- Email: [email protected]
Genetic Association of Refractive Error and Axial Length with 15q14 but Not 15q25 in the Blue Mountains Eye Study Cohort
Genetic Association of Refractive Error and Axial Length with 15q14 but Not 15q25 in the Blue Mountains Eye Study Cohort Maria Schache, BSc(Hons), PhD,1 Andrea J. Richardson, BSc(Hons),1 Paul Mitchell, MD, PhD,2 Jie Jin Wang, MMed, PhD,1,2 Elena Rochtchina, BSc, MApplStat,2 Ananth C. Viswanathan, MD, PhD,3 Tien Y. Wong, MD, PhD,1,4 Seang Mei Saw, MBBS, PhD,5 Fotis Topouzis, MD, PhD,6 Jing Xie, MD, PhD,1 Xueling Sim, BSc, PhD,7 Elizabeth G. Holliday, MBiostatistics, PhD,8,9 John Attia, MD, PhD,8,9 Rodney J. Scott, BSc, PhD,9,10,11 Paul N. Baird, BSc, PhD1* Purpose: Myopia is a common complex condition influenced by genetic and environmental factors. Two recent genome-wide association studies have identified loci on chromosomes 15q25 and 15q14 associated with refractive error in Caucasian populations. Our study aimed to assess the association of these 2 loci with refractive error and ocular biometric measures in an independent ethnically matched Caucasian cohort. Design: Genetic association study using unrelated individuals. Participants: Blue Mountains Eye Study (BMES) cohort. A total of 1571 individuals were included in this study. Methods: Single nucleotide polymorphism (SNP) genotype data were collected from the BMES cohort as part of the Wellcome Trust Case Control Consortium 2. Imputation was performed using MACH version 1.1.16, and statistical analysis was conducted using PLINK. Association tests were performed at both loci using refractive error (spherical equivalent), axial length, corneal curvature, and anterior chamber depth as the phenotypes. Main Outcome Measures: Refractive error, axial length, corneal curvature, and anterior chamber depth. Results: A total of 1571 individuals were available from the BMES for analysis. A statistically significant association for refractive error was evident for SNPs at the 15q14 locus, with P values ranging from 1.5⫻10⫺2 at rs685352 to 6.4⫻10⫺4 at rs560764, whereas association could not be confirmed for SNPs at the 15q25 locus, with P values ranging from 8.0⫻10⫺1 to 6.4⫻10⫺1. Ocular biometric analysis revealed that axial length was the most likely trait underlying the refractive error association at the 15q14 locus for SNPs rs560766 (P⫽0.0054), rs634990 (P⫽0.0086), and rs8032019 (P⫽0.0081). Conclusions: Our results confirm the association with refractive error at the 15q14 locus but do not support the association observed at the 15q25 locus. Axial length seemed to be a major parameter at the 15q14 locus, underscoring the importance of this locus in myopia and future clinical treatment. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Ophthalmology 2013;120:292–297 © 2013 by the American Academy of Ophthalmology. *Group members listed online in Appendix 1 (available at http://aaojournal.org)
Myopia (short sightedness) is a common ocular disorder resulting in visual impairment. Myopia is clinically defined using spherical equivalent measured in diopters (D), typically defined as less than ⫺0.5 D. Myopia occurs when light rays entering the eye are focused in front of rather than on the retina. In the majority of cases, this is due to an elongated eyeball (axial length), but it can also result from increased steepness of the cornea (corneal curvature) or increased lens density. Myopia is a complex trait known to be influenced by genetic and environmental factors. Although the cause of myopia is not fully understood, it is known that environmental factors, such as excessive exposure to near work (reading), lack of time spent outdoors, high educational
292
© 2013 by the American Academy of Ophthalmology Published by Elsevier Inc.
attainment, intelligence quotient, and urbanization, are contributory risk factors.1–3 However, known environmental factors so far account for a small proportion of the total phenotypic variance seen in myopia; thus, the remaining variance is likely to be explained by so far unknown environmental or genetic factors or through their interaction.4 In the pursuit of genetic factors for refractive error, numerous linkage and candidate gene– based studies have identified a number of genetic loci associated with myopia. Despite more than 30 candidate genes and 27 chromosomal loci having been reported, few of these have been independently confirmed; thus, few strong candidate genes currently exist for myopia.5 One approach that has been undertaken to identify the “missing heritability” has been ISSN 0161-6420/13/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.08.006
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES through the use of genome-wide association studies (GWAS). Several such GWAS have recently been undertaken in Asian populations identifying loci at chromosomes 11q24.1, 5p15, 4q25, and 13q12.12.6 – 8 In addition, 2 GWAS in European Caucasians have been undertaken for common myopia, and susceptibility loci have been shown to be associated with refractive error at the single nucleotide polymorphism (SNP) rs634990 at 15q14 (odds ratio [OR], 1.83; 95% confidence interval [CI], 1.42–2.36) and SNP rs8027411 at 15q25 (OR, 1.16; 95% CI, 1.02–1.28).9,10 These 2 studies also showed associations in meta-analyses based on 5 and 7 cohorts, respectively. A replication study of these 2 loci in Japanese individuals with high myopia and 2 control cohorts (cataract and healthy Japanese) confirmed association at the 15q14 locus but showed weak to no association at the 15q25 locus.9,10 The Australian Blue Mountains Eye Study (BMES) is a longitudinal, ethnically homogeneous cohort of individuals of mainly European descent in whom both refractive error and ocular biometric measures have been collected. By using this cohort, we wanted to replicate the findings obtained at the 15q14 and 15q25 loci for refractive error and assess whether any of the ocular biometric measures of corneal curvature, anterior chamber depth, and ocular axial length were also associated with SNPs at these 2 loci.
performed if the initial visual acuity was less than 54 letters read correctly (Snellen equivalent 6/6), using the Beaver Dam Eye Study modification of the Early Treatment Diabetic Retinopathy Study protocol.13 Spherical equivalent was calculated using the standard formula: spherical equivalent ⫽ sphere ⫹ (cylinder/2). Spherical equivalent measurements from the second follow-up of the BMES were used because this was when blood samples for DNA extraction were taken. At the third BMES follow-up visit, axial length and anterior chamber depth were measured in millimeters. The average of the horizontal and vertical K readings was used to obtain corneal curvature measures. Analysis was performed using the spherical equivalent and ocular biometric measures averaged over both eyes.
Genotyping and Imputation Genotype data for SNPs at the 15q14 and 15q25 were collected as part of the Wellcome Trust Case Control Consortium 2. An Illumina Human 670-Quad custom genotyping array v1 was used with genotyping performed at the Wellcome Trust Sanger Institute (Cambridge, UK). Imputation of SNPs was performed using MACH version 1.1.16 (http://www.sph.umich.edu/csg/abecasis/ MACH/, accessed April 30, 2012) for autosomes only and calculated with reference to HapMap release 24 using the CEU (CEPH [Centre d’Etude du Polymorphisme Humain] Utah Residents with Northern and Western European ancestry) population panel (National Center for Biotechnology Information Build 36.1).
Materials and Methods Ethics Statement This study was conducted after approval from the Human Ethics Research Committee at the University of Sydney and the Western Sydney Area Health Service. All participants provided written, informed consent in accordance with the Declaration of Helsinki.
Participants and Examination The BMES is a population-based longitudinal survey of vision and common eye diseases that has followed residents in an area west of Sydney, New South Wales, Australia, over a 15-year time period. This suburban area has a stable and homogeneous population representative of Australia for socioeconomic status. A total of 3654 residents of predominately European ancestry (99%) aged 49 to 97 years participated in the baseline survey in 1992–1994. A second follow-up was performed in 1997–1999, a third follow-up was performed in 2002–2004, and a fourth follow-up was performed in 2007–2009. At each visit, a comprehensive questionnaire was administered in addition to a detailed ophthalmic examination. The BMES used a protocol for phenotyping similar to that used by Solouki et al10 and Hysi et al.9 Participants underwent a comprehensive eye examination including best-corrected visual acuity, objective and subjective refraction, slit-lamp examination, and an extensive lifestyle questionnaire together with height and weight measures.11,12 In addition, axial length, anterior chamber depth, and keratometry readings were obtained at the third follow-up of the BMES.11,12 Visual acuity of each eye was measured with current glasses, if worn, and with pinhole, using a logarithm of the minimum angle of resolution chart.13 A Humphrey autorefractor (Model 530; Humphrey Instruments Inc., San Leandro, CA) was used to obtain an objective refraction. Subjective refraction was then
Data Quality Control Numerous quality-control filters were applied before association testing for both SNP and sample filtering. Three filters were applied to the SNPs: (1) SNPs with a genotyping rate of ⬍0.95, (2) Hardy– Weinberg P value ⬍1⫻10 – 6, and (3) minor allele frequency ⬍0.001. Four filters were applied to the individuals: (1) genotype call frequency ⬍0.99, (2) discrepancy between clinical and genotypic gender, (3) evidence of cryptic relatedness based on a pair sharing 0 allele identity by descent at ⬍0.95 of their genome, and (4) evidence of non–European ancestry based on principle component analysis and mixture modeling for outlier detection.
Statistical Analysis Further filtering of individuals was undertaken to exclude those who had undergone cataract surgery, had severe visual impairment, had any known ocular pathologies that may affect refraction measures (e.g., macular degeneration and nuclear cataracts), or had evidence of anisometropia ⬎2 D between eyes. In addition, individuals identified as having refraction changes of ⬎0.5 D between the second and third follow-up studies were excluded because this may indicate the advent of subclinical nuclear sclerosis or hyperopic shift. Spherical equivalent and ocular biometric measures were considered to be continuous traits, and linear regression was performed using PLINK (http://pngu.mgh.harvard.edu/⬃purcell/ plink/, accessed April 30, 2012) with SNP allele dosage specified as the predictor variable and age and sex included as covariates. A P value of 0.05/5⫽0.01 for the 15q25 region and 0.05/14⫽0.004 for the 15q14 region were considered statistically significant for refractive error after Bonferroni correction. Power calculations were performed using QUANTO version 1.2 (http://hydra.usc. edu/gxe, accessed April 30, 2012).
293
Ophthalmology Volume 120, Number 2, February 2013 Table 1. Association Study Results for Refractive Error in the Blue Mountains Eye Study Cohort at the 15q25 Locus Blue Mountains Eye Study SNP
Minor Allele
Allele Frequency
Beta
SEM
P
rs939661 rs939658 rs17175798* rs8027411† rs8033963
A G C T T
0.49 0.49 0.49 0.49 0.49
⫺0.05 ⫺0.04 ⫺0.03 ⫺0.03 ⫺0.03
0.10 0.10 0.10 0.10 0.10
0.6403 0.7239 0.7951 0.7391 0.7391
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism. *Imputed SNP. SNP with best P value from the study by Hysi et al.9
†
Results Clinical Information All tested SNPs at the 15q14 and 15q25 loci remained in the analysis after application of the SNP filtering criteria. A total of 1571 individuals remained in our final analysis after application of the sample filtering criteria. In the trimmed cohort, 43.0% were female and 57.0% were male. The average age of participants was 73.84⫾7.81 years, and the average spherical equivalent measure was ⫹0.59⫾2.17 D (range, ⫺13.56 to ⫹8.56 D). The average axial length was 23.44⫾1.0 mm (range, 28.45–19.94 mm).
Replication Assessment of Loci in the Blue Mountains Eye Study Cohort We next sought to determine whether the association detected at the 15q14 and 15q25 loci reported by Solouki et al10 and Hysi et al9 could be replicated in the BMES. Results from the BMES are shown in Tables 1 and 2. For the 15q25 locus reported by Hysi et al,9 there were no statistically significant associations observed for any SNP. For the 15q14 locus reported by Solouki et al,10 the lowest P value in the BMES was 6.4⫻10⫺4 at SNP rs560766, whereas SNP rs688220,
the strongest SNP in the study Solouki et al,10 showed a P value of 2.3⫻10⫺3 in the BMES. The effect (beta) was in the same direction as previously described, with SNPs explaining between ⫺0.24 and ⫺0.33 D of the effect at this locus. Considering the replication of the 15q14 region for refractive error in the BMES, we then assessed whether any ocular biometric traits also may show association at these SNPs because the phenotypes of refractive error and ocular biometric components, particularly axial length, have been shown to be highly correlated.14 Table 3 indicates that 3 SNPs (rs560766, rs634990, and rs8032019) at the 15q14 have a statistically significant association with axial length (P⫽0.0054, P⫽0.0086, and P⫽0.0081, respectively). No associations with SNPs for the other ocular biometric traits of corneal curvature and anterior chamber depth were evident at this locus.
Discussion The data presented confirm an association of refractive error with the 15q14 locus initially reported by both Solouki et al,10 with SNPs rs524952 and rs634990 showing the strongest association. In the case of the 15q25 locus reported by Hysi et al,9 we were not able to demonstrate association for any SNPs at
Table 2. Association Study Results for Refractive Error in the Blue Mountains Eye Study 15q14 Locus Blue Mountains Eye Study SNP
Minor Allele
Allele Frequency
Beta
SEM
P
rs560766 rs4924134* rs11073058* rs11073059* rs11073060* rs7163001* rs524952* rs634990* rs7176510* rs688220*† rs619788* rs580839 rs8032019* rs685352*
A G T A A A A C T A A A G G
0.44 0.44 0.44 0.44 0.44 0.44 0.48 0.49 0.44 0.44 0.44 0.44 0.39 0.44
⫺0.33 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.30 ⫺0.29 ⫺0.30 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.28 ⫺0.24
0.0960 0.0957 0.0956 0.0956 0.0956 0.0956 0.0959 0.0960 0.0955 0.0960 0.0958 0.0958 0.0953 0.0973
0.0006445 0.002189 0.002635 0.002635 0.002635 0.002635 0.002075 0.002558 0.002022 0.002299 0.002275 0.002275 0.003276 0.01552
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism. *Imputed SNPs. † SNP with best P value from the study by Solouki et al.10
294
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES Table 3. Association Study Results for Axial Length, Corneal Curvature, and Anterior Chamber Depth in the Blue Mountains Eye Study at the 15q14 Locus Axial Length
Corneal Curvature
Anterior Chamber Depth
SNP
Minor Allele
Allele Frequency
Beta
SEM
P
Beta
SEM
P
Beta
SEM
P
rs560766 rs4924134 rs11073058 rs11073059 rs11073060 rs7163001 rs524952 rs634990 rs7176510 rs688220 rs619788 rs580839 rs8032019 rs685352
A G T A A A A C T A A A G G
0.44 0.44 0.44 0.44 0.44 0.44 0.48 0.49 0.44 0.44 0.44 0.44 0.39 0.44
0.13 0.12 0.12 0.12 0.12 0.12 0.11 0.12 0.12 0.11 0.12 0.12 0.12 0.10
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
0.0054 0.0143 0.0139 0.0139 0.0139 0.0139 0.0146 0.0086 0.0122 0.0158 0.0134 0.0134 0.0081 0.0372
⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 0.02 ⫺0.01 ⫺0.03 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.01 ⫺0.01
0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07
0.7754 0.8258 0.7494 0.7494 0.7494 0.7494 0.7893 0.8839 0.6868 0.7397 0.7286 0.7286 0.8950 0.8923
0.003 0.003 0.004 0.004 0.004 0.004 0.006 0.009 0.003 0.003 0.002 0.002 0.020 0.017
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
0.8722 0.8409 0.7888 0.7888 0.7888 0.7888 0.7161 0.5624 0.8475 0.8699 0.8917 0.8917 0.1999 0.2900
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism.
this locus using the BMES cohort. In addition, we were able to show that the ocular biometric trait of axial length seems to be a contributor to the effect seen at the 15q14 locus, implying that SNPs identified for refractive error may harbor associations for other phenotypic traits that underlie myopia. The BMES cohort is a good choice to undertake replication because it consists of ethnically similar individuals (Caucasians of European descent) with a phenotype value (mean spherical equivalent ⫹0.6 D) similar to that of the studies by both Solouki et al10 and Hysi et al.9 In addition, there is no evidence of gender bias in the BMES cohort because 57.3% were female, whereas 54.1% to 92.6% were female in the previous studies. The BMES is comparable to the studies by Solouki et al10 and Hysi et al9 in terms of both phenotyping protocols and allele frequencies. Although replication was confirmed in the BMES cohort for the 15q14 region, our analysis did not replicate that of the 15q25 locus by Hysi et al.9 The beta effect was in the same direction and overlapped in magnitude (range, ⫺0.03 to ⫺0.05 D) compared with the majority of the discovery and replication studies (range, ⫺0.02 to ⫺0.29 D) in the original analysis by Hysi et al9; thus, a discussion of why there was no replication in the BMES is warranted. One possibility is that myopia represents a complicated trait composed of different endophenotypes, such as axial length and corneal curvature. Thus, when myopia is considered as a single phenotype using spherical equivalent measures, true associations may be masked. Our finding of axial length as an important component at the 15q14 locus underscores the importance of considering other endophenotypes that might underlie refractive error and myopia. Indeed, we previously showed through a heritability analysis that a number of genes are likely shared between refractive error and axial length.14 This suggests that the underlying genetic influences for refractive error may be related to axial length, and thus there may be overlapping genes. This does not mean that all refractive error loci will necessarily present with an ocular biometric component, because independent loci for ocular biometric components also are likely to exist as indi-
cated by the report of a quantitative trait locus for axial length on the long arm of chromosome 5 identified in twins of European ethnicity.15 A second possibility is that the current study is underpowered to detect an OR of 1.16 identified in the study by Hysi et al.9 A power calculation based on the beta (effect) value for BMES indicated that to replicate the finding at the 15q25 locus we would have approximately 29% power (2-sided t test), whereas to obtain 80% power would require a study size of approximately 5000 individuals. Thus, the BMES is most likely underpowered to detect association at this locus on its own. In the study by Hysi et al,9 no association was evident in the Rotterdam 2 (P⫽0.08), the 1958 British Birth Cohort (P⫽0.04 – 0.16), and the Australian Twin (P⫽0.4 – 0.9) cohorts. These cohorts all presented with a sample size of 1000 to 2157 individuals and would fit with the predicted nonsignificant P value findings reported in the current study. However, the Rotterdam 3 study in the study by Hysi et al9 consisted of 2082 individuals but did result in a significant P value (0.01– 0.02). Thus, a small sample size is not necessarily an accurate predictor of nonsignificant association. The Forest plot in the study by Hysi et al shows that the CIs do not overlap across many studies, indicating a high degree of heterogeneity at this locus. A recent Japanese study16 of a cohort with high myopia undertook replication of 4 SNPs (rs634990, rs524952, rs8027411, and rs17175798) from the 15q14 and 15q25 loci and reported findings similar to the data presented in the current study. In that study, a total of 1125 unrelated Japanese patients with high myopia (⬎26.1 mm in both eyes) and 2 independent control groups consisting of 366 patients with cataract (axial length ⬍25.0 mm in both eyes) and 929 healthy Japanese were evaluated. An association with high myopia was observed at the 15q14 locus for both the cataract control group (P⫽0.0035 for rs634990 and P⫽0.0017 for rs524952) and the healthy control group (P⫽1.91⫻10⫺6 for rs634990 and P⫽8.78⫻10⫺7 for rs524952). However, no statistically significant association was evident for SNPs at the 15q25
295
Ophthalmology Volume 120, Number 2, February 2013 region in either of the control groups. These findings and the current study indicate that the 15q14 association would seem to be associated across all levels of refractive error, as well as being present in different ethnicities. In contrast, the 15q25 locus has a weaker, less robust association. Our results suggest that the association at the 15q25 locus is weak. However, our data indicate that the 15q14 locus represents a strong and robust locus for refractive error and axial length. We have shown that rather than SNP rs688220 at 15q14 being the most highly associated for refractive error as previously reported by Solouki et al,10 the neighboring SNPs, rs524952 and rs634990, are more strongly associated with refractive error. The most associated SNPs for axial length are rs560766, rs634990, and rs8032019, with only SNP rs634990 overlapping between refractive error and axial length. In conclusion, there exists a genetic driver for myopia at the 15q14 locus, but we were unable to provide evidence to support a locus for refractive error at 15q25. The association at the 15q14 locus seems robust; thus, the next step is to search for candidate genes in the region. To date, there are no known refractive error or axial length genes in this region, so the exact nature of the genetic drivers in the region and how they influence refractive error remain to be elucidated. This may come through further genetic analysis, such as targeted sequencing across the region to identify causal variants, or through functional studies of known genes to identify the biological mechanism behind these findings.
References 1. Saw SM, Chua WH, Hong CY, et al. Nearwork in early-onset myopia. Invest Ophthalmol Vis Sci 2002;43:332–9. 2. Morgan I, Rose K. How genetic is school myopia? Prog Retin Eye Res 2005;24:1–38. 3. Rose KA, Morgan IG, Smith W, et al. Myopia, lifestyle, and schooling in students of Chinese ethnicity in Singapore and Sydney. Arch Ophthalmol 2008;126:527–30.
4. Saw SM, Tan SB, Fung D, et al. IQ and the association with myopia in children. Invest Ophthalmol Vis Sci 2004;45:2943–8. 5. Baird PN, Schache M, Dirani M. The GEnes in Myopia (GEM) study in understanding the aetiology of refractive errors. Prog Ret Eye Res 2010;29:520 – 42. 6. Li Z, Qu J, Xu X, et al. A genome-wide association study reveals association between common variants in an intergenic region of 4q25 and high-grade myopia in the Chinese Han population. Hum Mol Genet 2011;20:2861– 8. 7. Nakanishi H, Yamada R, Gotoh N, et al. A genome-wide association analysis identified a novel susceptible locus for pathological myopia at 11q24.1. PLoS Genet 2009;5:e1000 660. 8. Shi Y, Qu J, Zhang D, et al. Genetic variants at 13q12.12 are associated with high myopia in the Han Chinese population. Am J Hum Genet 2011;88:805–13. 9. Hysi PG, Young TL, Mackey DA, et al. A genome-wide association study for myopia and refractive error identifies a susceptibility locus at 15q25. Nat Genet 2010;42: 902–5. 10. Solouki AM, Verhoeven VJ, van Duijn CM, et al. A genomewide association study identifies a susceptibility locus for refractive errors and myopia at 15q14. Nat Genet 2010;42: 897–901. 11. Foran S, Rose K, Wang JJ, et al. Correctable visual impairment in an older population: the Blue Mountains Eye Study. Am J Ophthalmol 2002;134:712–9. 12. Mitchell P, Smith W, Attebo K, et al. Prevalence of openangle glaucoma in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:1661–9. 13. Attebo K, Mitchell P, Smith W. Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:357– 64. 14. Dirani M, Shekar SN, Baird PN. Evidence of shared genes in refraction and axial length: the Genes in Myopia (GEM) twin study. Invest Ophthalmol Vis Sci 2008;49:4336 –9. 15. Zhu XS, Lin T, Stone RA, et al. Sex-differences in chick eye growth and experimental myopia. Exp Eye Res 1995;61: 173–9. 16. Hayashi H, Yamashiro K, Nakanishi H, et al. Association of 15q14 and 15q25 with high myopia in Japanese. Invest Ophthalmol Vis Sci 2011;52:4853– 8.
Footnotes and Financial Disclosures Originally received: March 4, 2012. Final revision: August 2, 2012. Accepted: August 3, 2012. Available online: November 3, 2012.
7
Centre for Molecular Epidemiology, National University of Singapore, Singapore.
8
Manuscript no. 2012-314.
1
Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, Melbourne, Australia.
2
Centre for Vision Research, Department of Ophthalmology and Westmead Millennium Institute, University of Sydney, NSW Australia.
School of Medicine and Public Health, University of Newcastle, Newcastle, Australia.
9
Hunter Medical Research Institute, Newcastle, Australia.
10
The Centre for Information Based Medicine and the School of Biomedical Sciences and Pharmacy, University of Newcastle, Newcastle, Australia. 11
3
National Institute for Health Research Biomedical Research Centre for Ophthalmology, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, United Kingdom.
4
Singapore Eye Research Institute, Singapore National Eye Centre, National University of Singapore, Singapore.
5
Department of Epidemiology and Public Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
6
Department of Ophthalmology, School of Medicine, Aristotle University of Thessaloniki, AHEPA Hospital, Thessaloniki, Greece.
296
The Division of Genetics, Hunter Area Pathology Service, John Hunter Hospital, Newcastle, Australia. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Funding: National Health and Medical Research Council, Canberra, Australia (Project Grants: 974159, 211069, 457349, 512423, 475604, and 529912); Centre for Clinical Research Excellence in Translational Clinical Research in Eye Diseases; National Health & Medical Research Council research fellowships to P.N.B. (No. 1028444) and J.J.W. (No. 358702, 632909); Wellcome Trust as part of the Wellcome Trust
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES Case Control Consortium 2 (No. 085475B08Z, 08547508Z, 076113); Medical Research Council (UK) (G0401527); Fight for Sight (London); and National Institute for Health Research grant for a Biomedical Research Centre for Ophthalmology. This study used genotyping data generated and funded by the Wellcome Trust Case-Control Consortium 2 using samples from the BMES.
A full list of contributing investigators to the generation of these data is available online in Appendix 1 (http://aaojournal.org). Correspondence: Maria Schache, BSc, PhD, Centre for Eye Research Australia, 32 Gisborne St., Melbourne, Australia 3002. E-mail: [email protected].
297
Myopia (short sightedness) is a common ocular disorder resulting in visual impairment. Myopia is clinically defined using spherical equivalent measured in diopters (D), typically defined as less than ⫺0.5 D. Myopia occurs when light rays entering the eye are focused in front of rather than on the retina. In the majority of cases, this is due to an elongated eyeball (axial length), but it can also result from increased steepness of the cornea (corneal curvature) or increased lens density. Myopia is a complex trait known to be influenced by genetic and environmental factors. Although the cause of myopia is not fully understood, it is known that environmental factors, such as excessive exposure to near work (reading), lack of time spent outdoors, high educational
292
© 2013 by the American Academy of Ophthalmology Published by Elsevier Inc.
attainment, intelligence quotient, and urbanization, are contributory risk factors.1–3 However, known environmental factors so far account for a small proportion of the total phenotypic variance seen in myopia; thus, the remaining variance is likely to be explained by so far unknown environmental or genetic factors or through their interaction.4 In the pursuit of genetic factors for refractive error, numerous linkage and candidate gene– based studies have identified a number of genetic loci associated with myopia. Despite more than 30 candidate genes and 27 chromosomal loci having been reported, few of these have been independently confirmed; thus, few strong candidate genes currently exist for myopia.5 One approach that has been undertaken to identify the “missing heritability” has been ISSN 0161-6420/13/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.08.006
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES through the use of genome-wide association studies (GWAS). Several such GWAS have recently been undertaken in Asian populations identifying loci at chromosomes 11q24.1, 5p15, 4q25, and 13q12.12.6 – 8 In addition, 2 GWAS in European Caucasians have been undertaken for common myopia, and susceptibility loci have been shown to be associated with refractive error at the single nucleotide polymorphism (SNP) rs634990 at 15q14 (odds ratio [OR], 1.83; 95% confidence interval [CI], 1.42–2.36) and SNP rs8027411 at 15q25 (OR, 1.16; 95% CI, 1.02–1.28).9,10 These 2 studies also showed associations in meta-analyses based on 5 and 7 cohorts, respectively. A replication study of these 2 loci in Japanese individuals with high myopia and 2 control cohorts (cataract and healthy Japanese) confirmed association at the 15q14 locus but showed weak to no association at the 15q25 locus.9,10 The Australian Blue Mountains Eye Study (BMES) is a longitudinal, ethnically homogeneous cohort of individuals of mainly European descent in whom both refractive error and ocular biometric measures have been collected. By using this cohort, we wanted to replicate the findings obtained at the 15q14 and 15q25 loci for refractive error and assess whether any of the ocular biometric measures of corneal curvature, anterior chamber depth, and ocular axial length were also associated with SNPs at these 2 loci.
performed if the initial visual acuity was less than 54 letters read correctly (Snellen equivalent 6/6), using the Beaver Dam Eye Study modification of the Early Treatment Diabetic Retinopathy Study protocol.13 Spherical equivalent was calculated using the standard formula: spherical equivalent ⫽ sphere ⫹ (cylinder/2). Spherical equivalent measurements from the second follow-up of the BMES were used because this was when blood samples for DNA extraction were taken. At the third BMES follow-up visit, axial length and anterior chamber depth were measured in millimeters. The average of the horizontal and vertical K readings was used to obtain corneal curvature measures. Analysis was performed using the spherical equivalent and ocular biometric measures averaged over both eyes.
Genotyping and Imputation Genotype data for SNPs at the 15q14 and 15q25 were collected as part of the Wellcome Trust Case Control Consortium 2. An Illumina Human 670-Quad custom genotyping array v1 was used with genotyping performed at the Wellcome Trust Sanger Institute (Cambridge, UK). Imputation of SNPs was performed using MACH version 1.1.16 (http://www.sph.umich.edu/csg/abecasis/ MACH/, accessed April 30, 2012) for autosomes only and calculated with reference to HapMap release 24 using the CEU (CEPH [Centre d’Etude du Polymorphisme Humain] Utah Residents with Northern and Western European ancestry) population panel (National Center for Biotechnology Information Build 36.1).
Materials and Methods Ethics Statement This study was conducted after approval from the Human Ethics Research Committee at the University of Sydney and the Western Sydney Area Health Service. All participants provided written, informed consent in accordance with the Declaration of Helsinki.
Participants and Examination The BMES is a population-based longitudinal survey of vision and common eye diseases that has followed residents in an area west of Sydney, New South Wales, Australia, over a 15-year time period. This suburban area has a stable and homogeneous population representative of Australia for socioeconomic status. A total of 3654 residents of predominately European ancestry (99%) aged 49 to 97 years participated in the baseline survey in 1992–1994. A second follow-up was performed in 1997–1999, a third follow-up was performed in 2002–2004, and a fourth follow-up was performed in 2007–2009. At each visit, a comprehensive questionnaire was administered in addition to a detailed ophthalmic examination. The BMES used a protocol for phenotyping similar to that used by Solouki et al10 and Hysi et al.9 Participants underwent a comprehensive eye examination including best-corrected visual acuity, objective and subjective refraction, slit-lamp examination, and an extensive lifestyle questionnaire together with height and weight measures.11,12 In addition, axial length, anterior chamber depth, and keratometry readings were obtained at the third follow-up of the BMES.11,12 Visual acuity of each eye was measured with current glasses, if worn, and with pinhole, using a logarithm of the minimum angle of resolution chart.13 A Humphrey autorefractor (Model 530; Humphrey Instruments Inc., San Leandro, CA) was used to obtain an objective refraction. Subjective refraction was then
Data Quality Control Numerous quality-control filters were applied before association testing for both SNP and sample filtering. Three filters were applied to the SNPs: (1) SNPs with a genotyping rate of ⬍0.95, (2) Hardy– Weinberg P value ⬍1⫻10 – 6, and (3) minor allele frequency ⬍0.001. Four filters were applied to the individuals: (1) genotype call frequency ⬍0.99, (2) discrepancy between clinical and genotypic gender, (3) evidence of cryptic relatedness based on a pair sharing 0 allele identity by descent at ⬍0.95 of their genome, and (4) evidence of non–European ancestry based on principle component analysis and mixture modeling for outlier detection.
Statistical Analysis Further filtering of individuals was undertaken to exclude those who had undergone cataract surgery, had severe visual impairment, had any known ocular pathologies that may affect refraction measures (e.g., macular degeneration and nuclear cataracts), or had evidence of anisometropia ⬎2 D between eyes. In addition, individuals identified as having refraction changes of ⬎0.5 D between the second and third follow-up studies were excluded because this may indicate the advent of subclinical nuclear sclerosis or hyperopic shift. Spherical equivalent and ocular biometric measures were considered to be continuous traits, and linear regression was performed using PLINK (http://pngu.mgh.harvard.edu/⬃purcell/ plink/, accessed April 30, 2012) with SNP allele dosage specified as the predictor variable and age and sex included as covariates. A P value of 0.05/5⫽0.01 for the 15q25 region and 0.05/14⫽0.004 for the 15q14 region were considered statistically significant for refractive error after Bonferroni correction. Power calculations were performed using QUANTO version 1.2 (http://hydra.usc. edu/gxe, accessed April 30, 2012).
293
Ophthalmology Volume 120, Number 2, February 2013 Table 1. Association Study Results for Refractive Error in the Blue Mountains Eye Study Cohort at the 15q25 Locus Blue Mountains Eye Study SNP
Minor Allele
Allele Frequency
Beta
SEM
P
rs939661 rs939658 rs17175798* rs8027411† rs8033963
A G C T T
0.49 0.49 0.49 0.49 0.49
⫺0.05 ⫺0.04 ⫺0.03 ⫺0.03 ⫺0.03
0.10 0.10 0.10 0.10 0.10
0.6403 0.7239 0.7951 0.7391 0.7391
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism. *Imputed SNP. SNP with best P value from the study by Hysi et al.9
†
Results Clinical Information All tested SNPs at the 15q14 and 15q25 loci remained in the analysis after application of the SNP filtering criteria. A total of 1571 individuals remained in our final analysis after application of the sample filtering criteria. In the trimmed cohort, 43.0% were female and 57.0% were male. The average age of participants was 73.84⫾7.81 years, and the average spherical equivalent measure was ⫹0.59⫾2.17 D (range, ⫺13.56 to ⫹8.56 D). The average axial length was 23.44⫾1.0 mm (range, 28.45–19.94 mm).
Replication Assessment of Loci in the Blue Mountains Eye Study Cohort We next sought to determine whether the association detected at the 15q14 and 15q25 loci reported by Solouki et al10 and Hysi et al9 could be replicated in the BMES. Results from the BMES are shown in Tables 1 and 2. For the 15q25 locus reported by Hysi et al,9 there were no statistically significant associations observed for any SNP. For the 15q14 locus reported by Solouki et al,10 the lowest P value in the BMES was 6.4⫻10⫺4 at SNP rs560766, whereas SNP rs688220,
the strongest SNP in the study Solouki et al,10 showed a P value of 2.3⫻10⫺3 in the BMES. The effect (beta) was in the same direction as previously described, with SNPs explaining between ⫺0.24 and ⫺0.33 D of the effect at this locus. Considering the replication of the 15q14 region for refractive error in the BMES, we then assessed whether any ocular biometric traits also may show association at these SNPs because the phenotypes of refractive error and ocular biometric components, particularly axial length, have been shown to be highly correlated.14 Table 3 indicates that 3 SNPs (rs560766, rs634990, and rs8032019) at the 15q14 have a statistically significant association with axial length (P⫽0.0054, P⫽0.0086, and P⫽0.0081, respectively). No associations with SNPs for the other ocular biometric traits of corneal curvature and anterior chamber depth were evident at this locus.
Discussion The data presented confirm an association of refractive error with the 15q14 locus initially reported by both Solouki et al,10 with SNPs rs524952 and rs634990 showing the strongest association. In the case of the 15q25 locus reported by Hysi et al,9 we were not able to demonstrate association for any SNPs at
Table 2. Association Study Results for Refractive Error in the Blue Mountains Eye Study 15q14 Locus Blue Mountains Eye Study SNP
Minor Allele
Allele Frequency
Beta
SEM
P
rs560766 rs4924134* rs11073058* rs11073059* rs11073060* rs7163001* rs524952* rs634990* rs7176510* rs688220*† rs619788* rs580839 rs8032019* rs685352*
A G T A A A A C T A A A G G
0.44 0.44 0.44 0.44 0.44 0.44 0.48 0.49 0.44 0.44 0.44 0.44 0.39 0.44
⫺0.33 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.30 ⫺0.29 ⫺0.30 ⫺0.29 ⫺0.29 ⫺0.29 ⫺0.28 ⫺0.24
0.0960 0.0957 0.0956 0.0956 0.0956 0.0956 0.0959 0.0960 0.0955 0.0960 0.0958 0.0958 0.0953 0.0973
0.0006445 0.002189 0.002635 0.002635 0.002635 0.002635 0.002075 0.002558 0.002022 0.002299 0.002275 0.002275 0.003276 0.01552
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism. *Imputed SNPs. † SNP with best P value from the study by Solouki et al.10
294
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES Table 3. Association Study Results for Axial Length, Corneal Curvature, and Anterior Chamber Depth in the Blue Mountains Eye Study at the 15q14 Locus Axial Length
Corneal Curvature
Anterior Chamber Depth
SNP
Minor Allele
Allele Frequency
Beta
SEM
P
Beta
SEM
P
Beta
SEM
P
rs560766 rs4924134 rs11073058 rs11073059 rs11073060 rs7163001 rs524952 rs634990 rs7176510 rs688220 rs619788 rs580839 rs8032019 rs685352
A G T A A A A C T A A A G G
0.44 0.44 0.44 0.44 0.44 0.44 0.48 0.49 0.44 0.44 0.44 0.44 0.39 0.44
0.13 0.12 0.12 0.12 0.12 0.12 0.11 0.12 0.12 0.11 0.12 0.12 0.12 0.10
0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05
0.0054 0.0143 0.0139 0.0139 0.0139 0.0139 0.0146 0.0086 0.0122 0.0158 0.0134 0.0134 0.0081 0.0372
⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.02 0.02 ⫺0.01 ⫺0.03 ⫺0.02 ⫺0.02 ⫺0.02 ⫺0.01 ⫺0.01
0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07 0.07
0.7754 0.8258 0.7494 0.7494 0.7494 0.7494 0.7893 0.8839 0.6868 0.7397 0.7286 0.7286 0.8950 0.8923
0.003 0.003 0.004 0.004 0.004 0.004 0.006 0.009 0.003 0.003 0.002 0.002 0.020 0.017
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
0.8722 0.8409 0.7888 0.7888 0.7888 0.7888 0.7161 0.5624 0.8475 0.8699 0.8917 0.8917 0.1999 0.2900
SEM ⫽ standard error of the mean; SNP ⫽ single nucleotide polymorphism.
this locus using the BMES cohort. In addition, we were able to show that the ocular biometric trait of axial length seems to be a contributor to the effect seen at the 15q14 locus, implying that SNPs identified for refractive error may harbor associations for other phenotypic traits that underlie myopia. The BMES cohort is a good choice to undertake replication because it consists of ethnically similar individuals (Caucasians of European descent) with a phenotype value (mean spherical equivalent ⫹0.6 D) similar to that of the studies by both Solouki et al10 and Hysi et al.9 In addition, there is no evidence of gender bias in the BMES cohort because 57.3% were female, whereas 54.1% to 92.6% were female in the previous studies. The BMES is comparable to the studies by Solouki et al10 and Hysi et al9 in terms of both phenotyping protocols and allele frequencies. Although replication was confirmed in the BMES cohort for the 15q14 region, our analysis did not replicate that of the 15q25 locus by Hysi et al.9 The beta effect was in the same direction and overlapped in magnitude (range, ⫺0.03 to ⫺0.05 D) compared with the majority of the discovery and replication studies (range, ⫺0.02 to ⫺0.29 D) in the original analysis by Hysi et al9; thus, a discussion of why there was no replication in the BMES is warranted. One possibility is that myopia represents a complicated trait composed of different endophenotypes, such as axial length and corneal curvature. Thus, when myopia is considered as a single phenotype using spherical equivalent measures, true associations may be masked. Our finding of axial length as an important component at the 15q14 locus underscores the importance of considering other endophenotypes that might underlie refractive error and myopia. Indeed, we previously showed through a heritability analysis that a number of genes are likely shared between refractive error and axial length.14 This suggests that the underlying genetic influences for refractive error may be related to axial length, and thus there may be overlapping genes. This does not mean that all refractive error loci will necessarily present with an ocular biometric component, because independent loci for ocular biometric components also are likely to exist as indi-
cated by the report of a quantitative trait locus for axial length on the long arm of chromosome 5 identified in twins of European ethnicity.15 A second possibility is that the current study is underpowered to detect an OR of 1.16 identified in the study by Hysi et al.9 A power calculation based on the beta (effect) value for BMES indicated that to replicate the finding at the 15q25 locus we would have approximately 29% power (2-sided t test), whereas to obtain 80% power would require a study size of approximately 5000 individuals. Thus, the BMES is most likely underpowered to detect association at this locus on its own. In the study by Hysi et al,9 no association was evident in the Rotterdam 2 (P⫽0.08), the 1958 British Birth Cohort (P⫽0.04 – 0.16), and the Australian Twin (P⫽0.4 – 0.9) cohorts. These cohorts all presented with a sample size of 1000 to 2157 individuals and would fit with the predicted nonsignificant P value findings reported in the current study. However, the Rotterdam 3 study in the study by Hysi et al9 consisted of 2082 individuals but did result in a significant P value (0.01– 0.02). Thus, a small sample size is not necessarily an accurate predictor of nonsignificant association. The Forest plot in the study by Hysi et al shows that the CIs do not overlap across many studies, indicating a high degree of heterogeneity at this locus. A recent Japanese study16 of a cohort with high myopia undertook replication of 4 SNPs (rs634990, rs524952, rs8027411, and rs17175798) from the 15q14 and 15q25 loci and reported findings similar to the data presented in the current study. In that study, a total of 1125 unrelated Japanese patients with high myopia (⬎26.1 mm in both eyes) and 2 independent control groups consisting of 366 patients with cataract (axial length ⬍25.0 mm in both eyes) and 929 healthy Japanese were evaluated. An association with high myopia was observed at the 15q14 locus for both the cataract control group (P⫽0.0035 for rs634990 and P⫽0.0017 for rs524952) and the healthy control group (P⫽1.91⫻10⫺6 for rs634990 and P⫽8.78⫻10⫺7 for rs524952). However, no statistically significant association was evident for SNPs at the 15q25
295
Ophthalmology Volume 120, Number 2, February 2013 region in either of the control groups. These findings and the current study indicate that the 15q14 association would seem to be associated across all levels of refractive error, as well as being present in different ethnicities. In contrast, the 15q25 locus has a weaker, less robust association. Our results suggest that the association at the 15q25 locus is weak. However, our data indicate that the 15q14 locus represents a strong and robust locus for refractive error and axial length. We have shown that rather than SNP rs688220 at 15q14 being the most highly associated for refractive error as previously reported by Solouki et al,10 the neighboring SNPs, rs524952 and rs634990, are more strongly associated with refractive error. The most associated SNPs for axial length are rs560766, rs634990, and rs8032019, with only SNP rs634990 overlapping between refractive error and axial length. In conclusion, there exists a genetic driver for myopia at the 15q14 locus, but we were unable to provide evidence to support a locus for refractive error at 15q25. The association at the 15q14 locus seems robust; thus, the next step is to search for candidate genes in the region. To date, there are no known refractive error or axial length genes in this region, so the exact nature of the genetic drivers in the region and how they influence refractive error remain to be elucidated. This may come through further genetic analysis, such as targeted sequencing across the region to identify causal variants, or through functional studies of known genes to identify the biological mechanism behind these findings.
References 1. Saw SM, Chua WH, Hong CY, et al. Nearwork in early-onset myopia. Invest Ophthalmol Vis Sci 2002;43:332–9. 2. Morgan I, Rose K. How genetic is school myopia? Prog Retin Eye Res 2005;24:1–38. 3. Rose KA, Morgan IG, Smith W, et al. Myopia, lifestyle, and schooling in students of Chinese ethnicity in Singapore and Sydney. Arch Ophthalmol 2008;126:527–30.
4. Saw SM, Tan SB, Fung D, et al. IQ and the association with myopia in children. Invest Ophthalmol Vis Sci 2004;45:2943–8. 5. Baird PN, Schache M, Dirani M. The GEnes in Myopia (GEM) study in understanding the aetiology of refractive errors. Prog Ret Eye Res 2010;29:520 – 42. 6. Li Z, Qu J, Xu X, et al. A genome-wide association study reveals association between common variants in an intergenic region of 4q25 and high-grade myopia in the Chinese Han population. Hum Mol Genet 2011;20:2861– 8. 7. Nakanishi H, Yamada R, Gotoh N, et al. A genome-wide association analysis identified a novel susceptible locus for pathological myopia at 11q24.1. PLoS Genet 2009;5:e1000 660. 8. Shi Y, Qu J, Zhang D, et al. Genetic variants at 13q12.12 are associated with high myopia in the Han Chinese population. Am J Hum Genet 2011;88:805–13. 9. Hysi PG, Young TL, Mackey DA, et al. A genome-wide association study for myopia and refractive error identifies a susceptibility locus at 15q25. Nat Genet 2010;42: 902–5. 10. Solouki AM, Verhoeven VJ, van Duijn CM, et al. A genomewide association study identifies a susceptibility locus for refractive errors and myopia at 15q14. Nat Genet 2010;42: 897–901. 11. Foran S, Rose K, Wang JJ, et al. Correctable visual impairment in an older population: the Blue Mountains Eye Study. Am J Ophthalmol 2002;134:712–9. 12. Mitchell P, Smith W, Attebo K, et al. Prevalence of openangle glaucoma in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:1661–9. 13. Attebo K, Mitchell P, Smith W. Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology 1996;103:357– 64. 14. Dirani M, Shekar SN, Baird PN. Evidence of shared genes in refraction and axial length: the Genes in Myopia (GEM) twin study. Invest Ophthalmol Vis Sci 2008;49:4336 –9. 15. Zhu XS, Lin T, Stone RA, et al. Sex-differences in chick eye growth and experimental myopia. Exp Eye Res 1995;61: 173–9. 16. Hayashi H, Yamashiro K, Nakanishi H, et al. Association of 15q14 and 15q25 with high myopia in Japanese. Invest Ophthalmol Vis Sci 2011;52:4853– 8.
Footnotes and Financial Disclosures Originally received: March 4, 2012. Final revision: August 2, 2012. Accepted: August 3, 2012. Available online: November 3, 2012.
7
Centre for Molecular Epidemiology, National University of Singapore, Singapore.
8
Manuscript no. 2012-314.
1
Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, Melbourne, Australia.
2
Centre for Vision Research, Department of Ophthalmology and Westmead Millennium Institute, University of Sydney, NSW Australia.
School of Medicine and Public Health, University of Newcastle, Newcastle, Australia.
9
Hunter Medical Research Institute, Newcastle, Australia.
10
The Centre for Information Based Medicine and the School of Biomedical Sciences and Pharmacy, University of Newcastle, Newcastle, Australia. 11
3
National Institute for Health Research Biomedical Research Centre for Ophthalmology, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, United Kingdom.
4
Singapore Eye Research Institute, Singapore National Eye Centre, National University of Singapore, Singapore.
5
Department of Epidemiology and Public Health, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
6
Department of Ophthalmology, School of Medicine, Aristotle University of Thessaloniki, AHEPA Hospital, Thessaloniki, Greece.
296
The Division of Genetics, Hunter Area Pathology Service, John Hunter Hospital, Newcastle, Australia. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Funding: National Health and Medical Research Council, Canberra, Australia (Project Grants: 974159, 211069, 457349, 512423, 475604, and 529912); Centre for Clinical Research Excellence in Translational Clinical Research in Eye Diseases; National Health & Medical Research Council research fellowships to P.N.B. (No. 1028444) and J.J.W. (No. 358702, 632909); Wellcome Trust as part of the Wellcome Trust
Schache et al 䡠 Refractive Error and AL with 15q14 in the BMES Case Control Consortium 2 (No. 085475B08Z, 08547508Z, 076113); Medical Research Council (UK) (G0401527); Fight for Sight (London); and National Institute for Health Research grant for a Biomedical Research Centre for Ophthalmology. This study used genotyping data generated and funded by the Wellcome Trust Case-Control Consortium 2 using samples from the BMES.
A full list of contributing investigators to the generation of these data is available online in Appendix 1 (http://aaojournal.org). Correspondence: Maria Schache, BSc, PhD, Centre for Eye Research Australia, 32 Gisborne St., Melbourne, Australia 3002. E-mail: [email protected].
297