Abstract: TH-OR024
Mapping Kidney Trait Heritability to Individual Cells Reveals Disease-Specific Remodeling of Genetic Risk Architecture
Session Information
- Artificial Intelligence in Kidney Care: From Pathology to Prediction
October 22, 2026 | Location: Room 711, Convention Center
Abstract Time: 05:20 PM - 05:30 PM
Category: Artificial Intelligence, Digital Health, and Data Science
- 300 Artificial Intelligence, Digital Health, and Data Science
Authors
- Hu, Huiqian, The University of Utah Department of Pharmaceutics and Pharmaceutical Chemistry, Salt Lake City, Utah, United States
- Paul, Anindya S., University of Florida Division of Nephrology Hypertension & Renal Transplantation, Gainesville, Florida, United States
- Rubin, Jeremy, University of Maryland Department of Epidemiology and Biostatistics, College Park, Maryland, United States
- Sarder, Pinaki, University of Florida Division of Nephrology Hypertension & Renal Transplantation, Gainesville, Florida, United States
Background
Genome-wide association studies (GWAS) have identified hundreds of loci associated with kidney function and disease, yet the cell-type-specific mechanisms through which these variants act remain unclear. We built the Kidney Genetic Disease Cell Atlas by mapping GWAS signals for six kidney traits onto single-cell transcriptomic data across multiple clinical conditions.
Methods
We applied single-cell disease relevance scoring (scDRS) to map GWAS signals for estimated glomerular filtration rate (eGFR), cystatin C-based eGFR (eGFRcys), blood urea nitrogen (BUN), urinary albumin-to-creatinine ratio (UACR), type 2 diabetes (T2D), and IgA nephropathy (IgAN) onto a single-nucleus RNA-seq atlas of 304,652 kidney cells spanning five conditions: healthy reference, acute kidney injury (AKI), COVID-19-associated AKI (COV-AKI), diabetic kidney disease (DKD), and hypertensive chronic kidney disease (H-CKD). Findings were validated using Slide-seqV2 spatial transcriptomics (920,088 beads, 44 pucks). scDRS rank shifts were integrated with druggability databases to nominate therapeutic targets.
Results
Cross-platform validation showed concordance between snRNA-seq and spatial data (Spearman ρ = 0.72–0.89). Disease-condition analysis revealed remodeling of genetic risk across cell types, with fibroblasts gaining T2D enrichment in DKD (Δ = +1.07) and immune cells dominating IgAN risk (Cohen's d = 1.40). Gene-level correlation identified condition-specific programs, including mitochondrial gene dominance for eGFRcys and PDE4D emergence for T2D/UACR. Three therapeutic targets were nominated—PDE4D (roflumilast), ITGB6 (STX-100), and SPP1 (anti-OPN antibody)—each showing disease-specific upregulation in distinct cell populations.
Conclusion
The Kidney Genetic Disease Cell Atlas provides a resource to define the cellular basis of kidney disease heritability and identify condition-specific therapeutic opportunities.
GWAS summary statistics for six kidney traits are processed via MAGMA v1.10; the top 1,000 genes per trait feed scDRS v1.0.2 to score 304,652 cells across five conditions
Funding
- NIDDK Support