Identification of CDK2 as a key apoptotic gene for predicting cervical cancer prognosis using bioinformatics and machine learning
Year: 2025
Type: article
Abstract: OBJECTIVES: This study aimed to identify apoptosis - related genes with diagnostic and prognostic value in cervical cancer (CC) using integrated bioinformatics and machine learning approaches. METHODS: Gene expression datasets were obtained from the National Center for Biotechnology Information Gene Expression Omnibus (GEO) and the Cancer Genome Atlas (TCGA), with GSE192897 used as the training set. A total of 451 differentially expressed genes (DEGs) were identified, including 221 upregulated and 230 downregulated genes. Eleven apoptosis - related upregulated DEGs were selected for further analysis using three machine learning algorithms: random forest, logistic regression, and support vector machine. Validation was performed using GSE192897, GSE166466, and TCGA-CESC datasets. RESULTS: Among the evaluated genes, cyclin-dependent kinase 2 (CDK2) consistently achieved an AUC > 0.8 in all three validation datasets and had a weighted sum rank > 10, meeting stringent selection criteria. In a CC mouse model, CDK2 expression was significantly elevated and positively correlated with squamous cell carcinoma antigen, carcinoembryonic antigen, vascular endothelial growth factor, and heparanase. siRNA-mediated knockdown of CDK2 reduced cell proliferation and migration while promoting apoptosis. Mice with high CDK2 expression showed significantly lower 4-week survival rates, indicating poor prognosis. CONCLUSIONS: This study identified CDK2 as a key apoptosis - related gene with strong diagnostic and prognostic value in cervical cancer. CDK2 promotes tumor progression and is associated with poor survival, suggesting its potential as a biomarker and therapeutic target for personalized treatment strategies in CC.
DOI: 10.62347/ruxj3980
Author: Miao-Miao Li
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4411919378