Purchase this article with an account.
Robert Wojciechowski, Pirro G Hysi, Milly S. Tedja, Virginie JM Verhoeven, Jeremy A Guggenheim, Stuart MacGregor, Christopher J Hammond, Caroline Klaver; Meta-analyses of two large GWAS show a high genetic correlation between myopia age-at-onset and refractive error during adulthood and older age.. Invest. Ophthalmol. Vis. Sci. 2017;58(8):1227.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Genes and environment play important roles in refractive development. Myopia age-at-onset (MAO) is a strong determinant of the magnitude of myopia throughout life. Hence, delaying the onset of myopia has become a priority of public health efforts and treatment strategies for myopia. We investigated the genetic correlation between MAO and ocular refraction in two large meta-analyses (MA) of genomewide association studies (GWAS).
We conducted whole-genome meta-analyses (MA) of two independent studies: CREAM, an international consortium of 30 studies (n=56,368); and two datasets from 23andMe (n=104,293). MA of CREAM GWAS were performed using summary statistics from linear regressions of refractive error (RE) on allele dosage. GWAS of MAO in 23andMe were performed using Cox proportional hazards models. All GWAS were conducted on marker sets imputed to 1000 Genomes reference panels (>10 million markers per study). Normalized CREAM and 23andMe association results were then combined in a MA using equal variances on marker-specific Z-scores.We calculated cross-trait genetic correlations (r) using several methods: 1) LD-score regression (LDr); 2) a thresholded regression of beta statistics (BREGr); 3) a variable threshold bootstrapped regression of beta coefficients from near-independent (r2<0.1) markers (iBREGr); 4) a trans-ethnic method implemented in the POPCORN software (POPr).
We identified ~200 loci for MAO and RE. Estimated genetic correlations between MAO and RE in European cohorts were high using all methods: LDr=0.94 (with 6.6M SNPs); BREGr=0.95 (8,434 SNPs at p<0.0011 in both MA); iBREGr=0.67 (8,456 SNPs with p<0.0001 in the stage 3 MA). Inter-ethnic POPr for CREAM-Asian (n=12,839) vs 23andMe and CREAM-Asian vs CREAM-European (n=43,529) were 0.79 and 0.80, respectively. We estimated that a 1.1-fold increase in the genetic risk (hazard) of myopia is associated with a -0.15 diopter mean shift in myopia in adulthood.
Using two independent GWAS (n=160,661), we show a high genetic correlation between MAO and RE in adulthood, indicating that the complex polygenic architecture of myopia onset significantly determines relative refractive status later in life. Despite marked differences in myopia incidence between East Asian and European populations, the underlying genetic causes do not differ significantly.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only