Abstract
Background
The cost efficient twostage design is often used in genomewide association studies (GWASs) in searching for genetic loci underlying the susceptibility for complex diseases. Replicationbased analysis, which considers data from each stage separately, often suffers from loss of efficiency. Joint test that combines data from both stages has been proposed and widely used to improve efficiency. However, existing joint analyses are based on test statistics derived under an assumed genetic model, and thus might not have robust performance when the assumed genetic model is not appropriate.
Results
In this paper, we propose joint analyses based on two robust tests, MERT and MAX3, for GWASs under a twostage design. We developed computationally efficient procedures and formulas for significant level evaluation and power calculation. The performances of the proposed approaches are investigated through the extensive simulation studies and a real example. Numerical results show that the joint analysis based on the MAX3 test statistic has the best overall performance.
Conclusions
MAX3 joint analysis is the most robust procedure among the considered joint analyses, and we recommend using it in a twostage genomewide association study.
Background
The twostage design is often adopted in genomewide association studies (GWASs) to search for genetic variants underlying susceptibility for complex diseases. The advantages of the twostage design have been investigated extensively (see e.g., [112]). In a typical twostage design for GWASs, a proportion of the available samples are genotyped at the initial stage on a large number of single nucleotide polymorphisms (SNPs) using a commercial genotyping platform. Based on association test results obtained at this stage, a small percentage of SNPs are selected and further genotyped on the remaining samples in the second stage. To analyze data generated from such a twostage design, the joint analysis strategy has been recommended, which combines the test statistics from both stages as the final test statistic, and is shown to be more powerful than the replicationbased analysis that only utilizes the second stage data [12].
The efficiency of joint analysis based on the allelefrequencydifferencebased test (AFDT) was evaluated in detail in comparison to the replicationbased analysis [12]. It is commonly adopted as a single marker test in GWASs. The AFDT is valid when HardyWeinberg equilibrium (HWE) holds in the target population, and is powerful when the underlying genetic models are additive or multiplicative. The CochranArmitage trend test (CATT) [13,14] derived under the additive (in log scale) genetic risk model is also used in singlemaker analysis, which is optimal when the underlying additive genetic model is true. However, both tests are not so powerful compared with other methods such as MAX3 [15] when the underlying genetic model is not additive. Since in most cases there is no evidence suggesting that the additive risk model is most appropriate for the underlying disease model, especially in the typical GWASs where we most likely evaluate only the tagging SNPs, but not the causal SNPs directly. Thus, it is advantageous to adopt a more robust single marker test that has a relatively good performance under all possible disease models. To this end, two types of such robust tests, the MERT (maximin efficiency robust test) [15,16] and MAX3 (the maximum values of CATTs under recessive, additive and dominant models) have been recently considered [15,17]. Nevertheless, their performances under the twostage design have not been thoroughly investigated.
In this report we propose two types of joint test statistics for the twostage design based on the two robust tests, MERT and MAX3. We derive closedform formula to calculate the power of the MERTbased joint analysis, and propose a computationally efficient Monte Carlo procedure to evaluate the significance level of the MAX3based joint analysis. Facilitated by these two procedures, we evaluate the performances of the two robust test based joint analyses, in comparison with the ones based on AFDT, under various twostage design setups and disease models.
Methods
Notations
Suppose that r cases and s controls are randomly sampled from the source population in a GWAS. Denote the number of SNPs genotyped and the proportion of the subjects in Stage 1 by m and π, respectively. Throughout, we only consider biallelic SNPs with two alleles G and g, with G being the risk allele. Then there are three genotypes: gg, Gg, and GG. Using the disease risk at gg as the baseline, we define the relative risks of Gg and GG as λ_{1 }= f_{1}/f_{0 }and λ_{2 }= f_{2}/f_{0}, respectively, where f_{0 }= Pr(casegg) > 0, f_{1 }= Pr(caseGg), f_{2 }= Pr(caseGG) are the penetrances. Let K = Pr(case) be the disease prevalence. Denote the genotype frequencies in case population as p_{0 }= Pr(ggcase) = Pr(gg)f_{0}/K, p_{1 }= Pr(Ggcase) = Pr(Gg)f_{1}/K, p_{2 }= Pr(GGcase) = Pr(GG)f_{2}/K and in control population as q_{0 }= Pr(ggcontrol) = Pr(gg)(1f_{0})/(1K), q_{1 }= Pr(Ggcontrol) = Pr(Gg)(1f_{1})/(1K), q_{2 }= Pr(GGcontrol) = Pr(GG)(1f_{2})/(1K). Then the null hypothesis of no association is H_{0 }: p_{i }= q_{i}, i = 0, 1, 2, which is equivalent to H_{0 }: λ_{1 }= λ_{2 }= 1. The alternative hypothesis is H_{1 }: λ_{2 }≥ λ_{1 }≥ 1 with λ_{2 }> 1. The commonly used three genetic models, recessive, additive and dominant models are corresponding to λ_{2 }>λ_{1 }= 1, 2λ_{1 }= λ_{2 }+1 and λ_{1 }= λ_{2 }> 1, respectively. We assume that SNPs with pvalues less than γ in Stage 1 will be further investigated in Stage 2 and α be the whole genomewide type I error.
The notations for genotype frequencies in case population and control population of Stage 1 and Stage 2 are given in Table 1. It should be noted that p_{1i }= p_{2i }and q_{1i }= q_{2i }for i = 0, 1, 2 in the table using the first subscript on behalf of Stage 1 or Stage 2 since they are the population parameters. However, the estimates of p_{1i }and q_{1i }for i = 0, 1, 2 based on the data of Stage 1 and those of p_{2i }and q_{2i }for i = 0, 1, 2 based on the data of Stage 2 might be different although the data of Stage 1 and Stage 2 are drawn from the same source population.
Table 1. Genotype frequencies in case population and control population for both stages
AlleleFrequencyDifferenceBased Joint Analysis
Denote the risk allele frequencies in case population and control population by θ and ϖ, respectively. Let
The threshold for selecting SNPs in Stage 1 is b_{1 }= Φ^{1}(1γ/2), where Φ(·) is the cumulative standard normal distribution function. Similarly,
we can get the maximum likelihood estimates of the risk allele frequencies in case
population and control population using the data from Stage 2, denoted by
The joint statistic is
where
The calculation of
The above approach is slightly different from the one considered in [12], where the authors constructed the test statistics by estimating the variance of the differences of allele frequency between case population and control population using the cases and controls separately under the null hypothesis. In our joint analysis, we estimated the variance using the combined data of case sample and control sample. Results (not show here) show that the two approaches have very similar performance.
CochranArmitage Trend Test under the Additive ModelBased Joint Analysis
CochranArmitage trend test under the additive model (CATTA) (see e.g., [13,15]) is often used in the genetic association studies including GWASs. Denote CATTA for
both stages by
Additional file 1. Appendix for the main text. The file (including Appendix A, B, C) is a Microsoft Word document. Appendix A gives
a detailed description of the joint distribution of the additive trend test statistic
Format: DOC Size: 190KB Download file
This file can be viewed with: Microsoft Word Viewer
MERTBased Joint Analysis
MERT was originally proposed in [16] to find robust test statistic in situations when multiple alternative models are
plausible. It was used to define a robust test for singlemarker analysis [15]. Here we apply the test to twostage design. Similar to
The power of the test is given by
MAX3Based Joint Analysis
MAX3, the maximal value of CATT under three genetic models, is another commonly used
robust test in the current GWASs (see e.g., [7,15,17]). Once we have
According to Chapter 6 of [21], we have
1) Generate B identical and independently distributed bivariate normal random variates
2) Generate B_{1 }identical and independently distributed bivariate normal random variates
3) Find v_{J }that yields
Once we have v_{1 }and v_{J}, we generate the data under the alternative hypothesis to calculate the power empirically.
In the simulation studies, we generate 10,000 data sets under the alternative hypothesis.
For the i^{th }data set (i = 1, 2. ⋯, 10000), we calculate
Results
Simulation Setup
In order to mimic the real GWAS, we choose the simulation parameters similar to [12,22]. In a typical GWAS, there are thousands of individuals randomly chosen from the source population and the number of SNPs being examined in Stage 1 is usually from 0.1 million to 1 million. Based on the results of Stage 1 (pvalues), the number of SNPs to be genotyped in Stage 2 is in tens or hundreds. For example, in a diabetes mellitus GWAS [7], there were 392,935 SNPs genotyped on 1,363 subjects in Stage 1, and 57 SNPs were genotyped in Stage 2 after removing those SNPs with pvalues greater than 0.0001 based on the data of Stage 1. In a GWAS, the significance level in the whole genome is often set to be 0.05, and the Bonferronicorrection is often used to adjust for multiple comparisons and to control the false positive rate. So, in our simulation studies, we set the number of SNPs at Stage 1 m = 500,000 and the pvalue threshold for significant SNPs to be 0.05/m = 1 ×10^{7}. The proportion of subjects genotyped in Stage 1 is set to be 0.5, 0.4 and 0.3, and the pvalue threshold for SNPs selection at the end of Stage 1 be 0.0001 and 0.0002. The disease prevalence is set to be K = 0.1. Throughout our simulation procedures, we assume that HardyWeinberg equilibrium (HWE) holds in the general population. Furthermore, the risk allele is assumed to be the minor allele, with frequency (MAF) equal to 0.15, 0.25, 0.35 and 0.45. The considered genetic models are the recessive, additive, and dominant models. We specified different genotype relative risks λ_{1 }and λ_{2 }for the three genetic models (see details in Table 2, 3, 4 and 5). The critical values for MAX3 joint analysis are simulated, while thresholds for other three joint analysis are exactly calculated based on their asymptotic distributions under the null hypothesis where the genotype probabilities (p_{0}, p_{1}, p_{2}) for cases and (q_{0}, q_{1}, q_{2}) for controls are calculated by p_{0 }= q_{0 }= Pr(gg) = (1MAF)^{2}, p_{1 }= q_{1 }= Pr(Gg) = 2 × MAF ×(1MAF) and p_{2 }= q_{2 }= Pr(GG) = MAF^{2}. Under the alternative hypothesis, the genotype frequencies can be obtained using the formulas given in the Notations Subsection and f_{0 }= K/[Pr(gg) + λ_{1 }Pr(Gg) + λ_{2 }Pr(GG)]. More details could be referred to [23] and [24]. The genotype counts in case sample and control sample were generated from a multinomial distribution.
Table 2. Power comparison for MAF = 0.15 (K = 0.1, α = 0.05, m = 5 × 10^{5})
Table 3. Power comparison for MAF = 0.25 (K = 0.1, α = 0.05, m = 5 × 10^{5})
Table 4. Power comparison for MAF = 0.35 (K = 0.1, α = 0.05, m = 5 × 10^{5})
Table 5. Power comparison for MAF = 0.45 (K = 0.1, α = 0.05, m = 5 × 10^{5})
Simulation Results
For convenience, we refer to the aforementioned four joint analysis approaches as, respectively, ALLEJ (allelefrequencydifferencebased joint analysis), CATAJ (CochranArmitage trend test under the additive modelbased joint analysis), MERTJ (MERTbased joint analysis), MAX3J (MAX3based joint analysis). Table 2, 3, 4 and 5 report powers of the four joint analysis methods corresponding to MAF equal to 0.15, 0.25, 0.35 and 0.45, respectively. From these tables, we have the following observations. Under the recessive model, MERTJ and MAX3J are more powerful than ALLEJ and CATAJ, with MAX3J being most powerful among the four methods under consideration. In some cases, the advantage of MAX3J is quite impressive. For example, in Table 2, with π = 0.5, γ = 0.0001, the powers of ALLEJ, CATAJ, MERTJ and MAX3J are 0.070, 0.058, 0.385 and 0.759, respectively. Under the additive model, CATAJ and ALLEJ have comparable power and are more powerful than the other two tests. However, the power difference between CATAJ and MAX3J is mostly at the level of 6.6%, with the largest discrepancy of 7%. Under the dominant model, CATAJ and MAX3J are more powerful than ALLEJ and MERTJ. Both tests have comparable power when MAF = 0.15, and MAX3J is much more powerful than CATAJ when MAF = 0.25, 0.35 and 0.45. In summary, it appears that MAX3J has the best overall performance.
A Real Example: Type 2 Diabetes Mellitus
Type 2 diabetes mellitus is one of the most common diseases, and has been found to be associated with environmental factors and genetic variants. A twostage GWAS for type 2 diabetes mellitus was reported in [7]. In this study, 392,935 SNPs were genotyped on 1,363 subjects in Stage 1. Based on the statistical significance level of 1 × 10^{4}, 57 SNPs were selected and further screened on 2,617 cases and 2,894 controls in Stage 2. We applied the above four considered methods to two SNPs, rs1005316 and rs2876711, which were not reported in their Table 1, but were shown in their Appendix. Table 6 gives the genotype counts and pvalues of these two SNPs. We found a genomewide significant association between rs2876711 and the outcome. Although the association between rs2876711 and type 2 diabetes mellitus has not been reported by [7], our results show that we should be concerned with this SNP and its neighborhood area. Additional experiments should be further conducted to validate this association.
Table 6. Genotype counts and pvalues of SNPs rs1005316 and rs2876711 for type 2 diabetes mellitus
Discussion and Conclusions
In genetic association studies, the underlying genetic inheritance model is often unknown, and thus hinders the use of methods such as CATT, which has to be derived under an assumed genetic model. Robust tests, such as MERT and MAX3, had been proposed to relax the dependence on the underlying genetic models. Extending these tests to a twostage setting, we construct two robust joint analyses based on MERT and MAX3. Numerical results show that MAX3J has the best overall performance among the four considered joint analysis approaches. For type 2 diabetes mellitus, based on MAX3J, we found that SNP rs2876711 was significantly associated with type 2 diabetes mellitus besides their findings.
Pearson Chisquare test is a robust test that was used in genetic association studies (see e.g., [25]). Recently, a comprehensive power comparison between MAX3 and Pearson Chisquare test and CochranArmitage trend test under the additive model was conducted in [17]. They reported that MAX3 has the most robust performances. The proposed joint analysis combing the test statistics of both stages considers the betweenstage heterogeneity. It is intractable for Pearson Chisquare test to consider the relative risk heterogeneity of both stages, especially when the relative risk in Stage 1 is larger than one and that in Stage 2 is less than one.
Recently, a joint analysis based on genetic model selection [26] to overcome the genetic model uncertainty was proposed in [22]. Based on the data in Stage 1, they used HardyWeinberg disequilibrium trend test studied in [27] to determine a score that corresponds to a genetic model. This score was then used to construct the trend test based on the data of Stage 2. Results (not shown here) show that the proposed joint analysis has comparable power. Therefore, the proposed MAX3J can be used as an alternative procedure in twostage genomewide association studies.
Abbreviations
GWAS: genomewide association study; SNP: single nucleotide polymorphism; MAF: minor allele frequency; AFDT: allelefrequencydifferencebased test; CATT: CochranArmitage trend test; MERT: maximin efficiency robust test; MAX3: maximum values of CochranArmitage trend tests under recessive, additive and dominant models; ALLEJ: allelefrequencydifferencebased joint analysis; CATAJ: CochranArmitage trend test under the additive modelbased joint analysis; MERTJ: MERTbased joint analysis; MAX3J: MAX3based joint analysis.
Authors' contributions
PD and LQ implemented the model. All authors read and approved the final manuscript.
Acknowledgements
We would like to thank the editor and three anonymous reviewers for their very constructive comments and suggestions, which significantly improved our presentation. This work is partially support by the National Young Science Foundation of China, No. 10901155.
References

Zuo YJ, Zou GH, Zhao HY: Twostage designs in casecontrol association analysis.
Genetics 2006, 173:17471760. PubMed Abstract  Publisher Full Text  PubMed Central Full Text

Goll A, Bauer P: Twostage designs applying methods differing in costs.
Bioinformatics 2007, 23:15191526. PubMed Abstract  Publisher Full Text

Muller HH, Pahl R, Schafer H: Including sampling and phenotyping costs into the optimization of twostage designs for genomewide association studies.
Genetic Epidemiology 2007, 31:844852. PubMed Abstract  Publisher Full Text

Satagopan JM, Elston RC: Optimal twostage genotyping in populationbased association studies.
Genetic Epidemiology 2003, 25:149157. PubMed Abstract  Publisher Full Text

Satagopan JM, Venkatraman ES, Begg CB: Twostage designs for genedisease association studies with sample size constraints.
Biometrics 2004, 60:589597. PubMed Abstract  Publisher Full Text

Skol AD, Scott LJ, Abecasis GR, Boehnke M: Optimal designs for twostage genomewide association studies.
Genetic Epidemiology 2007, 31:776788. PubMed Abstract  Publisher Full Text

Sladek R, Rocheleau G, Rung J, Dina C, Shen L, Serre D, Boutin P, Vincent D, Belisle A, Hadjadj S, Balkau B, Heude B, Charpentier G, Hudson TJ, Montpetit A, Pshezhetsky AV, Prentki M, Posner BI, Balding DJ, Meyre D, Polychronakos C, Froguel P: A genomewide association study identifies novel risk loci for type 2 diabetes.
Nature 2007, 445:881885. PubMed Abstract  Publisher Full Text

Thomas D, Xie R, Gebregziabher M: Twostage sampling designs for gene association studies.
Genetic Epidemiology 2004, 27:401414. PubMed Abstract  Publisher Full Text

Wang H, Thomas DC, Pe'er I, Stram DO: Optimal twostage designs for genomewide association scans.
Genetic Epidemiology 2006, 30:356368. PubMed Abstract  Publisher Full Text

Yu K, Chatterjee N, Wheeler W, Li Q, Wang S, Rothman N, Wacholder S: Flexible design for following up positive findings.
American Journal of Human Genetics 2007, 81:540551. PubMed Abstract  Publisher Full Text  PubMed Central Full Text

Zheng G, Meyer M, Li W, Yang Y: Comparison of twophase analyses for casecontrol genetic association studies.
Statistics in Medicine 2008, 27:50545075. PubMed Abstract  Publisher Full Text

Skol AD, Scott LJ, Abecasis GR, Boehnke M: Joint analysis is more efficient than replicationbased analysis for twostage genomewide association studies.
Nature Genetics 2006, 38:209213. PubMed Abstract  Publisher Full Text

Sasieni PD: From genotypes to genes: Doubling the sample size.
Biometrics 1997, 53:12531261. PubMed Abstract  Publisher Full Text

Zheng G, Gastwirth JL: On estimation of the variance in CochranArmitage trend tests for genetic association using casecontrol studies.
Statistics in Medicine 2006, 25:31503159. PubMed Abstract  Publisher Full Text

Zheng G, Friedlin B, Gastwirth JL: Comparison of robust tests for genetic association using casecontrol studies. In Optimality: The Second Erich L. Lehmann Symposium. Edited by Rojo J. Beachwood: Institute of Mathematical Statistics; 2006:253265.
[DasGupta A (Series Editor): Lecture NotesMonograph Series, vol 49.]

Gastwirth JL: The use of maximin efficiency robust tests in combining contingency tables and survival analysis.
Journal of the American Statistical Association 1985, 80:380384. Publisher Full Text

Li Q, Zheng G, Liang X, Yu K: Robust tests for singlemarker analysis in casecontrol genetic association studies.
Annals of Human Genetics 2009, 73:245252. PubMed Abstract  Publisher Full Text

Genz A: Numerical computation of multivariate normal probabilities.
Journal of Computational and Graphical Statistics 1992, 1:141150. Publisher Full Text

Genz A: Comparison of methods for the computation of multivariate normal probabilities.

Tong YL: The multivariate normal distribution. New York: SpringerVerlag; 1990.

Zheng G, Yang Y, Zhu X, Elston RC: Casecontrol studies of genetic association. New York: Springer; 2010.

Kwak M, Joo J, Zheng G: A robust test for twostage design in genomewide association studies.
Biometrics 2009, 65:12881295. PubMed Abstract  Publisher Full Text

Schaid DJ, Sommer SS: Genotype relative risks: methods for design and analysis of candidategene association studies.
American Journal of Human Genetics 1993, 53:11141126. PubMed Abstract  PubMed Central Full Text

Terwilliger JD, Ott J: Handbook of human genetic linkage. Baltimore: Johns Hopkins University Press; 1994.

Yeager M, Orr N, Hayes RB, Jacobs KB, Kraft P, Wacholder S, Minichiello MJ, Fearnhead P, Yu K, Chatterjee N, Wang Z, Welch R, Staats BJ, Calle EE, Feigelson HS, Thun MJ, Rodriguez C, Albanes D, Virtamo J, Weinstein S, Schumacher FR, Giovannucci E, Willett WC, CancelTassin G, Cussenot O, Valeri A, Andriole GL, Gelmann EP, Tucker M, Gerhard DS, Fraumeni JF Jr, Hoover R, Hunter DJ, Chanock SJ, Thomas G: Genomewide association study of prostate cancer identifies a second risk locus at 8q24.
Nature Genetics 2007, 39:645649. PubMed Abstract  Publisher Full Text

Zheng G, Ng HKT: Genetic model selection in twophase analysis for case control association studies.
Biostatistics 2008, 9:391399. PubMed Abstract  Publisher Full Text

Song K, Elson RC: A powerful method of combining measures of association and HardyWeinberg disequilibrium for finemapping in casecontrol studies.
Statistics in Medicine 2006, 25:105126. PubMed Abstract  Publisher Full Text