Screening peripheral circulation diagnostic markers for preeclampsia based on multi-omics and machine learning methods
10.3760/cma.j.cn121382-20231020-00207
- VernacularTitle:基于多组学及机器学习方法筛选子痫前期的外周循环诊断标志物
- Author:
Xiaolu WANG
1
;
Ronghui LIU
;
Qian YAN
Author Information
1. 烟台市烟台山医院产科，烟台 264003
- Keywords:
Eclampsia;
Multiomics;
Machine learning;
Marker;
COL17A1 gene;
DIO2 gene
- From:
International Journal of Biomedical Engineering
2024;47(2):149-155
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To identify early diagnostic biomarkers for preeclampsia by analyzing the placental and peripheral circulatory transcriptomic data of patients.Methods:Clinical information and microarray expression profiles of preeclampsia patients were sourced from high-throughput gene expression databases. Multi-omics approaches, including differential gene expression analysis, enrichment analysis, and weighted gene co-expression network analysis (WGCNA), were utilized to identify candidate diagnostic markers and explore potential mechanisms of preeclampsia. Subsequently, a combination of machine learning techniques, including random forest, support vector machine, and least absolute shrinkage and selection operator (LASSO), were employed for further screening of these candidates. Finally, the selected diagnostic markers were validated using a peripheral circulation dataset.Results:Differential gene expression analysis revealed 71 upregulated and 21 downregulated genes in preeclampsia. WGCNA linked the onset of preeclampsia with blue and teal modules. Enrichment analysis of candidate biomarkers suggested changes in cell cycle, cellular senescence, and immune-related pathways as primary drivers of preeclampsia. Further refinement through machine learning identified significant upregulation of COL17A1 and DIO2 genes in the peripheral blood of patients, demonstrating robust diagnostic potential. Conclusions:COL17A1 and DIO2 genes can be used as peripheral circulating diagnostic markers for the early diagnosis of eclampsia.
