Background:
Diabetic nephropathy (DN) is a major cause of end-stage renal disease and affects up to 40% of individuals with diabetes. Early detection is challenging because current clinical indicators often appear only after substantial kidney damage. Large-scale molecular and imaging resources in UK Biobank provide an opportunity to identify early biomarkers and potential therapeutic targets.
Research questions:
Which clinical, molecular, and imaging features are associated with DN and its early stages?
Which circulating proteins or metabolites show potential causal effects on DN?
Can kidney MRI features improve early prediction of DN or serve as intermediate phenotypes?
Objectives:
Define DN outcomes using diabetes status, albuminuria, eGFR decline, and ICD-coded diagnoses.
Characterize associations between DN traits and clinical, metabolic, proteomic, and imaging phenotypes.
Identify candidate biomarkers using regression and machine-learning approaches.
Apply Mendelian randomization and colocalization to evaluate causal effects of proteins, metabolites, and gene expression on DN.
Integrate kidney MRI features with omics and clinical data to improve risk stratification.
Scientific rationale:
Previous studies have highlighted metabolic and protein markers associated with CKD and diabetes complications, but few have systematically explored DN-specific biomarkers across multi-omics layers or combined these with kidney MRI. Leveraging the scale and depth of UK Biobank, this student-led project aims to generate a comprehensive profile of DN risk markers and causal pathways. Findings will help prioritize proteins, metabolites, and genes for functional validation in ongoing laboratory studies, with the goal of identifying early indicators and potential therapeutic targets for DN.