Unraveling the Genetic Architecture and Phenotypic Features of Diabetic Kidney Disease Through Integrative Genomic and Cluster-Based Approaches

Abstract

Introduction: Diabetic kidney disease (DKD) is the leading cause of end-stage renal disease (ESRD) worldwide. It is a major contributor to the global healthcare and socioeconomic burden. However, its pathogenesis remains only partially understood, warranting integrative research approaches. By combining quantitative meta-analysis, genetic association study, next-generation sequencing, and data-driven clustering, this thesis offers an in-depth investigation of the multifactorial mechanisms underlying DKD. Methods: This study integrated meta-analytic, genomic, and data-driven approaches. The meta-analytic component synthesized published evidence on genetic variants in the PPARγ (Pro12Ala) and ApoE genes, both of which play key roles in lipid metabolism. Eighteen studies on PPARγ and twenty on ApoE were identified through systematic searches in MEDLINE Complete, Web of Science, Embase, and PubMed. The pooled effect estimates were calculated using quantitative meta-analytic methods. In a case–control genetic association study, 170 patients with type 2 diabetes mellitus (T2DM) and DKD, 157 patients with T2DM without DKD, and 118 conditionally healthy controls were genotyped for four single-nucleotide polymorphisms (SNPs) in vitamin D metabolism genes, CYP27A1 (rs17470271), CYP2R1 (rs1074165), and GC (rs4588, rs7041), using quantitative real-time polymerase chain reaction (qPCR). Whole-exome sequencing (WES) was performed in 40 individuals (20 with T2DM and DKD and 20 T2DM controls without DKD) using the Illumina NovaSeq 6000 platform. Standard bioinformatics pipelines were applied for sequence processing, variant calling, annotation, and post-filtering to identify high-confidence variants potentially associated with DKD. The phenotypic component involved an unsupervised cluster analysis of clinical data from 558 patients with diabetes in real-world clinical settings. K-means clustering was applied to identify distinct diabetes subtypes characterized by unique clinical profiles...