| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review |
1 The Center for Applied Genomics and Division of Human Genetics, the Abramson Research Center of the Joseph Stokes Jr. Research Institute, The Children’s Hospital of Philadelphia; 2 Department of Pediatrics, The University of Pennsylvania School of Medicine, Philadelphia, PA.
Address correspondence to the authors at: Struan F.A. Grant, Center for Applied Genomics, 1216F Abramson Research Center, 3615 Civic Center Blvd., Philadelphia, PA 19104-4318. Fax 267-426-0363; e-mail grants{at}chop.edu; or Hakon Hakonarson, Center for Applied Genomics, 1216E Abramson Research Center, 3615 Civic Center Blvd., Philadelphia, PA 19104-4318. Fax 267-426-0363; e-mail hakonarson{at}chop.edu
 |
Abstract |
|---|
 Top Abstract IntroductionDiscussionReferences |
|---|
Content: The GWA approach serves the critical need for a comprehensive and unbiased strategy to identify causal genes related to complex disease, and is rapidly replacing the more traditional candidate gene studies and microsatellite-based linkage mapping approaches that have dominated gene discovery attempts for common diseases. As a consequence of employing array-based technologies, over the last 3 years dramatic discoveries of key variants involved in multiple complex diseases and related traits have been reported in the top scientific literature and, most importantly, have been largely replicated by independent investigator groups. As a consequence, several novel genes have been identified, most notably in the metabolic, cardiovascular, autoimmune, and oncology disease areas, that are clearly rooted in the biology of these disorders. These discoveries have opened up new avenues for investigators to address novel molecular pathways that were not previously linked to or thought of in relation with these diseases.
Summary: This review provides a synopsis of recent advances and what we may expect to still emerge from this field.
|
Introduction |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
This progress was made possible by key developments in human genomics over the last decade. The completion of the human genome sequence was a crucial prerequisite. The subsequent International HapMap Project (1)(2) yielded insights into human genetic diversity. During this same period advances in single-base extension biochemistry and hybridization/detection to synthetic oligonucleotides have made it possible to accurately genotype hundreds of thousands of genetic variants, known as single nucleotide polymorphisms (SNPs),1 in tandem (3).
The spirit that drove the genome sequencing projects has lived on in the era of genome-wide association (GWA) studies, where the intense competition between groups has driven forward genetic research at an impressive pace in the last 36 months, producing dramatic genetic associations that often explain large proportions of disease pathogenesis. These studies have been subsequently robustly replicated, leading to substantial consensus in the genetics community, for the first time, on what variants really underlie many phenotypes.
previous genetic approaches
In the past, investigators generally worked with their favorite gene, whose identity derived from a particular biological hypothesis on the pathogenesis of a given complex disease. This candidate gene approach was understandable, as there was biological reasoning to focus on a given gene. In addition, technology has not until very recently allowed for questions that involved querying across the entire genome. Candidate gene studies have been plagued with the "winner’s curse" (4), in which an initial report of association is made but is not subsequently replicated by independent investigators.
Overall, family studies using linkage methodologies conducted to date have achieved only limited success in identifying genetic determinants of complex disease for various reasons, importantly including the generic problem that the linkage analysis approach is generally poor in identifying common genetic variants that have modest effects.
Therefore it is discouraging, but not entirely surprising, that to date relatively few of the many variants that were reported to influence the risk for common diseases before the advent of GWA have been replicated (4).
genome-wide association
By taking advantage of the knowledge gained from the HapMap project, which revealed that the genome is organized into discrete linkage disequilibrium (LD) blocks with limited haplotype diversity within each block, the minimal set of SNPs necessary for detecting all major haplotypes can be determined, thus improving genotyping accuracy and reducing genotyping cost. As such, genome-wide genotyping of hundreds of thousands of SNPs can now be readily achieved in an efficient and highly accurate manner to capture (tag) the bulk of the diversity in the genome (3). Therefore such a strategy has a very high likelihood of tagging a disease causing mutation if the study is sufficiently powered.
Currently there are 2 major vendors of this high-throughput genome-wide SNP genotyping technology, namely Affymetrix and Illumina. Assays from both companies can scale to as many SNPs as can be represented on the array and are readily adaptable to automation.
With the Affymetrix technology, input DNA is digested with the restriction enzymes NspI and StyI and ligated to adaptors that recognize the 4-bp overhangs. All fragments resulting from restriction enzyme digestion, regardless of size, represent substrates for adaptor ligation; a generic primer that recognizes the adaptor sequence is used to amplify adaptor-ligated DNA fragments. PCR is then used to preferentially amplify fragments in the 200- to 1100-bp size range. Amplification products for each restriction enzyme digest are combined and purified using polystyrene beads, and the amplified DNA is fragmented, labeled, and hybridized to a GeneChip®. The samples are scanned, and the genotypes are scored using software developed by the company.
The Illumina technology represents a random assembly of oligonucleotide-containing beads in microwells located at the end of optical fiber bundles (3). The InfiniumTM technology uses single-tube whole-genome amplification followed by primer extension without PCR or ligation. The 3-day protocol begins with genomic DNA being amplified 1000- to 1500-fold and fragmented to approximately 300 to 600 bp. This genome representation is precipitated, resuspended, and hybridized onto a BeadChip. Single-base extension (SBE) uses a single probe sequence approximately 50 bp long designed to hybridize immediately adjacent to the SNP query site. After targeted hybridization to the bead array, the arrayed SNP locus-specific primers (attached to beads) are extended with a single hapten-labeled dideoxynucleotide in the SBE reaction. The haptens are detected by a multilayer immunohistochemical sandwich assay, and the genotypes are called using software developed by the company.
At the Center for Applied Genomics at the Children’s Hospital of Philadelphia, we use genome-wide SNP genotyping technology from both vendors, with the specific goal of determining the key genetic factors underlying pediatric diseases (5). Our experience, based on the genotyping of approximately 60 000 DNA samples using high-density SNP arrays, is that the cost for an association study in terms of reagents and chips is comparable between the 2 vendors. The cost per SNP chip has dropped substantially over the past 2 years, making GWA studies affordable to most research groups. It is now also feasible to genotype >1 individual on the same chip. The relatively recent availability of the Illumina’s duo and quad chips, where 2 and 4 individuals, respectively, can be genotyped in parallel on the same chip, has further reduced processing costs and increased throughput.
It is reasonable to estimate that the cost to genotype an individual for a GWA study on either the Affymetrix or Illumina high-throughput SNP arrays may be in the range of $400 to $800 (US), subject to the genotyping platform used and the number of SNPs genotyped. The first chips launched for this purpose were in the 50K- to 100K-SNP range, but as technology has developed so has the number of SNPs on the platform. The Illumina HH317K and HH550K BeadChips were designed based on tag-SNP selection, whereas the Affymetrix platform available at that time, the 5.0 GeneChip, was limited by restriction enzyme chemistry so the information content with respect to tagging was lower. However, both vendors now offer in excess of 1 million SNPs on each of their chip products, so the difference in information content is negligible. Furthermore, the SNP density is now also sufficient to detect copy number variations in the genome, which are known to play a role in the pathogenesis of complex disease. Both companies provide user-friendly software for downstream appraisal of the data generated, namely BeadStudio from Illumina and Birdseed from Affymetrix.
High accuracy and high yields with the SNP genotyping platform of choice are crucial to decrease false-positive signals resulting from disparities in the quality and information content of the data between patients and controls. This problem is much more acute in GWA studies, which have a number of tests that are 3 to 4 orders of magnitude greater than most candidate gene studies. Besides the discovery groups, an adequately designed study needs independent groups of patients and controls for replication and validation studies.
The relative size of the discovery cohort is often based on how strong the evidence is for a genetic component to the disease. For example, the sibling risk for type 1 diabetes is about 15 times that of the general population, whereas for type 2 diabetes it is only approximately 3.5-fold. This approach has been undermined, however, as more GWA runs have been reported with a number of strong associations for diseases with low heritability, such as type 2 diabetes, and a paucity of signals in high heritability traits, such as body mass index (BMI). As such, an investigator should design a GWA study based on classic power calculations, for example using CaTS (http://www.sph.umich.edu/csg/abecasis/CaTS), assuming a particular disease prevalence for given models of relative risk and allele frequency.
For all patient and control comparisons, especially in diverse communities such as the United States, it is crucial to match control groups to patient groups as much as possible in terms of genetic background. A spurious association may be obtained or true association overlooked if allele frequencies are represented unequally in different cohorts. Population structure can impact the most carefully designed studies and needs to be assessed to make reliable conclusions in association studies (6)(7)(8). Most study designs sample individuals from groups that share the same nationality or self-reported ethnic background, with the assumption that no substructure exists within such groups. Results indicate, however, that self-reported ethnicity is often inaccurate. Ancestry informative markers (AIMs), which are chosen from 3 populations and have relatively common alleles, based on the HapMap project, can be used to assess ethnicity. One can further use the information generated on the SNP chips to more thoroughly evaluate evidence for population stratification using a combination of 3 strategies: a) the structure approach (9), which attempts to assign individuals to a small number of underlying clusters; b) genomic control methods (10), which check for inflation of null test statistics; and c) the Eigenstrat method (11), which uses principal components to model ancestry differences between individuals. Both the structure and the genomic control methods require a set of markers in linkage equilibrium. In contrast, Eigenstrat can be applied to hundreds of thousands of markers, some of which might be in linkage disequilibrium.
Once investigators are content with a design and have addressed any potential population stratification issues, they are ready to carry out an association analysis. In a standard GWA analysis, the data are simply queried for which alleles of all the SNPs on the chip are significantly over- or underrepresented in the patients. A commonly used software package for this is Plink (http://pngu.mgh.harvard.edu/purcell/plink), and the resulting data can be visualized with respect to P value, Q-Q plots, and linkage disequilibrium plots using software such as WGAViewer from Duke University (http://www.genome.duke.edu/centers/pg2/downloads/wgaviewer.php) or Haploview from the Broad Institute (http://www.broad.mit.edu/mpg/haploview).
There are continuing concerns regarding the performance of association studies in complex traits, and independent replication efforts are now considered mandatory (12). With the many errors and biases that can blight any individual study, replication by others can ensure that the original findings are robust and can also provide a more accurate estimate of the likely effect size. The replication cohorts in original published reports tend to be larger than the discovery cohort; this is to get a better handle on the effect size. Generally a customized genotyping platform is employed to follow up on these cherry-picked signals. It is apparent from the way these studies have been published that the first reports tend to describe phase I results, where the identification of the strong ("low-hanging fruit") signals are described, i.e., they are significant genome-wide (P approximately 10–7 to 10–8; corrected P < 0.05 for multiple testing). These represent a handful of signals that are followed up with platforms, such as TaqMan from Applied BioSystems or iPlex from Sequenom, which are designed to genotype a small number of SNPs on a large number of individuals. In phase II, those signals that were just below the strict threshold for significance, which can number in the hundreds, can be genotyped on a platform such as Illumina’s GoldenGate, where a few hundred SNPs can be genotyped in parallel on a large volume of samples. In both replication models, the threshold for significance in this effort is set lower, as there is substantially less testing being carried out.
Thus, the GWA approach serves the critical need for a more comprehensive and unbiased strategy to identify causal genes related to complex disease. Here we outline, by medical discipline, the advances made in the arena of complex disease genetics, where some historic findings have already been accomplished over the last 3 years, and perhaps more importantly, generally replicated with ease (see also Table 1
).
|
ophthalmology
The first reported GWA was also one of the most impressive. Three separate reports that appeared in the same issue of Science in 2005 all concluded that a key gene for age-related macular degeneration (AMD), a leading cause of blindness in the developed world, was CFH (complement factor H).2
The GWA study (13) reported a relatively low-resolution genome-wide screen (by 2008 standards) with just over 100 000 SNPs in only 96 AMD cases and 50 controls. The association was centered ultimately over a coding variant in the gene, where having at least 1 histidine at amino acid position 402 increased an individual’s risk of developing AMD 2.7-fold and, staggeringly, was estimated to account for approximately 50% of the attributable risk of AMD.
Even more breathtaking was a later report of a GWA analysis of exfoliation glaucoma, which revealed association to 2 nonsynonymous SNPs in exon 1 of LOXL1 (lysyl oxidase-like 1) (14). These variants explained the entire genetic basis to this phenotype, as the population attributed risk was estimated to be in excess of 99%—i.e., they effectively solved the genetics of this subform of glaucoma, a leading cause of irreversible blindness.
endocrinology
Type 2 diabetes (T2D) has been the focus of more GWAs than any other disorder studied to date. From the first batch of such studies, recently published in Nature (15)(16) and Science (17)(18)(19), the strongest association by far of the approximately 20 established loci was with a gene established in 2006 by Grant et al. (20) in Nature Genetics as playing a role in the disease, namely the Wnt-signaling pathway member, TCF7L2 [transcription factor 7–like 2 (T-cell specific, HMG-box)]. This is now considered the most significant genetic finding in T2D to date and represents one of the most important findings in the arena of complex disease. Indeed, other investigators have already independently replicated the association of variation in TCF7L2 with T2D in individuals of European, Asian, and African descent.
Obesity is also an important risk factor for T2D. Variation in INSIG2 (insulin-induced gene 2) was shown to be associated with both adult and childhood obesity from the first GWA published for this phenotype (21); however, this report has proven controversial, with 3 highly regarded groups already refuting the observation. The publication of a second obesity gene, FTO (fat mass and obesity associated) (22) almost a year later was made indirectly as a consequence of T2D GWA studies, but turned out to be operating through insulin resistance. This finding has been subsequently observed elsewhere and has been more readily replicated.
A component of BMI is height. Weedon et al. (23) reported association to common variation in the HMGA2 (high-mobility group AT-hook 2) oncogene for this trait. They estimated that the observed variation in this gene explains approximately 0.3% of population variation in height. The same group (24) subsequently showed that common variants at the GDF5-UQCC locus [growth differentiation factor 5 and ubiquinol-cytochrome c reductase complex chaperone, CBP3 homolog (yeast)] also contribute to variation in height.
cardiology
Multiple studies have identified a region on chromosome 9p21, near CDKN2A and CDKN2B [cyclin-dependent kinase inhibitor 2A (melanoma, p 16, inhibits CDK4) and 2B (p15, inhibits CDK4)], that is associated with coronary heart disease (16)(25)(26)(27). Six other loci have also been implicated on 6q25, 2q36, 1p13, 1q41, 10q11, and 15q22 (27), but these are still subject to further characterization. Meanwhile, a GWA scan of atrial fibrillation (AF) found association with 2 SNPs residing nearby PITX2 (paired-like homeodomain 2) (28), which has been previously functionally implicated in left–right asymmetry of the heart.
Identifying novel genes responsible for their serum concentrations of lipoproteins and lipids could potentially aid in the discovery of novel therapeutic targets for cardiovascular disease. Two groups showed that multiple loci were reproducibly associated with lipid concentrations (29)(30). Unsurprisingly, they also noted that many of these loci also showed association with coronary artery disease.
oncology
After an initial report in 2006 of association between prostate cancer and a variant on 8q24 based on linkage results (31), the investigators went on to discover a second variant in the same region as a consequence of their GWA scan (32). When these 2 variants were combined, they accounted for approximately 11%–13% of cases of prostate cancer in whites, and amazingly almost one third of cases in African Americans. The newer variant was also reported to have stronger association with earlier age at diagnosis. Others also reported multiple variants at the same locus (33)(34). It should be noted that none of these variants reside in a known gene or alter the coding sequence of an encoded protein, so further work is required to elucidate the function of this neighborhood with respect to prostate cancer pathogenesis. Interestingly, the 8q24 locus has also been clearly implicated in colon cancer (35)(36)(37); however, it has been shown that the risk conferred among the variants differs significantly between the 2 cancers and that the variants at this locus have different effects on cancer development and depend on tissue type (38).
GWA studies have also revealed 5 additional, independent breast cancer loci beyond those already established in the literature (39)(40)(41). On the pediatric side, we (42) have reported highly significant and replicated association between neuroblastoma and variation on 6p22 containing the predicted genes FLJ22536 and FLJ44180.
immunology
Inflammatory bowel disease is a relatively common inflammatory disorder, consisting primarily of Crohn disease and ulcerative colitis. In late 2006, Duerr et al. (43) reported highly significant association between Crohn disease and a coding variant in IL23R (interleukin-23 receptor). Shortly thereafter, Hampe et al. (44) provided a crucial addition by reporting a nonsynonymous SNP scan in the same disease, in which they identified a significant association of a coding SNP in ATG16L1 [ATG16 autophagy-related 16-like 1] (44).
One of the very few pediatric GWA studies to date (45) reported that multiple SNPs in a region containing multiple genes on 17q21 was strongly associated with childhood asthma. To elucidate the responsible gene in the region, the authors observed that these SNPs were consistently and strongly associated with transcript levels of ORMDL3 [ORM1-like 3] derived from Epstein-Barr virus–transformed lymphoblastoid cell lines.
A number of strong genetic determinants of type 1 diabetes (T1D) have been established through candidate gene studies, primarily within the major histocompatibility complex (MHC) but also with other loci. We recently reported the outcome of our GWA for T1D (46). In addition to confirming previously identified loci, we observed highly significant association with CLEC16A [C-type lectin domain family 16, member A (formerly KIAA0350)], the gene product of which is predicted to be a sugar binding C-type lectin. Subsequent follow-up of our data revealed a locus on 12q13 (47). The Wellcome Trust Case Control Consortium (16)(48) demonstrated association to the same regions plus reported other loci on 12q24 and 18p11.
Four GWA studies for systemic lupus erythematosus were published on the same day in early 2008, reporting a number of novel loci other than the previously observed strong association within the HLA region and with IRF5 (interferon regulatory factor 5) (49)(50)(51)(52). Meanwhile, the Wellcome Trust Case Control and Australo-Anglo-American Spondylitis Consortiums genotyped nonsynonymous SNPs and MHC tag SNPs in an ankylosing spondylitis cohort, identifying 2 new loci, ERAP1 [endoplasmic reticulum aminopeptidase 1 (formerly ARTS1)] and IL23R (53).
HLA-DRB1 (MHC class II, DR beta 1) and PTPN22 [protein tyrosine phosphatase, nonreceptor type 22 (lymphoid)] are well-established susceptibility loci for rheumatoid arthritis. Undeterred, Plenge et al. (54) carried out a GWA analysis to find other loci associated with an increased risk for the disease. In addition to the usual suspects, a signal was observed that resides in a region that harbors 2 genes relevant to chronic inflammation, namely TRAF1 (TNF receptor–associated factor 1) and C5 (complement component 5). A few weeks later, the same group described the identification of 2 SNPs located only 3.8 kb apart, but statistically independent, on 6q23 near TNFAIP3 (TNF alpha–induced protein 3) and OLIG3 (oligodendrocyte transcription factor 3), through a strategy of cross-comparison of their dataset with public datasets (55).
A GWA study of multiple sclerosis using family trios revealed extremely strong association to the HLA-DRA (MHC class II, DR alpha) locus (P = approximately 10–81) and 2 additional signals within IL2RA and IL7RA (interleukin-2 receptor 
and interleukin-7 receptor ) (56). Celiac disease is strongly associated with the MHC as well, but an additional signal has now been determined at the KIAA1109-TENR-IL2-IL21 locus (57).
Some individuals establish and maintain effective control of HIV-1, and others do not. These observations could have dramatic implications for the development of new treatments for HIV/AIDS. Fellay et al. (58) identified variants that explain almost 15% of the variation in viral load during the asymptomatic set-point period of infection. One of these variants was in an endogenous retroviral element on the HLA-B*5701 background, whereas a second variant was in the neighborhood of HLA-C (MHC class I, C). A secondary analysis, looking at time to HIV disease progression, revealed 2 further genes, 1 of which encodes an RNA polymerase I subunit.
neurology
Neurological disorders have proven the most challenging complex disease to address using genome-wide SNP approaches, primarily as a consequence of the need for strict, uniform phenotyping across very large, multicenter cohorts.
A controversial GWA study (59) for Parkinson’s disease (PD) using a 2-stage GWA approach still has to be confirmed through independent replication efforts. When the authors combined the data from the 2 stages, a SNP residing in SEMA5A [sema domain, 7 thrombospondin repeats (type 1 and type 1–like), transmembrane domain (TM), and short cytoplasmic domain (semaphorin) 5A] yielded the lowest combined P value, although it was not significant genome-wide. Similarly for schizophrenia, the strongest signal was also not strictly significant [near CSF2RA—colony-stimulating factor 2 receptor alpha, low-affinity (granulocyte-macrophage)—and IL3RA—interleukin-3 receptor alpha (low affinity)] (60). Research on restless legs syndrome (RLS), a common neurologic disorder involving the involuntary movement of legs that causes severe sleep disruption, has borne more obvious fruit. Recent studies reported genome-wide significant association with a common variant in an intron of BTBD9 [BTB (POZ) domain–containing 9] (61)(62).
Of all the neurological disorders, amyotrophic lateral sclerosis (ALS) has received the most attention with respect to GWA studies. Dunckley et al. (63) reported the first GWA analysis; however, their approach was somewhat different in that they used pooled DNA, presumably for cost-saving purposes. The most significant association, although far from significant genome-wide, was found near the uncharacterized gene FLJ10986. This somewhat controversial study is subject to extensive follow-up before solid conclusions can be drawn. There is, however, a growing consensus that the primary genetic association for ALS is in fact to variation in DPP6 (dipeptidyl-peptidase 6), as a consequence of 2 further published GWA studies of the disease (64)(65).
other disease areas
With respect to progress in the liver metabolism field, Buch et al. (66) reported a GWA study for gallstones, in which they identified a coding variant in ABCG8 [ATP-binding cassette, subfamily G (WHITE), member 8 (sterolin 2)]. Interestingly, the association strengthened when the analysis was restricted specifically to cholesterol gallstones.
Menzel et al. (67) used the GWA approach in the arena of hematology to identify an F cell quantitative trait–associated variant in BCL11A [B-cell CLL/lymphoma 11A (zinc finger protein)]. F-cell levels are an indication of fetal hemoglobin presence, the variation of which accounts in large part for the phenotypic diversity of sickle cell disease and β-thalassemia. The authors estimated that this locus accounts for 15.1% of the variance of this trait. Uda et al. (68) subsequently reported similar findings in their GWA study of the same trait.
consortiums
The Wellcome Trust Case Control Consortium publication in Nature (16) described a joint GWA study (using the Affymetrix GeneChip 500K Mapping Array Set) carried out in the British population, in which they examined approximately 2000 individuals for each of 7 major diseases and a shared set of approximately 3000 controls. As a result, 24 independent association signals at the 10–7 threshold were observed: 1 in bipolar disorder, 1 in coronary artery disease, 9 in Crohn disease, 3 in rheumatoid arthritis, 7 in T1D, and 3 in T2D. They observed association at many of the previously identified loci (as described above), and all novel observations were subject to future replication attempts. Their data, results, and software are now widely available to other investigators.
The Framingham Heart Study (FHS) was initiated in 1948 to investigate the epidemiology of cardiovascular disease. Using 100K Affymetrix GeneChip on various disease-state cohorts, FHS published selected findings in a series of 17 reports in a dedicated issue of BMC Medical Genetics, summarized by Cupples et al. (69). Their results are now posted on the NCBI dbGaP Web site (view.ncbi.nlm.nih.gov/dbgap). Their main findings, on the whole, were not novel, which may be partly due to the timing of the release, but also to the resolution of the genotyping platform they selected.
|
Discussion |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
It should also be noted that all these discoveries were made using white cohorts, since the information capture in this ethnic group is optimal with these SNP chips, and many more SNPs are required for the same degree of capture in populations of African ancestry. As more work is carried out in this area, full elucidation will eventually be established of key loci globally. This elucidation will also be vital for the new era of consumer-based genetics, in which companies such as 23andMe and Navigenics will want to offer content to consumers of all races.
After the discovery of variants underlying discrete traits, it is exciting to see studies now come out on human quantitative traits, such as BMI, HIV titer, height, and lipids. These results represent the first consistently replicated associations with quantitative traits.
The "Holy Grail" for GWA now is to determine the genetic variants underlying neurological disorders. Apart from RLS and ALS, no convincing results to date have been published in this disease discipline, with only controversial results available for Parkinson’s and schizophrenia.
Of course the discovery of key genetic factors involved in the pathogenesis of complex disease is only the first step in a much longer process. The scientists in the spotlight at the moment are the geneticists who reported these signals, but the new stars will be the molecular and cell biologists who can elucidate the mechanism of action of the underlying mutations represented by these association signals, so that therapeutic agents can be raised to these targets that will in turn lead to more efficacious treatments.
Ultimately, determining the variants influencing variable pharmacological response using GWA will revolutionize the drug market, which may well lead to a more partitioned market, but one that offers more tailored drug therapy.
|
Acknowledgments |
|---|
Financial Disclosures: None declared.
|
Footnotes |
|---|
2 Human genes: CFH, complement factor H; LOXL1, lysyl oxidase-like 1; TCF7L2, transcription factor 7–like 2 (T-cell specific, HMG-box); INSIG2, insulin-induced gene 2; FTO, fat mass and obesity associated; HMGA2, high-mobility group AT-hook 2; GDF5, growth differentiation factor 5; UQCC, ubiquinol-cytochrome c reductase complex chaperone, CBP3 homolog (yeast); CDKN2A, cyclin-dependent kinase inhibitor 2A (melanoma, p 16, inhibits CDK4); CDKN2B, cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4); PITX2, paired-like homeodomain 2; IL23R, interleukin-23 receptor; ATG16L1, ATG16 autophagy-related 16-like 1 (S. cerevisiae); ORMDL3 [ORM1-like 3 (S. cerevisiae); CLEC16A, C-type lectin domain family 16, member A (formerly KIAA0350); IRF5, interferon regulatory factor 5; ERAP1, endoplasmic reticulum aminopeptidase 1 (formerly ARTS1); HLA-DRB1, MHC, class II, DR beta 1; PTPN22, protein tyrosine phosphatase, nonreceptor type 22 (lymphoid); TRAF1, TNF receptor–associated factor 1; C5, complement component 5; TNFAIP3, TNF alpha–induced protein 3; OLIG3 (oligodendrocyte transcription factor 3; HLA-DRA, MHC class II, DR alpha; IL2RA, interleukin-2 receptor ; IL7RA, interleukin-7 receptor ; HLA-C, MHC class I, C; SEMA5A, sema domain, 7 thrombospondin repeats (type 1 and type 1–like), transmembrane domain (TM), and short cytoplasmic domain (semaphorin) 5A; CSF2RA, colony-stimulating factor 2 receptor alpha, low-affinity (granulocyte-macrophage); IL3RA, interleukin-3 receptor alpha (low affinity); BTBD9, BTB (POZ) domain–containing 9; DPP6, dipeptidyl-peptidase 6; ABCG8, ATP-binding cassette, subfamily G (WHITE), member 8 (sterolin 2); BCL11A, B-cell CLL/lymphoma 11A (zinc finger protein); CDKAL1, CDK5 regulatory subunit–associated protein 1–like 1; MLXIPL, MLX-interacting protein-like; SMAD7, SMAD family, member 7; FGFR2, fibroblast growth factor receptor 2 (bacteria-expressed kinase, keratinocyte growth factor receptor, craniofacial dysostosis 1, Crouzon syndrome, Pfeiffer syndrome, Jackson-Weiss syndrome); TOX3, TOX high-mobility group box family, member 3 (formerly TNRC9); ITGAM, integrin, alpha M (complement component 3 receptor 3 subunit); BANK1, B-cell scaffold protein with ankyrin repeats 1; IL2, interleukin-2; IL21, interleukin-21.
|
References |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. E. Ritchie, B. S. Carvalho, K. N. Hetrick, S. Tavare, and R. A. Irizarry R/Bioconductor software for Illumina's Infinium whole-genome genotyping BeadChips Bioinformatics, October 1, 2009; 25(19): 2621 - 2623. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
