Abstract: FR-PO915
Imputation-Powered Whole-Exome Analysis Identifies Rare Coding Variants and Genes Associated With Kidney Function and Disease in the UK Biobank
Session Information
- CKD: Epidemiology, Risk Factors, Prevention - II
November 04, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2201 CKD (Non-Dialysis): Epidemiology‚ Risk Factors‚ and Prevention
Authors
- Katsara, Maria-Alexandra, Institute of Genetic Epidemiology, Uniklinik Freiburg, Freiburg, Germany
- König, Eva, Institute for Biomedicine, Bolzano, Italy
- Wuttke, Matthias, Institute of Genetic Epidemiology, Uniklinik Freiburg, Freiburg, Germany
Group or Team Name
- on behalf of all authors
Background
Chronic kidney disease (CKD) is a major public health concern affecting ~10% of the global adult population. Genome-wide association studies on imputed common genotypes have identified variants associated with kidney function and CKD, but cannot comprehensively investigate rare coding variation. We aimed to identify rare pathogenic variants and genes impacting human kidney function and disease.
Methods
A genotype imputation approach was applied to whole exome sequencing data of European ancestry participants from the UK Biobank to increase our sample size from 166,891 to 408,511. We used exome-wide association studies (ExWAS) and gene-level Burden tests in order to identify rare variants and genes associated with different kidney phenotypes including eGFR, UACR, urea, urate and clinically diagnosed CKD. We performed a phenome-wide association study of the resulting genes.
Results
We imputed ~7.5 million exonic variants for 241,620 individuals without WES data. In a validation sample, the overall concordance of sequenced and imputed genotypes was >0.98.
We identified 158 rare variants and 106 genes significantly associated with kidney function. Among these genes, there were known monogenic kidney disease genes such as CUBN and PKD2, for which we highlight novel putatively disease-causing mutations. Furthermore, we implicate genes currently not linked to kidney function and disease in humans, such as FNIP1 and EPB41L5. Disruption of Fnip1 in mice is sufficient for renal cyst formation, and Epb41l5-deficient mice are known to develop focal-segmental glomerulosclerosis. Our results not only establish human relevance, but also identify potentially causal variants.
Across the phenome, we identified numerous instances where the assumed loss-of-function of implicated genes resulted in increased disease risk, often reflecting known clinical signs and symptoms. For example, carriers of rare damaging variants in COL4A3 showed higher odds of hematuria.
Conclusion
Our study underscores the value of imputation-powered analysis and reveals both novel and confirms known variants associated with kidney disease. It generates a comprehensive resource to direct future functional and clinical studies. Results will be shared via an interactive website.