Implications for health and disease in the genetic signature of the Ashkenazi Jewish population
© Guha et al.; licensee BioMed Central Ltd. 2012
Received: 20 June 2011
Accepted: 25 January 2012
Published: 25 January 2012
Relatively small, reproductively isolated populations with reduced genetic diversity may have advantages for genomewide association mapping in disease genetics. The Ashkenazi Jewish population represents a unique population for study based on its recent (< 1,000 year) history of a limited number of founders, population bottlenecks and tradition of marriage within the community. We genotyped more than 1,300 Ashkenazi Jewish healthy volunteers from the Hebrew University Genetic Resource with the Illumina HumanOmni1-Quad platform. Comparison of the genotyping data with that of neighboring European and Asian populations enabled the Ashkenazi Jewish-specific component of the variance to be characterized with respect to disease-relevant alleles and pathways.
Using clustering, principal components, and pairwise genetic distance as converging approaches, we identified an Ashkenazi Jewish-specific genetic signature that differentiated these subjects from both European and Middle Eastern samples. Most notably, gene ontology analysis of the Ashkenazi Jewish genetic signature revealed an enrichment of genes functioning in transepithelial chloride transport, such as CFTR, and in equilibrioception, potentially shedding light on cystic fibrosis, Usher syndrome and other diseases over-represented in the Ashkenazi Jewish population. Results also impact risk profiles for autoimmune and metabolic disorders in this population. Finally, residual intra-Ashkenazi population structure was minimal, primarily determined by class 1 MHC alleles, and not related to host country of origin.
The Ashkenazi Jewish population is of potential utility in disease-mapping studies due to its relative homogeneity and distinct genomic signature. Results suggest that Ashkenazi-associated disease genes may be components of population-specific genomic differences in key functional pathways.
Since the advent of genomewide SNP microarrays for disease mapping, considerable attention has been paid to the potentially confounding role of population stratification [1, 2]. In addition to variation introduced by major continental ancestry, substantial intra-continental clines have been reliably demonstrated, typically mapping onto geographic patterns of historic migration [3–5]. By contrast, population isolates and relatively small founder populations demonstrate less background diversity, which may provide increased power to detect disease-related alleles [6, 7]. Nevertheless, even these populations tend to reveal very subtle patterns of genetic structure that reflect demographic history and may affect interpretation of disease association studies [8–10].
The Ashkenazi Jewish (AJ) population is one such founder cohort, composed of Jewish individuals whose ancestors are thought to have advanced from the Rhine valley to populate Eastern Europe and beyond, beginning approximately 1,000 years ago . The AJ population has been associated with very specific genetically derived predispositions to disease, primarily monogenic recessive disorders , but more recent studies also demonstrate increased frequency of certain alleles associated with complex diseases [13–15]. Despite the interest in the AJ population for disease mapping, however, population genetic studies in AJ cohorts to date have not focused on the relevance of genetic results to the study of complex disease.
Classic population genetic studies of Jewish cohorts, based on uniparental markers, have provided strong evidence of founder effects for the AJ population in both the mitochondrial and Y-chromosome lineage [16, 17]. Such studies typically have shown reduced variability within AJ samples, and a greater degree of resemblance to other Levantine-derived populations (including Arabs and non-Ashkenazi Jews) than to the host European populations; moreover, these studies have concluded that genetic drift has played a primary role in the heightened frequency of certain parental lineages that are rare or virtually absent in other populations [18, 19]. More recent studies have also demonstrated the ability of SNP microarrays to differentiate AJ samples embedded within larger non-AJ European-American cohorts [20–22]; these studies placed AJ samples along a dimension intermediate to European and Middle Eastern populations. Most recently, three genomewide studies of autosomal markers in Jewish samples of varying origins have yielded results indicating: 1) considerable similarity between AJ and (most) non-Ashkenazi Jewish cohorts; and 2) Jewish populations (except those from India and Ethiopia) can be viewed as a mixture of European and Middle Eastern genetic ancestry [23–25]. However, two of these studies [23, 24] were limited to relatively small sample sizes of AJ individuals, which may have restricted their ability to detect AJ-specific patterns of genetic variation. Moreover, these studies did not specifically test for geographic or other structure within the AJ population, and no attempt was made to characterize AJ-related variation with respect to disease susceptibility.
The present study was designed to examine these issues using genomewide SNP markers in a very large (n = 1,394) cohort of unselected AJ individuals from Israel. First, we sought to identify an AJ-specific allelic pattern from autosomal markers, using both clustering and principal components approaches as applied to AJ samples and non-Jewish samples derived from European, Middle Eastern, and Central/South Asian origins. Next, we tested whether genetic distance measures placed AJ in an intermediate position relative to European and Middle Eastern populations. Additionally, we examined whether the AJ population demonstrated internal structure, and whether any such structure would correlate with geographical region of origin. Next, we used genome wide association study (GWAS) methods to examine the relationship of AJ-specific variation to the biology of health and disease. Finally, we provide an optimized and cross-validated list of AJ-related ancestry informative markers (AIMs) for future disease-mapping studies.
As shown in Figure 1, for K = 7, pea green remains the dominant AJ color, accounting for as much as 87.5% of the ancestry of AJ individuals. Across all AJ samples, the median degree of contribution of this component was 64.6%; the mean (57.9 ± 19.1%) was somewhat lower than the median due to the presence of a number of subjects in the cohort with virtually no AJ contribution. Amongst non-AJ samples, sharing of this ancestry component was quite limited, with the greatest amount of overlap (approximately 10%) seen for Palestinians, Adygei (from the northern Caucasus), and Italians/Tuscans/Sardinians, respectively. At the same time, most of the AJ samples demonstrated little overlap with other specific ancestry components, with the exception of a subset of approximately 16% of the sample that had significant contributions from the European ancestry component (red). These admixed individuals will be examined in greater detail below.
Results did not change when we re-ran these ADMIXTURE analyses with larger subsamples of the Ashkenazi cohort (n = 350, n = 700, and n = 1,050 AJ individuals); for each of these analyses, K = 7 provided the optimal solution. Results changed slightly when the full Ashkenazi cohort was compared to the neighboring HGDP populations; as depicted in Additional file 1, the K = 8 solution was marginally (but not significantly) better than the K = 7 solution, which was also indistinguishable from the K = 9 solution. Compared to the K = 7 results, however, neither of these solutions introduced substantive changes into the AJ population ancestry component.
Additional file 3 further demonstrates similar results when all HGDP samples are included in the ADMIXTURE analysis. Cross-validation analysis (ten runs) indicated that model fit is optimized at K = 11, with second-best fit obtained at K = 8, which coincides with the emergence of the AJ-specific ancestry component (colored brown in Additional file 3). Moreover, at K = 11, there is virtually no evidence of this AJ component in any of the other populations.
Principal components analysis
Genetic distances between populations
Pairwise genetic distances (Fst) between Ashkenazi Jewish, European and Middle Eastern populations
Residual intra-population structure
Implications for health and disease
Next, we sought to identify which genetic variants were contributing to the AJ-specific ancestry factor identified in Figure 1. Allelic contributions to the ADMIXTURE-based cluster 3 (C3) scores were examined using quantitative GWAS (additive model comparing C3 against approximately 739 K high-quality SNPs) in all (n = 1,312 after all quality control procedures) AJ samples. A total of approximately 13,841 SNPs were strongly (P < 10-6) associated with C3 (Additional file 6).
Gene Ontology categories significantly over-represented (P < 0.001) in ALIGATOR analysis
Genes in category
Genes on list
Expected on list
Expected hits per study
Response to nutrient levels
Response to nutrient
Response to extracellular stimulus
Transepithelial chloride transport
COPII vesicle coat
ER to Golgi vesicle membrane
Intriguingly, several of the statistically significant GO process categories in Table 2 include autosomal recessive disease-causing genes marked by relatively high-frequency Ashkenazi-specific mutations. For example, five of the six genes involved in transepithelial chloride transport (GO:0030321) are significantly associated with C3 scores; these include CFTR, a gene that harbors characteristic mutations that cause the increased prevalence of cystic fibrosis in the Ashkenazi population . Similarly, six out of eight genes involved in equilibrioception (GO:0050597) are on the C3 GWAS list, including PCDH15 and CLRN1. Specific founder mutations in these two genes are responsible for increased prevalence of Usher syndrome (types I and III) in the Ashkenazi population [27, 28]. Notably, both of these GO categories also were significant in a complementary gene-set enrichment analysis using GSA-SNP (details in Materials and methods). In the GSA-SNP analysis, the equilibrioception category demonstrated enrichment at nearly four standard deviations beyond the mean of all GO categories (Z = 3.99, P = 3.37E-05, false discovery rate (FDR) = 0.002); transepithelial chloride transport was also enriched more than 3 standard deviations beyond the mean (Z = 3.09; P = 9.98E-04; FDR = 0.029). However, the other categories listed in Table 2 did not achieve corrected significance levels (FDR > 0.05, P > 0.002) on the GSA-SNP enrichment list.
Fifteen coding variants with functionally characterized SNPs crossing the threshold (P < 10-6) for association with C3 scores
Amino acid position
Amino acid change
V | M
A | V
A | V
P | A
V | L
L | P
R | W
W | R
R | G
V | I
The relative over-representation of the minor allele at rs1801133 (also known as MTHFR C677T) in AJ populations has been previously noted ; homozygosity at this allele is associated with hyperhomocysteinemia. Several novel findings are also apparent from Table 3, with potential impact on disease risk within the AJ population. For example, SH2B3 regulates cytokine activity, and rs3184504 within this gene has been replicably associated with risk for type 1 diabetes and celiac disease [32, 33]. The AJ population has a lower frequency of the protective C allele (that is, a higher frequency of the disease-associated T allele) than any other HapMap population. Similarly, the AJ cohort has a reduced frequency of the Ala12 variant (G allele at rs1801282) in the PPARG gene; the Ala12 variant, while rare in all populations, reduces risk for type 2 diabetes by a factor of 0.86 . The V60L variant at MC1R, also quite common in the AJ cohort, has been associated with melanoma in some, but not all, populations . By contrast, the AJ population has a reduced frequency of the T allele at rs2227564, which has been associated with Alzheimer's disease . Similarly, the AJ population has a reduced frequency of the R15W variant of the AIF1 gene, which has been strongly (odds ratio > 2) associated with rheumatoid arthritis .
We also performed a GWAS on scores derived from PC1 of the intra-population PCA depicted in Additional file 5. As shown in Additional file 7, this source of population variance was strictly accounted for by allelic differences in the major histocompatibility complex (MHC). Notably, the MHC alleles associated with intra-AJ population structure are completely different from the MHC component associated with the inter-population analysis (Additional file 6). For example, the AJ population is differentiated from neighboring non-AJ populations by a reduced frequency of the A allele at rs3135391 (Table 3), which tags the HLA-DRB*1501 allele. This allele has been associated with susceptibility to multiple sclerosis and other autoimmune diseases . By contrast, the intra-population principal component (PC1) is most strongly correlated with alleles in the class I region of the MHC - for example, rs9260952 in the region of HLA-A (P = 2.46 × 10-106) and rs3828875 (P = 4.76 × 10-106), which has been correlated with HLA-B *6701 and *3802 alleles .
Ancestry informative markers
Classification of AJ individuals derived from PCA clustering using 1,357 SNPs obtained from ADMIXTURE analysis
Ashkenazi Jewish grandparents (self-report)
Classification of AJ individuals derived from PCA clustering using 121,834 SNPs in Need et al. 
Ashkenazi Jewish grandparents (self-report)
Classification of AJ individuals derived from PCA clustering using AIM 103 SNPs obtained from ADMIXTURE analysis
Ashkenazi Jewish grandparents (self-report)
While there have been several population genetics studies of Jewish cohorts published in the past two years [21–25], the findings of the present study are novel in several ways. First, prior studies have emphasized commonalities amongst Jewish sub-populations, as well as relative proximity to European and Levantine populations. By contrast, the present study took the complementary approach of defining the spectrum of autosomal variation that is AJ-specific. Moreover, using novel pathway analyses, the present study related population genetic variation to patterns of disease propensity in the Ashkenazi population. Second, the present study examined intra-Ashkenazi variation. Finally, we provide a robust yet compact list of AIMs for the Ashkenazi population.
The primary result of the present study is the specification of the allelic content of an autosomal genetic signature that can distinguish the Ashkenazi Jewish population from both its host populations in Europe and other populations that originate in the same geographic area of the Levant. To our knowledge, ours is the first study of the Ashkenazi population to utilize cross-validation metrics to identify the optimal solution to the assignment of population ancestry scores. Previous studies using similar approaches have demonstrated the ability of genomic information to differentiate Ashkenazi samples from those drawn from other populations [1, 20–25]. However, each of these studies has suggested that AJ samples represent an intermediate position or admixture between European and Levantine populations. Although one recent paper suggested 30 to 60% European admixture in Ashkenazi and other Jewish samples , the present study found relatively little (≤10%) overlap of AJ genetic ancestry components in non-AJ Levantine populations. In the statistically optimal ADMIXTURE result in our study, European admixture followed a pattern indicative of second-generation admixture rather than deeper mingling with the host populations. Moreover, pairwise genetic distances were not consistent with an intermediate positioning of the AJ population relative to the European and Levantine populations.
It should be emphasized that these results do not suggest an independent (for example, Khazar or non-Levantine) lineage for the AJ population, a hypothesis that has generally been ruled out by prior literature [16, 17, 24]. Rather, Table 1 demonstrates relative proximity amongst several populations with Mediterranean heritage, including the AJ, Palestinians, and Italians, suggestive of an ancient common deme. Additionally, the FST data indicate approximately equal genetic distances between the AJ and western (French), eastern (Adygei), and Middle Eastern (Palestinian) cohorts, consistent with the suggestion that founder effects and subsequent drift account for the data more strongly than substantial local in-mixture with the European host populations in the last 1,000 years.
Moreover, the present study is the first to examine residual intra-population variance in AJ samples in comparison to host European populations. Results of our intra-AJ principal components analysis indicated that residual structure was minimal, was not related to geographic origin within Europe, and did not map onto differences in host population. Taken together, these data most likely reflect the unique contributions of the AJ founder population to the genetic make-up of present-day Ashkenazim. At the same time, it is acknowledged that our autosomal data may not capture certain components of ancestry that are accessible to mitochondrial DNA and Y-chromosome studies, such as sex differences in origin and number of founders [16–18].
Having identified this AJ-specific signature, we then sought to characterize its primary allelic content in order to determine potential relevance to future disease mapping studies. We developed a robust yet compact set of AIMs that can be applied to refine studies of European or European-American cohorts, which are still the most commonly used in disease mapping GWASs. These AIMs will also be useful in future GWASs of AJ cohorts, insofar as they can identify individuals with varying degrees of recent European admixture, thereby reducing residual intra-population structure (Figure 6). The lack of significant intra-population structure suggests that the AJ population may be useful for disease-mapping studies, with the possibility of enhanced signal-to-noise for the detection of (at least a subset) of disease-related alleles .
Alleles within the MHC were the most substantial contributors to both inter-population and intra-population variance. MHC markers comprised approximately 6% of all approximately 13,841 SNPs that were correlated with the AJ-specific signature, including polymorphisms in both class I and class II genes. Prior research has consistently demonstrated the MHC to be most sensitive to population differences , typically due to geographic differences in exposure history . These population differences have implications for susceptibility to autoimmune diseases , and may account for the increased rate of pemphigus vulgaris in AJ individuals . Recent studies associating SNPs in the MHC with serious drug-induced side effects , viral load in HIV  and psychiatric illness  also indicate the clinical relevance of more extensive elaboration of population differences in MHC alleles.
Characterization of the AJ-specific component also resulted in the identification of several coding variants known to be associated with disease, and was able to detect markers in CFTR and NOD2 that are relevant to increased prevalence of cystic fibrosis and Crohn's disease in the Ashkenazi population [47, 48]. Perhaps the most surprising result from the present study, however, was the over-representation of GO categories containing disease-bearing genes commonly associated with the AJ population. For example, the AJ cohort did not merely differ from other populations in CFTR allele frequencies, but also in allelic frequencies in most other genes associated with transepithelial chloride transport. However, it should be noted that these data do not provide specific evidence of causality between the existence of AJ-prevalent disease-causing mutations in these pathways and the over-representation of certain common alleles in related genes. Speculatively, these results suggest the possibility that deleterious recessive alleles may persist at relatively high frequencies in the AJ population due to epistatic effects with other genes in the same biological pathway, which also display altered allelic frequencies in the AJ population.
The present study characterized statistically significant components of autosomal variation specific to the AJ population. By focusing on common variants available on a dense GWAS platform, results add to prior literature on rare, disease-causing mutations that are over-represented in the Ashkenazi population. GO analysis points to significant allele frequency differences in multiple genes in pathways implicated by AJ-associated diseases such as cystic fibrosis and Usher's syndrome. However, it will be important for future research to determine which elements of this genetic signature are shared with non-AJ populations, and may therefore be reflective of ancient founder effects, as opposed to more recent founder effects specific to the introduction and expansion of the Jewish people into Europe.
Materials and methods
The AJ cohort consisted of 1,394 volunteers (986 male, 408 female) recruited from the Israeli blood bank. Each subject self-reported that all four grandparents were of AJ origin, and all subjects provided written, informed consent. Subsequent to genomic DNA extraction from blood samples through use of the Nucleon kit (Pharmacia, Piscataway, NJ, USA), all samples were fully anonymized prior to genotyping and analysis, under protocols approved by the National Genetic Committee of the Ministry of Health (Israel) and the Institutional Review Board of the North Shore-LIJ Health System.
The HGDP genome-wide genotype data containing 1,043 individuals from 51 worldwide population groups were obtained from the HGDP database . The sample sizes for many individual groups were very small and grouped together based on their geographical distribution and ethnicity for comparison analysis as suggested .
Additional genotype data on 611 Caucasian subjects recruited at Duke University, including 94 individuals who self-reported having one or more AJ grandparents, were from Need at al. .
Genotyping and quality control
Genotyping of AJ samples was performed using Illumina HumanOmni1-Quad arrays according to the manufacturer's specifications. The samples were subjected for genotyping quality control filters, for example, samples call rate > 97%, SNP call rate > 98%, Hardy-Weinberg exact test P < 0.000001. The resulting individuals were tested for gender mismatch based on X chromosome genotype using Sex check estimation at PLINK (v1.07) . Cryptic identity and first-degree relatedness within individuals were examined using pairwise IBD estimation in PLINK performed on 128 K LD (linkage disequilibrium) pruned (r2 > 0.2) genomewide SNPs; one individual in each pair was randomly excluded. The final dataset contains 1,312 individuals with 739,409 SNPs with 99.86% average call rates.
The HGDP samples were genotyped on the Illumina HumanHap 650 k bedchip as previously described and filtered based on a sample call rate > 98.5%, resulting in 1,043 individuals with 660,918 SNPs. This dataset was again filtered based on a SNP call rate > 95%. The filtered AJ dataset was merged with the HGDP dataset and the resulting merged dataset contained 281,232 SNPs common to the two cohorts with an average call rate of 99.5%.
To perform inter-population comparison analysis (for example, ancestry estimation and PCA) the AJ and HGDP merged dataset was pruned using a LD threshold of r2 > 0.2 at PLINK (v1.07). The resulting dataset contained 95,600 unlinked genomewide SNPs shared by the AJ and HGDP samples with an average call rate of 99.7%.
Genotyping of the Duke samples was performed on Illumina Infinium HumanHap550 version 1, version 3 and 610-quad chips. The dataset contains information on 121,834 LD-pruned (r2 > 0.3) SNPs and was used to validate an AIM panel specific to AJ ancestry.
The population structure analysis was performed using the maximum likelihood based ADMIXTURE program . The maximum likelihood approaches are as accurate as Bayesian-based estimations while being computationally tractable with genomewide markers within a reasonable time. This algorithm is also considered to be more accurate and faster than the expectation-maximization-based program FRAPPE .
To detect underlying ancestral population clustering, AJ samples were compared with members of three neighboring population groups derived from the HGDP: EU (n = 159), ME (n = 163), and CAS (n = 177, excluding Kalash as per Behar et al. ). We performed ancestry estimation by randomly selecting n = 175 AJ subjects of varying national origins, in order to maintain a roughly equal sample size with each of the other three HGDP groups. Briefly, the ADMIXTURE algorithm models the genomic data from each subject as a combination of K ancestral populations, where K can be any number ≥2. ADMIXTURE output results were systematically plotted using the Distruct program , which permits visual determination of similarities and differences in ancestral make-up of each population. More formally, ten-fold cross-validation 'C' scores were computed for each K separately to determine the best fit model for ancestry estimation. We then re-performed ADMIXTURE and cross-validation analyses ten times to develop statistical confidence intervals around fit scores for each of the ancestry estimates. In order to test the effect of varying sample sizes on the analyses, and to exploit the full sample size of AJ individuals available, these analyses were repeated using n = 350, n = 700, n = 1,050, and all n = 1,312 AJ samples. The ancestry estimation was also performed with all HGDP groups using both randomly selected n = 175 AJ and all n = 1,312 AJ individuals. This analysis was carried out with 95,600 LD-pruned unlinked SNPs for K = 2 to 20, where K is the prior assumption of theoretical ancestral population.
Principal component analysis
PCA was performed to examine the inter- and intra-population distribution using EIGENSTRAT  as implemented in SNP & Variation Suite v7.3 (Golden Helix, Bozeman, MT, USA). The previously described 95,600 LD-pruned unlinked SNPs were used to perform inter-population PCA with a randomly selected subset of n = 175 AJ samples with members of the three neighboring population groups used in the ADMIXTURE analysis. Intra-population PCA was performed for all AJ individuals who clustered strongly with the AJ cohort (based on admixture analysis), using all 739,409 high quality SNPs.
Calculation of distances between populations
Pairwise FST values for all pairs of populations were estimated using GENEPOP v4.1  by a weighted analysis of variance . For each locus, an unbiased estimate of the P-value was also computed using Fisher's exact probability test, and the significance of each pairwise distance was empirically tested using a permutation algorithm (n = 5,000 runs) as previously described .
Quantitative genome-wide association study
To identify which genetic variants were contributing to the AJ-specific ancestry dimension, a quantitative GWAS was performed based on ADMIXTURE-derived AJ-specific cluster scores for all AJ samples (n = 1,312) with 739,409 high quality SNPs. A quantitative GWAS was also performed on scores derived from PC1 of the intra-population PCA to identify the source of this genetic variation.
Gene Ontology enrichment analysis
To determine whether any biologically relevant pathways were over-represented amongst this list of associated SNPs, we utilized Association LIst Go AnnoTatOR (ALIGATOR) . Like the Database for Annotation, Visualization, and Integrated Discovery (DAVID) , ALIGATOR characterizes lists of genes with respect to their relative inclusion of the various GO categories. However, ALIGATOR is specifically designed for analysis of SNP data (as opposed to gene expression data), controlling for the size of each gene and the number of SNPs present on the array.
After assigning each SNP to the closest gene, and calculating the number of genes in each GO category appearing above a specified threshold (for example, P < 10-6) in the quantitative GWAS analysis, the degree of over-representation of specific GO categories is then tested using two sets of permutations. First, the SNPs appearing above and below the GWAS cutoff are permuted (50,000 times), to determine the likelihood that a given GO category is over-represented in the list of significant SNPs. Thus, each GO category is assigned an empirically determined P-value. Second, simulated studies are permuted (10,000 times) in order to determine whether the number of categories designated as 'over-represented' (that is, category-specific P-values < 0.05, < 0.01, and < 0.001) is statistically unlikely given the number of genes on the list. Note that the initial threshold boundary (P < 10-6) is not, strictly speaking, a statistical threshold for significance. Rather, it is selected based on the assumption, intrinsic to the polygenic model approach, that true associations exist below the threshold of strict genomewide significance [61, 62]. Thus, the purpose of the ontology enrichment analysis is to identify biologically relevant signals emerging from the pattern of observed associations, irrespective of strict statistical significance, and even if no SNPs achieved strict genomewide significance . Following the suggestions of the software developer, the algorithm tends to be most robust when approximately 10% of all genes appear on the list (P Holmans, personal communication); consequently, we selected a threshold that resulted in 12.4% (2,349 out of 19,011 genes with GO annotations and a minimum set size of 2) of all genes submitted to ALIGATOR.
While ALIGATOR was the primary method of pathway analysis, due to its unique two-stage approach to control for study-wide significance, it is acknowledged that there are many ways to evaluate aggregation of the associated SNPs within biological pathways . Consequently, we sought to validate results using the recently developed GSA-SNP program , which utilizes a fundamentally different approach. The essential difference between ALIGATOR and GSA-SNP is that the first method uses overrepresentation based analysis, whereas the second uses gene-set enrichment-based analysis. Overrepresentation based analysis defines significant SNPs by a pre-specified P-value threshold, then counts significant genes in each pathway, whereas gene-set enrichment analysis considers all the SNPs in the analysis and then ranks the gene sets in order of significance . Moreover, ALIGATOR bases its analysis on the single most strongly associated SNP in each gene, whereas GSA-SNP permits the use of the kth (k = 1, 2, 3, 4 or 5) best P-value to represent each gene. We utilized the authors' recommended default of the second best P-value within each gene, which removes singleton false-positive signals and provides a more symmetric distribution to the gene scores . Significant gene set enrichment was determined by the z-statistic, with FDR < 0.05 based on Benjamini-Hochberg correction.
Ancestry informative markers
A potential set of AIMs specific to AJ was selected based on the quantitative GWAS of the AJ-specific component derived from ADMIXTURE analysis. This set of candidate AIMs was reduced and validated using a publicly available dataset previously used for identification of AJ-specific allelic variation . After identification of overlapping markers, PCA was performed on the Need et al. dataset  using the candidate AIMs, and results were compared to self-reported AJ ancestry.
ancestry informative marker
Association LIst Go AnnoTatOR
false discovery rate
genome-wide association study
human genome diversity panel
major histocompatibility complex
first principal component
second principal component
third principal component
principal components analysis
single nucleotide polymorphism.
The authors would like to thank Michael Ryan of the Feinstein Institute Biorepository for assistance with sample handling and preparation. This work was supported by the North Shore-LIJ Health System Foundation and the National Institutes of Health (RC2 MH089964 to TL).
- Tian C, Gregersen PK, Seldin MF: Accounting for ancestry: population substructure and genome-wide association studies. Hum Mol Genet. 2008, 17: R143-R150. 10.1093/hmg/ddn268.PubMedPubMed CentralView ArticleGoogle Scholar
- Price AL, Zaitlen NA, Reich D, Patterson N: New approaches to population stratification in genome-wide association studies. Nat Rev Genet. 2010, 11: 459-463.PubMedPubMed CentralView ArticleGoogle Scholar
- Novembre J, Johnson T, Bryc K, Kutalik Z, Boyko AR, Auton A, Indap A, King KS, Bergmann S, Nelson MR, Stephens M, Bustamante CD: Genes mirror geography within Europe. Nature. 2008, 456: 98-101. 10.1038/nature07331.PubMedPubMed CentralView ArticleGoogle Scholar
- Reich D, Thangaraj K, Patterson N, Price AL, Singh L: Reconstructing Indian population history. Nature. 2009, 461: 489-494. 10.1038/nature08365.PubMedPubMed CentralView ArticleGoogle Scholar
- Abdulla MA, Ahmed I, Assawamakin A, Bhak J, Brahmachari SK, Calacal GC, Chaurasia A, Chen CH, Chen J, Chen YT, Chu J, Cutiongco-de la Paz EM, De Ungria MC, Delfin FC, Edo J, Fuchareon S, Ghang H, Gojobori T, Han J, Ho SF, Hoh BP, Huang W, Inoko H, Jha P, Jinam TA, Jin L, Jung J, Kangwanpong D, Kampuansai J, Kennedy GC, et al: Indian Genome Variation Consortium. Mapping human genetic diversity in Asia. Science. 2009, 326: 1541-1545.PubMedView ArticleGoogle Scholar
- Shifman S, Darvasi A: The value of isolated populations. Nat Genet. 2001, 28: 309-310. 10.1038/91060.PubMedView ArticleGoogle Scholar
- Bonnen PE, Pe'er I, Plenge RM, Salit J, Lowe JK, Shapero MH, Lifton RP, Breslow JL, Daly MJ, Reich DE, Jones KW, Stoffel M, Altshuler D, Friedman JM: Evaluating potential for whole-genome studies in Kosrae, an isolated population in Micronesia. Nat Genet. 2006, 38: 214-217. 10.1038/ng1712.PubMedView ArticleGoogle Scholar
- Hunter-Zinck H, Musharoff S, Salit J, Al-Ali KA, Chouchane L, Gohar A, Matthews R, Butler MW, Fuller J, Hackett NR, Crystal RG, Clark AG: Population genetic structure of the people of Qatar. Am J Hum Genet. 2010, 87: 17-25. 10.1016/j.ajhg.2010.05.018.PubMedPubMed CentralView ArticleGoogle Scholar
- Jakkula E, Rehnström K, Varilo T, Pietiläinen OP, Paunio T, Pedersen NL, deFaire U, Järvelin MR, Saharinen J, Freimer N, Ripatti S, Purcell S, Collins A, Daly MJ, Palotie A, Peltonen L: The genome-wide patterns of variation expose significant substructure in a founder population. Am J Hum Genet. 2008, 83: 787-794. 10.1016/j.ajhg.2008.11.005.PubMedPubMed CentralView ArticleGoogle Scholar
- Helgason A, Yngvadóttir B, Hrafnkelsson B, Gulcher J, Stefánsson K: An Icelandic example of the impact of population structure on association studies. Nat Genet. 2005, 37: 90-95.PubMedGoogle Scholar
- Ben-Sasson HH: History of the Jewish People. 1976, Cambridge: Harvard University PressGoogle Scholar
- Klugman S, Gross SJ: Ashkenazi Jewish screening in the twenty-first century. Obstet Gynecol Clin North Am. 2010, 37: 37-46. 10.1016/j.ogc.2010.01.001.PubMedView ArticleGoogle Scholar
- Thaler A, Ash E, Gan-Or Z, Orr-Urtreger A, Giladi N: The LRRK2 G2019S mutation as the cause of Parkinson's disease in Ashkenazi Jews. J Neural Transm. 2009, 116: 1473-1482. 10.1007/s00702-009-0303-0.PubMedView ArticleGoogle Scholar
- Kaklamani VG, Wisinski KB, Sadim M, Gulden C, Do A, Offit K, Baron JA, Ahsan H, Mantzoros C, Pasche B: Variants of the adiponectin (ADIPOQ) and adiponectin receptor 1 (ADIPOR1) genes and colorectal cancer risk. JAMA. 2008, 300: 1523-1531. 10.1001/jama.300.13.1523.PubMedPubMed CentralView ArticleGoogle Scholar
- Bronstein M, Pisanté A, Yakir B, Darvasi A: Type 2 diabetes susceptibility loci in the Ashkenazi Jewish population. Hum Genet. 2008, 124: 101-104. 10.1007/s00439-008-0520-x.PubMedView ArticleGoogle Scholar
- Behar DM, Metspalu E, Kivisild T, Achilli A, Hadid Y, Tzur S, Pereira L, Amorim A, Quintana-Murci L, Majamaa K, Herrnstadt C, Howell N, Balanovsky O, Kutuev I, Pshenichnov A, Gurwitz D, Bonne-Tamir B, Torroni A, Villems R, Skorecki K: The matrilineal ancestry of Ashkenazi Jewry: portrait of a recent founder event. Am J Hum Genet. 2006, 78: 487-497. 10.1086/500307.PubMedPubMed CentralView ArticleGoogle Scholar
- Hammer MF, Redd AJ, Wood ET, Bonner MR, Jarjanazi H, Karafet T, Santachiara-Benerecetti S, Oppenheim A, Jobling MA, Jenkins T, Ostrer H, Bonne-Tamir B: Jewish and Middle Eastern non-Jewish populations share a common pool of Y-chromosome biallelic haplotypes. Proc Natl Acad Sci USA. 2000, 97: 6769-6774. 10.1073/pnas.100115997.PubMedPubMed CentralView ArticleGoogle Scholar
- Behar DM, Garrigan D, Kaplan ME, Mobasher Z, Rosengarten D, Karafet TM, Quintana-Murci L, Ostrer H, Skorecki K, Hammer MF: Contrasting patterns of Y chromosome variation in Ashkenazi Jewish and host non-Jewish European populations. Hum Genet. 2004, 114: 354-365. 10.1007/s00439-003-1073-7.PubMedView ArticleGoogle Scholar
- Ostrer H: A genetic profile of contemporary Jewish populations. Nat Rev Genet. 2001, 2: 891-898.PubMedView ArticleGoogle Scholar
- Price AL, Butler J, Patterson N, Capelli C, Pascali VL, Scarnicci F, Ruiz-Linares A, Groop L, Saetta AA, Korkolopoulou P, Seligsohn U, Waliszewska A, Schirmer C, Ardlie K, Ramos A, Nemesh J, Arbeitman L, Goldstein DB, Reich D, Hirschhorn JN: Discerning the ancestry of European Americans in genetic association studies. PLoS Genet. 2008, 4: e236-10.1371/journal.pgen.0030236.PubMedPubMed CentralView ArticleGoogle Scholar
- Need AC, Kasperaviciute D, Cirulli ET, Goldstein DB: A genome-wide genetic signature of Jewish ancestry perfectly separates individuals with and without full Jewish ancestry in a large random sample of European Americans. Genome Biol. 2009, 10: R7-10.1186/gb-2009-10-1-r7.PubMedPubMed CentralView ArticleGoogle Scholar
- Tian C, Kosoy R, Nassir R, Lee A, Villoslada P, Klareskog L, Hammarström L, Garchon HJ, Pulver AE, Ransom M, Gregersen PK, Seldin MF: European population genetic substructure: further definition of ancestry informative markers for distinguishing among diverse European ethnic groups. Mol Med. 2009, 15: 371-383.PubMedPubMed CentralView ArticleGoogle Scholar
- Behar DM, Yunusbayev B, Metspalu M, Metspalu E, Rosset S, Parik J, Rootsi S, Chaubey G, Kutuev I, Yudkovsky G, Khusnutdinova EK, Balanovsky O, Semino O, Pereira L, Comas D, Gurwitz D, Bonne-Tamir B, Parfitt T, Hammer MF, Skorecki K, Villems R: The genome-wide structure of the Jewish people. Nature. 2010, 466: 238-242. 10.1038/nature09103.PubMedView ArticleGoogle Scholar
- Atzmon G, Hao L, Pe'er I, Velez C, Pearlman A, Palamara PF, Morrow B, Friedman E, Oddoux C, Burns E, Ostrer H: Abraham's children in the genome era: major Jewish diaspora populations comprise distinct genetic clusters with shared Middle Eastern ancestry. Am J Hum Genet. 2010, 86: 850-859. 10.1016/j.ajhg.2010.04.015.PubMedPubMed CentralView ArticleGoogle Scholar
- Bray SM, Mulle JG, Dodd AF, Pulver AE, Wooding S, Warren ST: Signatures of founder effects, admixture, and selection in the Ashkenazi Jewish population. Proc Natl Acad Sci USA. 2010, 107: 16222-16227. 10.1073/pnas.1004381107.PubMedPubMed CentralView ArticleGoogle Scholar
- Lerer I, Sagi M, Cutting GR, Abeliovich D: Cystic fibrosis mutations delta F508 and G542X in Jewish patients. J Med Genet. 1992, 29: 131-133. 10.1136/jmg.29.2.131.PubMedPubMed CentralView ArticleGoogle Scholar
- Ben-Yosef T, Ness SL, Madeo AC, Bar-Lev A, Wolfman JH, Ahmed ZM, Desnick RJ, Willner JP, Avraham KB, Ostrer H, Oddoux C, Griffith AJ, Friedman TB: A mutation of PCDH15 among Ashkenazi Jews with the type 1 Usher syndrome. N Engl J Med. 2003, 348: 1664-1670. 10.1056/NEJMoa021502.PubMedView ArticleGoogle Scholar
- Ness SL, Ben-Yosef T, Bar-Lev A, Madeo AC, Brewer CC, Avraham KB, Kornreich R, Desnick RJ, Willner JP, Friedman TB, Griffith AJ: Genetic homogeneity and phenotypic variability among Ashkenazi Jews with Usher syndrome type III. J Med Genet. 2003, 40: 767-772. 10.1136/jmg.40.10.767.PubMedPubMed CentralView ArticleGoogle Scholar
- Pompei F, Ciminelli BM, Bombieri C, Ciccacci C, Koudova M, Giorgi S, Belpinati F, Begnini A, Cerny M, Des GM, Claustres M, Ferec C, Macek M, Modiano G, Pignatti PF: Haplotype block structure study of the CFTR gene. Most variants are associated with the M470 allele in several European populations. Eur J Hum Genet. 2006, 14: 85-93.PubMedGoogle Scholar
- Ciminelli BM, Bonizzato A, Bombieri C, Pompei F, Gabaldo M, Ciccacci C, Begnini A, Holubova A, Zorzi P, Piskackova T, Macek M, Castellani C, Modiano G, Pignatti PF: Highly preferential association of NonF508del CF mutations with the M470 allele. J Cyst Fibros. 2007, 6: 15-22. 10.1016/j.jcf.2006.04.003.PubMedView ArticleGoogle Scholar
- Rady PL, Tyring SK, Hudnall SD, Vargas T, Kellner LH, Nitowsky H, Matalon RK: Methylenetetrahydrofolate reductase (MTHFR): the incidence of mutations C677T and A1298C in the Ashkenazi Jewish population. Am J Med Genet. 1999, 86: 380-384. 10.1002/(SICI)1096-8628(19991008)86:4<380::AID-AJMG13>3.0.CO;2-9.PubMedView ArticleGoogle Scholar
- Hunt KA, Zhernakova A, Turner G, Heap GA, Franke L, Bruinenberg M, Romanos J, Dinesen LC, Ryan AW, Panesar D, Gwilliam R, Takeuchi F, McLaren WM, Holmes GK, Howdle PD, Walters JR, Sanders DS, Playford RJ, Trynka G, Mulder CJ, Mearin ML, Verbeek WH, Trimble V, Stevens FM, O'Morain C, Kennedy NP, Kelleher D, Pennington DJ, Strachan DP, McArdle WL, et al: Newly identified genetic risk variants for celiac disease related to the immune response. Nat Genet. 2008, 40: 395-402. 10.1038/ng.102.PubMedPubMed CentralView ArticleGoogle Scholar
- Smyth DJ, Plagnol V, Walker NM, Cooper JD, Downes K, Yang JH, Howson JM, Stevens H, McManus R, Wijmenga C, Heap GA, Dubois PC, Clayton DG, Hunt KA, van Heel DA, Todd JA: Shared and distinct genetic variants in type 1 diabetes and celiac disease. N Engl J Med. 2008, 359: 2767-2777. 10.1056/NEJMoa0807917.PubMedPubMed CentralView ArticleGoogle Scholar
- Gouda HN, Sagoo GS, Harding AH, Yates J, Sandhu MS, Higgins JP: The association between the peroxisome proliferator-activated receptor-gamma2 (PPARG2) Pro12Ala gene variant and type 2 diabetes mellitus: a HuGE review and meta-analysis. Am J Epidemiol. 2010, 171: 645-655. 10.1093/aje/kwp450.PubMedPubMed CentralView ArticleGoogle Scholar
- Scherer D, Nagore E, Bermejo JL, Figl A, Botella-Estrada R, Thirumaran RK, Angelini S, Hemminki K, Schadendorf D, Kumar R: Melanocortin receptor 1 variants and melanoma risk: a study of 2 European populations. Int J Cancer. 2009, 125: 1868-1875. 10.1002/ijc.24548.PubMedView ArticleGoogle Scholar
- Riemenschneider M, Konta L, Friedrich P, Schwarz S, Taddei K, Neff F, Padovani A, Kölsch H, Laws SM, Klopp N, Bickeböller H, Wagenpfeil S, Mueller JC, Rosenberger A, Diehl-Schmid J, Archetti S, Lautenschlager N, Borroni B, Müller U, Illig T, Heun R, Egensperger R, Schlegel J, Förstl H, Martins RN, Kurz A: A functional polymorphism within plasminogen activator urokinase (PLAU) is associated with Alzheimer's disease. Hum Mol Genet. 2006, 15: 2446-2456. 10.1093/hmg/ddl167.PubMedView ArticleGoogle Scholar
- Pawlik A, Kurzawski M, Szczepanik T, Dziedziejko V, Safranow K, Borowiec-Chłopek Z, Giedrys-Kalemba S, Drozdzik M: Association of allograft inflammatory factor-1 gene polymorphism with rheumatoid arthritis. Tissue Antigens. 2008, 72: 171-175. 10.1111/j.1399-0039.2008.01086.x.PubMedView ArticleGoogle Scholar
- Schmidt H, Williamson D, Ashley-Koch A: HLA-DR15 haplotype and multiple sclerosis: a HuGE review. Am J Epidemiol. 2007, 165: 1097-1109. 10.1093/aje/kwk118.PubMedView ArticleGoogle Scholar
- de Bakker PI, McVean G, Sabeti PC, Miretti MM, Green T, Marchini J, Ke X, Monsuur AJ, Whittaker P, Delgado M, Morrison J, Richardson A, Walsh EC, Gao X, Galver L, Hart J, Hafler DA, Pericak-Vance M, Todd JA, Daly MJ, Trowsdale J, Wijmenga C, Vyse TJ, Beck S, Murray SS, Carrington M, Gregory S, Deloukas P, Rioux JD: A high-resolution HLA and SNP haplotype map for disease association studies in the extended human MHC. Nat Genet. 2006, 38: 1166-1172. 10.1038/ng1885.PubMedPubMed CentralView ArticleGoogle Scholar
- Hughes LB, Morrison D, Kelley JM, Padilla MA, Vaughan LK, Westfall AO, Dwivedi H, Mikuls TR, Holers VM, Parrish LA, Parrish LA, Alarcón GS, Conn DL, Jonas BL, Callahan LF, Smith EA, Gilkeson GS, Howard G, Moreland LW, Patterson N, Reich D, Bridges SL: The HLA-DRB1 shared epitope is associated with susceptibility to rheumatoid arthritis in African Americans through European genetic admixture. Arthritis Rheum. 2008, 58: 349-358. 10.1002/art.23166.PubMedPubMed CentralView ArticleGoogle Scholar
- Prugnolle F, Manica A, Charpentier M, Guégan JF, Guernier V, Balloux F: Pathogen-riven selection and worldwide HLA class I diversity. Curr Biol. 2005, 15: 1022-1027. 10.1016/j.cub.2005.04.050.PubMedView ArticleGoogle Scholar
- Gregersen PK, Olsson LM: Recent advances in the genetics of autoimmune disease. Annu Rev Immunol. 2009, 27: 363-391. 10.1146/annurev.immunol.021908.132653.PubMedPubMed CentralView ArticleGoogle Scholar
- Mobini N, Yunis EJ, Alper CA, Yunis JJ, Delgado JC, Yunis DE, Firooz A, Dowlati Y, Bahar K, Gregersen PK, Ahmed AR: Identical MHC markers in non-Jewish Iranian and Ashkenazi Jewish patients with pemphigus vulgaris: possible common central Asian ancestral origin. Hum Immunol. 1997, 57: 62-67. 10.1016/S0198-8859(97)00182-1.PubMedView ArticleGoogle Scholar
- Daly AK, Donaldson PT, Bhatnagar P, Shen Y, Pe'er I, Floratos A, Daly MJ, Goldstein DB, John S, Nelson MR, Graham J, Park BK, Dillon JF, Bernal W, Cordell HJ, Pirmohamed M, Aithal GP, Day CP, DILIGEN Study; International SAE Consortium: HLA-B*5701 genotype is a major determinant of drug-induced liver injury due to flucloxacillin. Nat Genet. 2009, 41: 816-819. 10.1038/ng.379.PubMedView ArticleGoogle Scholar
- Fellay J, Ge D, Shianna KV, Colombo S, Ledergerber B, Cirulli ET, Urban TJ, Zhang K, Gumbs CE, Smith JP, Castagna A, Cozzi-Lepri A, De Luca A, Easterbrook P, Günthard HF, Mallal S, Mussini C, Dalmau J, Martinez-Picado J, Miro JM, Obel N, Wolinsky SM, Martinson JJ, Detels R, Margolick JB, Jacobson LP, Descombes P, Antonarakis SE, Beckmann JS, O'Brien SJ, et al: Common genetic variation and the control of HIV-1 in humans. PLoS Genet. 2009, 5: e1000791-10.1371/journal.pgen.1000791.PubMedPubMed CentralView ArticleGoogle Scholar
- International Schizophrenia Consortium, Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, Sullivan PF, Sklar P: Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009, 460: 748-752.PubMed CentralGoogle Scholar
- Kerem B, Chiba-Falek O, Kerem E: Cystic fibrosis in Jews: frequency and mutation distribution. Genet Test. 1997, 1: 35-39.PubMedGoogle Scholar
- Zhou Z, Lin XY, Akolkar PN, Gulwani-Akolkar B, Levine J, Katz S, Silver J: Variation at NOD2/CARD15 in familial and sporadic cases of Crohn's disease in the Ashkenazi Jewish population. Am J Gastroenterol. 2002, 97: 3095-3101. 10.1111/j.1572-0241.2002.07105.x.PubMedView ArticleGoogle Scholar
- Li JZ, Absher DM, Tang H, Southwick AM, Casto AM, Ramachandran S, Cann HM, Barsh GS, Feldman M, Cavalli-Sforza LL, Myers RM: Worldwide human relationships inferred from genome-wide patterns of variation. Science. 2008, 319: 1100-1104. 10.1126/science.1153717.PubMedView ArticleGoogle Scholar
- Rosenberg NA, Pritchard JK, Weber JL, Cann HM, Kidd KK, Zhivotovsky LA, Feldman MW: Genetic structure of human populations. Science. 2002, 298: 2381-2385. 10.1126/science.1078311.PubMedView ArticleGoogle Scholar
- Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, Maller J, Sklar P, de Bakker PIW, Daly MJ, Sham PC: PLINK: a toolset for whole-genome association and population-based linkage analysis. Am J Hum Genet. 2007, 81: 559-575. 10.1086/519795.PubMedPubMed CentralView ArticleGoogle Scholar
- Alexander DH, Novembre J, Lange K: Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009, 19: 1655-1664. 10.1101/gr.094052.109.PubMedPubMed CentralView ArticleGoogle Scholar
- Tang H, Peng J, Wang P, Risch NJ: Estimation of individual admixture: analytical and study design considerations. Genet Epidemiol. 2005, 28: 289-301. 10.1002/gepi.20064.PubMedView ArticleGoogle Scholar
- Rosenberg NA: DISTRUCT: a program for the graphical display of population structure. Mol Ecol Notes. 2004, 4: 137-138.View ArticleGoogle Scholar
- Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D: Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006, 38: 904-909. 10.1038/ng1847.PubMedView ArticleGoogle Scholar
- Rousset F: genepop'007: a complete re-implementation of the genepop software for Windows and Linux. Mol Ecol Resour. 2008, 8: 103-106. 10.1111/j.1471-8286.2007.01931.x.PubMedView ArticleGoogle Scholar
- Weir BS, Cockerham CC: Estimating F-statistics for the analysis of population structure. Evolution. 1984, 38: 1358-1370. 10.2307/2408641.View ArticleGoogle Scholar
- Raymond M, Rousset F: An exact test for population differentiation. Evolution. 1995, 49: 1283-1286.View ArticleGoogle Scholar
- Holmans P, Green EK, Pahwa JS, Ferreira MA, Purcell SM, Sklar P, Wellcome Trust Case-Control Consortium, Owen MJ, O'Donovan MC, Craddock N: Gene ontology analysis of GWA study data sets provides insights into the biology of bipolar disorder. Am J Hum Genet. 2009, 85: 13-24. 10.1016/j.ajhg.2009.05.011.PubMedPubMed CentralView ArticleGoogle Scholar
- Dennis G, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA: DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003, 4: P3-10.1186/gb-2003-4-5-p3.PubMedView ArticleGoogle Scholar
- International Schizophrenia Consortium, Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, Sullivan PF, Sklar P: Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009, 460: 748-752.PubMed CentralGoogle Scholar
- Yang J, Benyamin B, McEvoy BP, Gordon S, Henders AK, Nyholt DR, Madden PA, Heath AC, Martin NG, Montgomery GW, Goddard ME, Visscher PM: Common SNPs explain a large proportion of the heritability for human height. Nat Genet. 2010, 42: 565-569. 10.1038/ng.608.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang K, Li M, Hakonarson H: Analysing biological pathways in genome-wide association studies. Nat Rev Genet. 2010, 11: 843-854. 10.1038/nrg2884.PubMedView ArticleGoogle Scholar
- Holmans P: Statistical methods for pathway analysis of genome-wide data for association with complex genetic traits. Adv Genet. 2010, 72: 141-179.PubMedView ArticleGoogle Scholar
- Nam D, Kim J, Kim SY, Kim S: GSA-SNP: a general approach for gene set analysis of polymorphisms. Nucleic Acids Res. 2010, 38: W749-54. 10.1093/nar/gkq428.PubMedPubMed CentralView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.