Quantifying variant contributions in cystic kidney disease using national-scale whole genome sequencing.

Abstract

BackgroundCystic kidney disease (CyKD) is a predominantly familial disease in which gene discovery has been led by family-based and candidate gene studies, an approach that is susceptible to ascertainment and other biases.MethodsUsing whole genome sequencing data from 1,209 cases and 26,096 ancestry-matched controls participating in the 100,000 Genomes Project, we adopted hypothesis-free approaches to generate quantitative estimates of disease risk for each genetic contributor to CyKD, across genes, variant types and allelic frequencies.ResultsIn 82.3% of cases, a qualifying potentially disease-causing rare variant in an established gene was found. There was an enrichment of rare coding, splicing, and structural variants in known CyKD genes, with novel statistically significant gene-based signals in COL4A3 and (monoallelic) PKHD1. Quantification of disease risk for each gene (with replication in the separate UK BioBank study) revealed substantially lower risk associated with genes more recently associated with autosomal dominant polycystic kidney disease, with odds ratios for some below what might usually be regarded as necessary for classical Mendelian inheritance. Meta-analysis of common variants did not reveal significant associations but suggested this category of variation contributes 3-9% to the heritability of CyKD across European ancestries.ConclusionBy providing unbiased quantification of risk effects per gene, this research suggests that not all rare variant genetic contributors to CyKD are equally likely to manifest as a Mendelian trait in families. This information may inform genetic testing and counselling in the clinic.Keywords: genomics, cystic kidney disease, renal, ADPKD, WGS.