort in GEO. To minimize the possibility of losing potential biomarkers of endometriosis, we incorporate DEGs from two distinct DEGs datasets for further analysis. Then, to identify the genes most relevant to endometriosis, we performed WGCNA on these DEGs and identified three modules, among which turquoise module had the most significant association with endometriosis. By analyzing the connectivity degree of each node in the co-expression module, the hub genes were tentatively identified. Multi-gene diagnostic model involved a crowd of genes was too complicated for clinical application. To establish a user-friendly model with limited number of genes for diagnosis of endometriosis, we performed LASSO regression analysis and identified three genes, including ELOVL6、UTP20 and ARHGAP18, correlating with endometriosis. Currently, there was no reports to directly prove that these genes are associated with endometriosis, but there were some clues indicating their association with endometriosis. It was recognized that there was an association between endometriosis and ovarian carcinoma. A group found the level of ELOVL6 mRNA was significantly lower in high-grade serous ovarian carcinoma comparing with normal ovarian tissue[ 17 ]. ARHGAP18 was a negative regulator of RhoC and had an anti-inflammatory effect[ 18 , 19 ]. While RhoC had been reported to be involved in the origin and the maintenance of endometriosis[ 20 ]. UTP20 has not been previously reported to be associated with endometriosis. For the first time, we predicted that UTP20 may involve in the processes of endometriosis, which deserve further investigation. Then,we employed ROC curve to assess the performance of the three-gene-based diagnosis model. The results showed that AUC of the ROC curve were 0.75 and 0.73 in training cohort and validation cohort, respectively, which indicated that this model had a good performance in endometriosis diagnosis. However, there were certain limitations in this study. Firstly, endometrium displayed different expression pattern in particular stages of menstrual cycle[ 20 , 21 ]. But the present analysis didn't consider the impact of this factor, which may limit the diagnostic power of this model. Secondly, our analysis was descriptive and we did not collect clinical specimen to validate the performance of this model. Conclusions In conclusion, we established a novel model for the diagnosis of endometriosis through integrated method. Although this model still needs to be validated in clinical practice, it may provide promising targets and new research ideas for the diagnosis of endometriosis. Methods Expression datasets RNA expression data for endometrium were downloaded from the Genotype-Tissue Expression (GTEx) and the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/ ) datasets (Table 1). 148 of endometrium samples from GSE51981[11](comprising 77 endometriosis cases and 71 control samples) and 78 of endometrium samples from merged dataset of GSE135485 and GTEx (comprising 55 cases and 23 control samples) were referred to as the ‘training set’ in this study. For assessing reproducibility, a subset of 20 of endometrium samples from GSE7305[12] (comprising 10 cases and 10 control samples) was referred to as ‘validation set’. A summary of data analysis workflow are provided as follow (Figure 1). Table 1 Information of the Expression datasets GSE dataset GPL Case Control GSE51981 GPL570 77 71 GSE135485 NA 55 3 GSE7305 GPL570 10 10 GTEx NA 0 20 Differentially expressed genes and pathway analyses To identify the differentially expressed genes (DEGs) between the endometriosis cases and control, the limma package and DESeq2 package in R were applied on the cases from GSE51981 dataset(Case: Control= 77:71), and merged dataset of GSE135485 and GTEx(Case: Control= 55:23) respectively. The significance of the genes was defined by using |log2FC| set at 1 and an adjusted p-value cutoff set at 0.05. Subsequently, the union of DEGs from two distinct DEGs datasets were applied for further analysis. Weighted gene co-expression network analysis The expression profile of 166 DEGs was obtained from GSE51981. Then, these data were used to construct a co-expression network using the WGCNA package in R. WGCNA was implemented in accordance to the protocol of WGCNA package in R[13]Screening of Hub Genes The co-expression network was visualized with Cytoscape (https://cytoscape.org/). To identify hub nodes in the co-expression network, the degree of each node was calculated by CytoHubba, a plugin in Cytoscape. In this study, the genes with the top 10 degree were considered as hub genes. Diagnostic classifier The Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression algorithm could be applied to prioritize variables in high-dimensional microarray data with prognostic value. First, we performed LASSO Cox regression analysis to identify best prognostic markers among the hub genes in the GSE51981 training cohort. Then, we constructed a multi-gene based risk model to predict prognosis of endometriosis based on the best prognostic markers. The receiver operating characteristic (ROC) curve was performed to evaluate the predictive accuracy of this model. Finally, the prognostic value of the regression risk model was further validated in an independent testing cohort (GSE7305). Declarations Ethics approval and consent to participate： Not applicable. Consent for publication： Not applicable. Availability of data and materials： The datasets used during the current study are available from Genotype-Tissue Expression (GTEx) and the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/ ) datasets. Competing interests： The authors declare that they have no conflict of interest. Funding： This work was supported by Special Fund for Clinical Research of Chinese Medical Association (No. 17020130682) and Natural Science Foundation of Hubei Province of China (No. 2018CFB472). Authors' contributions： YXZ and FFF designed the research. WQP carried out data collection. XY and YXF data analysis. YXZ obtained funding. FFF, YXZ wrote the paper. All authors read and approved the final manuscript. Acknowledgements： Not applicable. References Giudice LC, Kao LC. Endometriosis Lancet. 2004;364(9447):1789–99. Bulun SE, et al. Endometriosis. Endocr Rev. 2019;40(4):1048–79. Chapron C, et al. Rethinking mechanisms, diagnosis and management of endometriosis. Nat Rev Endocrinol. 2019;15(11):666–82. Horne AW, Saunders PTK. SnapShot: Endometriosis Cell. 2019;179(7):1677–7.e1. Agarwal SK, et al. Clinical diagnosis of endometriosis: a call to action. Am J Obstet Gynecol. 2019;220(4):354. .e1-354.e12.. Giudice LC. Clinical practice. Endometriosis. N Engl J Med. 2010;362(25):2389–98. Dunselman GA, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29(3):400–12. Matson BC, et al. Elevated levels of adrenomedullin in eutopic endometrium and plasma from women with endometriosis. Fertil Steril. 2018;109(6):1072–8. Manousopoulou A, et al. Integrated Eutopic Endometrium and Non-Depleted Serum Quantitative Proteomic Analysis Identifies Candidate Serological Markers of Endometriosis. PROTEOMICS – Clinical Applications. 2018;13(3):1800153. Evans-Hoeker E, et al. Endometrial BCL6 Overexpression in Eutopic Endometrium of Women With Endometriosis. Reproductive Sciences. 2015;23(9):1234–41. Tamaresis JS, et al. Molecular Classification of Endometriosis and Disease Stage Using High-Dimensional Genomic Data. Endocrinology. 2014;155(12):4986–99. Hever A, et al., Human endometriosis is associated with plasma cells and overexpression of B lymphocyte stimulator. Proceedings of the National Academy of Sciences - PNAS, 2007. 104(30): p. 12451–12456. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559. Kralickova M, et al. Endometriosis and risk of ovarian cancer: what do we know? Arch Gynecol Obstet. 2020;301(1):1–10. Anglesio MS, Yong PJ. Endometriosis-associated Ovarian Cancers. Clin Obstet Gynecol. 2017;60(4):711–27. Matias-Guiu X, Stewart C. Endometriosis-associated ovarian neoplasia. Pathology. 2018;50(2):190–204. Li FJ, et al. [Expression and clinical significance of ELOVL6 gene in high-grade serous ovarian carcinoma]. Zhonghua Fu Chan Ke Za Zhi. 2016;51(3):192–7. Coleman PR, et al. YAP and the RhoC regulator ARHGAP18, are required to mediate flow-dependent endothelial cell alignment. Cell Commun Signal. 2020;18(1):18. Chang GH, et al. ARHGAP18: an endogenous inhibitor of angiogenesis, limiting tip formation and stabilizing junctions. Small GTPases. 2014;5(3):1–15. Meola J, et al. RHOC: a key gene for endometriosis. Reprod Sci. 2013;20(8):998–1002. Totorikaguena L, et al. Mu opioid receptor in the human endometrium: dynamics of its expression and localization during the menstrual cycle. Fertil Steril. 2017;107(4):1070–7.e1. Wang W, et al. Single-cell transcriptomic atlas of the human endometrium during the menstrual cycle. Nat Med. 2020;26(10):1644–53. 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-824061","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research","associatedPublications":[],"authors":[{"id":48508948,"identity":"5343fa9d-b0c6-425e-9f29-70ab7c9bed29","order_by":0,"name":"Fangfang Fan","email":"","orcid":"","institution":"Xiangyang Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Fangfang","middleName":"","lastName":"Fan","suffix":""},{"id":48508949,"identity":"930be496-7622-46f2-a7bb-fbcf5dbc4eac","order_by":1,"name":"Qipeng Wei","email":"","orcid":"","institution":"Xiangyang Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Qipeng","middleName":"","lastName":"Wei","suffix":""},{"id":48508950,"identity":"042f7265-5542-4f95-881b-39f4fcfea798","order_by":2,"name":"Yue Xie","email":"","orcid":"","institution":"Xiangyang Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Xie","suffix":""},{"id":48508951,"identity":"b00f9383-3bed-4329-b75d-4163116e8416","order_by":3,"name":"Xiaofang Yin","email":"","orcid":"","institution":"Xiangyang Central Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xiaofang","middleName":"","lastName":"Yin","suffix":""},{"id":48508952,"identity":"d316d6c9-04e8-484f-b96b-540fcd8d2db9","order_by":4,"name":"Xuezhou Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAw0lEQVRIiWNgGAWjYDCCA1Can5n58APStEi2s6UZkKbF4DyPggRROvhunzH8XPDrsL3xYR4GA4Yam2iCWiTP5RhLz+xLYzY7zHvgAcOxtNwGQloMzvAYSPP22LCZHeZLMGBsOEyUFuPfvD0SPMbNPAYSxGoxk+b5YSNhwEysFskzbGXWvA1pBhKHgYGcQIxf+M4wb77N8+ewPX//4cMPPtTYENbCwMBhwMDYBmUnEFYOAuwPGBj+EKd0FIyCUTAKRigAAE9SOz4xA8dTAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-4535-1723","institution":"Xiangyang Central Hospital","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Xuezhou","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2021-08-18 10:13:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-824061/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-824061/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":12889659,"identity":"689e580f-72f5-4ea5-b873-4658c7f7dbf0","added_by":"auto","created_at":"2021-08-30 13:36:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":65364,"visible":true,"origin":"","legend":"The workflow of data analysis.","description":"","filename":"OnlineFig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/42b3b2c576a2af86ae94bfee.png"},{"id":12889663,"identity":"fab0f63b-2c10-42aa-94bf-c54ecf2d46fe","added_by":"auto","created_at":"2021-08-30 13:36:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":728004,"visible":true,"origin":"","legend":"DEGs in endometriosis cases and controls\n(A) Volcano plot of DEGs between endometriosis cases and controls in the merged dataset of GSE135485 and GTEx. (B) Volcano plot of DEGs between endometriosis cases and controls in GSE51981 dataset. DEGs with log2FC \u003e1 were shown in red dots; DEGs with log2FC \u003c −1 were in blue dots (P \u003c 0.05). No significantly changed genes are marked as gray dots.\n","description":"","filename":"OnlineFig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/1ba8a94cdd0eaba4e87831a4.png"},{"id":12889662,"identity":"8c69d4d2-768b-4263-b925-b5d24f65043d","added_by":"auto","created_at":"2021-08-30 13:36:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":124390,"visible":true,"origin":"","legend":"Identification of modules associated with endometriosis by WGCNA. \n(A) Analysis of the scale-free fit index and the mean connectivity for various soft-thresholding powers. (B) Dendrogram of all differentially expressed genes clustered based on a dissimilarity measure. (C) Heatmap of the correlation between module Eigengenes and clinical traits (normal and endometriosis). \n","description":"","filename":"OnlineFig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/a8af1231eb41d962bdeaace1.png"},{"id":12889995,"identity":"09780fb8-24e6-4ddf-83db-20eb25e9a9d4","added_by":"auto","created_at":"2021-08-30 13:39:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2298559,"visible":true,"origin":"","legend":"The network analysis of the turquoise module genes. The nodes represent the selected genes, and the edges represent the interactions between two genes.","description":"","filename":"OnlineFig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/629ebb8f98bef87f52fdd9c5.png"},{"id":12889660,"identity":"e05d9f4b-6712-451b-8cfb-357fbde88001","added_by":"auto","created_at":"2021-08-30 13:36:57","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":524264,"visible":true,"origin":"","legend":"The network analysis of the top ten hub genes in endometriosis. The nodes represent the selected genes, and the edges represent the interactions between two genes.","description":"","filename":"OnlineFig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/b93b8aefac96af0b6aa45846.png"},{"id":12889665,"identity":"442ae569-89f7-4b0b-8d56-c9e35c13724f","added_by":"auto","created_at":"2021-08-30 13:36:57","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":681119,"visible":true,"origin":"","legend":"The construction of LASSO Cox regression model\n(A,B) Lasso Cox analysis identified three genes at lambda with minimum partial likelihood deviance that correlated with endometriosis. (C) The coefficients of the three genes identified by Lasso Cox analysis. (D) The expression level of the three identified genes in GSE5198 cohort.\n","description":"","filename":"OnlineFig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/547a8e083a350dd9db646b53.png"},{"id":12889994,"identity":"f6f41528-d036-44fb-b30f-d8ce798287ec","added_by":"auto","created_at":"2021-08-30 13:39:57","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":56300,"visible":true,"origin":"","legend":"Diagnostic performance of the diagnostic prediction model for endometriosis. \n(A) Receiver operating characteristic (ROC) curve evaluated the diagnostic prediction efficiency of the diagnostic prediction model for endometriosis in the training cohort(AUC=0.75). (B) Receiver operating characteristic (ROC) curve analyzed the diagnostic prediction efficiency of the diagnostic prediction model for endometriosis in the validation cohort(AUC=0.73).\n","description":"","filename":"OnlineFig.7.png","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/0dafbdc9c09b333fe242cfc1.png"},{"id":13711614,"identity":"1d2848c9-f8eb-4277-9a24-df2b64712fde","added_by":"auto","created_at":"2021-09-17 14:23:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1522849,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-824061/v1/4fd19203-e63c-4d7f-8cb8-443ceb2ff7e1.pdf"}],"financialInterests":"","formattedTitle":"\u003cp\u003eA Three-Gene Based Diagnos Tic Model for Endometriosis\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eEndometriosis is characterized by the aberrant presence of estrogen-sensitive endometrium-like tissue outside the uterine cavity. It is a chronic inflammatory disease that affects 6\u0026ndash;10% of reproductive-aged women[\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. It usually has a 5\u0026ndash;10 years of delay from the presentation of symptoms to the definitive diagnosis, which have a long-term negative effect on their daily life and work[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEndometriosis varies significantly in clinical presentations. For endometrial lesions frequently located on the surrounding organs situated in the pelvic cavity, patients with endometriosis present with a collection of symptoms, including chronic pelvic pain, urinary tract symptoms and gastrointestinal symptoms[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These symptoms resemble other pain-associated disorders, which makes great challenge for doctors to distinguish whether these symptoms are caused by endometriosis or other diseases associated with chronic pain. This in part lead to the delay in diagnosis.\u003c/p\u003e \u003cp\u003eFor now, exploratory laparoscopy under general anesthesia and subsequent histological confirmation of suspicious lesions remains the gold standard for definitive diagnosis of endometriosis. However, it has to be weighted against the anesthetic risks, the potential decreases in ovarian reserve and the cost. These disadvantages of laparoscopy also contribute to the aforementioned diagnosis delay. Thus, a non-invasive or minimal invasive, safe, and cost-effective approach, such as endometrial biopsy, is needed to shorten time to the diagnosis.\u003c/p\u003e \u003cp\u003eIt has been reported that the eutopic endometrium of endometriosis patients displays a number of molecular alterations. It has been reported that adrenomedullin (ADM) was elevated in the eutopic endometrium of women with endometriosis compared with healthy donors[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. An integrated quantitative proteomic analysis of eutopic endometrium of women with endometriosis versus controls had identified 1214 differential expressed proteins[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Endometrial BCL6 over-expressed in eutopic endometrium of women with endometriosis[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. These altered molecules could distinguish endometriosis from healthy women. Eutopic endometrium is easily obtainable by the semi-invasive sampling procedure, thus it is an excellent source for biomarker discovery and developing a less invasive diagnostic approach for endometriosis.\u003c/p\u003e \u003cp\u003eIn this work, we integrated 132 endometriosis cases and 94 normal endometrium controls with gene expression data from 3 independent cohorts, including GSE51981, GSE135485 and Genotype-Tissue Expression (GTEx), to develop a highly accurate diagnostic predictive model for classifying endometriosis samples from the control. We also validated this diagnostic predictive model on an independent cohort, GSE7305.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eDifferentially expressed genes and pathway analyses\u003c/h2\u003e\n\u003cp\u003eDifferentially expressed genes (DEGs) among GSE51981 and merged dataset of GSE135485 and GTEx were analyzed. In total, 97 and 73 DEGs (|log2FC|\u0026gt;1 and adj.Pvalue\u0026lt;0.05) were identified (Figure2).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eIdentification of the Most Relevant Co-Expression Modules for Endometriosis\u003c/h2\u003e\n\u003cp\u003eA total of 166 DEGS from the GSE51981 dataset were included to construct co-expression network. According to the scale-free topology criterion, the soft threshold power of \u0026beta;= 6 was selected (Figure 3A). Modules with a minimum size of 30 were identified by dynamic tree cut method. After the highly similar modules had been merged, a total of 3 modules were obtained (Blue, Turquoise, and Grey). Correlation analysis between different co-expression modules and endometriosis sample traits indicated that the turquoise module had the most significant association with endometriosis (correlation coefficient=\u0026minus;0.43, p= 4e-08, respectively) (Figure 3B and 3C). Hence, genes in turquoise module were considered to be the most significant module gene for further analysis.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eHub Genes Identification\u003c/h2\u003e\n\u003cp\u003eFor the turquoise module gene interaction inspection, we constructed a PPI network of 64 nodes and 1336 edges by Cytoscape software (Figure 4). The intra-modular connectivity of each gene was analyzed by CytoHubba and the top 10 genes with the highest connectivity, including MORC4, ELOVL6, JUNB, UTP20, CCDC146, MSH2, ARHGAP18, ANKRD12, SP100 and SRGAP2B, were considered as hub genes (Figure 5).\u003c/p\u003e\n\u003ch2\u003eEstablishment of the three-gene-based prognostic gene model\u003c/h2\u003e\n\u003cp\u003e3 out of the 10 hub genes, including ELOVL6, UTP20 and ARHGAP18, were selected for endometriosis model construction in GSE5198 (training cohort) by the minimizing \u0026lambda; method of LASSO Cox regression analysis (Figure 6A and 6B). According to the weighted coefficients in the training cohort, the three-gene-based classifier for predicting endometriosis was built as follows: Risk score =\u0026nbsp;EXP\u003csub\u003eELOVL6\u003c/sub\u003e * (-0.303243) + EXP\u003csub\u003eUTP20\u003c/sub\u003e * (-0.163774) + EXP\u003csub\u003eARHGAP18\u0026nbsp;\u003c/sub\u003e* (-0.296289) (Figure 6C and 6D). \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eROC analysis was applied to estimate the performance of the three-gene-based model in the endometriosis classification and the result showed that this classifier exhibited good classification accuracy (area under the curve (AUC) =0.75) (Figure 7A). Then, the classifier was validated in an independent validation cohort and achieved a similar predictive accuracy (AUC=0.73) (Figure 7B).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe lack of easily accessible and non-invasive tool is responsible for the delayed diagnosis of endometriosis. Therefore, there is an urgent need to develop a novel strategy for the earlier diagnosis of endometriosis. Gene expression of eutopic endometrium of endometriosis patients are different from healthy women and eutopic endometrium is easily obtainable. Thus identifying altered genes from eutopic endometrium is a way to explore potential biomarkers for endometriosis.\u003c/p\u003e \u003cp\u003eEndometriosis is a highly heterogeneous disease. The molecular heterogeneity will negatively affect the diagnostic performance of individual biomarkers. Compared with single biomarkers, a combination of multiple biomarkers has a more accurate and effective diagnostic capacity for endometriosis. Consequently, we tried to establish a multi-gene diagnostic model for endometriosis via integrative bioinformatics analysis.\u003c/p\u003e \u003cp\u003eIn this study, we identified DEGs between endometriosis cases and controls from a combined cohort of GSE135485 (an endometriosis cohort including 55 endometriosis samples and 3 control samples) and 20 normal endometrial tissue samples from GTEx. Meanwhile, we also identified DEGs by analyzing the gene expression profiles of GSE51981, another endometriosis cohort in GEO. To minimize the possibility of losing potential biomarkers of endometriosis, we incorporate DEGs from two distinct DEGs datasets for further analysis. Then, to identify the genes most relevant to endometriosis, we performed WGCNA on these DEGs and identified three modules, among which turquoise module had the most significant association with endometriosis. By analyzing the connectivity degree of each node in the co-expression module, the hub genes were tentatively identified. Multi-gene diagnostic model involved a crowd of genes was too complicated for clinical application. To establish a user-friendly model with limited number of genes for diagnosis of endometriosis, we performed LASSO regression analysis and identified three genes, including ELOVL6、UTP20 and ARHGAP18, correlating with endometriosis. Currently, there was no reports to directly prove that these genes are associated with endometriosis, but there were some clues indicating their association with endometriosis. It was recognized that there was an association between endometriosis and ovarian carcinoma. A group found the level of ELOVL6 mRNA was significantly lower in high-grade serous ovarian carcinoma comparing with normal ovarian tissue[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. ARHGAP18 was a negative regulator of RhoC and had an anti-inflammatory effect[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. While RhoC had been reported to be involved in the origin and the maintenance of endometriosis[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. UTP20 has not been previously reported to be associated with endometriosis. For the first time, we predicted that UTP20 may involve in the processes of endometriosis, which deserve further investigation.\u003c/p\u003e \u003cp\u003eThen,we employed ROC curve to assess the performance of the three-gene-based diagnosis model. The results showed that AUC of the ROC curve were 0.75 and 0.73 in training cohort and validation cohort, respectively, which indicated that this model had a good performance in endometriosis diagnosis.\u003c/p\u003e \u003cp\u003eHowever, there were certain limitations in this study. Firstly, endometrium displayed different expression pattern in particular stages of menstrual cycle[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. But the present analysis didn't consider the impact of this factor, which may limit the diagnostic power of this model. Secondly, our analysis was descriptive and we did not collect clinical specimen to validate the performance of this model.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn conclusion, we established a novel model for the diagnosis of endometriosis through integrated method. Although this model still needs to be validated in clinical practice, it may provide promising targets and new research ideas for the diagnosis of endometriosis.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eExpression datasets\u003c/h2\u003e\n\u003cp\u003eRNA expression data for endometrium were downloaded from the Genotype-Tissue Expression (GTEx) and the Gene Expression Omnibus (GEO,\u0026nbsp;\u003ca href=\"https://www.ncbi.nlm.nih.gov/geo/\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/a\u003e) datasets (Table 1). 148 of endometrium samples from GSE51981[11](comprising 77 endometriosis cases and 71 control samples) and 78 of endometrium samples from merged dataset of GSE135485 and GTEx (comprising 55 cases and 23 control samples) were referred to as the \u0026lsquo;training set\u0026rsquo; in this study. For assessing reproducibility, a subset of 20 of endometrium samples from GSE7305[12]\u0026nbsp;(comprising 10 cases and 10 control samples) was referred to as \u0026lsquo;validation set\u0026rsquo;.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA summary of data analysis workflow are provided as follow\u0026nbsp;(Figure 1).\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eTable 1\u003c/p\u003e\n\u003cp style=\"text-align: center;\"\u003eInformation of the\u0026nbsp;Expression datasets\u003c/p\u003e\n\u003ctable border=\"0\" cellpadding=\"0\" cellspacing=\"0\" style=\"border-collapse: collapse; margin: 0px auto;\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGSE dataset\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGPL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eCase\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eControl\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGSE51981\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e71\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGSE135485\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGSE7305\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGPL570\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eGTEx\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"25%\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2\u003eDifferentially expressed genes and pathway analyses\u003c/h2\u003e\n\u003cp\u003eTo identify the differentially expressed genes (DEGs) between the endometriosis cases and control, the limma package and DESeq2 package in R were applied on the cases from GSE51981 dataset(Case: Control= 77:71), and merged dataset of GSE135485 and GTEx(Case: Control= 55:23) respectively. The significance of the genes was defined by using |log2FC| set at 1 and an adjusted p-value cutoff set at 0.05. Subsequently, the union of DEGs from two distinct DEGs datasets were applied for further analysis.\u003c/p\u003e\n\u003ch2\u003eWeighted gene co-expression network analysis\u003c/h2\u003e\n\u003cp\u003eThe expression profile of 166 DEGs was obtained from GSE51981. Then, these data were used to construct a co-expression network using the WGCNA package in R. WGCNA was implemented in accordance to the protocol of WGCNA package in R[13]Screening of Hub Genes\u003c/p\u003e\n\u003cp\u003eThe co-expression network was visualized with Cytoscape (https://cytoscape.org/). To identify hub nodes in the co-expression network, the degree of each node was calculated by CytoHubba, a plugin in Cytoscape. In this study, the genes with the top 10 degree were considered as hub genes.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eDiagnostic classifier\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThe Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression algorithm could be applied to prioritize variables in high-dimensional microarray data with prognostic value. First, we performed LASSO Cox regression analysis to identify best prognostic markers among the hub genes in the GSE51981 training cohort. Then, we constructed a multi-gene based risk model to predict prognosis of endometriosis based on the best prognostic markers. The receiver operating characteristic (ROC) curve was performed to evaluate the predictive accuracy of this model. Finally, the prognostic value of the regression risk model was further validated in an independent testing cohort (GSE7305).\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate：\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eConsent for publication：\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials：\u003c/h2\u003e\n\u003cp\u003eThe datasets used during the current study are available from Genotype-Tissue Expression (GTEx) and the Gene Expression Omnibus (GEO, \u003ca href=\"https://www.ncbi.nlm.nih.gov/geo/\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/a\u003e) datasets.\u003c/p\u003e\n\u003ch2\u003eCompeting interests：\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003ch2\u003eFunding：\u003c/h2\u003e\n\u003cp\u003eThis work was supported by Special Fund for Clinical Research of Chinese Medical Association (No. 17020130682) and Natural Science Foundation of Hubei Province of China (No. 2018CFB472).\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026apos; contributions：\u003c/h2\u003e\n\u003cp\u003eYXZ and FFF designed the research. WQP carried out data collection. XY and YXF data analysis. YXZ obtained funding. FFF, YXZ wrote the paper. All authors read and approved the final manuscript.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAcknowledgements：\u003c/h2\u003e\n\u003cp\u003eNot applicable. \u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eGiudice LC, Kao LC. Endometriosis Lancet. 2004;364(9447):1789\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulun SE, et al. Endometriosis. Endocr Rev. 2019;40(4):1048\u0026ndash;79.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChapron C, et al. Rethinking mechanisms, diagnosis and management of endometriosis. Nat Rev Endocrinol. 2019;15(11):666\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHorne AW, Saunders PTK. SnapShot: Endometriosis Cell. 2019;179(7):1677\u0026ndash;7.e1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgarwal SK, et al. Clinical diagnosis of endometriosis: a call to action. Am J Obstet Gynecol. 2019;220(4):354. .e1-354.e12..\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiudice LC. Clinical practice. Endometriosis. N Engl J Med. 2010;362(25):2389\u0026ndash;98.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDunselman GA, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29(3):400\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatson BC, et al. Elevated levels of adrenomedullin in eutopic endometrium and plasma from women with endometriosis. Fertil Steril. 2018;109(6):1072\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManousopoulou A, et al. Integrated Eutopic Endometrium and Non-Depleted Serum Quantitative Proteomic Analysis Identifies Candidate Serological Markers of Endometriosis. PROTEOMICS \u0026ndash; Clinical Applications. 2018;13(3):1800153.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEvans-Hoeker E, et al. Endometrial BCL6 Overexpression in Eutopic Endometrium of Women With Endometriosis. Reproductive Sciences. 2015;23(9):1234\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTamaresis JS, et al. Molecular Classification of Endometriosis and Disease Stage Using High-Dimensional Genomic Data. Endocrinology. 2014;155(12):4986\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHever A, et al., Human endometriosis is associated with plasma cells and overexpression of B lymphocyte stimulator. Proceedings of the National Academy of Sciences - PNAS, 2007. 104(30): p.\u0026nbsp;12451\u0026ndash;12456.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLangfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKralickova M, et al. Endometriosis and risk of ovarian cancer: what do we know? Arch Gynecol Obstet. 2020;301(1):1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnglesio MS, Yong PJ. Endometriosis-associated Ovarian Cancers. Clin Obstet Gynecol. 2017;60(4):711\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatias-Guiu X, Stewart C. Endometriosis-associated ovarian neoplasia. Pathology. 2018;50(2):190\u0026ndash;204.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi FJ, et al. [Expression and clinical significance of ELOVL6 gene in high-grade serous ovarian carcinoma]. Zhonghua Fu Chan Ke Za Zhi. 2016;51(3):192\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColeman PR, et al. YAP and the RhoC regulator ARHGAP18, are required to mediate flow-dependent endothelial cell alignment. Cell Commun Signal. 2020;18(1):18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang GH, et al. ARHGAP18: an endogenous inhibitor of angiogenesis, limiting tip formation and stabilizing junctions. Small GTPases. 2014;5(3):1\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeola J, et al. RHOC: a key gene for endometriosis. Reprod Sci. 2013;20(8):998\u0026ndash;1002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTotorikaguena L, et al. Mu opioid receptor in the human endometrium: dynamics of its expression and localization during the menstrual cycle. Fertil Steril. 2017;107(4):1070\u0026ndash;7.e1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang W, et al. Single-cell transcriptomic atlas of the human endometrium during the menstrual cycle. Nat Med. 2020;26(10):1644\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Endometriosis, Machine Learning, weighted gene co-expression network analysis, Least Absolute Shrinkage and Selection Operator","lastPublishedDoi":"10.21203/rs.3.rs-824061/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-824061/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eEndometriosis\u0026nbsp;is\u0026nbsp;a common gynecologic pathology among reproductive-aged women. The lack of a definitive clinical symptom and a minimally invasive diagnostic method lead to a diagnostic latency of endometriosis. In this study, we aim to establish a gene-based diagnostic model that can easily diagnose endometriosis. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eGene expression profiles of 226 tissue samples from the Gene Expression Omnibus and Genotype tissue expression databases were used to construct model for diagnosis of endometriosis. Using differential gene expression analysis, a total of 170 genes were identified. The weighted gene co-expression network analysis was utilized to identify hub genes. Lasso Cox regression was performed to identify diagnostic biomarkers and construct diagnosis model. Three genes including ELOVL6, UTP20 and ARHGAP18 were screened in the hub genes as the promising targets used for diagnosis of endometriosis. The three genes based diagnostic model was established by Least absolute shrinkage and selection operator and then validated in an independent validation cohort. Receiver operating characteristic curve analysis inferred the diagnostic model has a good performance in both training and validation cohort (The area under the curve of ROC achieved 0.75 and 0.73 in training and validation cohort, respectively). \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eTherefore, our study established a novel three gene-based diagnostic model for endometriosis, which may assist clinicians in diagnosis of endometriosis.\u0026nbsp;\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"A Three-Gene Based Diagnos Tic Model for Endometriosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-08-30 13:36:55","doi":"10.21203/rs.3.rs-824061/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":"b5f4848a-f86e-4286-9f6b-4a74c2f65c17","owner":[],"postedDate":"August 30th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":6817322,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2021-08-30T13:36:57+00:00","versionOfRecord":[],"versionCreatedAt":"2021-08-30 13:36:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-824061","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-824061","identity":"rs-824061","version":["v1"]},"buildId":"ehx78VzkSd0WSzXnipQa-","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}