Volume 3 Supplement 7
Genetic Analysis Workshop 16
Analysis of genomewide association data by largescale Bayesian logistic regression
 Yuanjia Wang^{1}Email author,
 Nanshi Sha^{1} and
 Yixin Fang^{2}
DOI: 10.1186/175365613S7S16
© Wang et al; licensee BioMed Central Ltd. 2009
Published: 15 December 2009
Abstract
Singlelocus analysis is often used to analyze genomewide association (GWA) data, but such analysis is subject to severe multiple comparisons adjustment. Multivariate logistic regression is proposed to fit a multilocus model for casecontrol data. However, when the sample size is much smaller than the number of singlenucleotide polymorphisms (SNPs) or when correlation among SNPs is high, traditional multivariate logistic regression breaks down. To accommodate the scale of data from a GWA while controlling for collinearity and overfitting in a high dimensional predictor space, we propose a variable selection procedure using Bayesian logistic regression. We explored a connection between Bayesian regression with certain priors and L_{1} and L_{2} penalized logistic regression. After analyzing large number of SNPs simultaneously in a Bayesian regression, we selected important SNPs for further consideration. With much fewer SNPs of interest, problems of multiple comparisons and collinearity are less severe. We conducted simulation studies to examine probability of correctly selecting disease contributing SNPs and applied developed methods to analyze Genetic Analysis Workshop 16 North American Rheumatoid Arthritis Consortium data.
Background
Singlelocus analysis is a widely used approach to analyze genomewide association (GWA) data, but it may not be adequate to capture complex pattern of disease etiology [1] and is subject to severe multiple comparisons adjustment, especially in a GWA, in which the typical number of comparisons made is hundreds of thousands. Methods to handle large number of singlenucleotide polymorphisms (SNPs) simultaneously are in demand. Logistic regression is a popular tool to assess association between a dichotomous trait and SNP genotypes. To analyze multiple SNPs simultaneously by logistic regression, one can include all SNPs of interest as predictors. A challenge of applying such approaches to GWA data is that the sample size is usually much smaller than the number of SNPs. Traditional multivariate logistic regression breaks down in this case. Another disadvantage of such an approach is that when the correlation between SNPs is high due to linkage disequilibrium (LD), the estimated coefficients are highly variable and the method performs poorly.
To accommodate large number of SNPs from a GWA while controlling for collinearity and overfitting in a high dimensional predictor space, we propose a variable selection procedure using Bayesian logistic regression. We explored a connection between certain priors and penalized logistic regression. After analysing large number of SNPs simultaneously in a Bayesian logistic regression, we selected important SNPs for further consideration. With much fewer selected SNPs of interest, problems of multiple comparisons and collinearity are less severe. We conducted simulation studies to examine the probability of correctly selecting disease contributing SNPs. Finally, we applied the methods to analyze Genetic Analysis Workshop (GAW) 16 Problem 1 chromosome 9 data.
Methods
Maximum likelihood is used to estimate parameters in the model. When the number of predictors exceeds the sample size, traditional logistic regression breaks down. In addition, when the predictors are high correlated, the maximum likehood estimate from Eq. (1) is of poor quality.
Gaussian prior and L_{2}penalty
where l is the log likelihood of the data. Choosing prior variances σ^{2} is equivalent to choosing smoothing parameter λ. This is also the ridge regression.
Laplace prior and L_{1}penalty
While L_{2} penalized regression shrinks coefficients towards zero, it does not favor them to be exactly zero. In contrast, L_{1} penalized regression provides sparse solutions when a large number of coefficients will be zero. Here we assume the prior parameter τ_{j} to take the common value τ. This is also the LASSO regression.
Selecting prior parameters
Choosing prior variance of the parameters in a Bayesian regression, or equivalently, the regularization parameter in a penalized regression, is important for variable selection. A small prior variance provides more shrinkage towards zero or favors more coefficients to be zero. A large prior variance reflects more uncertainty of the prior information. The prior variance was chosen by 10fold cross validation. The sample was split randomly into 10 parts. The model was fit on 9 out of the 10 parts and the log likelihood function was computed using the remaining one part of the data. This procedure was done for each of the 10 parts and the average log likelihood was calculated. The prior variance was chosen as the one that maximizes the "crossvalidated" average log likelihood.
Simulations
We performed simulation studies to examine the effectiveness of Bayesian logistic regression as a variable selection procedure. We simulated 100 dichotomous predictors from a Bernoulli distribution. The probability of the predictor being one is generated from a uniform distribution, U(0.25, 0.45). Ten of the hundred predictors jointly determine a subject's disease status. The remaining 90 predictors are not used in simulating subjects' disease status. We simulated two settings of sample sizes (n = 150 and n = 250) and two settings of odds ratios. The odds ratios are simulated from a uniform distribution, U(1.5, 2), or U(2, 2.5).
We fit Bayesian logistic regression with Gaussian and Laplace priors using software BBRBMR [4]. BBRBMR can fit largescale regressions with tens of thousands of predictors in a timely fashion. The algorithms used find posterior mode of a logistic likelihood efficiently [4]. We chose the prior variances by 10fold cross validation. The logistic regression with Gaussian prior does not do variable selection directly. After performing the Bayesian analysis of all SNPs together, we selected SNPs for the second stage analysis by ranking their estimated regression coefficients from the first stage simultaneous SNP analysis. We simulated 30 sets of data under each of the four combinations of sample size and odds ratio. The effectiveness of proposed methods is evaluated by 1) the average number of diseasecontributing predictors selected (out of the ten); and 2) how consistent each of the ten predictors is selected. The consistency is defined as the average percent times of each diseasecontributing variable being selected across simulation data sets.
NARAC data analysis
All analyses were performed on the GAW16 Problem 1 North American Rheumatoid Arthritis Consortium (NARAC) data. We analyzed 2705 SNPs on chromosome 9, ranging from 91,730,970 kb to 138,303,776 kb with minor allele frequency greater than 0.01 and no missing genotypes. This area covers the location where the most significant SNP (rs3761847) was reported by Plenge et al. [5]. We checked all 2705 SNPs for HardyWeinberg equilibrium (HWE) in the controls using PLINK [6] and did not find any SNP significantly violate HWE assumption after using the Bonferroni adjustment for multiple comparisons. The SNPs were coded in two ways: dominant and additive.
We divided the sample into a discovering sample (N = 1031) and a replication sample (N = 1031). First, we fit Bayesian logistic regression with a Gaussian prior using BBRBMR software on the training sample. We bootstrapped 100 times to provide standard error of the estimated coefficients. Second, we selected the top 300 SNPs according to two criteria: 1) the absolute value of the coefficients, and 2) the ratio of the coefficients to their bootstrapped standard errors (z scores). Selecting variables based on the absolute value of the coefficients instead of z scores may provide more reproducible results [7]. Especially for the SNPs with large signals and large variability, the z score may be low, but the coefficient may be large. We compare results using these two selection criteria. Third, we conducted chisquare tests on the 300 selected topranking SNPs using the independent testing sample. We analyzed data under both a dominant and additive model.
Results
Simulations
For the Gaussian prior with sample size 250 and high odds ratio (odds ratio ranging from 2 to 2.5), the average number of correctly identified SNPs in the top 20 SNPs selected by the magnitude of the regression coefficients is 8.3 (out of the 10 diseaseassociated SNPs). For the same prior and the sample size but with moderate odds ratio (odds ratio ranging from 1.5 to 2), the average number of correctly identified SNPs is 6.7. When decreasing the sample size to 150, in the high and moderate odds ratio model, the average number of correctly identified SNPs is 7.4 and 6.4, respectively. The consistencies (the average percent times of each diseasecontributing variable being selected across simulation data sets) in the above four settings ranges from 0.73 to 0.97, 0.53 to 0.77, 0.6 to 0.87, and 0.57 to 0.73. For the Laplace prior, the average numbers of SNPs correctly identified in each of the four settings were: 6.7, 4.5, 4.2, and 4.0, respectively. The consistencies were lower than the Gaussian prior.
NARAC data analysis
Bayesian logistic regression of 2705 SNPs on chromosome 9
Additive model  Dominant model  

Rank  SNP  Position  abs (zscore)  SNP  Position  abs (zscore) 
1  rs1407869  101353456  8.02  rs7864653  100860678  7.16 
2  rs4437724  113188649  7.76  rs10989329  100794635  7.16 
3  rs10120479  111426956  6.97  rs4237190  97922972  6.58 
4  rs9697192  116879138  6.97  rs6478644  123942505  6.42 
5  rs3824535  122763410  6.90  rs1407869  101353456  6.03 
6  rs10491578  116463442  6.39  rs2229594  101204219  5.97 
7  rs10121681  111718477  6.37  rs10820559  103716588  5.87 
8  rs694428  117692812  6.13  rs1536705  126851425  5.86 
9  rs2900180  120785936  5.96  rs2564362  123365200  5.74 
10  rs11243755  132287257  5.96  rs10978456  106155366  5.73 
SingleSNP analysis of the top 300 selected SNPs
Additive model  Dominant model  

Rank  SNP  Position  pValue  SNP  Position  pValue 
1  rs2900180  120785936  6.24 × 10^{9}  rs2900180  120785936  6.24 × 10^{9} 
2  rs1953126  120720054  2.76 × 10^{8}  rs11787779  114820894  6.89 × 10^{5} 
3  rs942152  121031239  3.94 × 10^{6}  rs17148869  132180015  1.00 × 10^{4} 
4  rs7858974  91959665  1.26 × 10^{5}  rs7862566  117133575  2.00 × 10^{4} 
5  rs11787779  114820894  6.89 × 10^{5}  rs4978629  107708375  3.00 × 10^{4} 
6  rs6478300  117115323  7.12 × 10^{5}  rs4978890  110046695  3.00 × 10^{4} 
7  rs989980  106309592  1.00 × 10^{4}  rs1333914  119662788  4.00 × 10^{4} 
8  rs17148869  132180015  1.00 × 10^{4}  rs1332408  122271713  4.00 × 10^{4} 
9  rs7862566  117133575  2.00 × 10^{4}  rs2095069  94782055  0.001 
10  rs945246  119953710  2.00 × 10^{4}  rs4743420  100567644  0.0011 
Discussion
We propose a Bayesian logistic regression procedure to select important SNPs based on the z scores or the regression coefficient estimates for further analysis. From the simulation studies, when using a Gaussian prior, the percentage of causal SNPs correctly selected ranges from 64% to 83% among the top 20% SNPs. For the Laplace prior, the percentage of correctly identified causal SNPs ranges from 40% to 67%. The Gaussian prior outperforms Laplace prior, which could be attributable to a less stringent feature selection criterion employed for the Gaussian prior.
Among the top 300 SNPs selected by the z scores for the dominance model, three are significant after adjusting for multiple comparisons (see Table 2). For the additive model, five additional SNPs are significant after multiple comparisons adjustment. These SNPs lie in a region from 91,959,665 kb to 132,180,015 kb on chromosome 9 (LD plots not included due to space limitations). Three of the eight SNPs are in the region reported in Plenge et al. [5] (rs1953126, rs2900180, and rs942152), and two of them are in LD (rs1953126 and rs2900180). One of these SNPs, rs1953126, was reported in a study of 475 Caucasian patients [8] to be significantly associated with rheumatoid arthritis (odds ratio 1.28, CI 1.161.40, trend pvalue = 1.45 × 10^{6}). The other five SNPs are not in the candidate region and are not in LD with SNPs in the region. The significance of other SNPs deserves further investigation in an independent sample.
An alternative onestep approach would be reporting permutation pvalues of Bayesian logistic regression with all SNPs on the whole sample. However, it is well known that increasing number of predictors, and therefore the number of parameters, in a multivariate analysis may reduce power. The twostep approach provides a balance between the need to reduce multiple comparisons and the loss of power due to increasing number of parameters.
We only analyzed SNPs with no missing data due to the incapability of handling missing covariates data of the BBRBMR software. One solution is to first impute the missing genotypes and then run the Bayesian regression on the imputed data. An alternative is to handle missing data directly in a Bayesian analysis by data augmentation.
Here the priors are assumed to be independent and their variances are assumed to be the same. We choose prior variance by crossvalidation. An alternative strategy would be specifying a hyperprior distribution (such as noninformative prior). To incorporate prior knowledge such as physical distance between the SNPs, one can specify prior distribution to have distancebased correlation. How to specify such a correlation for a large scale regression is worth further attention.
Conclusion
Large scale Bayesian logistic regression is useful to analyze genome wide casecontrol data with large number of SNPs. Coefficient estimates or z scores from such regression can be used to select important SNPs for further genetic analysis. Such procedure reduces number of tests performed and alleviates problem of multiple comparisons.
List of abbreviations used
 GAW:

Genetic Analysis Workshop
 GWA:

Genomewide association
 HWE:

HardyWeinberg equilibrium
 LD:

Linkage disequilibrium
 NARAC:

North American Rheumatoid Arthritis Consortium
 SNP:

Singlenucleotide polymorphism
Declarations
Acknowledgements
The Genetic Analysis Workshops are supported by NIH grant R01 GM031575 from the National Institute of General Medical Sciences. YW (PI) and YF (subcontract PI) are supported by NIH grant AG03111301A2.
This article has been published as part of BMC Proceedings Volume 3 Supplement 7, 2009: Genetic Analysis Workshop 16. The full contents of the supplement are available online at http://www.biomedcentral.com/17536561/3?issue=S7.
Authors’ Affiliations
References
 Hoh J, Ott J: Mathematical multilocus approaches to localizing complex human trait genes. Nat Rev Genet. 2003, 4: 701709. 10.1038/nrg1155.View ArticlePubMedGoogle Scholar
 Santner T, Duffy D: The Statistical Analysis of Discrete Data. 1989, New York, SpringerView ArticleGoogle Scholar
 Park MY, Hastie T: Penalized logistic regression for detecting gene interactions. Biostat. 2008, 9: 3050. 10.1093/biostatistics/kxm010.View ArticleGoogle Scholar
 Genkin A, Lewis D, Madigan D: Largescale Bayesian logistic regression for text categorization. Technometrics. 2007, 49: 291304. 10.1198/004017007000000245.View ArticleGoogle Scholar
 Plenge RM, Seielstad M, Padyukov L, Lee AT, Remmers EF, Ding B, Liew A, Khalili H, Chandrasekaran A, Davies LR, Li W, Tan AK, Bonnard C, Ong RT, Thalamuthu A, Pettersson S, Liu C, Tian C, Chen WV, Carulli JP, Beckman EM, Altshuler D, Alfredsson L, Criswell LA, Amos CI, Seldin MF, Kastner DL, Klareskog L, Gregersen PK: TRAF1C5 as a risk locus for rheumatoid arthritisa genomewide study. N Engl J Med. 2007, 357: 11991209. 10.1056/NEJMoa073491.PubMed CentralView ArticlePubMedGoogle Scholar
 Purcell S, Neale B, ToddBrown K, Thomas L, Ferreira MAR, Bender D, Maller J, Sklar P, de Bakker PIW, Daly MJ, Sham PC: PLINK: a tool set for wholegenome association and populationbased linkage analyses. Am J Hum Genet. 2007, 81: 559575. 10.1086/519795.PubMed CentralView ArticlePubMedGoogle Scholar
 Guo L, Lobenhofer EK, Wang C, Shippy R, Harris SC, Zhang L, Mei N, Chen T, Herman D, Goodsaid FM, Hurban P, Phillips KL, Xu J, Deng X, Sun YA, Tong W, Dragan YP, Shi L: Rat toxicogenomic study reveals analytical consistency across microarray platforms. Nat Biotechnol. 2006, 24: 11621169. 10.1038/nbt1238.View ArticlePubMedGoogle Scholar
 Chang M, Rowland CM, Garcia VE, Schrodi SJ, Catanese JJ, Helmvan Mil van der AH, Ardlie KG, Amos CI, Criswell LA, Kastner DL, Gregersen PK, Kurreeman FA, Toes RE, Huizinga TW, Seldin MF, Begovich AB: A largescale rheumatoid arthritis genetic study identifies association at chromosome 9q33.2. PLoS Genet. 2008, 4: e100010710.1371/journal.pgen.1000107.PubMed CentralView ArticlePubMedGoogle Scholar
Copyright
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.