基于GEO数据库筛选子宫内膜异位症的诊断标志物及免疫浸润分析

打开文本图片集

【关键词】子宫内膜异位症;诊断标志物;加权基因共表达网络分析;机器学习;免疫浸润中图分类号：R711.71 文献标志码：A DOI：10.3969/j.issn.1003-1383.2025.10.002

Screening diagnostic biomarkers and immune infiltration analysisofendometriosisbasedonGEO

SONG Yanlun 1 ， CEN Changxue 1 ， YU Linjun 1 ，WANG Rong 2 ，WANG Junli 1s （1.School ofMedicalLaboratoryScience，YoujiangMedical UniversityforNationalities， Baise 533ooo，Guangxi，China；2. Department of Transfusion Medicine，Affiliated Hospital ofYoujiangMedical UniversityforNationalities，Baise533ooo，Guangxi，China）

【Abstract】ObjectiveToscreen potential biomarkers for endometriosis（EMs）basedon the gene expresion omnibus （GEO），so as to construct diagnostic and prognostic models，and elucidate the pathogenic mechanisms.MethodsGene expresion profiles wereobtained from GEO.Key modules were screened through differential expressonanalysis（DEGs） and weighted geneco-expresion network analysis（WGCNA）.Diagnostic linecharts and prognostic models wereconstructed byintegrating geneontology（GO）/Kyotoencyclopediaof genes and genomes（KEGG）enrichment analysis，machine learming models（GLM，RF，SVM，XGB）and Cox regression analysis.Inaddition，the expresions of keygenes werefurther validated byreverse transcription-quantitative polymerasechainreaction（RT-qPCR）.ResultsAtotalof14 DEGs were identified，and MEgray60（216 genes）and MEblack （786 genes）modules were screened.GO/KEGG enrichment analysis revealed pathwaysrelatedto growthfactorresponse，angiogenesis，and immunecellinfiltration.TheGLMmodel hashebest diagnostic performance （ AUC=1.000 ）.A diagnostic column chart was constructed based on the selected 5 key genes，including SF3B1 and KDR，etc，and the Hosmer-Lemeshow test showed good model fit （ P>0.05 ）.Immunohistochemical analysisshowed thatthedecreaseofmacrophageswas negativelycorrelatedwiththeexpressionofkeygenes.Experimental verificationshowedthatthediferentialexpressionof5genesinEMstissuewasconsistent withtheresultsof publicdatabases. ConclusionThisstudyidentifiesdiagnosticand prognosticbiomarkersforendometriosis，deciphersfunctionalrolesofkey gene modules，and provides novel targets for mechanism researches and clinical interventions.

【Keywords】endometriosis（EMs）；diagnostic biomarkers；weightedgeneco-expression network analysis（WGCNA）; machine learning；immune infiltration

子宫内膜异位症（endometriosis，EMs）是一种常见的慢性疾病，主要影响女性的生育能力和生活质量，其特点是子宫内膜组织在盆腔以外的部位生长和扩展。（剩余8051字）