目的：探索帕金森病（PD）小脑的差异表达枢纽基因及构建诊断模型，探讨PD的发病机制，寻求潜 在干预靶点。方法：通过GEO数据库获取PD小脑总RNA芯片GSE20314、GSE28894的数据，利用limma 和 WGCNA筛选出对照组人群和PD患者的差异表达基因；再通过VennDiagram寻找枢纽（Hub）基因。获 取Hub基因后再使用逻辑回归分析构建诊断模型，并通过血液样本数据集GSE18838和GSE6613进行验 证，计算受试者工作特征曲线（ROC）的曲线下面积（AUC）评估其诊断性能。结果：①获得CDKL2、DGKI、 ERO1B、SBNO1、SDK2、SYNRG、VPS13C共7个Hub基因。②逻辑回归分析获取SBNO1+VPS13C构建诊 断模型，ROC 曲线显示在 GSE20314+GSE2889、GSE18838、GSE6613 中的 AUC 分别为 0.9386、0.8663、 0.6564。结论：Hub基因SBNO1、VPS13C可能参与PD的发生、发展过程，未来或可作为PD的潜在干预靶 点，且基于SBNO1+VPS13C基因构建的诊断模型具有较好的诊断性能。

To explore the differentially expressed Hub genes in the cerebellum of Parkinson’s disease (PD) and to construct a diagnostic model, explore the pathogenesis of PD and seek the potential causes pretarget. Methods: The data of GSE20314 and GSE28894, PD cerebellar total RNA, were obtained from the GEO database. Limma and WGCNA were used to screen the differentially expressed genes between control population and PD patients. Then VennDiagram was used to search for Hub genes. Logistic regression was used to construct the diagnostic model after obtaining the hub genes, and the AUC of ROC curve was used to evaluate its diagnostic performance using data of GSE18838 and GSE6613, two data of PD blood samples. Results: (1) Seven Hub genes including CDKL2, DGKI, ERO1B, SBNO1, SDK2, SYNRG and VPS13C were obtained. (2) The diagnostic model was constructed based on SBNO1+VPS13C by logistic regression analysis. ROC curve showed that the AUC of GSE20314 + GSE2889, GSE18838 and GSE6613 were 0.9386, 0.8663 and 0.6564, respectively. Conclusion: Hub genes such as SBNO1 and VPS13C may be involved in the occurrence and development of PD, which may be used as potential intervention targets for PD in the future. The diagnostic model based on SBNO1+VPS13C gene has good diagnostic performance.