Zheng Yihan, Fang Zhuanji, Wu Xizhu, Zhang Huale, Sun Pengming
Department of Anesthesiology, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou, Fujian, 350001, China.
Department of Obstetrics, Fujian Maternity and Child Health Hospital, College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fuzhou, Fujian, 350001, China.
BMC Pregnancy Childbirth. 2024 Dec 21;24(1):847. doi: 10.1186/s12884-024-07028-3.
This study aimed to identify novel biomarkers for preeclampsia (PE) diagnosis by integrating Weighted Gene Co-expression Network Analysis (WGCNA) with machine learning techniques.
We obtained the PE dataset GSE25906 from the gene expression omnibus (GEO) database. Analysis of differentially expressed genes (DEGs) and module genes with Limma and Weighted Gene Co-expression Network analysis (WGCNA). Candidate hub genes for PE were identified using machine learning. Subsequently, we used western-blotting (WB) and real-time fluorescence quantitative (qPCR) to verify the expression of F13A1 and SCCPDH in preeclampsia patients. Finally, we estimated the extent of immune cell infiltration in PE samples by employing the CIBERSORT algorithms.
Our findings revealed that F13A1 and SCCPDH were the hub genes of PE. The nomogram and two candidate hub genes had high diagnostic values (AUC: 0.90 and 0.88, respectively). The expression levels of F13A1 and SCCPDH were verified by WB and qPCR. CIBERSORT analysis confirmed that the PE group had a significantly larger proportion of plasma cells and activated dendritic cells and a lower portion of resting memory CD4 + T cells.
The study proposes F13A1 and SCCPDH as potential biomarkers for diagnosing PE and points to an improvement in early detection. Integration of WGCNA with machine learning could enhance biomarker discovery in complex conditions like PE and offer a path toward more precise and reliable diagnostic tools.
本研究旨在通过将加权基因共表达网络分析（WGCNA）与机器学习技术相结合，鉴定用于子痫前期（PE）诊断的新型生物标志物。
我们从基因表达综合数据库（GEO）中获取了PE数据集GSE25906。使用Limma和加权基因共表达网络分析（WGCNA）对差异表达基因（DEG）和模块基因进行分析。利用机器学习确定PE的候选枢纽基因。随后，我们采用蛋白质免疫印迹法（WB）和实时荧光定量聚合酶链反应（qPCR）验证子痫前期患者中F13A1和SCCPDH的表达。最后，我们采用CIBERSORT算法评估PE样本中免疫细胞浸润程度。
我们的研究结果显示，F13A1和SCCPDH是PE的枢纽基因。列线图和两个候选枢纽基因具有较高的诊断价值（AUC分别为0.90和0.88）。通过WB和qPCR验证了F13A1和SCCPDH的表达水平。CIBERSORT分析证实，PE组浆细胞和活化树突状细胞的比例显著更高，而静息记忆CD4 + T细胞的比例更低。
本研究提出F13A1和SCCPDH作为诊断PE的潜在生物标志物，并指出早期检测有所改善。WGCNA与机器学习的整合可增强在PE等复杂情况下的生物标志物发现，并为更精确可靠的诊断工具提供一条途径。
