Zhang Demin, Luo Liqin, Lu Feng, Li Bo, Lai Xiaoyun
Department of Neurology, The 923rd Hospital of the Joint Logistics Support Force of the People's Liberation Army, Nanning, China.
Front Genet. 2023 Mar 27;14:1106359. doi: 10.3389/fgene.2023.1106359. eCollection 2023.
As one of the most common autoimmune diseases, myasthenia gravis (MG) severely affects the quality of life of patients. Therefore, exploring the role of dysregulated genes between MG and healthy controls in the diagnosis of MG is beneficial to reveal new and promising diagnostic biomarkers and clinical therapeutic targets. The GSE85452 dataset was downloaded from the Gene Expression Omnibus (GEO) database and differential gene expression analysis was performed on MG and healthy control samples to identify differentially expressed genes (DEGs). The functions and pathways involved in DEGs were also explored by functional enrichment analysis. Significantly associated modular genes were identified by weighted gene co-expression network analysis (WGCNA), and MG dysregulated gene co-expression modular-based diagnostic models were constructed by gene set variance analysis (GSVA) and least absolute shrinkage and selection operator (LASSO). In addition, the effect of model genes on tumor immune infiltrating cells was assessed by CIBERSORT. Finally, the upstream regulators of MG dysregulated gene co-expression module were obtained by Pivot analysis. The green module with high diagnostic performance was identified by GSVA and WGCNA. The LASSO model obtained NAPB, C5orf25 and ERICH1 genes had excellent diagnostic performance for MG. Immune cell infiltration results showed a significant negative correlation between green module scores and infiltration abundance of Macrophages M2 cells. In this study, a diagnostic model based on the co-expression module of MG dysregulated genes was constructed, which has good diagnostic performance and contributes to the diagnosis of MG.
重症肌无力（MG）作为最常见的自身免疫性疾病之一，严重影响患者的生活质量。因此，探索MG与健康对照之间失调基因在MG诊断中的作用，有助于揭示新的、有前景的诊断生物标志物和临床治疗靶点。从基因表达综合数据库（GEO）下载GSE85452数据集，并对MG和健康对照样本进行差异基因表达分析，以鉴定差异表达基因（DEG）。还通过功能富集分析探索了DEG所涉及的功能和通路。通过加权基因共表达网络分析（WGCNA）鉴定显著相关的模块基因，并通过基因集变异分析（GSVA）和最小绝对收缩和选择算子（LASSO）构建基于MG失调基因共表达模块的诊断模型。此外，通过CIBERSORT评估模型基因对肿瘤免疫浸润细胞的影响。最后，通过枢纽分析获得MG失调基因共表达模块的上游调节因子。通过GSVA和WGCNA鉴定出具有高诊断性能的绿色模块。LASSO模型获得的NAPB、C5orf25和ERICH1基因对MG具有优异的诊断性能。免疫细胞浸润结果显示绿色模块得分与巨噬细胞M2细胞浸润丰度之间存在显著负相关。在本研究中，构建了基于MG失调基因共表达模块的诊断模型，该模型具有良好的诊断性能，有助于MG的诊断。
