CPD Modules Available

Print this page

The Role of Genetics in School Myopia

2 CPD in Australia | 1G in New Zealand | 1 December 2018

By Dr. Seyhan Yazar   

By 2050, half of the world’s population may be myopic.1 Our ability to control or prevent this from occurring depends on how much of myopia is genetic, and whether we can identify those susceptible to its onset or rapid progression...

Various study methodologies have been used to determine the role of genetics in myopia. Earlier studies surveyed individuals as in traditional epidemiology and later, genetic epidemiology studies targeted groups of individuals with increasing unit size: sibling-pairs or twins, families, and populations. Furthermore, various study methods including segregation, linkage, and association analysis have been applied. While segregation and linkage analysis are preferred in studies of high myopia, association studies are commonly used in investigations of low-to-moderate myopia genetics due to ease in data collection. Although the familial studies in high myopia and pathogenic (syndromic) myopia have provided our fundamental knowledge on genetics of myopia, in this review we will focus on ‘juvenile-onset’ or ‘school myopia’, which often develops and progresses between the ages of six and 17 years.     

Evidence from Twin Studies

Classical twin studies compare the similarity between monozygotic (identical) and dizygotic (non-identical) twins for different traits, with the assumption that monozygotic twins share a common environment as well as genes. If monozygotic twins are significantly more similar than dizygotic twins, then this will suggest that genes play an important role in these traits. This model has been used to investigate the heritability of myopia, refraction, and biometric components of refraction including axial length, corneal curvature, corneal power, lens thickness, and lens power. These studies used data from large twin registries such as TwinsUK, the Finnish Twin Cohort Study and some twin cohort studies including Genes in Myopia, Twins Eye Study in Tasmania, Brisbane Adolescent Twin Study, and the Guangzhou Twins Eye Study.

In one of the earliest studies on myopia heritability, Karlsson identified that 94 per cent of the identical twins who participated had similar myopic refractive power compared to 29 per cent in non-identical twins.3 Verifying this work, larger studies estimated the heritability of refraction from 50 per cent to 91 per cent and myopia inheritance between 11 per cent to 98 per cent. In a systematic review of ocular biometry inheritance, heritability of axial length and corneal curvature were shown to range from 27 per cent to 94 per cent and 16 per cent to 95 per cent, respectively. Heredity accounted for 69 per cent variation in corneal power, 92 per cent in lens thickness, and 65 per cent in lens power.4

In a cohort of 64 monozygotic twin pairs, using the discordant identical twin model, where monozygotic twins with discordant refractive error were included as a matched genetic control, Ramessur et al.5 studied environmental effects and found that differences in life choices before the age of 25 years have a sustained effect. This finding suggests that epigenetic changes in monozygotic twins may alter ocular growth and homeostasis.

Evidence From Familial Aggregation Studies

In familial aggregation studies, it is assumed that occurrence of a given trait shared in genetically related members of a family cannot be readily accounted by chance. In these studies, heritability can be estimated from degree of resemblance between siblings, parent-child, second, and third-degree relatives.

Many family studies have indicated that there is a positive correlation between myopia in parents and children. For example, examining 2,888 children in China, Yap et al.6 showed that children with myopic parents were more likely to have myopia than their peers without myopic parents. Similarly, in the Sydney Myopia Study in Australia, examination of 2,353 children aged seven years revealed myopia rates increased dose dependently; a child’s risk of developing myopia was 7.6 per cent for no myopic parents, 14.9 per cent for one myopic parent, and 43.6 per cent for two myopic parents. In addition, severity of myopia correlated with the severity of myopia in either parent.7

Similar observations were made in older populations. The estimated heritability of refractive error in siblings over 70 years was 61 per cent in the Salisbury Eye Evaluation Study (SEES) in Maryland and on average, odds of having myopia was 2.72 times higher in siblings of myopic individuals than in siblings of nonmyopic participants.8 This was lower than the likelihood calculated in the Beaver Dam Eye Study (mean odd ratio of 3.42 ranging from 2.82 to 4.25).9 Both in the Framingham Offspring Eye Study (FOES) and SEES, recurrence of myopia was related to the age difference of siblings. The chance of having myopia ranged from 2.50 for siblings whose ages differed by >10 years to 5.13 for siblings whose age differences were within two years in FOES.10 A strong familial aggregation of myopia among siblings (risk ratio [RR] ranging from 2.09 to 3.86) and parent–offspring pairs (RR from 1.82 to 3.81) was also identified in a cohort of 1,259 nuclear families with the average size of 3.6 in Tehran.11

Evidence From Genome-wide Association Studies

Genome-wide association (GWA) studies have immensely improved our understanding of common complex conditions including school myopia. The principles and methodologies underlying GWA studies are briefly introduced in breakout box one. Due to the difficulty in standardisation of a myopia definition across multiple populations, and the advantage of increased statistical power with the use of continuous data, refractive error is the frequently preferred outcome measure in GWA studies. In this section, after summarising the findings of GWA studies on refractive error, we will review the GWA studies of refractive-related ocular biometry.

GWA Studies of Refractive Error

The first two genome-wide studies of refractive error were led by British and Dutch researchers simultaneously.12,13 Both studies identified single nucleotide polymorphisms (SNPs) located on chromosome 15 using a dataset of approximately 15,000 individuals. While the discovery cohorts were unique to each study, the controls were exchanged. Study results from the Dutch cohort found genetic variants near RASGRF1 (on the region 14 of the long arm of chromosome 15 [15q14]) gene, whereas the British study – TwinsUK – reported SNPs near GJD2 and ACTC1 genes (on the region 25 of the short arm of chromosome 15[15q25]). Following these separate attempts, because the statistical power to detect associations between SNPs and a trait highly depends on size sample, the British and Dutch groups, along with researchers across the globe including Australia, established the Consortium of Refractive Error and Myopia (CREAM) project. The first collaborative work from this effort comprised 31 cohorts representing four different continents with 55,177 individuals; 42,845 Caucasians and 12,332 Asians.14 In this study, it was confirmed that genetic variants on chromosome 15q14 influence susceptibility for myopia in Caucasian and Asian populations worldwide. A total of 16 novel genetic variants were associated with refractive error in Europeans, of which eight were shared with Asians. Additionally, a further eight new associations were discovered in the combined analysis. While the previous association of RASGRF1 and GDJ2 were confirmed, the new candidate genes were implicated in pathways of neurotransmission (GRIA4), ion transport (KCNQ5), retinoic acid metabolism (RDH5), extracellular matrix remodelling (LAMA2 and BMP2) and eye development (SIX6 and PRSS56). At that stage, the cumulative predictive value of these variants in determining whether one has myopia vs. hyperopia was 67 per cent.

In cohort studies, participants often go through extensive examination protocols and, when the aim is to analyse data from geographically separated studies, participating groups try to standardise the protocols as performed in CREAM. Taking an unusual approach at the time, 23andMe, the well-known direct-to-consumer genotyping company, asked its participants whether they had been diagnosed with myopia, and if so, at what age. In this way, by collecting data from over 45,000 people of European ancestry, Kiefer et al15 have conducted the largest GWAS of the time and reported association of 22 genetic variants with refractive error; 16 of which were novel. Although the CREAM and 23andMe studies had different analysis designs (the outcome variable was the degree of refractive error in dioptres for the CREAM study and the age of myopia onset, as measured by hazard ratios for the 23andme study), both GWA studies independently identified the same 25 genetic variants across the genome. Thirteen of the variants reached the genome-wide significant level. A study of approximately six million SNPs in 40,036 Europeans replicated 12 previously implicated genetic regions and revealed four newly reported genetic variants associated with spherical equivalent. These four SNPs were located in or near FAM150B-ACP1, LINC00340, FBN1, and DIS3L-MAP2K1 genes. Variants were found in three genes – AREG, GABRR1, and PDE10A – in an Asian cohort of 10,315 individuals.16

In a genome-wide meta-analysis of myopia as a dichotomous outcome, only one of the myopia-related genetic regions (chr 8q12) was replicated from Kiefer et al.’s study.17 Additionally, using age of examination and years of education as a proxy for age onset, the same study showed a replication of 10 additional genetic regions associated with myopia from the Kiefer et al. study. Notably, the 15q14 region was associated with hyperopia, which represents the other end of the refractive error continuum. However, the direction of SNP effects was exactly the opposite of what has been identified in myopia studies, suggesting that these regions are involved in refractive error variability and have an impact on development of both myopia and hyperopia. Of SNPs in the 51 genes reported by CREAM and Kiefer et al., 15 were replicated in a cohort of 3,792 Japanese individuals.18

Most recently, CREAM and 23andMe combined their efforts and conducted a discovery meta-analysis involving 160,420 participants and replication in 95,505 participants from the UKBiobank.19 This study increased the total number of known associated SNPs from 37 to 161. More than 78 per cent of these associations were present both in European and Asian cohorts. Moreover, these 161 genetic variants together explain 7.8 per cent of the variation in refractive error that we observe in epidemiological studies. One aim of the genetic studies is to identify those who are susceptible to disease. To achieve this, a ‘polygenic risk score’ (PGRS) of an individual is calculated by summing the weighted effect sizes of associated genetic variants. A PGRS estimation, constructed using genetic information from the Rotterdam Study III, had a 77 per cent power to accurately identify whether someone is myopic or hyperopic. Although this estimation is 10 per cent better than the previous attempts, its usefulness is still very limited.

GWA Studies of Axial Length and Corneal Curvature

Axial length and corneal curvature are the key determinants of refractive error. The first GWA study of axial length involved 1,118 cases and 5,433 controls from Chinese and Malay populations. This study identified the association between SNP rs4373767 in ZC3H11B and axial length.20 A larger GWA study comprising a total of 12,531 Europeans and 8,216 Asians from the CREAM replicated this association and reported an additional eight genetic variants associated with axial length.21 Of these, SNPs near GJD2 and CD55 were identified in the GWA study of refractive error, as described above. This overlap of signals between studies of axial length and refractive error suggests a possible shared biological mechanism. The first published GWA study for corneal curvature was from a Singaporean Asian population, in which the significant associations of loci in FRAP1 and PDGFRA genes were reported.22 The association between corneal curvature and the PDGFRA gene has been validated in two separate GWA studies of Europeans23,24 and one study of Asians which identified two novel SNPs in CMPK1 and RBP3.25 PDGFRA is important for regulating cellular growth factors and hence thought to influence the eyeball size while maintaining its scaling. In addition to the role of this gene in corneal curvature, PDGFRA has also been implicated in corneal astigmatism in the Singaporean Asian population. In a Japanese study, an SNP in the WNT7B gene, found to be associated with axial length and corneal curvature, was downregulated in the cornea and upregulated in the retina, suggesting a role in myopia development.27

Gene-Environment Interactions

In addition to the evidence of a genetic contribution to risk, epidemiological studies have identified environmental riskfactors for the development of myopia. These are reviewed in the feature written by Dr. Samantha Lee and Mr Gareth Lingham in this issue of mivision. Although studying the genetic and environmental influences separately is useful to understand the independent effects of these factors in myopia development and progression, gene-environment interaction studies are necessary to understand how environmental influences affect individuals differently, depending on their genetic composition. These types of studies are now of interest to myopia researchers. For example, Fan et al22 showed that more educated participants have greater genetic predisposition to myopia compared to participants with lower education levels in Asian cohorts. However, no interactions were identified in the European cohort. There could be many reasons for this difference as suggested by the paper’s authors:

  1. Slight differences in genetic architecture between the populations,
  2. Ideally, near work intensity before the onset of myopia is the determinant of disease development, however cumulative lifetime education is used as a surrogate to this measure, and
  3. There are differences in near work intensity in education systems in various populations.16 Furthermore, a causal role of educational attainment on refractive error was shown in a Mendelian randomisation study.28 In another study, CREAM showed that 39 genetic variants associated with refractive error exert their effects at different periods of childhood. However, there was no strong evidence showing that effects of these variants, individually or combined, are altered with time outdoors or near work.29 Increased association of PGRS with age was replicated in a larger cohort. Genes with the strongest effects (GJD2 and LAMA2) were found to play a role early with increasing age.30

New genetic studies will continue to help us unravel the genetic architecture of school myopia and develop better risk scores for those who have a greater predisposition to myopia. Ultimately, we will be able to provide interventions, such as spending time outdoors or atropine treatment in a dose-response manner, based on an individual’s genetic susceptibility level.  

About The Genome-Wide Association Study (GWAS)

What is Genetic Variation?

Genetic variation is a term that describes the subtle differences in the DNA sequence in our genome. It is what makes each of us unique; not only with our physical characteristics such as hair or eye colour, but also our susceptibility to disease. A single nucleotide polymorphism (SNP) occurs when the DNA is altered at a single position among individuals. It is the most common genetic variation in the human genome.

What Is A Genome-wide Association Study?

A genome-wide association study (GWAS) is an examination of large numbers of SNPs from different individuals to identify any association with a disease or a trait. Identification of these associations allows us to develop better strategies to detect, treat, and prevent diseases, and these studies are particularly useful in finding genetic variations that contribute to common, complex diseases such as myopia.

How Is A GWAS Conducted?

GWA studies compare the DNA of participants with varying phenotypes for a particular trait or disease. The two groups can be individuals with a disease being studied (cases) and individuals from the same population without the disease (controls) or they can be individuals with different phenotypes for a particular trait, for example refractive error.

First, each individual’s complete DNA is genotyped (sequenced). Genotyping can be done by methods, which vary in experimental time and cost. A DNA microarray (also known as a DNA chip) is a tool that searches the known variants (alleles) one at a time. After millions of SNPs are read through a DNA microarray for each individual, these are aligned to each other and surveyed for the most common variation. If a certain SNP is found to be significantly more frequent in individuals with the disease, then this SNP is said to be ‘associated’ with the disease. The associated SNPs are considered to point to a region of the genome involved in the risk of disease. These are usually depicted in Manhattan plots. Sometimes, an identified association may not directly contribute to a disease cause but just be ‘tagging along’ with the true causal variants, which can be determined by sequencing DNA base pairs in that particular region of the genome. Success of a GWAS in identifying new associations depends on a number of factors:

  1. How many genetic regions affecting the trait segregate (inherited) in the population,
  2. The distribution of effect size and allele frequency of each variant within the region,
  3. The sample size of the study,
  4. The panel of genome-wide SNPs used in the DNA microarray, and
  5. How much variation is observed in the trait or diseases of interest. (Visscher 2017) 

An example of a Manhattan plot. Each dot represents a single nucleotide polymorphism (SNP). SNPs are plotted on the x-axis based on their chromosomal position. The Y-axis denotes the negative logarithm of the associated p-values for each SNP. SNPs that achieve genome-wide significance (p<5x10-8) are shown above the red line. SNPs above the blue line represent the suggestive genetic variants that are associated with disease or phenotype.



Key Definitions

Association analysis: a method to determine whether a genetic variant is associated with a disease or a phenotypic trait.

Chromosome: the microscopic thread-like molecules within the cell that carry hereditary information in the form of genes. Each chromosome has two sections (arms) based on the location of narrowing called the centromere. The shorter arm is called p and the longer arm is called q.

Genetic epidemiology: the study of the role of the genetic factors in determining health and disease in a defined population. It differs from traditional epidemiology by its explicit focus on genetic factors, and it differs from medical genetics by its emphasis on population-based studies.

Genetic variant: an alteration in the DNA nucleotide sequence within a population.

Genome-wide meta-analysis: a method of combining GWA studies across distinct cohorts. Often the pooled cohorts are separated into a discovery and a replication cohort. First, associated genetic variants are identified in discovery cohorts then confirmed if they are present in the replication cohort.

Genome-wide significant level: a commonly accepted statistical significance threshold to be able to differentiate a true association from a false one in GWA studies.

Genotyping: the technology that detects the genetic variants an individual possesses.

Heritability: a measure of how well variation in people's genes account for phenotypic differences.

Linkage analysis: a method that is used in establishing linkage between the genes. Linkage is the tendency for genetic variants or genes to be inherited together because of their proximity to each other on the same chromosome.

Mendelian randomisation study: a method of using measured variation in genes of known function to examine the causal effect of a modifiable exposure on disease or phenotypic trait.

Phenotype: all the observable characteristics of an organism that result from the interactions of its genes with the environment.

Polygenic risk score: an estimation of genetic susceptibility to a disease or a phenotypic trait that is calculated by using GWAS results.

Segregation analysis: a technique within genetic epidemiology to determine whether there is evidence that a major gene forms the foundation of a phenotype distribution.

Single nucleotide polymorphism (SNP): the variation in a genetic sequence that affects one of the four bases - adenine (A), guanine (G), thymine (T), or cytosine (C) –  in a segment of a DNA molecule. It has to occur in more than 1 per cent of a population.

Traditional epidemiology: the study and analysis of distribution and determinants of health-related states or events (including diseases) in a population. In traditional epidemiology, researchers are interested in measuring or assessing the relationship of exposure with a disease or an outcome.



Dr. Seyhan Yazar is a Postdoctoral Research Fellow within the Centre for Ophthalmology and Visual Science at the University of Western Australia and the Lions Eye Institute. She received a Bachelor of Medical Science from the University of New South Wales, a Masters in Orthoptics from the University of Sydney and a PhD in Ophthalmic Genetics and Epidemiology from the University of Western Australia.

In her thesis, Dr. Yazar investigated the genetic and environmental influences on ocular disease development and progression with a particular interest in myopia, corneal dystrophies and glaucoma through exploring datasets from large population-based studies including the Western Australian Pregnancy Cohort (Raine) Study.

Dr. Yazar was awarded an NHMRC CJ Martin Overseas Biomedical Fellowship to undertake further research training in the UK. Her current research interests include computer programming, statistics, engineering sequence analysis pipelines and performing well-validated and reproducible computational eye research


1. Holden BA, Fricke TR, Wilson DA, et al. Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050. Ophthalmology 2016;123:1036–1042.
2. Chen Y, Wang W, Han X, et al. What Twin Studies Have Taught Us About Myopia. Asia Pac J Ophthalmol (Phila) 2016;5:411–414.
3. Karlsson JI. Concordance rates for myopia in twins. Clin Genet 1974;6:142–146.
4. Sanfilippo PG, Hewitt AW, Hammond CJ, Mackey DA. The heritability of ocular traits. Survey of Ophthalmology 2010;55:561–583.
5. Ramessur R, Williams KM, Hammond CJ. Risk factors for myopia in a discordant monozygotic twin study. Ophthalmic Physiol Opt 2015;35:643–651.
6. Yap M, Wu M, Liu ZM, et al. Role of heredity in the genesis of myopia. Ophthalmic Physiol Opt 1993;13:316–319.
7. Ip JM, Huynh SC, Robaei D, et al. Ethnic Differences in the Impact of Parental Myopia: Findings from a Population-Based Study of 12-Year-Old Australian Children. Invest Ophthalmol Vis Sci 2007;48:2520–9.
8. Wojciechowski R, Congdon N, Bowie H, et al. Heritability of Refractive Error and Familial Aggregation of Myopia in an Elderly American Population. Invest Ophthalmol Vis Sci 2005;46:1588–5.
9. Lee KE, Klein BE, Klein R, Fine JP. Aggregation of refractive error and 5-year changes in refractive error among families in the Beaver Dam Eye Study. Arch Ophthalmol 2001;119:1679–1685.
10. Framingham Offspring Eye Study Group. Familial aggregation and prevalence of myopia in the Framingham Offspring Eye Study. Arch Ophthalmol 1996;114:326–332.
11. Fotouhi A, Etemadi A, Hashemi H, et al. Familial aggregation of myopia in the Tehran eye study: estimation of the sibling and parent offspring recurrence risk ratios. Br J Ophthalmol 2007;91:1440–1444.
12. 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;42:902–905.
13. Solouki AM, Verhoeven VJM, van Duijn CM, et al. A genome-wide association study identifies a susceptibility locus for refractive errors and myopia at 15q14. Nat Genet. 2010;42:897–901.
14. Verhoeven VJM, Hysi PG, Wojciechowski R, et al. Genome-wide meta-analyses of multiancestry cohorts identify multiple new susceptibility loci for refractive error and myopia. Nat Genet. 2013;45:314–318.
15. Kiefer AK, Tung JY, Do CB, et al. Genome-wide analysis points to roles for extracellular matrix remodeling, the visual cycle, and neuronal development in myopia. Gibson G, ed. PLoS Genet 2013;9:e1003299.
16. Fan Q, Verhoeven VJM, Wojciechowski R, et al. Meta-analysis of gene&ndash;environment-wide association scans accounting for education level identifies additional loci for refractive error. Nature Comm 2016;7:1–12.
17. Simpson CL, Wojciechowski R, Oexle K, et al. Genome-wide meta-analysis of myopia and hyperopia provides evidence for replication of 11 Loci. Miao X, ed. PLoS ONE 2014;9:e107110.
18. Yoshikawa M, Yamashiro K, Miyake M, et al. Comprehensive replication of the relationship between myopia-related genes and refractive errors in a large Japanese cohort. Invest Ophthalmol Vis Sci 2014;55:7343–7354.
19. Tedja MS, Wojciechowski R, Hysi PG, et al. Genome-wide association meta-analysis highlights light-induced signaling as a driver for refractive error. Nat Genet 2018:1–20.
20. Fan Q, Barathi VA, Cheng C-Y, et al. Genetic Variants on Chromosome 1q41 Influence Ocular Axial Length and High Myopia Wiggs JL, ed. PLoS Genet 2012;8:e1002753–14.
21. Cheng C-Y, Schäche M, Ikram MK, et al. Nine loci for ocular axial length identified through genome-wide association studies, including shared loci with refractive error. Am J Hum Genet 2013;93:264–277.
22. Fan Q, Zhou X, Khor CC, et al. Genome-Wide Meta-Analysis of Five Asian Cohorts Identifies PDGFRA as a Susceptibility Locus for Corneal Astigmatism Wiggs J, ed. PLoS Genet 2011;7:e1002402.
23. Mishra A, Yazar S, Hewitt AW, et al. Genetic variants near PDGFRA are associated with corneal curvature in Australians. Invest Ophthalmol Vis Sci 2012;53:7131–7136.
24. Guggenheim JA, McMahon G, Kemp JP, et al. A genome-wide association study for corneal curvature identifies the platelet-derived growth factor receptor α gene as a quantitative trait locus for eye size in white Europeans. Mol Vis 2013;19:243–253.
25. Chen P, Miyake M, Fan Q, et al. CMPK1 and RBP3 are associated with corneal curvature in Asian populations. Hum Mol Genet 2014;23:6129–6136.
26. Fan Q, Zhou X, Khor CC, et al. Genome-Wide Meta-Analysis of Five Asian Cohorts Identifies PDGFRA as a Susceptibility Locus for Corneal Astigmatism Wiggs J, ed. PLoS Genet 2011;7:e1002402–10.
27. Miyake M, Tabara Y, Suda K, et al. Identification of myopia-associated WNT7B polymorphisms provides insights into the mechanism underlying the development of myopia. Nat Comm 2015;6:1–9.
28. Cuellar-Partida G, Lu Y, Kho PF, et al. Assessing the Genetic Predisposition of Education on Myopia: A Mendelian Randomization Study. Genet Epidemiol 2015;40:66–72.
29. Fan Q, Guo X, Tideman JWL, et al. Childhood gene-environment interactions and age-dependent effects of genetic variants associated with refractive error and myopia: The CREAM Consortium. Sci Rep 2016;6:25853.
30. Tideman JWL, Fan Q, Polling JR, et al. When do myopia genes have their effect? Comparison of genetic risks between children and adults. Genet Epidemiol 2016;40:756–766.

' a child’s risk of developing myopia was 7.6 per cent for no myopic parents, 14.9 per cent for one myopic parent, and 43.6 per cent for two myopic parents '