基于WGCNA分析和SVM建模对轻度认知功能障碍患者血液基因生物标志物的筛选研究

Screening blood genetic biomarkers of patients with mild cognitive impairment based on weighted gene co-expression network analysis and support vector machine modeling

摘要:

目的： 分析轻度认知功能损害（mild cognitive impairment，MCI）患者的外周血表达谱数据，寻找MCI的基因生物标志物。 方法： 从GEO数据库下载GSE63063表达谱数据，采用加权基因共表达网络分析（weighted gene co-expression network analysis，WGCNA）明确与MCI相关的共表达模块，对最显著的模块进行功能富集分析，利用STRING数据库构建模块的蛋白互作（protein-protein interaction，PPI）网络，识别网络中的Hub基因，基于支持向量机（support vector machine，SVM）建立MCI的诊断模型，并进行受试者工作特征曲线（receiver operating characteristic curve，ROC）分析，检测其诊断能力。 结果： 通过WGCNA分析，发现9个与MCI相关的共表达模块，brown模块与MCI的相关性最强。通过MCC算法筛选出每个模块的前15个Hub基因，其中brown模块的Hub基因诊断能力最高，在训练集及验证集中的ROC曲线下面积分别为0.864、0.789。 结论： brown模块的Hub基因可成为诊断MCI的潜在生物学标志物。

Abstract:

Objective: To analyze the expression profile of peripheral blood of patients with mild cognitive impairment (MCI) and search for genetic biomarkers of MCI. Methods: The expression profile data of GSE63063 was downloaded from GEO database. Weighted gene co-expression network analysis (WGCNA) was used to identify the co-expression modules related to MCI. Functional enrichment analysis was performed on the most significant module. Then, the protein-protein interaction (PPI) network of the module was constructed by STRING database, and the Hub genes in the network were identified to establish the diagnosis model of MCI which was established based on the support vector machine (SVM). Finally, the receiver operating characteristic (ROC) analysis was carried out to detect its diagnostic ability. Results: Through WGCNA analysis, 9 co-expression modules related to MCI were found and brown module had the strongest correlation with MCI. The top 15 Hub genes of each module were screened out by MCC algorithm, among which Hub gene of brown module had the highest diagnosis ability, and the area under the ROC curve in the training set and the verification set was 0.864 and 0.789 respectively. Conclusion: Hub gene of brown module can be potential biomarkers for diagnosis of MCI.