Abstract
Purpose :
Myopia is an eye condition in which distant objects appear out of focus. Within recent years, the incidence and prevalence of Myopia have increased in most populations. While population based association studies are well-powered for the identification of common variants, they are not well suited for the identification of rare variants. We sought to identify both population-specific and general rare variants that increase risk for Myopia.
Methods :
We performed a family-based association test (FBAT) of Myopia using Exome Chip genotyping (Illumina Human Exome v1.1 array plus 24,263 custom SNPs) in five family cohorts for a total of 1718 subjects in 261 multiplex Myopia families. Individuals in the families were defined as myopic if their average refractive error was <= -1 Diopter (D) and were considered unaffected if it was > 0.0 D. After quality control, there were ~127,000 polymorphic SNPs. We used FBAT software to test for associations at the variant and gene-level.
Results :
FBAT analysis resulted in a significant association in a region upstream of two gamma-Aminobutryric Acid (GABA) receptor genes (GABRA6; GABRB2). GABA is a neurotransmitter that has been implicated in refractive development. The associated SNP, rs1373602, is not found in the Genotype-Tissue Expression (GTEx) project, but a nearby SNP, rs62381591, has been identified as an expression quantitative trait locus for the GABRA6 gene. As the significant variant is common, we wondered why the larger, population-based association studies of Myopia have not detected an association in this region. We found that this variant is not in high linkage disequilibrium with any other variants in our dataset (highest r2 was ~0.002) and is indicated as triallelic in the 1000 genomes dataset (although it was biallelic in our dataset).
Conclusions :
Potentially, this region harbors a missed signal that is tagged by a variant which is filtered out before GWAS analysis due to it being triallelic and not tagged by any other variant in the region. Additionally, using a family-based study may have increased our power to detect the effect. We plan to follow-up this analysis by collaborating with groups that have performed genome-wide genotyping studies of Myopia to determine if this signal was missed due to the aforementioned reasons.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.