SEARCH

SEARCH BY CITATION

Keywords:

  • age related macular degeneration;
  • epigenetics;
  • genetic variants;
  • gene-environment;
  • next generation sequencing

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

The last 10 years have seen an unprecedented explosion in our knowledge regarding the genomic basis of age-related macular degeneration. This has come about through major advances in computing power, microfabrication of large numbers of molecular markers on chips and improved statistical algorithms for analysis. In tandem, it has become clear that age-related macular degeneration appears to be a multifactorial disease with influences from genetic and structural variants, as well as epigenetic involvement. The combination of these factors with known environmental determinants indicates the highly complex nature of this disease, but at the same time also offers insights into risk prediction and disease stratification through genotype profiling.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Age-related macular degeneration (AMD) presents as a progressive neurodegenerative disease detected in one in seven (14%) Australians over the age of 50, but increasing to one in four by the age of 90.[1, 2] It accounts for nearly half of all legal blindness in elderly Australians (24 000 individuals) and is a leading cause of blindness in the developed world.[3]

AMD is a complex disease where contributions come from both genetic and environmental risk factors. A number of environmental influences have been suggested as influencing AMD. These include diet, where there is an inverse association with omega-3 fatty acid intake, antioxidants and white meat,[4] whereas there is a direct association with increased intake of trans-unsaturated fat, red meat, alcohol and high total serum cholesterol.[5, 6] Additionally, other environmental factors, including history of cataract surgery, hyperopia, a light iris colour, cardiovascular disease, hypertension, smoking, sunlight exposure, exposure to chronic Chlamydia pneumoniae infection and physiological changes associated with normal ageing, have also been found to be associated.[7-9] Of these factors, only smoking has been shown to be consistently associated with AMD across various studies.[9]

A genetic contribution to AMD is now well established and was first described through family and twin studies.[10-12] Until recently, detection of genetic involvement was limited to three main approaches. The first of these being the use of genetic linkage studies, with the assumption that a disease gene segregates in a Mendelian fashion in a family. In these cases, markers that co-segregate with the disease through families could be used to determine the approximate position of a disease gene on a chromosome. However, in the case of late-onset diseases, such as AMD, large families are typically not available, as many elderly relatives would have passed away and the younger generations may be presymptomatic. Thus, the linked regions identified via this approach would typically be large, extending over many megabases on the chromosome and contain potentially hundreds of genes. Identification of the disease-causative gene, thus, required substantiation through the use of a second approach, which was typically through Sanger DNA sequencing, referred to as first-generation sequencing. This was time-consuming and relatively expensive, but did lead to the identification of a number of putative linkage regions across the human genome through the use of genome-wide linkage analysis.[13] Two of the most promising linked regions that were replicated in several studies occurred on the long arms of chromosome 1 and chromosome 10.[13]

The second approach used in the early 2000s was based on searching for disease causing mutations in genes previously identified in other macular diseases. For example, the Arg345Trp mutation in the EFEMP1 gene had been described as causing Doyne honeycomb retinal dystrophy and Malattia Leventinese, whereas multiple mutations had been described in the RDS/peripherin gene that were associated with a range of macular dystrophies, including retinitis pigmentosa, adult-onset vitelliform macular dystrophy, pattern butterfly dystrophy and bulls-eye maculopathy. Although much effort was invested in the study of genetic alterations in many such candidate genes, only a few showed promising associations with AMD, such as the TIMP3 (mutated in Sorsby's fundus dystrophy)[14, 15] and the ABCA4 gene (identified in Stargardt disease), that appeared to be associated in a subset of AMD patients.[16]

A third approach involved the screening of variants in candidate genes in case control association studies where the gene was deemed to have a plausible biological function in AMD. There were some early successes, such as the apolipoprotein E gene (APOE)[17] on chromosome 19, which was subsequently confirmed in multiple independent studies[18] and also showed similar effect in meta-analysis.[19] However, many of the early targeted candidate genes presented as false positives and their associations could not be replicated in other cohorts.[20] These studies typically involved the limited analysis of a handful of genetic variants or single nucleotide polymorphisms (SNPs) in a gene, limited sample size, in terms of the number of cases and control samples available for the study, as well as differing methodological approaches including phenotype definition.

Genome-wide association studies (GWAS)

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

A fourth approach resulted from the coalescence of advances in computer applications, technological advancement in terms of miniaturization of devices and improved statistical algorithms. In 2005, following an extensive analysis of the chromosome 1 region that was first identified through linkage analysis, variants in the complement factor H (CFH) gene, a gene involved in the complement pathway, were identified as being significantly associated with AMD.[21-24] Three of these studies assessed a series of variants across the chromosome 1 linked region[22-24] to eventually identify the potential AMD-associated SNPs in the CFH gene. The fourth study represented the first successful attempt of a genome-wide association study (GWAS) in identifying common genetic variants in a gene associated with an ocular disease.[21] This GWAS assessed over 100 000 SNPs across the genome through their placement on a prefabricated SNP chip, thereby allowing screening in an unbiased manner. Initially, two SNPs were identified in the CFH gene as being significantly associated with AMD following Bonferroni correction for multiple testing (in this case, 0.05/103 611 = 4.8 × 10−7). Further screening of SNPs and Sanger sequencing across the CFH gene revealed several SNPs as being commonly associated with AMD; some of which also tended to be inherited, together defining a risk haplotype for AMD. The non-synonymous SNP, rs1061170 (also known as Y402H) in exon 9 of the CFH gene was identified as being the SNP most significantly associated with disease within this risk haplotype. In addition, this SNP also appeared to occur in a region containing a binding site for heparin and C-reactive protein (CRP), as well as for a number of pathogens, thus suggesting a possible functional role in AMD.[24] A reduction in binding of CFH to CRP at this site has been shown, further implicating its role in the immune system.[25]

Based on the success of identifying CFH with AMD through the use of linkage and association studies, further studies were conducted to identify candidate genes associated with the disease. These studies were modelled on either assessing other genes, aside the CFH gene, in the complement pathway, or through assessment of regions that had previously been genetically linked with AMD. These studies were incredibly successful at identifying a number of genes associated with AMD. These included complement factor B (BF) and the complement component 2 (C2) genes localized to the major histocompatibility complex class III region on chromosome 6,[26] the complement C3 gene on chromosome 19[27] and the CFH paralogous genes (CFHR3, CFHR1, CFHR4, CFHR2 and CFHR5) on chromosome 1,[28] which presented as partial duplications of the CFH gene.[28, 29] Another major genetic locus was also identified as occurring on the long arm of chromosome 10 comprising the LOC387715 (ARMS2)[30, 31] and HtrA serine peptidase 1 (HTRA1) genes.[32, 33] Although its functional role is still not entirely clear, it has been reported that the ARMS2 protein is present on the outer membrane of the mitochondria and may play a role in mitochondrial dysfunction.[34]

The initial success of GWAS in identifying CFH in AMD also spurred the use of this technique in identifying other genetic associations, both in AMD as well as in other complex diseases. As a result, over 1350 GWAS studies have been reported up to the middle of 2012 for a multitude of different diseases (http://www.genome.gov/gwastudies/). Of interest, the original AMD GWAS used only 96 AMD cases and 50 controls, with approximately 100 000 SNPs to demonstrate an association of the CFH. However, since 2005, the number of samples and the number of SNPs used in GWAS has steadily grown. The largest number of reported case and control samples is now in excess of over 200 000 individuals, for instance in the study of blood pressure and cardiovascular disease.[35] Likewise the number of SNPs available for association testing has now reached in excess of four million SNPs, with the ability to add a further 500 000 SNPs based on targeted genetic variants derived from the International HapMap and 1000 Genomes Projects. In AMD, several GWAS have been conducted with increasing number of patient samples with increasing SNP densities through technological advancement allowing for greater coverage of these SNPs on chips. These GWAS have identified additional significantly associated AMD genes, including complement factor I (CFI) on chromosome 4,[36] TIMP3, LIPC, CETP genes on chromosomes 22, 15 and 16, respectively,[14, 15, 37] the FRK/COL10A1 and VEGFA genes on chromosome 6[38] and the TNFRSF10A/LOC389641 gene locus on chromosome 8.[39]

Typically, larger sample sizes used in GWAS have the ability to not only confirm but also identify additional novel genetic associations. These associations may be uncovered due to the increase in study power compared with smaller studies, thereby allowing the detection of genetic associations with a smaller risk or protective effect. To extend our understanding of genetic associations in AMD, an international consortium (the AMD Gene Consortium) was established with the support of the National Eye Institute (NEI) in the USA. The consortium was established in early 2010 to identify additional genetic loci that might be associated with AMD. Eighteen AMD research groups from across the world coordinated their efforts to accelerate the discovery of susceptibility genes in AMD. This involved a meta-analysis of GWAS data that was undertaken on over 7600 cases with advanced disease (either geographic atrophy [GA], neovascularization or both) and almost 52 000 controls, representing the most comprehensive genetic association study of macular degeneration yet conducted. A total of 32 variants representing loci with promising evidence of association were then genotyped in an additional 9500 cases and 8200 controls. This culminated in the identification of 19 loci (Table 1) significantly associated with advanced AMD (P < 5 × 10−8), of which seven were novel and implicated the nearby genes: COL8A1-FILIP1L, IER3-DDR1, SLC16A8, TGFBR1, RAD51B, ADAMTS9-MIR548A2 and B3GALTL.[40] The other 12 loci corresponded to genes previously reported as AMD susceptibility genes, with the most significantly associated genes being those at the ARMS2/HTRA1 locus (odds ratio [OR] 2.76, P = 4 × 10−540) and the CFH genes (OR 2.43, P = 1 × 10−434). The 19 loci identified in this study showed significant enrichment for genes involved in a number of different biological pathways. These include the regulation of complement and coagulation cascades, lipid metabolism, atherosclerotic signalling, collagen and extracellular matrix remodelling, regulation of apoptosis, as well as angiogenesis. These findings indicate the complex nature of the genes, pathways and mechanisms that underlie AMD.

Table 1. Summary of loci reaching genome-wide significance identified by the International AMD Gene Consortium
SNPChromosomeNearest geneJoint meta P valueJoint odds ratioGene pathway or potential function
  1. Loci reaching at least P = 5 × 10−8 as described by the AMD Gene Consortium.[40] AMD, age-related macular degeneration; SNP, single nucleotide polymorphism.

Loci previously reported with genome-wide significance (P < 5 × 10−8) 
rs1049092410ARMS2/HTRA14 × 10−5402.76Mitochondrial/serine protease/cell growth
rs107376801CFH1 × 10−4342.43Complement system/immune-related
rs4296086C2/CFB4 × 10−891.74Complement system/immune-related
rs223019919C31 × 10−411.42Complement system/immune-related
rs574948222SYN3/TIMP32 × 10−261.31Synaptogenesis/neurotransmission/matrix metalloproteinase inhibitor
rs442063819APOE2 × 10−201.30Atherosclerosis signalling
rs186416316CETP7 × 10−161.22Atherosclerosis/longevity
rs9430806VEGFA9 × 10−161.15Inhibition of angiogenesis by TSP1
rs132780628TNFRSF10A/LOC3896413 × 10−151.15Cell death/apoptosis
rs92091515LIPC3 × 10−111.13Phospholipid degradation
rs46987754CFI7 × 10−111.14Complement system/immune-related
rs38121116FRK/COL10A12 × 10−81.10Atherosclerosis signalling/dendritic cell maturation
AMD loci reaching P < 5 × 10−8 for the first time 
rs130818553COL8A1/FILIP1L4 × 10−131.23Connective tissue/regulator of antiangiogenic activity on endothelial cells
rs31307836IER3/DDR12 × 10−111.16Cell death protection/dendritic cell maturation/cell communication
rs813566522SLC16A83 × 10−111.15Lactate membrane transfer
rs3343539TGFBR13 × 10−111.13Protein kinase
rs801730414RAD51B9 × 10−111.11DNA repair
rs67957353ADAMTS9MIR548A25 × 10−91.10Proteoglycan cleavage, control of organ shape, inhibition of angiogenesis
rs954223613B3GALTL2 × 10−81.10Glucose transporter

Previous twin studies have indicated that the heritability or the proportion of variance attributable to a genetic component in AMD ranges from 45% for early stage disease to up to 70% for advanced stage disease.[12, 41] The 19 loci described by the AMD Gene Consortium have been estimated to account for between 10% and 30% of the total variance of AMD, and up to 65% of the genetic contribution. Thus, these findings confirm previous heritability estimates and reflect the observation that the majority of the variance is explained through previously identified loci, while the newly described loci and secondary signals identified by the AMD Gene Consortium contribute an additional 2–3%.[40] Although these findings explain a great proportion of the variance of AMD and are far in excess of explanations of variance in most other complex diseases, it still leaves a proportion of variance that is unexplained that may come from either other genetic or environmental sources.

Other potential sources of genetic involvement in AMD

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Gene–gene and gene–environment interactions

Assessment of gene–gene and gene–environment interactions potentially allows the uncovering of synergistic actions between different genes or their interactions with other environmental factors.[42] Findings from such interactions may allow insights into mechanisms underlying various biological pathways. The advantage of detecting such interactions is that it potentially allows for the detection of effects that can either exacerbate disease risk or identify factors that afford a protective effect, dependent on the genetic susceptibility of an individual. This has important implications for personalized medicine, in both treatment regimens as well as disease prediction models.[43] Major challenges affecting identification of gene–gene and gene–environment effects are the need for large sample sizes, longitudinal cohorts, exposure assessment and heterogeneity issues related to finding such associations. Typically, at least four times the number of individuals are required for the detection of a main effect compared with a single effect, and may thus require tens of thousands of individuals in GWAS to meet the stringent significance levels.[44]

Interestingly, gene–gene and gene–environment effects have been reported for AMD. In the AMD Gene Consortium, the strongest observed and statistically significant gene interaction involved the risk alleles at rs10737680 (near the CFH gene) and rs429608 (near the C2-CFB gene), with individuals carrying risk alleles at both these loci being at a slightly higher risk of disease than expected. Several other nominally significant gene–gene interactions were also detected, but given the large sample size in this study these findings suggest that such interactions do not appear to exert a major effect in AMD, but rather that gene effects affecting risk tend to be independent of each other. With regard to gene-environment interactions, it has been reported in a disease progression study of AMD that an interaction exists between smokers and the CC risk genotype of the Y402H variant of the CFH gene.[45] However, other studies in advanced AMD have failed to identify such a finding.[46, 47] Similarly, the analysis of an interaction between SNPs in the ARMS2/HTRA1 region and smoking has been inconsistent. In the case of the SNP A69S (rs10490924) in exon 1 of the ARMS2 gene, a significant interaction between this variant with smoking has been shown.[48, 49] Risk increases threefold, from OR 7.07 (95%CI, 3.30–15.13) in individuals with the TT risk genotype at this variant who were non-smokers to OR 20.49 (95%CI, 10.36–40.52) in individuals with the TT risk genotype who were smokers.[49] In contrast, no such interaction was shown in smokers when a genetic haplotype across this region was considered.[46] A gene–environment interaction has also been shown for exposure to infectious agents,[45] while other studies have demonstrated that healthy behaviour or through diet such as omega-3 fatty acids intake can modify genetic risk of progression to GA.[50] However, these results have not been universally replicated, with some studies reporting no apparent interaction. Thus, the influence that gene–gene and gene–environment interactions have in AMD is not totally clear and is an area of intense investigation at this moment. This should not distract from the important health message to patients that an effective reduction in risk of AMD can be achieved through alteration of environmental exposure, such as quitting smoking or by healthy diet.

Epigenetic involvement in AMD

Given that various environmental components have been suggested as risk factors in AMD, including smoking, several dietary components and infection, it is plausible that these may lead to an alteration in gene expression in an individual. It is also likely that the underlying pathology of AMD will result in altered gene expression patterns compared with none affected individuals. Although gene–environment interactions are not consistently defined for known AMD risk genes and environmental factors at present, one mechanism that may influence gene expression is through epigenetics. Epigenetics describes the mechanisms that enable cells to respond to environmental changes without a change in DNA sequence. These mechanisms include DNA methylation, covalent modification of histones, nucleosome remodelling, nuclear dynamics and chromatin interaction with regulatory noncoding RNAs. The National Human Genome Research Institute launched a public research consortium named ENCODE, the Encyclopedia Of DNA Elements, in September 2003, to identify all functional elements in the human genome sequence to map the human ‘epigenome’ (http://www.genome.gov/10005107). One of the most actively studied areas of epigenetics has come through the study of changes in methylation patterns. In vertebrates, approximately one third of DNA methylation occurs in repetitive sequences in the genome defined as Alu or LINE-1 elements. However, the majority of DNA methylation occurs in the context of CpG dinucleotides where addition of methyl groups to cytosine allows formation of 5-methyl-cytosine (5mC).

Recent findings using a genome-wide methylation microarray chip have identified significant methylation changes in discordant AMD twins, which was subsequently confirmed through analysis of discordant AMD sib pairs and case controls samples. In this study, DNA collected from peripheral blood was labelled and hybridized to Roche-NimbleGen Human DNA Methylation 2.1 million Deluxe Promoter Arrays (Madison, WI, USA). Probes were designed to encompass localized regions of transcription start sites of all genes, and microRNAs were also able to be assessed. Significant hypo- and hyper-methylated differences were identified in ∼1.5% of potential methylation (CpG) sites within 231 gene promoters in the examined AMD twins.[51] These genes were identified in immunological disease, gastrointestinal disorders, endocrine system disorders, metabolic disease and genetic disease pathways. One gene that was chosen for further study was the interleukin 17 receptor (IL17RC) gene, which exhibited significant hypomethylation in its promoter region in AMD patients compared with non-AMD individuals, suggesting increased gene expression in patients. IL-17RC mediates signal transduction and proinflammatory activities of IL-17A and IL-17F, and IL17A was demonstrated as being elevated only in the serum and macular tissues of AMD patients, suggesting a potentially crucial role for IL-17A and IL-17RC in the pathogenesis of AMD.[51] Interestingly, the methylation changes identified in AMD subjects showed very little overlap with the published gene variants identified as significantly associated with AMD detected by the AMD Gene Consortium. These findings potentially suggest another level of complexity in terms of genetic involvement in AMD.

Copy number variants

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Although the majority of genetic association studies of AMD have been based on the use of SNPs, there are other levels of genetic variation that can occur in the genome. The next most common genetic variation after SNPs is due to small insertion and deletion polymorphisms (indels) that can vary in size from 1 base pair (bp) to 49 bps in size. These are followed by larger genomic rearrangements including copy number variants, varying in size from 50 bp to several megabases (Mb), and through to larger chromosomal rearrangements including translocations, inversions and chromosome rearrangements, deletions, and duplications.

Several copy number variants have been found to be associated with complex diseases. Many of these have been identified indirectly as a result of genome-wide SNP association studies and were typically found due to their very high or complete linkage disequilibrium (LD) with SNPs. Studies of genetic architecture have indicated that certain chromosomal regions are prone to genomic rearrangements and have been termed segmental duplications (SDs). They comprise approximately 5% of the genome, and there is a tenfold enrichment for structural rearrangements flanked by SDs.

Copy number variants are enriched towards genes that are more likely to interact with environment stimuli, such as immune and inflammatory response genes, cell signalling and cell adhesion molecules, structural proteins, and ion channels.[52, 53] Certainly, there is a precedent for the existence of copy number variants in AMD. The CFH gene and its paragolous genes (CFHR1–5) represent a group of highly related genes that encompass several SDs and encode proteins involved in the inflammatory response. Extended haplotype analysis of the CFH region identified a large deletion encompassing the CFHR1 and CFHR3 genes that was associated with protection in AMD.[28, 54] This deletion was reported in 6–20% of chromosomes of control individuals and 1–8% of chromosomes of cases resulting in a 4 fold risk reduction of AMD.[28, 54] In addition, the proteins encoded by these genes are absent in the serum of individuals homozygous for this deletion.[28] The finding of both heterozygous and homozygous copy number variant loss has now been confirmed in several other studies, with significant association identified with both neovascular disease and GA, compared with controls.[55] A second copy number variant involving deletion of the CFHR1–4 genes has been inconclusive, with some studies reporting protection for AMD,[56] while other studies have found no association with this variant.[57, 58] However, one study suggested that the CFHR1–4 copy number change presents as a rare event in neovascular AMD at 0.74%, but is seven times more frequent at 4.7%, and significantly associated, in individuals with bilateral GA.[55] This may represent enrichment in individuals with this phenotype, potentially allowing for prediagnostic identification of individuals who may go on to develop this end stage of disease.[55] The mechanism of action of copy number variants at the CFH locus most likely occurs through maintenance of a homeostatic balance between CFHR1, CFHR3 and CFH, with CFHR3 and CFHR1 competing with CFH for C3b binding in the complement cascade. Thus, homozygous deletions of CFHR3/1 result in increased binding of CFH to C3b, thereby regulating CFH-mediated complement activity.[59]

A second structural variant has also been identified at the chromosome 10 locus, as a common indel (ARMS2 del443ins54).[60] The AMD Gene Consortium identified the SNP rs10490924 coding variant in the ARMS2 gene as the peak of association at 10q26 that showed strong association (r2 = 0.99) with this 3′ indel.[40] However, this SNP was also in strong LD (r2 = 0.90) with a nearby SNP, rs11200638, which regulates the HTRA1 gene,[33] and thus it is still unclear as to whether HTRA1 or ARMS2 is the causal gene at this locus. Although a role for the ARMS2 protein has been suggested through mitochondrial homeostasis,[34] it is proposed that the indel is associated with both a downregulation of ARMS2 through a destabilization of its message, as well as a concomitant upregulation of HTRA1. A search for other potential copy numbers variants at the 10q locus did not identify any additional changes.[55] These findings, therefore, support the idea that copy number variants likely play a functional role in AMD pathogenesis through gene dosage effects on genes that are involved in regulation of different pathways, such as the complement cascade.

Rare genetic variants

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Large-scale or second-generation sequencing efforts have identified approximately 15 million SNPs in multiple ethnic populations through the 1000 Genomes and the International HapMap Consortiums. Although most studies to date, including the AMD Gene Consortium, have assessed common genetic variants, another class of genetic variants also exist that are also likely to contribute to common complex diseases, these being rare variants. These typically have a minor allele frequency in the order of 1%, and as such are not in LD with common variants, and thus will not usually be detected by GWAS.[61] Common variants are typically ancient, while rare variants have a more recent common origin, estimated to be in the order of 10–20 generations ago and likely to have an effect size larger than that of common variants. As a consequence, they are likely to be population-specific, and thus careful phenotyping and ethnic information is required.

A recent study was undertaken to identify rare highly penetrant variants at the CFH locus.[62] A total of 20 common SNPs were genotyped across the CFH/CFHR1–3 region and the CFHR1–3 deletion (CFHR1-3Δ) in 711 advanced AMD cases and 1041 controls. A total of 11 haplotypes were defined, each of >0.3% frequency, that explained 97.4% of all possible haplotypes. One haplotype (H1) was present in 59% of case chromosomes and 37% of control chromosomes and exclusively contained the Y402H risk allele. A rare haplotype (H5) was identified in approximately 1% of the population, but 10 of the 11 individuals identified were heterozygote for this haplotype and had AMD. A targeted sequencing approach to discover the causal variant was undertaken by selecting 84 samples representing all 11 haplotypes, including all 10 affected individuals with the H5 haplotype. A 107-kb region containing CFH introns, exons and promoter region was sequenced to 20× average depth. Of a total of 623 variants identified, only six were non-synonymous and only one of these, R1210C (altering an arginine to cysteine), found in exon 22 of the CFH gene was present in all sequenced individuals who had the H5 haplotype. Further genotyping of R1210C in 2423 AMD cases and 1122 controls demonstrated the high penetrance of this rare variant, being present in 40 cases versus 1 control (P = 7.0 × 10−6) and resulting in a 6-year earlier onset of disease (P = 2.3 × 10−6). This result represents one of the first instances where knowledge of common variants has allowed discovery of a rare penetrant mutation.[62]

Genetic variant considerations and AMD

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

One of the most frequently described gene variants in the CFH gene is that of rs1061170 (also known as Y402H), with over 170 publications so far published (http://www.ncbi.nlm.nih.gov/pubmed/?term=y402h+macular). The majority of these studies have assessed genetic associations rather than function, and so its role in advanced AMD is not totally clear. Although this SNP is associated with the most frequent risk haplotype detected at this gene,[24] there are other SNPs in the CFH gene that appear more strongly associated with risk of AMD than Y402H.[63] In an effort to identify the causal variant in the CFH gene, extensive association and haplotype studies have been undertaken that identify two SNPs (rs1329428 and rs203687) in a 32-kb region that is downstream of Y402H (rs1061170) as being critical for AMD.[56] In the AMD Gene Consortium, the most strongly associated SNP in the CFH region was SNP rs10737680 that was not in strong LD with Y402H but also 20-kb downstream of it, and also occurred in the previously described 32-kb critical region. These data suggest that this region deserves more attention in establishing its role in AMD. Although the Y402H variant has been convincingly associated with AMD in European populations, studies in Asian cohorts from Japan, China and Korea are inconsistent, with many showing little association to AMD,[64] while another Japanese study has shown significant association.[65]

Similarly, issues regarding which genetic variant might be causally associated with AMD at the ARMS2/HTRA1 locus have not been clearly shown. In all studies, including the AMD Gene Consortium, it has not been possible so far to convincingly separate the association of the variant rs10490924 (a coding variant in the ARMS2 gene) from that of the nearby SNP rs11200638 (which regulates HTRA1), as both are in very high LD (r2 = 0.90).

Assessment of the chromosome 1 and 10 loci has provided some insights into the two major subtypes of AMD. Although both these regions are strongly associated with neovascular disease and GA, from two GWAS, it appears that the ARMS2/HTRA1 locus preferentially associates with risk of neovascular disease compared with GA (OR, 1.38; P = 7.4 × 10−14)[66] and neovascular disease (OR = 2.97) and GA (OR = 2.50).[40] In the case of CFH risk alleles, these preferentially associate with risk of GA (OR = 2.80) compared with neovascular disease (OR = 2.34).[40]

Risk prediction models

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

The information garnered from epidemiological and genetic studies has enabled a number of investigators to undertake risk prediction models in AMD. However, there are considerable challenges in identifying individuals with early-stage disease who are most at risk of developing the end-stage disease of either GA or CNV. These arise due to the progressive nature of the disease with age, and thus the potential for currently misclassifying individuals with no disease or ‘controls’ who may become incident cases later in life. AMD also presents as a heterogenous disease with the involvement of multiple genetic and environmental factors implicated in its aetiology. Thus, identifying the suite of factors ascribed to predictive risk has led to the reporting of several different risk factor models. Although some of these are based on the use of both lifestyle factors, such as body mass index and smoking, as well as inclusion of genetic factors, other models have been described based solely on the use of an increasing number of genetic risk factors. These models have led to improvements in predictability with the majority of studies, including our own, reporting a sensitivity of approximately 80%, specificity of approximately 70% and area under the curve of approximately 80%.[67] Newer models have now also begun to incorporate fundus changes alongside genotype to increase the area under the curve to over 90%.[68] Although this analysis was based on a single cohort (Age-Related Eye Disease Study), it does provide a prospective route whereby inclusion of multiple facets of information that incorporate disease features, lifestyle, environmental as well as genetic risk factors, together with improved statistical modelling, will further improve our predictive ability. This is likely to lead to stratification of patients based on different risk profiles with the potential for more targeted clinical trials.

Anti-vascular endothelial growth factor A (VEGF) treatments and genotype

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

A fundamental outcome of genetic knowledge is in being able to utilize this information in treatment response. In the advanced stage of AMD (neovascular or ‘wet’), abnormal growth of choroidal vessels in the macular area results in loss of central vision and ultimately legal blindness if left untreated. Several recombinant, humanized, monoclonal antibodies (Mabs) that neutralize all forms of vascular endothelial growth factor A (VEGF-A) are available for the treatment of the neovascularization process in AMD.[69] These include ranibizumab (Lucentis) bevacizumab (Avastin) and most recently aflibercept (Eylea), which neutralizes VEGF-A and B and have revolutionized the treatment for subfoveal neovascular AMD. Although all of them have been shown to improve visual acuity, approximately 10–25% of patients continue to lose vision.[70-72]

In an effort to identify whether common variants in known AMD genes influence treatment response to intraocular injections, a number of studies have been undertaken. These have had mixed results. For instance, a recent meta-analysis of the CFH CC risk genotype at Y402H showed a 1.6-fold decreased response to treatment compared with patients carrying the homozygous T-allele. We have also shown that individuals with the risk AA genotype at rs11200638 in the HTRA1 gene and the rs10490924 (A69S) SNP in the LOC387715/ARMS2 gene were significantly associated (P = 0.001 and P = 0.002, respectively) with poorer VA outcome at 12 months.[73] However, other studies have reported that genetic variants in the CFH and LOC387715/ARMS2/HTRA1 genes were not associated with treatment response. For instance, in the Comparison of AMD Treatment Trials study, no response to treatment was found in 834 patients for SNPs Y402H (CFH), rs10490924 (ARMS2), rs11200638 (HTRA1) and rs2230199 (C3) following bevacizumab treatment,[74] and another study reported similar findings following treatment with Lucentis.[75] Thus, consistent association with known AMD risk loci with treatment response is still lacking, and perhaps indicates either a minimal involvement of these genetic variants or that other genetic variants may be involved.

Next-generation sequencing – what is it and why is it needed?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Understanding the entire DNA sequence of the entire genome of an organism is pivotal in deciphering the underlying molecular mechanisms involved in various functions. The widely prevalent Sanger sequencing was based on the principles of capillary electrophoresis, and made a significant impact in understanding the structural and functional biology of genes and other non-gene elements in the genome. However, the technology was limited in providing high throughput within a reasonable amount of time and was not cost-effective.

Thus, the search for an improved method led to second-generation sequencing called next-generation sequencing (NGS), resulting in a revolution in genomic science. The NGS technique employs deep sequencing, wherein the same nucleotide is sequenced multiple times to ensure accuracy. This increases the depth of coverage of a given nucleotide and reduces the number of sequencing errors due to the presence of rare variations in a gene sequence.

The principle of NGS technology employs the fragmentation of genomic DNA into libraries of smaller segments that are sequenced in parallel. Following this, individual sequences are reassembled and aligned to a reference genome and the entire sequence is derived from the consensus of all the aligned reads. In the absence of a reference genome, the same principle is applied for de novo sequencing. In practice, NGS has enabled the scalability of sequencing from roughly a gigabyte of data using Sanger's method to terabyte(s) of data in a single run. Although the first complete human genome was sequenced in a decade, the same can now be achieved in less than 1 week and at a cost that is manyfold less.[76-78]

Further applications of the NGS technique

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Other immediate applications of the NGS technique have not only been achieved through complete whole genome sequencing, but also through other shorter read methods that can be used for targeting specific regions of interest, exomes (coding regions of the genome) and transcriptomes (the collection of all the transcripts in the genome). This has been further enhanced through the use of multiplexing of different samples and/or short sequences in a single run where samples can be readily separated by unique sequences or ‘bar codes’. These bar codes are effective in segregating different samples during the downstream analysis steps, thereby allowing the generation of high throughput data on multiple samples at the same time.

Sequencing of targeted regions and exomes

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

The targeted sequencing approach is used to characterize a defined region within the genome harbouring the gene(s) or intergenic regions of interest. Although large effect variants indicate a statistically significant association with the disease, several moderate to small effect variants often need further validation to determine their involvement. The NGS technique provides a useful approach for validating these small effect variants and also characterizing the region(s) of interest.[79] The advantage of NGS lies in its coverage of the genome (also called the depth of its sequencing capability), which allows for the identification of ‘rare variants’ that may otherwise have been missed in GWAS and already evident through the identification of the rare variant R1210C in the CFH gene.[62]

Complete exome sequencing emerged as an offshoot of NGS with the invention of the target enrichment technique, wherein specific regions of DNA can be sequenced per sample. This facilitated the sequencing of the entire exome for characterizing functional SNPs or rare disease-associated alleles in a population, and offered the advantage of producing data confined to the coding regions of a genome rather than its non-coding parts.

RNA sequencing

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

The NGS technique that was initially developed was designed to decipher the precise sequence of DNA, but its use has now been further extended to the assessment of RNA. Using a similar technique, RNA can now be screened to allow an understanding of the transcriptome in any given tissue or cell type. Counting the number of mRNAs that are encoded by individual genes provides an indicator of protein-coding potential, which is a major contributor to phenotype. Unlike microarrays that measure signal intensities, NGS quantifies digital sequencing read counts for capturing gene expression changes associated with biological processes.[80]

Choice of platforms and data analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Multiple platforms (Roche 454, ABI Solid, Illumina Hi Seq and Miseq, etc.) are available for NGS data capture and analysis. Each of these has its own unique chemistry and varies based on their ability to capture the high throughput sequence data. The chemistry used in data capture across these platforms is contingent upon the number of samples that can be analysed in a single run, read lengths, depth of coverage and their outputs.

Data analysis is very critical in any NGS experiment and requires a large amount of bioinformatics support by virtue of the huge amounts of data that are generated. The existing platforms are capable of generating the raw data and also undertaking some initial quality control measures. Further downstream quality control is needed before analysis is usually performed by several third-party software that are freely available on the web. Depending on the type of experiments performed, these include identifying common and rare variants, gene regulation analysis, quantitative and qualitative sequencing-based transcriptome analysis, SNP discovery and structural variation analysis, cytogenetic analysis, DNA–protein interaction analysis (ChIP-Seq), sequencing-based methylation analysis, small RNA discovery and analysis, de novo, metagenomics and metatranscriptomics.[81-83]

Future developments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

High throughput platform technologies are rapidly improving all the time. We will see a greater density of SNPs on gene chips, likewise for epigenetic chips. In addition, there will be further advances in bisulphite sequencing that will be beneficial to identification of epigenetic changes that affect regions of genes other than the promoter that are associated with AMD. Pricing of sequencing will continue to drop, and bench top sequencing machines will become more readily available in the future. The search for rare variants as currently underway through the NEI supported AMD Exome Chip Consortium will also bear fruit in terms of identifying patients where these variants have large effects but in subgroups of people. Statistical approaches to better capture SNP information or groups of SNPs and pathway interactions will also occur as will the ability to better detect copy number changes at the genome-wide level. Likewise gene–environment interactions will be better defined through improved statistical packages. All of these will be underpinned by improvements in imaging and a better understanding of changes in retinal function at the early and advanced stages of disease. There will in turn lead to improvements in better defining pharmacogenetic response influenced by genetic variants and how such variants may play a role in altering response to treatment.

Benefits for the patient

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

The arrival of high throughput platforms, such as GWAS, methylation chips or the ability to sequence data through either an exome/whole genome or targeted approach using NGS, will allow us to identify the genetic spectrum of variants in a patient. It will also allow us to assess whether these genetic variants, including copy number changes, are inherited in younger individuals who have as yet to develop disease, thereby building on existing predictive risk models and providing prediagnostic tests for particular variants of combinations of variants. Although there is no one causal variant or no one environmental factor that causes AMD, a combination of many such factors is clearly necessary. These factors may play differing roles at different stages of AMD, whether it be at the onset of early disease, disease progression or leading to onset of advanced disease. AMD research is leading the way in many aspects of teasing apart the factors involved in a complex disease, but there is still much to be done in terms of untangling the information that we have.

It is clear that the more information that we have, the better we will understand the disease especially when there are a number of different gene and environmental associations, as well as multiple pathways involved in its aetiology. The completion of this jigsaw will rely on being able to put these pieces together to identify at which stage of disease they are most important, how they are associated with likely sub-phenotypes of disease, as well as how and when different associated biological pathways play a role. Without the technological advances that we have witnessed in the last few years, little of this puzzle would have been available for us to assess. The immediate benefit to patients will be through improved risk prediction models and prediagnostic testing, as well as through better targeted or personalized drug treatments. This may come through clinical trials currently underway or planned, where optimal treatment may be achieved for one subgroup of patients whereas another subgroup of patients may not respond due to their different genetic profile, thus requiring a different drug therapy. It is also likely that given the different biological pathways so far identified in AMD, drug therapies currently being used for the treatment of other diseases may also be effective in the treatment of AMD, which again would be aided by an underlying knowledge of determinants of disease.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References

Since 2005, with the discovery of the CFH gene, the explosion in our knowledge of AMD has been incredible. Technological advances have identified many changes at different genetic levels, but these now need to be better married to other areas in a systems biology approach. Through this, we should be able to continue progress in understanding and better treating this disease.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genome-wide association studies (GWAS)
  5. Other potential sources of genetic involvement in AMD
  6. Copy number variants
  7. Rare genetic variants
  8. Genetic variant considerations and AMD
  9. Risk prediction models
  10. Anti-vascular endothelial growth factor A (VEGF) treatments and genotype
  11. Next-generation sequencing – what is it and why is it needed?
  12. Further applications of the NGS technique
  13. Sequencing of targeted regions and exomes
  14. RNA sequencing
  15. Choice of platforms and data analysis
  16. Future developments
  17. Benefits for the patient
  18. Conclusion
  19. References
  • 1
    Mitchell P, Smith W, Attebo K, Wang JJ. Prevalence of age-related maculopathy in Australia. The Blue Mountains Eye Study. Ophthalmology 1995; 102: 14501460.
  • 2
    VanNewkirk MR, Nanjan MB, Wang JJ, Mitchell P, Taylor HR, McCarty CA. The prevalence of age-related maculopathy: the visual impairment project. Ophthalmology 2000; 107: 15931600.
  • 3
    Friedman DS, O'Colmain BJ, Munoz B et al. Prevalence of age-related macular degeneration in the United States. Arch Ophthalmol 2004; 122: 564572.
  • 4
    Chong EW, Kreis AJ, Wong TY, Simpson JA, Guymer RH. Dietary omega-3 fatty acid and fish intake in the primary prevention of age-related macular degeneration: a systematic review and meta-analysis. Arch Ophthalmol 2008; 126: 826833.
  • 5
    Chong EW, Robman LD, Simpson JA et al. Fat consumption and its association with age-related macular degeneration. Arch Ophthalmol 2009; 127: 674680.
  • 6
    Chong EW, Simpson JA, Robman LD et al. Red meat and chicken consumption and its association with age-related macular degeneration. Am J Epidemiol 2009; 169: 867876.
  • 7
    Evans JR. Risk factors for age-related macular degeneration. Prog Retin Eye Res 2001; 20: 227253.
  • 8
    Guymer R, Robman L. Chlamydia pneumoniae and age-related macular degeneration: a role in pathogenesis or merely a chance association? Clin Experiment Ophthalmol 2007; 35: 8993.
  • 9
    Klein R. Overview of progress in the epidemiology of age-related macular degeneration. Ophthalmic Epidemiol 2007; 14: 184187.
  • 10
    Heiba IM, Elston RC, Klein BE, Klein R. Sibling correlations and segregation analysis of age-related maculopathy: the Beaver Dam Eye Study. Genet Epidemiol 1994; 11: 5167.
  • 11
    Klein ML, Mauldin WM, Stoumbos VD. Heredity and age-related macular degeneration. Observations in monozygotic twins. Arch Ophthalmol 1994; 112: 932937.
  • 12
    Hammond CJ, Webster AR, Snieder H, Bird AC, Gilbert CE, Spector TD. Genetic influence on early age-related maculopathy: a twin study. Ophthalmology 2002; 109: 730736.
  • 13
    Swaroop A, Branham KE, Chen W, Abecasis G. Genetic susceptibility to age-related macular degeneration: a paradigm for dissecting complex disease traits. Hum Mol Genet 2007; 16: R174182.
  • 14
    Chen W, Stambolian D, Edwards AO et al. Genetic variants near TIMP3 and high-density lipoprotein-associated loci influence susceptibility to age-related macular degeneration. Proc Natl Acad Sci U S A 2010; 107: 74017406.
  • 15
    Kaur I, Rathi S, Chakrabarti S. Variations in TIMP3 are associated with age-related macular degeneration. Proc Natl Acad Sci U S A 2010; 107: E112113.
  • 16
    Fritsche LG, Fleckenstein M, Fiebig BS et al. A subgroup of age-related macular degeneration is associated with mono-allelic sequence variants in the ABCA4 gene. Invest Ophthalmol Vis Sci 2012; 53: 21122118.
  • 17
    Klaver CC, Kliffen M, van Duijn CM et al. Genetic association of apolipoprotein E with age-related macular degeneration. Am J Hum Genet 1998; 63: 200206.
  • 18
    Baird PN, Guida E, Chu DT, Vu HT, Guymer RH. The epsilon2 and epsilon4 alleles of the apolipoprotein gene are associated with age-related macular degeneration. Invest Ophthalmol Vis Sci 2004; 45: 13111315.
  • 19
    McKay GJ, Patterson CC, Chakravarthy U et al. Evidence of association of APOE with age-related macular degeneration: a pooled analysis of 15 studies. Hum Mutat 2011; 32: 14071416.
  • 20
    Baird PN, Hageman GS, Guymer RH. New era for personalized medicine: the diagnosis and management of age-related macular degeneration. Clin Experiment Ophthalmol 2009; 37: 814821.
  • 21
    Klein RJ, Zeiss C, Chew EY et al. Complement factor H polymorphism in age-related macular degeneration. Science 2005; 308: 385389.
  • 22
    Edwards AO, Ritter R 3rd, Abel KJ, Manning A, Panhuysen C, Farrer LA. Complement factor H polymorphism and age-related macular degeneration. Science 2005; 308: 421424.
  • 23
    Haines JL, Hauser MA, Schmidt S et al. Complement factor H variant increases the risk of age-related macular degeneration. Science 2005; 308: 419421.
  • 24
    Hageman GS, Anderson DH, Johnson LV et al. A common haplotype in the complement regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular degeneration. Proc Natl Acad Sci U S A 2005; 102: 72277232.
  • 25
    Ormsby RJ, Ranganathan S, Tong JC et al. Functional and structural implications of the complement factor H Y402H polymorphism associated with age-related macular degeneration. Invest Ophthalmol Vis Sci 2008; 49: 17631770.
  • 26
    Gold B, Merriam JE, Zernant J et al. Variation in factor B (BF) and complement component 2 (C2) genes is associated with age-related macular degeneration. Nat Genet 2006; 38: 458462.
  • 27
    Yates JR, Sepp T, Matharu BK et al. Complement C3 variant and the risk of age-related macular degeneration. N Engl J Med 2007; 357: 553561.
  • 28
    Hageman GS, Hancox LS, Taiber AJ et al. Extended haplotypes in the complement factor H (CFH) and CFH-related (CFHR) family of genes protect against age-related macular degeneration: characterization, ethnic distribution and evolutionary implications. Ann Med 2006; 38: 592604.
  • 29
    Spencer KL, Hauser MA, Olson LM et al. Deletion of CFHR3 and CFHR1 genes in age-related macular degeneration. Hum Mol Genet 2008; 17: 971977.
  • 30
    Jakobsdottir J, Conley YP, Weeks DE, Mah TS, Ferrell RE, Gorin MB. Susceptibility genes for age-related maculopathy on chromosome 10q26. Am J Hum Genet 2005; 77: 389407.
  • 31
    Rivera A, Fisher SA, Fritsche LG et al. Hypothetical LOC387715 is a second major susceptibility gene for age-related macular degeneration, contributing independently of complement factor H to disease risk. Hum Mol Genet 2005; 14: 32273236.
  • 32
    Yang Z, Camp NJ, Sun H et al. A variant of the HTRA1 gene increases susceptibility to age-related macular degeneration. Science 2006; 314: 992993.
  • 33
    Dewan A, Liu M, Hartman S et al. HTRA1 promoter polymorphism in wet age-related macular degeneration. Science 2006; 314: 989992.
  • 34
    Kanda A, Chen W, Othman M et al. A variant of mitochondrial protein LOC387715/ARMS2, not HTRA1, is strongly associated with age-related macular degeneration. Proc Natl Acad Sci U S A 2007; 104: 1622716232.
  • 35
    International Consortium for Blood Pressure Genome-Wide Association Studies, Ehret GB, Munroe PB, Rice KM et al. Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk. Nature 2011; 478: 103109.
  • 36
    Fagerness JA, Maller JB, Neale BM, Reynolds RC, Daly MJ, Seddon JM. Variation near complement factor I is associated with risk of advanced AMD. Eur J Hum Genet 2009; 17: 100104.
  • 37
    Neale BM, Fagerness J, Reynolds R et al. Genome-wide association study of advanced age-related macular degeneration identifies a role of the hepatic lipase gene (LIPC). Proc Natl Acad Sci U S A 2010; 107: 73957400.
  • 38
    Yu Y, Bhangale TR, Fagerness J et al. Common variants near FRK/COL10A1 and VEGFA are associated with advanced age-related macular degeneration. Hum Mol Genet 2011; 20: 36993709.
  • 39
    Arakawa S, Takahashi A, Ashikawa K et al. Genome-wide association study identifies two susceptibility loci for exudative age-related macular degeneration in the Japanese population. Nat Genet 2011; 43: 10011004.
  • 40
    Fritsche LG, Chen W, Schu M et al. Seven new loci associated with age-related macular degeneration. Nat Genet 2013; 45: 433439.
  • 41
    Seddon JM, Cote J, Page WF, Aggen SH, Neale MC. The US twin study of age-related macular degeneration: relative roles of genetic and environmental influences. Arch Ophthalmol 2005; 123: 321327.
  • 42
    Kraft P, Yen YC, Stram DO, Morrison J, Gauderman WJ. Exploiting gene-environment interaction to detect genetic associations. Hum Hered 2007; 63: 111119.
  • 43
    Thomas D. Methods for investigating gene-environment interactions in candidate pathway and genome-wide association studies. Annu Rev Public Health 2010; 31: 2136.
  • 44
    Thomas D. Gene-environment-wide association studies: emerging approaches. Nat Rev Genet 2010; 11: 259272.
  • 45
    Baird PN, Robman LD, Richardson AJ et al. Gene-environment interaction in progression of AMD: the CFH gene, smoking and exposure to chronic infection. Hum Mol Genet 2008; 17: 12991305.
  • 46
    Hughes AE, Orr N, Patterson C et al. Neovascular age-related macular degeneration risk based on CFH, LOC387715/HTRA1, and smoking. Plos Med 2007; 4: e355.
  • 47
    Sofat R, Casas JP, Webster AR et al. Complement factor H genetic variant and age-related macular degeneration: effect size, modifiers and relationship to disease subtype. Int J Epidemiol 2012; 41: 250262.
  • 48
    Schmidt S, Hauser MA, Scott WK et al. Cigarette smoking strongly modifies the association of LOC387715 and age-related macular degeneration. Am J Hum Genet 2006; 78: 852864.
  • 49
    Chakravarthy U, McKay GJ, de Jong PT et al. ARMS2 increases the risk of early and late age-related macular degeneration in the European Eye Study. Ophthalmology 2013; 120: 342348.
  • 50
    Reynolds R, Rosner B, Seddon JM. Dietary omega-3 fatty acids, other fat intake, genetic susceptibility, and progression to incident geographic atrophy. Ophthalmology 2013; 120: 10201028.
  • 51
    Wei L, Liu B, Tuo J et al. Hypomethylation of the IL17RC promoter associates with age-related macular degeneration. Cell Rep 2012; 2: 11511158.
  • 52
    Sebat J, Lakshmi B, Troge J et al. Large-scale copy number polymorphism in the human genome. Science 2004; 305: 525528.
  • 53
    Tuzun E, Sharp AJ, Bailey JA et al. Fine-scale structural variation of the human genome. Nat Genet 2005; 37: 727732.
  • 54
    Hughes AE, Orr N, Esfandiary H, Diaz-Torres M, Goodship T, Chakravarthy U. A common CFH haplotype, with deletion of CFHR1 and CFHR3, is associated with lower risk of age-related macular degeneration. Nat Genet 2006; 38: 11731177.
  • 55
    Cantsilieris S, White SJ, Richardson AJ, Guymer RH, Baird PN. Comprehensive analysis of Copy Number Variation of genes at chromosome 1 and 10 loci associated with late age related macular degeneration. Plos ONE 2012; 7: e35255.
  • 56
    Sivakumaran TA, Igo RP Jr, Kidd JM et al. A 32 kb critical region excluding Y402H in CFH mediates risk for age-related macular degeneration. Plos ONE 2011; 6: e25598.
  • 57
    Kubista KE, Tosakulwong N, Wu Y et al. Copy number variation in the complement factor H-related genes and age-related macular degeneration. Mol Vis 2011; 17: 20802092.
  • 58
    Sawitzke J, Im KM, Kostiha B, Dean M, Gold B. Association assessment of copy number polymorphism and risk of age-related macular degeneration. Ophthalmology 2011; 118: 24422446.
  • 59
    Fritsche LG, Lauer N, Hartmann A et al. An imbalance of human complement regulatory proteins CFHR1, CFHR3 and factor H influences risk for age-related macular degeneration (AMD). Hum Mol Genet 2010; 19: 46944704.
  • 60
    Fritsche LG, Loenhardt T, Janssen A et al. Age-related macular degeneration is associated with an unstable ARMS2 (LOC387715) mRNA. Nat Genet 2008; 40: 892896.
  • 61
    Frazer KA, Murray SS, Schork NJ, Topol EJ. Human genetic variation and its contribution to complex traits. Nat Rev Genet 2009; 10: 241251.
  • 62
    Raychaudhuri S, Iartchouk O, Chin K et al. A rare penetrant mutation in CFH confers high risk of age-related macular degeneration. Nat Genet 2011; 43: 12321236.
  • 63
    Li M, Atmaca-Sonmez P, Othman M et al. CFH haplotypes without the Y402H coding variant show strong association with susceptibility to age-related macular degeneration. Nat Genet 2006; 38: 10491054.
  • 64
    Katta S, Kaur I, Chakrabarti S. The molecular genetic basis of age-related macular degeneration: an overview. J Genet 2009; 88: 425449.
  • 65
    Hayashi H, Yamashiro K, Gotoh N et al. CFH and ARMS2 variations in age-related macular degeneration, polypoidal choroidal vasculopathy, and retinal angiomatous proliferation. Invest Ophthalmol Vis Sci 2010; 51: 59145919.
  • 66
    Sobrin L, Ripke S, Yu Y et al. Heritability and genome-wide association study to assess genetic differences between advanced age-related macular degeneration subtypes. Ophthalmology 2012; 119: 18741885.
  • 67
    Hageman GS, Gehrs K, Lejnine S et al. Clinical validation of a genetic model to estimate the risk of developing choroidal neovascular age-related macular degeneration. Hum Genomics 2011; 5: 420440.
  • 68
    Perlee LT, Bansal AT, Gehrs K et al. Inclusion of genotype with fundus phenotype improves accuracy of predicting choroidal neovascularization and geographic atrophy. Ophthalmology 2013; 120: 18801892.
  • 69
    Otani A, Takagi H, Oh H et al. Vascular endothelial growth factor family and receptor expression in human choroidal neovascular membranes. Microvasc Res 2002; 64: 162169.
  • 70
    Rosenfeld PJ, Brown DM, Heier JS et al. Ranibizumab for neovascular age-related macular degeneration. N Engl J Med 2006; 355: 14191431.
  • 71
    Brown DM, Michels M, Kaiser PK, Heier JS, Sy JP, Ianchulev T. Ranibizumab versus verteporfin for neovascular age-related macular degeneration. N Engl J Med 2006; 355: 14321444.
  • 72
    Holz FG, Amoaku W, Donate J et al.; SUSTAIN Study Group. Safety and efficacy of a flexible dosing regimen of ranibizumab in patients with neovascular age-related macular degeneration: results from the 12-month, Phase III, multi-centre, open-label SUSTAIN trial. Ophthalmology 2011; 118: 663671.
  • 73
    Abedi F, Wickremasinghe S, Richardson AJ, Islam AF, Guymer RH, Baird PN. Genetic influences on the outcome of anti-vascular endothelial growth factor treatment in neovascular AMD. Ophthalmology 2013; 120: 16411648.
  • 74
    Hagstrom SA, Ying GS, Pauer GJ et al. Pharmacogenetics for genes associated with age-related macular degeneration in the comparison of AMD treatments trials (CATT). Ophthalmology 2013; 120: 593599.
  • 75
    Lee AY, Raya AK, Kymes SM, Shiels A, Brantley MA Jr. Pharmacogenetics of Complement Factor H (Y402H) and treatment of exudative age-related macular degeneration with ranibizumab. Br J Ophthalmol 2008; 93: 610613.
  • 76
    Majewski J, Schwartzentruber J, Lalonde E, Montpetit A, Jabado N. What can exome sequencing do for you? J Med Genet 2011; 48: 580589.
  • 77
    Bick D, Dimmock D. Whole exome and whole genome sequencing. Curr Opin Pediatr 2011; 23: 594600.
  • 78
    Harrison RJ. Understanding genetic variation and function – the applications of next generation sequencing. Semin Cell Dev Biol 2012; 23: 230236.
  • 79
    Gilissen C, Hoischen A, Brunner HG, Veltman JA. Disease gene identification strategies for exome sequencing. Eur J Hum Genet 2012; 20: 490497.
  • 80
    Ozsolak F, Milos PM. Single-molecule direct RNA sequencing without cDNA synthesis. Wiley Interdiscip Rev RNA 2011; 2: 565570.
  • 81
    Coonrod EM, Durtschi JD, Margraf RL, Voelkerding KV. Developing genome and exome sequencing for candidate gene identification in inherited disorders: an integrated technical and bioinformatics approach. Arch Pathol Lab Med 2013; 137: 415433.
  • 82
    Mutz KO, Heilkenbrinker A, Lönne M, Walter JG, Stahl F. Transcriptome analysis using next-generation sequencing. Curr Opin Biotechnol 2013; 24: 2230.
  • 83
    Gullapalli RR, Lyons-Weiler M, Petrosko P, Dhir R, Becich MJ, LaFramboise WA. Clinical integration of next-generation sequencing technology. Clin Lab Med 2012; 32: 585599.