Diabetic kidney disease (DKD) represents a major long-term complication of Type 2 Diabetes (T2D), increasing the risk of kidney failure and cardiovascular events. Yet, the relationship between T2D and DKD is complex, as it is difficult to accurately predict the degree of kidney damage a T2D patient will develop and whether it will eventually develop into DKD. This is largely driven by a lack of understanding of the precise molecular and cellular mechanisms underlying the association between DKD and T2D. This project aims to deepen our understanding of the development of DKD in T2D on a genetic and cellular level through the application of state-of-the-art single-cell multimodal sequencing assay and bioinformatics tools and deep learning models. By simultaneously profiling gene expression and genome-wide chromatin accessibility within the same kidney nuclei, we will construct a comprehensive molecular atlas derived from thirty kidney biopsies representing a spectrum of severity from non-diabetic kidney disease to DKD in T2D patients from the PRIMETIME2 Danish national cohort study.

This atlas will facilitate the generation of cell type-specific gene regulation networks and the integration of regulatory DNA atlases with disease genetic variants obtained from high-powered genome-wide association studies datasets. We will use this to calculate kidney cell type-specific polygenic risk scores (PRSs) to stratify large heterogenous patient groups and validate the predictive power of these cell type-specific PRSs in several large deeply genotyped cohorts.

Through this comprehensive analysis, we aim to gain novel insights into the shared genetic, cellular, and molecular basis of DKD and T2D. This understanding will enhance the prediction and precision treatment of DKD by stratifying the heterogeneous T2D patient group.