解放军医学杂志2026，Vol.51Issue(2)：219-231,13.DOI:10.11855/j.issn.0577-7402.1314.2025.1022
类风湿关节炎与代谢综合征共病基因的表达特征分析及诊断价值评估
Integrated multi-omics analysis of shared gene expression signatures and their diagnostic value in rheumatoid arthritis and metabolic syndrome comorbidity
摘要
Abstract
Objective To investigate the shared gene expression signatures of comorbid genes between rheumatoid arthritis(RA)and metabolic syndrome(MS)using bioinformatics approaches,identify potential biomarkers,and evaluate their diagnostic utility.Methods This study harnessed microarray datasets from the Gene Expression Omnibus(GEO)to explore gene expression patterns in RA and MS.Differentially expressed genes(DEGs)were identified,and weighted gene co-expression network analysis(WGCNA)was applied to uncover genes common to both conditions.Functional enrichment analyses,including Gene Ontology(GO)and the Kyoto Encyclopedia of Genes and Genomes(KEGG),alongside protein-protein interaction(PPI)network analysis,were employed to elucidate the biological roles of these shared genes.Key hub genes were subsequently screened using random forest and least absolute shrinkage and selection operator(LASSO)algorithms.Furthermore,Mendelian randomization(MR)analysis was utilized to verify causal relationships between these key genes and RA.To translate these findings into clinical application,a diagnostic prediction model was developed using the XGBoost machine learning framework.The CIBERSORT algorithm and gene set variation analysis(GSVA)were used to explore the correlations between hub genes and immune cell infiltration as well as metabolic pathway activities.Finally,the expression and potential roles of these hub genes were rigorously validated through single-cell RNA sequencing(scRNA-seq)data and clinical blood samples.Results Analysis of the GSE93777 and GSE98895 datasets using limma R package identified 259 and 280 DEGs,respectively.Integration with WGCNA revealed 88 genes co-expressed in both RA and MS.Functional enrichment analysis revealed that these genes were significantly enriched in biological processes related to immune response and metabolic regulation.Subsequent refinement using machine learning algorithms(LASSO and random forest)pinpointed 24 key hub genes,which were then used to construct a prognostic prediction model.These hub genes demonstrated significant associations with immune functions and metabolic activities in peripheral blood.Additionally,Mendelian randomization(MR)analysis suggested a potential causal relationship between signal transducer and activator of transcription 3(STAT3)and RA risk.Analysis of scRNA-seq data and clinical blood samples confirmed the diagnostic significance of two prominent hub genes:granzyme A(GZMA)and STAT3.Conclusions Key regulatory genes shared between RA and MS have been successfully identified.The GZMA and STAT3 genes are positively correlated with energy metabolism processes,suggesting that the metabolic pathways in which they participate may be closely associated with cellular activities.关键词
类风湿关节炎/生物信息学/机器学习/单细胞/代谢综合征Key words
rheumatoid arthritis/bioinformatics/machine learning/single-cell/metabolic syndrome分类
医药卫生引用本文复制引用
丁芸发,邓安霞,祁腾飞,张宏斌,余浩,吴良平..类风湿关节炎与代谢综合征共病基因的表达特征分析及诊断价值评估[J].解放军医学杂志,2026,51(2):219-231,13.基金项目
广东省科技计划项目(202002020069) This work was supported by the Guangdong Provincial Science and Technology Program(202002020069) （202002020069）
