Genetics of Allergic Diseases

Abstract

The allergic diseases are complex phenotypes for which a strong genetic basis has been firmly established. Genome-wide association studies (GWAS) has been widely employed in the field of allergic disease, and to date significant associations have been published for nearly 100 asthma genes/loci, in addition to multiple genes/loci for AD, AR and IgE levels, for which the overwhelming number of candidates are novel and have given a new appreciation for the role of innate as well as adaptive immune-response genes in allergic disease. A major outcome of GWAS in allergic disease has been the formation of national and international collaborations leading to consortia meta-analyses, and an appreciation for the specificity of genetic associations to sub-phenotypes of allergic disease. Molecular genetics has undergone a technological revolution, leading to next generation sequencing (NGS) strategies that are increasingly employed to hone in on the causal variants associated with allergic diseases. Unmet needs in the field include the inclusion of ethnically and racially diverse cohorts, and strategies for managing ‘big data’ that is an outcome of technological advances such as sequencing.

INTRODUCTION

Coca and Cooke were the first to describe asthma, atopic dermatitis (AD), allergic rhinitis (AR), food allergy, and urticaria as ‘phenomena of hypersensitiveness’ at the annual meeting of the American Association of Immunologists in 1922¹. Just prior to and following this discourse, there was considerable focus on the relative influence of the environment versus hereditary factors on allergic diseases, with family-based twin and migration studies providing the earliest and most compelling evidence for genetic contributions^2–6. Studies on the prevalence of allergic traits in relation to family history demonstrated incremental increases in risk of developing asthma, AR, or AD with the presence of one or both parents with allergic disease, and greater than three times the risk if allergic disease occurred in more than one first degree relative⁷. To this date, and despite the dramatic technological advances that have led to the identification of hundreds of genetic variants in genes associated with asthma, AD, and to a lesser degree, food allergy and AR, a positive family history remains one of the most reliable tools for prognosis of allergic disease.

Approaches for disentangling the genetic basis for the allergic diseases have evolved as technological tools for the field of molecular genetics have progressed. With the introduction of the polymerase chain reaction (PCR) in the 1980s, DNA fragments in the human genome could be amplified and then studied for variable fragment lengths of repeats, or ‘genetic fingerprinting’. With a catalog of microsatellite markers spanning the human genome, genome-wide linkage studies emerged as a robust approach for identifying genetic hot spots associated with complex traits. Nearly a dozen genome-wide linkage screens were performed on asthma and its associated phenotypes^8–18, for which multiple chromosomal regions provided significant evidence for linkage. From several of these family-based linkage genome-wide screens, six novel asthma genes were identified by positional cloning^18–23. Similarly, multiple linkage studies were performed for AD (summarized in Ref. ²⁴) and AR^25–29. It was frequently observed that loci overlapped across associated traits; for example, Daniels and colleagues observed overlapping linkage peaks with quantitative traits associated with asthma including total serum IgE, skin test index, and eosinophil counts, as well as atopy as a qualitative trait⁸. Alternatively, the multiethnic Collaborative Study on the Genetics of Asthma reported linkage peaks that were specific to different racial ethnic groups⁹.

With the publication of initial efforts in sequencing the human genome^30,31, the opportunity to genotype markers directly in genes of interest was greatly expanded as polymorphisms were identified in the approximately 20,000 to 25,000 genes across the 3 billion chemical base pairs that make up human DNA. Relying upon one of the simplest of these polymorphisms, single nucleotide polymorphisms (SNPs), and relatively simple structural variants, such as insertions/deletions and repeats, this advancement allowed researchers to expand genetic studies beyond linkage toward the genetic association study design. For asthma alone, literally hundreds of candidate genes have been elucidated, and eloquently summarized elsewhere^32–35, representing the relative success of this approach.

The GWAS Era

Following completion of the Human Genome Project, the International HapMap Project^36–38 cataloged genomes representing four biogeographical groups (whites from the United States with northern and western European ancestry; Yorubans from Ibadan, Nigeria [YRI]; Han Chinese from Beijing, China [CHB]; and Japanese from Tokyo, Japan [JPT]) to advance the development of new analytic methods and investigating patterns of genetic variation. Simultaneously, the technological capacity to rapidly (and cheaply) genotype >1M common (>5%) SNPs on thousands of DNA samples from patients phenotyped for various complex clinical traits took the spotlight, and the genome-wide association studies (GWAS) era took off. The content of commercially available GWAS chips grew exponentially with expansion of the human genome catalog through the Thousand Genomes Project (TGP)³⁹, and the capacity for discovery of genetic associations has likewise increased with the development of SNP genotype imputation methodologies^40,41, whereby genotyped content from the chip can be combined with the >35M sequenced variants cataloged in the TGP. In the span of only seven years, over 1,924 publications and 13,403 SNPs associated with various complex and quantitative traits^42,43 have been generated by GWAS (Figure 1, Panel A).

Figure 1.

Published genome-wide association studies (GWAS) to date according to ethnicity and race for all catalogued GWAS (Panel A) and asthma GWAS (Panel B). Data generated from the National Human Genome Research Institute’s GWAS catalog website (http://www.genome.gov/gwastudies/).

GWAS has been widely employed in the field of allergic disease. While the precise number of GWAS are difficult to determine, approximately 40 asthma, three atopy, and three AD GWAS (plus a study of >30,000 AD patients genotyped on the Immunochip⁴⁴) have been reported in the Catalog of Published Genome-Wide Association Studies^42,43 (Figure 1, Panel B and summarized in Table 1). A major outcome of GWAS in allergic disease has been the formation of national and international collaborations leading to consortia meta-analyses, which has greatly facilitated gene discovery owed to the increased power generated from larger sample sizes (which are necessary to detect true associations while adjusting for the multiple comparisons). For example, the first asthma GWAS only showed a significant association between childhood onset asthma and markers near the ORMDL3 gene on chromosome 17q21 (P<10⁻¹²) among European populations⁴⁵. When the study was expanded to include >26,000 cases and unaffected controls (e.g., the European-based GABRIEL Consortium⁴⁶), five additional genes plus the 17q locus were strongly associated with asthma⁴⁷. Following completion of 8 U.S.-based, independent asthma GWAS, the NHLBI-supported EVE Consortium was established, comprising >12,000 European American, African American and Hispanic cohorts plus >12,000 independent samples for replication⁴⁸. More recently, the Transnational Asthma Genetics Consortium (TAGC) was formed to perform a global meta-analysis for asthma, and to date TAGC includes 67 cohorts representing nearly 20 studies spanning the globe, representing data on over 100,000 asthma cases, controls and family members (Demenais, Nicolae, et al, unpublished data).

It can be argued that the huge research efforts and expense committed to GWAS on allergic disease have confirmed suspected genes and pathways, some of which were the focus following linkage study discoveries and a result of the many candidate gene studies undertaken. However, GWAS has, for the most part, generated novel candidate genes and a new appreciation for the role of innate as well as adaptive immune-response genes in allergic disease. In the European-based GABRIEL Consortium, six genes were strongly associated with asthma⁴⁷, of which three (IL33, ST2, and the IKZF3-ZPBP2-GSDMB-ORMDL3 region on chromosome 17q21) were replicated in the EVE Consortium⁴⁹. Independent GWAS have provided further support for these same loci^50,51,52. One of the strongest signals from the combined meta-analysis was for IL1RL1 (summarized in the Supplementary Figure 11 in Ref. ⁴⁸), even though the peak SNP differed across ethnic groups. The association between IL1RL1 SNPs among African samples was marginal, and might have been overlooked, but in light of evidence for association in other cohorts, IL1RL1 showed the strongest association overall (P=1.4×10⁻⁸).

Lessons learned from candidate gene and positional cloning studies included the specificity of genetic associations to sub-phenotypes of allergic disease. For example, two null mutations (R501X and 2282del4) in the gene encoding filaggrin (FLG) are arguably the most consistently associated polymorphisms with risk of AD, but numerous studies have also implicated a role for these mutations in the development of other atopic diseases, such as asthma and rhinitis, suggesting generalizability of FLG mutations to the allergic diathesis. However, it has been argued that the ‘atopic march’ (e.g., the tendency for AD to precede asthma, food allergy and AR) and the fact that ∼70% of severe AD patients also have asthma and AR later in life can account for this overlap⁵³. Similar observations have come from GWAS of allergic diseases. For example, the associations with the ORMDL3 locus has been strongest with childhood asthma⁵⁴, and associations between SNPs in IL1RL1 and IL33 have been strongest for atopic asthma as opposed to non-atopic asthma⁵⁰. From the GWAS performed total serum IgE levels, there has been relatively little overlap with genes contributing to risk of asthma (Table 1).

The Next Generation of Asthma Genetics

Despite its success, discoveries from GWAS to date have contributed relatively little to our understanding of the specific causal genetic mechanisms underlying allergic disease. For example, the cumulative genetic risk of the variants identified to date for asthma through GWAS (for which, among the allergic diseases, the most GWAS have been performed) is <15%³⁵. This is thought to be due, at least in part, to the fact that the most strongly associated SNPs in GWAS are generally not ‘directly causal’, but most likely tag SNPs in linkage disequilibrium (LD) with the true unobserved disease-causing SNPs. Moreover, the vast proportion of GWAS associations (>85%) involve variants in intergenic or intronic regions⁵⁵, which is likely a consequence of the array design; i.e., GWAS arrays are based on tag SNPs for common variants, and coding/exonic variation tends in general to be rare and therefore poorly tagged by a common variant, in contrast to intronic and intergenic regions that have a spectrum of variation that is common. Disappointment in GWAS is compounded by a paradigm shift away from the common disease—common variant hypothesis⁵⁶ towards the role of rare variants (unlikely to be identified by GWAS⁵⁷) in non-Mendelian diseases⁵⁸, particularly with the appreciation that rare variation constitutes the majority of polymorphisms across human populations^39,59.

Resequencing genes in individuals with well-characterized phenotypes is an alternative approach to assess the contribution that both rare and common variants make to disease and overcome the limitations of GWAS. Until recently, Sanger termination sequencing⁶⁰ was the only option for interrogating rare variants, but this approach is costly and cannot be done on a large scale. The emergence of massively parallel, second-generation DNA sequencing in 2005⁶¹ has made resequencing an affordable tool to study genetic variation, and in the past several years has been increasingly used either as a targeted approach to follow-up on specific genetic regions or as an unbiased approach towards gene discovery either by whole exome (WES; ∼30 Mb total) or whole genome sequencing (WGS)⁶². While rare coding variants may have a greater functional impact than common variants, their analysis must consider the low frequency of any variant since it will reduce the power to infer statistical associations (i.e., insufficient numbers of copies of the rare variant allele in a typical dataset). However, this can be overcome by evaluating the collective frequency of rare, nonsynonymous variants within one or more genes, or for a pathway(s), or the functional impact of the discovered variations, such as nonsense substitutions, frameshifts, and splice-site disruptions, that have important a priori evidence compared to other types of changes (reviewed in Ref. ⁶²).

To date, there are limited examples of the application of NGS technology to identify variants associated with risk of allergic disease, although efforts are underway. A recent example of success combined targeted array-based and in-solution enrichment with the SOLiD sequencing platform to accurately and simultaneously detect 161/170 mutations and deletions associated with primary immunodeficiency (PID) disorders ⁶³. NGS has also been applied to the study of airway inflammation, including asthma. A study by Leung et al utilized the next-generation sequencing technique called Roche 454 pyrosequencing on peak asthma association signals found in a large consortium-based study in European white subjects and a small group of Chinese children, and found substantial variation in haplotype structures across the populations, thus supporting the notion of potential sequence variations of asthma loci across different ethnic populations ⁶⁴. WES has been applied to a small family-based study⁶⁵ as well as asthmatics selected at both ends of a phenotype distribution (those with extreme severity phenotypes) ⁶⁶ with limited success, and a large WGS (>1,000 genomes) on asthma is currently underway⁶⁷.

Measuring the Transcriptome in Allergic Disease and its Application to Genetic Studies

Whole genome gene expression profiling, or transcriptomics, is a robust approach towards the quantitative and qualitative characterization of RNA expressed in a biological system. Since the development of synthetic oligonucleotide microarray platforms in 2003⁶⁸, transcriptomic profiling has been widely applied in allergic disease. For asthma and its associated traits alone, dozens of studies focusing on whole blood and target cells of the immune system and tissue from the upper and lower airways have been performed using these conventional platforms (reviewed in Ref. ⁶⁹).

The same robust NGS technology that has recently advanced genetics has similarly transformed transcriptomics. RNA-Sequencing (RNA-Seq) is a more powerful approach to interrogate the transcriptome compared to older microarray technology because of its smaller technical variation⁷⁰ and higher correlation with protein expression⁷¹. RNA-Seq has virtually unlimited dynamic range and permits digital quantification of transcript abundance, assessment of transcript isoforms and alternative splicing^72–74, and it allows for unbiased assembly of transcripts without relying on previous annotation (including non-coding RNAs). To date there are limited examples of applying RNA-Seq technology to allergic disease, but successes include the identification of transcriptomic changes in human airway smooth muscle (ASM) in asthmatics compared to non-asthmatics⁷⁵ and the identification of genes differentially expressed in response to glucocorticosteroid exposure (CRISPLD2⁷⁶, FAM129A and SYNPO2⁷⁷).

While it is ideal to measure the transcriptome of a primary cell specific to the disease of interest (i.e., cells from lung tissue in asthma), this is challenging when considering the large number of samples required given the demands of power. Recently, however, studies have demonstrated the value of focusing on surrogate target tissues/cells in predicting gene expression in tissues/cells that are challenging to access in large numbers (i.e., lung tissue), which have the potential to significantly move the field forward. For example, Poole and colleagues used whole-transcriptome sequencing (RNA-Seq) to demonstrate that the nasal airway epithelium mirrors the bronchial airway, and subsequent RNA sequencing of candidate airway biomarkers confirmed that children with asthma have an altered nasal airway transcriptome compared to healthy controls, and these changes are reflected by differential expression in the bronchial airway ⁷⁸.

Differential gene expression in humans is heritable^79,80 and GWAS of gene expression is an innovative approach for mapping functional non-coding variation. Referred to as expression quantitative trait locus (eQTL) mapping, this approach is predicated on the notion that abundance of a gene transcript (a quantitative trait) is directly modified by genetic polymorphisms in regulatory elements. The added value of eQTL is the ability to identify disease markers identified in GWAS that are also associated with gene transcripts, and several studies have integrated findings from asthma GWAS with cataloged genome-wide gene expression data^81,82, which can result in a ‘gain in power’⁸³. Because of limited access to human primary cell types from large populations, many of the human eQTL studies have focused on convenient and immortalized Epstein-Barr virus transformed lymphoblastoid cell lines (LCLs)^84–86, but this approach has had limited success in mapping eQTLs for more than a few of the known asthma genes. In one of the first asthma eQTL studies, SNPs associated with asthma in a subset of the GABRIEL sample were consistently and strongly associated (P<10⁻²²) with transcript levels of ORMDL3⁴⁵. Hao and colleagues performed an eQTL analysis using lung samples from transplant patients to identify variants affecting gene expression in human lung tissue, then integrated their lung eQTLs with GWAS data from GABRIEL to determine that one of their strongest eQTLs was, similar to the eQTL in LCLs study, a SNP in the chr. 17q21 region⁸². Murphy et al⁸⁷ identified common genetic variants influencing expression of 1,585 genes in peripheral blood CD4+ T cells from 200 asthmatics using conventional microarrays, but they acknowledged power was a major limitation. In mining a catalog of 285 published GWAS, however, they identified significant associations with variants in the ORMDL3 region. When performing tests for association on 6,706 cis-acting expression-associated variants (eSNPs) from a genome-wide eQTL survey of CD4⁺ T cells from asthmatics, the ORMDL3/GSDMB locus held up (P = 2.9 × 10⁻⁸)⁸⁸.

Common Genes in Common Diseases

Several reports have found that allergic diseases such as asthma, rhinitis, conjunctivitis and dermatitis as well as allergic reactions to drugs and foods, are more common in patients with the autoimmune disease systemic lupus erythematosus (SLE)^89–92. Furthermore, bronchial asthma was found to be the most common cause of cough in a small cohort of SLE patients from Bangladesh⁹³ and Taiwan⁹⁴. In addition, the inflammatory gene tumor necrosis factor α (TNFα) was found to be a common genetic risk factor for asthma, and autoimmune diseases juvenile rheumatoid arthritis and SLE⁹⁵. In a more recent study, PCR-based genotyping identified four FCRL3 single nucleotide polymorphisms associated with protection in either juvenile rheumatoid arthritis (JRA) or asthma, but no association was observed with childhood-onset SLE in male Mexican patients. The gene NRF2 has also been associated with various immunological pathologies including RA, acute lung injury, asthma, and emphysema⁹⁶, among others. There is a long-standing observation of common genetic determinants for both asthma and chronic obstructive pulmonary disease (COPD) identified both through candidate gene studies as well as GWAS^97,98. Recently, Hardin and colleagues performed a GWAS focusing specifically on patients from the COPD Gene Study with both asthma and COPD, referred to as the COPD-asthma overlap syndrome, and identified associations with variants in genes (i.e., GPR65) unique to this sub-phenotype⁹⁹. Finally, there is a large body of research associated with the ‘hygiene hypothesis’¹⁰⁰ addressing the potential beneficial role of microbial exposures for later development of asthma and allergies. Specifically, the underlying immunological mechanisms and the type of infectious/microbial stimuli relevant to helminth infection (i.e., schistosomiasis) are the same mechanisms that promote the Th2-mediated response in allergic disease^101,102, and common genetic mechanisms underlie both schistosomiasis and asthma have been reported from linkage and candidate gene studies¹⁰².

Other Omics and Allergic Disease

“Omics” refers to an experimental design in which large-scale datasets are acquired from a complete class of biomolecules with the aim of identifying the functional or pathological mechanisms of disease¹⁰³. Such data-dense technologies include: (1) DNA in the context of complete genomics; (2) gene regulation technologies (epigenomics); (3) global protein and/or post-transcriptional modifications (proteomics); and (4) all cellular metabolites (metabolomics) ¹⁰⁴.

Transcriptomics extended to microRNA is another burgeoning field. Several miRNAs have been identified as distinct profiles for the development and status of asthma, as well as other allergic phenotypes^105,106. Approximately 200 miRNAs are known to be altered in steroid naïve asthmatics, establishing a link between abnormal miRNA expression in asthmatic patients and inflammation^107–109. High-throughput data combined with sequence-based miRNA predictions have been successfully applied^110–114, and more recently, a transcriptome study on miRNA-long non coding RNA interactions suggests better understanding of lung disease regulation and progression¹¹⁵. NGS has been utilized to study microRNA expression and interactions with the phosphoinositide 3-kinase (PI3K) pathway in primary human airway smooth muscle (HASM) cells¹¹⁶.

Concordance rates for asthma and allergies of only ∼50% among monozygotic twins suggest differences in exposure to environmental triggers are critical in disease expression^2,117,118, and it has been demonstrated that genes and environmental factors contribute equally to asthma and its associated traits such as tIgE³. Similar to the other allergic diseases, the prevalence of asthma has increased dramatically within the 2–3 decades in relation to the deterioration of the environment, favoring a significant contribution of environmental factors¹¹⁹. Added to this complexity is the observation that associations with alleles at candidate genes and interactions between these genes might only be observed among certain subpopulations despite nearly identical environmental exposures and similar genetic backgrounds. For example, the CD14(-260)C>T variant was associated with low tIgE in school children living in urban/suburban Tucson, AZ¹²⁰, but the opposite association was reported in a farming community¹²¹. Alternatively, it has been shown that this same variant depends on the dose of endotoxin from household dust among African-ancestry asthmatics living in the tropics¹²², suggesting the role of endotoxin in allergic disease may be due to the combination of susceptibility genes and exposure. A large body of evidence implicates in utero and early life environmental tobacco smoke (ETS) exposure leads to impaired lung function and increased risk of asthma^123–126, and ETS exposure increases strength of the association between markers in candidate genes and atopic asthma^127–129. Indeed, environmental exposures such as smoking, air pollution and stress have been shown to cause changes in epigenetic modifications of genes as well as altered microRNA expression ¹³⁰.

Immune responses in allergic disease are dominantly initiated by the release of cytokines such as interleukin-4 (IL4), IL5 and IL13, which activate type 2 helper T cells (TH2) resulting in a decrease of TH1 cytokines and impaired regulatory T cell function, and up or down regulation of DNA methylation on Th-1/Th-2 cytokine genes may affect the sensitization of experimental asthma¹³¹. In addition, epigenetic changes in immune cells such as T cells, B cells, mast cells and dendritic cells exposed to environmental factors have also been shown to be associated with asthma¹³². A recent study found that DNA methylation in the β-2 adrenergic receptor (ADRB2) gene is associated with decreased asthma severity¹³³. In addition, an asthma mouse model found that microRNAs targeted genes involved in inflammatory responses and tissue remodeling, and demethylation status in the promoter of the IFN-γ changed in response to chronic antigen sensitization¹³⁴.

Environmental stimuli have been shown to directly influence epigenetic modifications, and thus epigenetic regulation may play a role in immune-mediated lung diseases like asthma. Epigenetic regulation maintains tolerance to self-antigens. Thus, abnormal epigenetic activity may lead to a deregulated immune response and thus an immune disorder¹³⁵. Epigenomics allows for the study of gene regulation at the chromosomal level using DNA methylation and CHIP technologies. As an example, 870 genes are differentially methylated in idiopathic pulmonary fibrosis (IPF) tissues¹³⁶, and changes in miRNAs and fibroblast signature for genes are known to regulate the extracellular matrix in IPF^137,138. While methylation decreases gene expression, acetylation of histones relaxes chromatin facilitating gene transcription and increasing expression. A recent study has implicated histone modifications in the decrease of Fas expression as well as resistance to apoptosis in fibrotic lung fibroblasts¹³⁹.

A novel example of this technology is a study in which methylated DNA immunoprecipitation-next generation sequencing (MeDIP-seq) on lung tissue DNA from saline and house dust mite (HDM)-exposed mice was performed and researchers found that chronic exposure to HDM increased airway reactivity and inflammation, as interpreted through increases in IL-4, IL-5 and serum IgE levels, resulting in structural remodeling and hyperresponsiveness consistent with allergic disease. In addition, mice that received HDM exposure had global changes in methylation and hydroxymethylation of approximately 213 genes, with TGFβ2 and SMAD3 having the most connected network¹⁴⁰. These findings demonstrate how allergen exposure could trigger epigenetic changes in the lung genome.

Clinical Implications & Personalized Medicine

Arguably the ultimate goal of genetic studies of allergic disease is to better match individualized treatments to specific genotypes to improve therapeutic outcomes and minimize side effects. For example, despite the relative success of conventional asthma therapies such as inhaled beta agonists and glucocorticoids, most cause adverse side effects^141–143 and a subset of asthmatics are refractory to anti-asthma therapies resulting in significant morbidity as well as a significant financial burden^144–146. Genetic variation determines drug response through various mechanisms including pharmacodynamics mechanisms, which determine drug metabolism¹⁴⁷.

Recent GWAS and studies of candidate genes related to the β2-adrenergic receptor pathway have attempted to identify specific variants associated with the response to inhaled beta agonists^148–150. The Arg¹⁶ allele in ADRB2 has been associated with greater post-bronchodilator FEV1 response to SABA asthma therapy in asthmatic children^151,152, while the Gly¹⁶ variant has been associated with changes in peak flow rate (PEFR)^153–155. In contrast, the Arg¹⁶ allele has been associated with worsening asthma symptom scores with LABA therapy compared to Gly¹⁶ homozygotes ¹⁵⁶. Other studies show no difference between the ADRB2 alleles and asthma symptoms after LABA therapy ^157,158. Further pharmacogenetic studies may achieve a more definitive characterization of the role of Gly¹⁶Arg after beta agonist exposure and determine whether receptor kinetics or pro inflammatory effects play a role in the contrasting effects of the genotypes. Additional candidate genes found to be associated with altered beta agonist response in asthmatic children include ADCY9 ¹⁵⁰ and ARG1 ¹⁵⁹ with FEV1 change, and CRH2 ¹⁴⁸ and SPATS2L ¹⁴⁹ with bronchodilator response. Additional candidate gene studies have also demonstrated altered asthma phenotypes in response to glucocorticoid therapies including CRH1 ¹⁶⁰, STIP1 ¹⁶¹, TBX21 ^162,163, ADCY9^{150 164}, and ORDML3 ¹⁶⁵.

Future Considerations/Summary

While GWAS has yielded promising results in the field of allergic disease, association does not imply biological functionality, and follow-up studies are needed to translate initial findings into the biological insights that ultimately will advance prognostics, diagnostics and therapeutics. While the vast amount of genomic data that is now available for a plethora of complex diseases, including allergic disease, has certainly facilitated follow-up association analyses to explore new hypotheses, meta-analyses, and replication of novel findings¹⁶⁶, the scientific community is facing a ‘big data’ crisis¹⁶⁷, as the size of genomic data sets today has begun to overwhelm the existing infrastructure and resources that allow researchers to share or use these data. For the genetics of allergic diseases specifically, there is increasing awareness of the need to design studies that are more inclusive of racially and ethnically diverse study participants¹⁶⁸. Consider that, in the field of pharmacogenetics, it has been clearly demonstrated that, as an example, African American asthmatics have an increased likelihood for treatment failures and overall differential response to treatment that may be caused by genetic variants specific to their ancestry ^169,170. Each of these needs will undoubtedly be addressed as clinicians and scientists in the field continue to move in a direction of collaboration and an appreciation for a multi-disciplinary approach, attributes that have already pushed the genetics of allergic disease into the genomic revolution, with promises of improved outcome for the patient.

KEY POINTS.

• Nearly 100 asthma genes/loci in addition to multiple genes/loci for AD, AR and IgE have been identified by genome-wide association studies (GWAS)

• Next generation sequencing (NGS) strategies are increasingly being used to hone in on the causal variants associated with allergic diseases

• A goal of the genetics of allergic disease is to better match individualized treatments to specific genotypes to improve therapeutic outcomes and minimize side effects

Appendix A

A summary of genome-wide association studies (GWAS) performed on allergic diseases (p-values on the discovery sample p<10-⁵).

Population	Location	Reported gene	Adjacent gene (L,R)	References
Asthma
European	1p13.1	IGSF3	CD58, MIR320B1	Ding et al 2013 1
European	1q25.3	XPR1	ACBD6, KIAA1614	Ding et al 20131
European	1q44	C1orf100	CEP170, HNRNPU	Forno et al 20122
European	1q21.3	IL6R	SHE, LOC101928101	Ferreira et al 20113
Mixed Ethnicities	1q23.1	PYHIN1	IFI16, LOC646377	Torgerson et al 20114
Mixed Ethnicities	1q21.3	CRCT1	LCE5A, LCE3E	Torgerson et al 20114
European	1q31.3	DENND1B	CRB1, C10rf53	Sleiman et al 20105
Korean	2p22.2	CRIM1	LOC10028911, FEZ2	Kim et al 20136
Korean	2q36.2	DOCK10	CUL3, MIR4439	Kim et al 20136
European	2p22.1	Intergenic	THUMPD2, SLC8A1-AS1	Ding et al 20131
European	2q34	CPS1	LOC102724820, ERBB4	Melen et al 20137
European	2p23.3	ADCY3	NCOA1, DNAJC27-AS1	Melen et al 20137
European	2p23.3	ADCY3	PTRHD1, DNAJC27	Melen et al 20137
European	2p23.3	EFR3B	DNAJC27, DNMT3A	Melen et al 20137
European	2p23.3	Intergenic	ADCY3, DNAJC27	Melen et al 20137
European	2q12.1	IL1RL1	IL1R1, IL18RAP	Ramasamy et al 20128
European	2q33.1	SPATS2L	TYW5, SGOL2	Himes et al 20129
European	2q12.1	IL1RL1, IL18R1	IL1R2, IL18RAP	Wan et al 201210
Mixed Ethnicities	2q12.1	IL1RL1	IL1R1, IL18RAP	Torgerson et al 20114
European	2q12.1	IL18R1	IL1RL1, IL18RAP	Moffatt et al 201011
European	3q13.2	ATG3	BTLA, SLC3A5	Ding et al 20131
European	3p22.3	Intergenic	LOC101928135, ARPP21	Ding et al 20131
European	3q26.32	Intergenic	LOC102724550, KCNB2	Ding et al 20131
European	3q12.2	ABI3BP	TFG, IMPG2	Ding et al 20131
European	3p26.2	IL5RA	CNTN4, LRRN1	Forno et al 20122
Korean	4q26	SYNPO2	SEC24D, MYOZ2	Kim JH et al 20136
European	4q12	Intergenic	IGFBP7, LPHN3	Ding et al 20131
European	4p14	KLHL5	TMEM156, WDR19	Ding et al 20131
European	4p15.1	Intergenic	PCDH7, ARAP2	Melen et al 20137
Japanese	4q31.21	LOC729675	INPP4B, USP38	Hirota et al 201112
Japanese	4q31.21	GAB1	USP38, SMARCA5	Hirota et al 201112
European	5q31.1	C5orf56	SLC22A5, IRF1	Wan et al 201210
European	5q31.3	NDFIP1	GNPDA1, NDFIP1	Wan et al 201210
Japanese	5q22.1	TSLP	SLC25A46, WDR36	Hirota et al 201112
Mixed Ethnicities	5q22.1	TSLP	SLC25A46, WDR36	Torgerson et al 20114
European	5q31.1	SLC22A5	LOC553103, C5orf56	Moffatt et al 201011
European	5q31.1	IL13	RAD50, IL4	Moffatt et al 201011
European	5q31.1	RAD50	IL5, IL13	Li et al 201013
European	5q12.1	PDE4D	RAB3C, PART1	Himes et al 200914
European	6p21.1	Intergenic	CDC5L, SUPT3H	Ding et al 20131
European	6q21	Intergenic	RFPL4B, LINC01268	Ding et al 20131
European	6p12.3	AL139097.1	TFA2B, PKHD1	Melen et al 20137
European	6p21.32	HLA-DQA1	HLA-DRB1, HLA-DQB1	Lasky-Su et al 201215
Korean	6p21.32	HLA-DPB1	HLA-DPA1, HLA-DPB2	Park et al 201316
European	6p21.32	BTNL2	HCG23, HLA-DRA	Ramasamy et al 20128
European	6q27	T	LINC00602, PRR18	Tantisira et al 201217
Japanese	6p21.32	PBX2	AGER, GPSM3	Hirota et al 201112
Japanese	6p21.32	NOTCH4	GPSM2, C6orf10	Hirota et al 201112
Japanese	6p21.32	C6orf10	NOTCH4, HCG23	Hirota et al 201112
Japanese	6p21.32	BTNL2	HCG23, HLA-DRA	Hirota et al 201112
Japanese	6p21.32	HLA-DRA	BTNL2, HLA-DRB5	Hirota et al 201112
Japanese	6p21.32	HLA-DQB1	HLA-DQA1, HLA-DQA2	Hirota et al 201112
Japanese	6p21.32	HLA-DQA2	HLA-DQB1, HLA-DQB2	Hirota et al 201112
Japanese	6p21.32	HLA-DOA	BRD2, HLA-DPA1	Hirota et al 201112
Japanese	6p21.32	HLA-DPB1	HLA-DPA1, HLA-DPB2	Noguchi et al 201118
European	6p21.32	HLA-DQB1	HLA-DQA1, HLA-DQA2	Moffatt et al 201011
European	7p15.3	Intergenic	NPY, STK31	Ding et al 20131
European	7q32.3	MKLN1	LINC-PINT, PODXL	Ding et al 20131
Korean	8q11.23	OPRK1	NPBWR1, ATP6V1H	Kim et al 20136
European	8p12	Intergenic	DUSP26, UNC5D	Ding et al 20131
European	8q24.23	COL22A1	FAM135B, KCNK9	Duan et al 201419
Japanese	8q24.11	SLC30A8	AARD, MED30	Noguchi et al 201118
Korean	9p13.3	TLN1	TPM2, MIR6852	Kim JH et al 20136
European	9p23	Intergenic	PTPRD-AS2, TYRP1	Ding et al 20131
European	9q21.33	Intergenic	ZCCHC6, GAS1	Ding et al 20131
European	9p22.1	SLC24A2	ACER2, MLLT3	Melen et al 20137
European	9q33.3	DENND1A	CRB2, LHX2	Melen et al 20137
European	9p21.1	ACO1	LINX01242, DDX58	Wan et al 201210
Mixed Ethnicities	9p24.1	IL33	RANBP6, TPD52L3	Torgerson et al 20114
European	9p24.1	IL33	RANBP6, TPD52L3	Moffatt et al 201011
Mexican	9q21.31	TLE4, CHCHD9	LOC101927450, LOC101927477	Hancock et al 200920
Korean	9p21.3	Intergenic	SLC24A2, MLLT3	Kim SH et al 200921
European	10q24.2	HPSE2	HPS1, CNNM1	Ding et al 20131
European	10q22.1	PSAP	CDH23, CHST3	Ding et al 20131
European	10p15.1	PRKCQ	LOC399715, PRKCQ-AS1	Melen et al 20137
European	10q26.11	EMX2	PDZD8, RAB11FIP2	Li et al 201322
European	10p15.1	PRKCQ	LOC101927964, LINC00702	Duan et al 201419
European	10q21.1	PRKG1	A1CF, PRKG1-AS1	Ferreira et al 20113
Japanese	10p14	LOC338591	LINC00708, LOC101928272	Hirota et al 201112
Korean	10q21.3	CTNNA3	LOC101928913,	Kim SH et al 200921
Korean	11q24.1	OR6X1	ZNF202, OR6M1	Kim JH et al 20136
European	11q13.4	P2RY2	FCHSD2, P2RY2	Melen et al 20137
European	11q24.2	NR	LOC101929497, ETS1	Forno et al 20122
European	11q13.5	LRRC32	C11orf30, GUCY2EP	Ferreira et al 20113
Mixed ethnicities	11q23.2	C11orf71	LOC101928940, RBM7	Torgerson et al 20114
Japanese	12q13.2	CDK2	PMEL, RAB5B	Hirota et al 201112
Japanese	12q13.2	IKZF4	SUOX, RPS26	Hirota et al 201112
European	13q13.1	STARD13, RP11-81F11.3	KL, RFC3	Melen et al 20137
European	13q13.3	NR	MIR548F5, DCLK1	Forno et al 20122
European	13q21.31	PCDH20	MIR3169, LINC00358	Ferreira et al 20113
Korean	13q12.13	Intergenic	GPR12, USP12	Kim SH et al 200921
Korean	14q32.2	LOC730217	C14orf64, C14orf177	Kim JH et al 20136
European	15q22.33	SMAD3	SMAD6, AAGAB	Moffatt et al 201011
European	15q22.2	RORA	LOC101928784, VPS13C	Moffatt et al 201011
European	15q21.2	SCG3	DMXL2, LYSMD2	Li et al 201013
Korean	16q23.3	CDH13	MPH0SPH6, MLYCD	Kim JH et al 20136
European	17q21.32	Intergenic	MIR196A1, PRAC1	Melen et al 20137
European	17q21.32	Intergenic	MIR196A1, PRAC1	Melen et al 20137
European	17q12	ORMDL3	GSDMB, LRRC3C	Wan et al 201210
European	17p12	NR	HS3ST3A1, COX10-AS1	Forno et al 20122
Mixed ethnicities	17q12	GSDMB	ZPBP2, ORMDL3	Torgerson et al 20114
European	17q12	ORMDL3	GSDMB, LRRC3C	Ferreira et al 201123
European	17q12	GSDMB	ZPBP2, ORMDL3	Moffatt et al 201011
European	17q21.1	GSDMA	LRRC3C, PSMD3	Moffatt et al 201011
European	17q12	ORMDL3	GSDMB, LRRC3C	Moffatt et al 201011
European	18p11.31	LPIN2	EMILIN2, MYOM1	Melen et al 20137
European	18p11.32	YES1	ENOSF1, ADCYAP1	Li et al 201322
Korean	19q13.43	ZNF71	ZNF470, SMIM17	Kim JH et al 20136
European	19p13.11	IL12RB1	AARDC2, MAST3	Li et al 201322
European	19q13.42	ZNF665	ZNF347, ZNF818P	Wan et al 201210
European	20p12.3	Intergenic	MIR8062, HA01	Ding et al 20131
European	20q13.2	Intergenic	LOC101927700, TSHZ2	Melen et al 20137
European	20p13	KIAA1271	AP5S1, MAVS	Li et al 201013
European	22q13.31	UPK3A	NUP50, FAM118A	Li et al 201322
European	22q12.3	IL2RB	TMPRSS6, C1QTNF6	Moffatt et al 201011
European	NR	Intergenic		Wan et al 201210
Atopic Dermatitis
European	1q21.3	FLG	HRNR, FLG2	Weidinger et al 201324
Chinese	1q21.3	FLG	HRNR, FLG2	Sun et al 201125
Japanese	2q12.1	IL1RL1, IL18R1, IL18RAP	IL1R1, IL18RAP, IL1R2	Hirota et al 201226
Japanese	2q13	LOC100505634	BCL2L11, MIR4435-1	Hirota et al 201226
Japanese	3p22.3	GLB1	CCR4, SUSD5	Hirota et al 201226
Japanese	3q13.2	CCDC80	LINC01279, LOC101929694	Hirota et al 201226
European	5q31.1	IL13	RAD50, IL4	Weidinger et al 201324
Japanese	5q31.1	IL13	RAD50, IL4	Hirota et al 201226
European	5q31.1	IL13	RAD50, IL4	Paternoster et al 201227
European	6p21.33	TNXB	CYP21A2, ATF6B	Weidinger et al 201324
Japanese	6p21.33	HLA-C	HCG27, HLA-B	Hirota et al 201226
Japanese	6p21.32	GPSM3	PBX2, NOTCH4	Hirota et al 201226
Japanese	6p21.32	C6orf10	NOTCH4, HCG23	Hirota et al 201226
European	6p21.33	BAT1	MCCD1, DDX39B	Paternoster et al 201227
Japanese	7p22.2	CARD11	GNA12, SDK1	Hirota et al 201226
Japanese	8q24.21	MIR1208	PVT1, LINC00977	Hirota et al 201226
European	8q21.13	ZBTB10	MIR5708, ZNF704	Paternoster et al 201227
Japanese	10q21.2	ZNF365	LOC283045, EGR2	Hirota et al 201226
Japanese	10q21.3	ADO, EGR2	ZNF365, NRBF2	Hirota et al 201226
European	11q13.5	C11orf30	LOC100506127, LRRC32	Weidinger et al 201324
Japanese	11p15.4	OR10A3, NLRP10	OR10A3, NLRP10	Hirota et al 201226
Japanese	11q13.5	C11orf30	LOC100506127, LRRC32	Hirota et al 201226
Japanese	11q13.1	OVOL1	AP5B1, SNX32	Hirota et al 201226
European	11q13.1	OVOL1	AP5B1, SNX32	Paternoster et al 201227
European	11q13.5	C11orf30	LOC100506127, LRRC32	200928
Japanese	16p13.13	CLEC16A	DEXI, SOCS1	Hirota et al 201226
European	19p13.2	ACTL9	ADAMTS10, OR2Z1	Paternoster et al 201227
Japanese	20q13.2	CYP24A1, PFDN4	CYP24A1, PFDN4	Hirota et al 201226
European	22q12.3	NCF4	PVALB, CSF2RB	Paternoster et al 201227
Atopy
European	2p21	SGK493	C2orf91, PKDCC	Castro-Giner et al 200929
Allergic Rhinitis
European	1p36.13	CROCC	MIR3675, MFAP2	Ramasamy et al 201130
European	5q22.1	TMEM232, SLCA25A46	LOC100289673, TSLP	Ramasamy et al 201130
European	5q22.1 5q23.1	TSLP	SLC25A46, WDR36 LOC101927190,	Ramasamy et al 201130
European		SEMA6A	LOC102467223	Ramasamy et al 201130
European	7p14.1	GLI3	INHBA-AS1, LINC01448	Ramasamy et al 201130
	11q13.5	C11orf30,	LOC100506127,
European		LRRC32	GUCY2EP	Ramasamy et al 201130
European	14q23.1	PPM1A, DHRS7	PCNXL4, C14orf39	Ramasamy et al 201130
European	16p13.13	CLEC16A	DEXI, SOCS1	Ramasamy et al 201130
European	20p11.21	ENTPD6	LOC101926889, PYGB	Ramasamy et al 201130
Total & Specific IgE
European	1p32.3	EPS15	TTC39A, OSBPL9	Ramasamy et al 201130
European	1q23.2	DARC	CADM3-AS1, ACKR1	Granada et al 201231
European	1q23.2	FCER1A	ACKR1, OR10J3	Weidinger et al 200832
European	1q23.2	FCER1A	Mus Olfr418-ps1,	Granada et al 201231
European	1q23.2	OR10J3	FCER1A, OR10J1	Granada et al 201231
European	1q25.2	ABL2	TOR3A, SOAT1	Ramasamy et al 201130
Korean	2p22.2	CRIM1	LOC100288911, FEZ2	Kim et al 20136
European	2p25.1	ID2	LOC100506299, MBOAT2	Granada et al 201231
Korean	2q36.2	DOCK10	CUL3, NYAP2	Kim et al 20136
Mixed ethnicities	3p14.1	SUCLG2	MIR4272, SUCLG2-AS1	Levin et al 201333
European	3q22.1	TMEM108	NPHP3-AS1, BFSP2	Ramasamy et al 201130
European	3q28	LPP	BCL6, TPRG1-AS1	Granada et al 201231
Korean	4q26	SYNPO2	SEC240, MYOZ2	Kim et al 20136
European	4q27	IL2	ADAD1, IL21	Ramasamy et al 201130
European	5p15.2	DNAH5 TMEM232,	LINC01194, TRIO	Ramasamy et al 201130
European	5q22.1	SLCA25A46	LOC100289673, TSLP	Ramasamy et al 201130
European	5q31.1	IL13	BC042122, IL4	Granada et al 201231
European	5q31.1	RAD50	IL5, IL13	Weidinger et al 200832
European	6p21.32	HLA region	HLA-DQB1, HLADQA2	Ramasamy et al 201130
European	6p21.32	HLA-DQA2	HLA-DQB1, HLA-DQB2	G ranada et al 201231
Mixed ethnicities	6p21.32	HLA-DQA2	HLA-DQB1, HLA-DQB2	Levin et al 201333
Mixed ethnicities	6p21.32	HLA-DQB1	HLA-DQA1, HLADQA2	Levin et al 201333
European	6p22.1	HLA-G	LOC554223, HLA-H	Granada et al 201231
European	6p22.1	HLA-A	HCG4B, HCG9	Granada et al 201231
Korean	8q11.23	OPRK1	NPBWR1, ATP6V1H	Kim et al 20136
Korean	9p13.3	TLN1	TPM2, CREB3	Kim et al 20136
Korean	11q24.1	OR6X1	ZNF202, ORM1	Kim et al 20136
European	12q13.3	STAT6, NAB2	TMEM194A, LRP1	Granada et al 201231
Korean	14q32.2	LOC730217	C14orf64, C14orf177	Kim et al 20136
European	16p12.1	IL4R	FLJ21408, IL21R	Granada et al 201231
European	16p13.2	Intergenic	MIR548X, MIR7641-2	Ramasamy et al 201130
Mixed ethnicities	16q22.1	WWP2	NOB1, PDXDC2P	Levin et al 201333
Korean	16q23.3	CDH13	MPHOSPH 6, LOC102724163	Kim et al 20136
Korean	19q13.43	ZNF71	ZNF470, SMIM17	Kim et al 20136
Airway hyperresponsiveness
European	2q36.3	AGFG1	TM4SF20, C2orf83	Himes et al 201334

Mixed ethnicities = African American/African Caribbean, Latino, European ancestry