基于核心基因集的CKD分类与进展预测模型研究

This study constructed core gene sets of varying sizes based on CKD-related transcriptomic datasets using WGCNA and random forest algorithms. It was found that the performance of different gene set scores in the classification diagnosis of CKD varied. By integrating other clinical features, we developed CKD progression prediction models based on Cox regression analysis, LASSO regression analysis, and Logistic regression analysis, clarifying the application value of core gene sets in these predictive models. Through comparisons between the mouse UUO model and human CKD transcriptomics, we validated the reliability of bioinformatics analysis of human CKD datasets and the feasibility of the mouse UUO model for functional studies of CKD-related risk genes. This study provides effective classification and progression risk prediction models for CKD diagnosis and prediction, laying a foundation for research into potential core genes associated with CKD.