Limits...
A unified framework for multi-locus association analysis of both common and rare variants.

Shriner D, Vaughan LK - BMC Genomics (2011)

Bottom Line: We identified a single risk haplotype with a directionally consistent effect in both samples in the gene GAK, which is involved in clathrin-mediated membrane trafficking.We also found suggestive evidence that directionally inconsistent marginal effects from single marker analysis appeared to result from risk being driven by different haplotypes in the two samples for the genes SYN3 and NGLY1, which are involved in neurotransmitter release and proteasomal degradation, respectively.These results illustrate the utility of our unified framework for genome-wide association analysis of common, complex diseases.

View Article: PubMed Central - HTML - PubMed

Affiliation: Center for Research on Genomics and Global Health, National Human Genome Research Institute, Bethesda, MD 20892, USA. shrinerda@mail.nih.gov

ABSTRACT

Background: Common, complex diseases are hypothesized to result from a combination of common and rare genetic variants. We developed a unified framework for the joint association testing of both types of variants. Within the framework, we developed a union-intersection test suitable for genome-wide analysis of single nucleotide polymorphisms (SNPs), candidate gene data, as well as medical sequencing data. The union-intersection test is a composite test of association of genotype frequencies and differential correlation among markers.

Results: We demonstrated by computer simulation that the false positive error rate was controlled at the expected level. We also demonstrated scenarios in which the multi-locus test was more powerful than traditional single marker analysis. To illustrate use of the union-intersection test with real data, we analyzed a publically available data set of 319,813 autosomal SNPs genotyped for 938 cases of Parkinson disease and 863 neurologically normal controls for which no genome-wide significant results were found by traditional single marker analysis. We also analyzed an independent follow-up sample of 183 cases and 248 controls for replication.

Conclusions: We identified a single risk haplotype with a directionally consistent effect in both samples in the gene GAK, which is involved in clathrin-mediated membrane trafficking. We also found suggestive evidence that directionally inconsistent marginal effects from single marker analysis appeared to result from risk being driven by different haplotypes in the two samples for the genes SYN3 and NGLY1, which are involved in neurotransmitter release and proteasomal degradation, respectively. These results illustrate the utility of our unified framework for genome-wide association analysis of common, complex diseases.

Show MeSH

Related in: MedlinePlus

Validity and power analysis of union testing. Each simulated data set consisted of 938 cases and 863 controls. For each experiment, 1,000 independent replicates were simulated. Coded genotypes were simulated by randomly sampling from a binomial distribution with the given frequencies. Case-control status was determined by a logistic model. Gray lines indicate the per comparison significance threshold of 0.05. A) Per comparison error rate. Unions of one, two, three, four, and five SNPs are represented by black, red, orange, green, and blue lines, respectively. B) Power for an odds ratio of 1.5 for each SNP. C) Power for an odds ratio of 2 for each SNP. D) Power for 2-marker unions with opposing effects. The black line represents odds ratios of 2 and 0.5 for the two markers, the red line represents 1.5 and 0.67, and the blue line represents 1 and 1. E) Power for unions consisting of one predictor with an odds ratio of 2 (black line), 1.5 (red line), or 1 (blue line), and four predictors with odds ratios of 1. F) Power for 2-marker unions with correlated predictors. Solid lines represent independent predictors and dotted lines represent predictors correlated at r2 = 0.8. Black lines represent odds ratios of 2, red lines represent 1.5, and blue lines represent 1. G) Power to detect epistasis for 2-marker unions. The black line represents odds ratios of 1 for both markers and 1 for the epistatic effect. The red line represents odds ratios of 1 for both markers and 2 for the epistatic effect. The blue line represents odds ratios of 1 for both markers and 0.5 for the epistatic effect. H) Power to detect differential correlation between 2-marker unions. Black lines represent r2 = 0 in controls and r2 = 0.8 in cases. Red lines represent r2 = 0.8 in controls and r2 = 0 in cases. Solid lines represent odds ratios of 2 and dotted lines represent 1.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3040731&req=5

Figure 1: Validity and power analysis of union testing. Each simulated data set consisted of 938 cases and 863 controls. For each experiment, 1,000 independent replicates were simulated. Coded genotypes were simulated by randomly sampling from a binomial distribution with the given frequencies. Case-control status was determined by a logistic model. Gray lines indicate the per comparison significance threshold of 0.05. A) Per comparison error rate. Unions of one, two, three, four, and five SNPs are represented by black, red, orange, green, and blue lines, respectively. B) Power for an odds ratio of 1.5 for each SNP. C) Power for an odds ratio of 2 for each SNP. D) Power for 2-marker unions with opposing effects. The black line represents odds ratios of 2 and 0.5 for the two markers, the red line represents 1.5 and 0.67, and the blue line represents 1 and 1. E) Power for unions consisting of one predictor with an odds ratio of 2 (black line), 1.5 (red line), or 1 (blue line), and four predictors with odds ratios of 1. F) Power for 2-marker unions with correlated predictors. Solid lines represent independent predictors and dotted lines represent predictors correlated at r2 = 0.8. Black lines represent odds ratios of 2, red lines represent 1.5, and blue lines represent 1. G) Power to detect epistasis for 2-marker unions. The black line represents odds ratios of 1 for both markers and 1 for the epistatic effect. The red line represents odds ratios of 1 for both markers and 2 for the epistatic effect. The blue line represents odds ratios of 1 for both markers and 0.5 for the epistatic effect. H) Power to detect differential correlation between 2-marker unions. Black lines represent r2 = 0 in controls and r2 = 0.8 in cases. Red lines represent r2 = 0.8 in controls and r2 = 0 in cases. Solid lines represent odds ratios of 2 and dotted lines represent 1.

Mentions: We first investigated the validity and power of our proposed method, exploring over sizes of unions of SNPs ranging from one to five. Note that analysis of a union of one SNP is identical to single marker analysis. Under the hypothesis of no association, analysis of unions has the expected per comparison error rate and is therefore valid (Figure 1A). For small effect sizes (odds ratios ≤ 2), analysis of unions is increasingly more powerful than single marker analysis as the frequency of the minor multi-locus genotype decreases (Figs. 1B and 1C). If a union consists of both risk-increasing and risk-decreasing predictors, effects will cancel and power will be lost (Figure 1D). Also, if a union consists of too many predictors with no effect on the outcome, then power to detect a predictor with an effect within the union will decrease (Figure 1E). Analysis of unions of correlated predictors can be more powerful than analysis of unions of independent predictors (Figure 1F). Epistasis (nonadditivity on the logit scale) can either increase or decrease power, depending on the directions of the effect sizes for epistatic effects vs. marginal effects (Figure 1G). For epistasis to be detectable, minor genotype frequencies must be large so that the joint genotype counts are reasonably large.


A unified framework for multi-locus association analysis of both common and rare variants.

Shriner D, Vaughan LK - BMC Genomics (2011)

Validity and power analysis of union testing. Each simulated data set consisted of 938 cases and 863 controls. For each experiment, 1,000 independent replicates were simulated. Coded genotypes were simulated by randomly sampling from a binomial distribution with the given frequencies. Case-control status was determined by a logistic model. Gray lines indicate the per comparison significance threshold of 0.05. A) Per comparison error rate. Unions of one, two, three, four, and five SNPs are represented by black, red, orange, green, and blue lines, respectively. B) Power for an odds ratio of 1.5 for each SNP. C) Power for an odds ratio of 2 for each SNP. D) Power for 2-marker unions with opposing effects. The black line represents odds ratios of 2 and 0.5 for the two markers, the red line represents 1.5 and 0.67, and the blue line represents 1 and 1. E) Power for unions consisting of one predictor with an odds ratio of 2 (black line), 1.5 (red line), or 1 (blue line), and four predictors with odds ratios of 1. F) Power for 2-marker unions with correlated predictors. Solid lines represent independent predictors and dotted lines represent predictors correlated at r2 = 0.8. Black lines represent odds ratios of 2, red lines represent 1.5, and blue lines represent 1. G) Power to detect epistasis for 2-marker unions. The black line represents odds ratios of 1 for both markers and 1 for the epistatic effect. The red line represents odds ratios of 1 for both markers and 2 for the epistatic effect. The blue line represents odds ratios of 1 for both markers and 0.5 for the epistatic effect. H) Power to detect differential correlation between 2-marker unions. Black lines represent r2 = 0 in controls and r2 = 0.8 in cases. Red lines represent r2 = 0.8 in controls and r2 = 0 in cases. Solid lines represent odds ratios of 2 and dotted lines represent 1.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3040731&req=5

Figure 1: Validity and power analysis of union testing. Each simulated data set consisted of 938 cases and 863 controls. For each experiment, 1,000 independent replicates were simulated. Coded genotypes were simulated by randomly sampling from a binomial distribution with the given frequencies. Case-control status was determined by a logistic model. Gray lines indicate the per comparison significance threshold of 0.05. A) Per comparison error rate. Unions of one, two, three, four, and five SNPs are represented by black, red, orange, green, and blue lines, respectively. B) Power for an odds ratio of 1.5 for each SNP. C) Power for an odds ratio of 2 for each SNP. D) Power for 2-marker unions with opposing effects. The black line represents odds ratios of 2 and 0.5 for the two markers, the red line represents 1.5 and 0.67, and the blue line represents 1 and 1. E) Power for unions consisting of one predictor with an odds ratio of 2 (black line), 1.5 (red line), or 1 (blue line), and four predictors with odds ratios of 1. F) Power for 2-marker unions with correlated predictors. Solid lines represent independent predictors and dotted lines represent predictors correlated at r2 = 0.8. Black lines represent odds ratios of 2, red lines represent 1.5, and blue lines represent 1. G) Power to detect epistasis for 2-marker unions. The black line represents odds ratios of 1 for both markers and 1 for the epistatic effect. The red line represents odds ratios of 1 for both markers and 2 for the epistatic effect. The blue line represents odds ratios of 1 for both markers and 0.5 for the epistatic effect. H) Power to detect differential correlation between 2-marker unions. Black lines represent r2 = 0 in controls and r2 = 0.8 in cases. Red lines represent r2 = 0.8 in controls and r2 = 0 in cases. Solid lines represent odds ratios of 2 and dotted lines represent 1.
Mentions: We first investigated the validity and power of our proposed method, exploring over sizes of unions of SNPs ranging from one to five. Note that analysis of a union of one SNP is identical to single marker analysis. Under the hypothesis of no association, analysis of unions has the expected per comparison error rate and is therefore valid (Figure 1A). For small effect sizes (odds ratios ≤ 2), analysis of unions is increasingly more powerful than single marker analysis as the frequency of the minor multi-locus genotype decreases (Figs. 1B and 1C). If a union consists of both risk-increasing and risk-decreasing predictors, effects will cancel and power will be lost (Figure 1D). Also, if a union consists of too many predictors with no effect on the outcome, then power to detect a predictor with an effect within the union will decrease (Figure 1E). Analysis of unions of correlated predictors can be more powerful than analysis of unions of independent predictors (Figure 1F). Epistasis (nonadditivity on the logit scale) can either increase or decrease power, depending on the directions of the effect sizes for epistatic effects vs. marginal effects (Figure 1G). For epistasis to be detectable, minor genotype frequencies must be large so that the joint genotype counts are reasonably large.

Bottom Line: We identified a single risk haplotype with a directionally consistent effect in both samples in the gene GAK, which is involved in clathrin-mediated membrane trafficking.We also found suggestive evidence that directionally inconsistent marginal effects from single marker analysis appeared to result from risk being driven by different haplotypes in the two samples for the genes SYN3 and NGLY1, which are involved in neurotransmitter release and proteasomal degradation, respectively.These results illustrate the utility of our unified framework for genome-wide association analysis of common, complex diseases.

View Article: PubMed Central - HTML - PubMed

Affiliation: Center for Research on Genomics and Global Health, National Human Genome Research Institute, Bethesda, MD 20892, USA. shrinerda@mail.nih.gov

ABSTRACT

Background: Common, complex diseases are hypothesized to result from a combination of common and rare genetic variants. We developed a unified framework for the joint association testing of both types of variants. Within the framework, we developed a union-intersection test suitable for genome-wide analysis of single nucleotide polymorphisms (SNPs), candidate gene data, as well as medical sequencing data. The union-intersection test is a composite test of association of genotype frequencies and differential correlation among markers.

Results: We demonstrated by computer simulation that the false positive error rate was controlled at the expected level. We also demonstrated scenarios in which the multi-locus test was more powerful than traditional single marker analysis. To illustrate use of the union-intersection test with real data, we analyzed a publically available data set of 319,813 autosomal SNPs genotyped for 938 cases of Parkinson disease and 863 neurologically normal controls for which no genome-wide significant results were found by traditional single marker analysis. We also analyzed an independent follow-up sample of 183 cases and 248 controls for replication.

Conclusions: We identified a single risk haplotype with a directionally consistent effect in both samples in the gene GAK, which is involved in clathrin-mediated membrane trafficking. We also found suggestive evidence that directionally inconsistent marginal effects from single marker analysis appeared to result from risk being driven by different haplotypes in the two samples for the genes SYN3 and NGLY1, which are involved in neurotransmitter release and proteasomal degradation, respectively. These results illustrate the utility of our unified framework for genome-wide association analysis of common, complex diseases.

Show MeSH
Related in: MedlinePlus