Abstract
PURPOSE: To identify novel candidates for cataract and evaluate the contribution of protein-coding variants to cataract susceptibility. METHODS: We first leveraged a publicly-available browser, Genebass, to extract significant gene-based and single-variant association results for cataract in UK Biobank exomes (30,550 cataract cases and 364,291 controls). We then validated findings using genome-wide association study (GWAS) summary statistics from the Genetic Epidemiology Research in Adult Health and Aging (GERA) cohort (28,092 cataract cases and 50,487 controls). Finally, we examined the expression of the prioritized genes in lens tissue using the iSyTE database. RESULTS: Gene-based association testing identified four genes (KDM5B, COL2A1, MIP and CRYBB2) that were associated with cataract (P < 2.50 × 10-6), of which one (KDM5B) was neither previously reported to be associated with congenital cataract nor reported in GWAS. Single-variant association testing identified seven variants within six genes (BFSP2, ZNF800, MIP, HERC2, TSPAN10 and CPAMD8) that were associated with cataract (P < 1.00 × 10-8). Among the identified cataract variants, we found four missense, one synonymous, one frameshift, and one stop-gained variant. Associations at COL2A1, HERC2, and ZNF800 were validated in GERA. Importantly, majority of prioritized cataract genes were robustly expressed in iSyTE lens data and were enriched in structural constituent of eye lens, lens development in camera-type eye, visual perception, and collagen type II trimer pathways. CONCLUSIONS: Our results demonstrate the value of gene-based and single-variant association testing for understanding cataract etiology and uncovering novel genetic risk factors. Our findings also show that cataract-associated genes are significantly expressed in lens tissues and lens-related biological pathways.