Bayesian variable selection regression for genome-wide association studies and other large-scale problems
Open accessThe DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Abstract
We consider applying Bayesian Variable Selection Regression, or BVSR, to genome-wide association studies and similar large-scale regression problems. Currently, typical genome-wide association studies measure hundreds of thousands, or millions, of genetic variants (SNPs), in thousands or tens of thousands of individuals, and attempt to identify regions harboring SNPs that affect some phenotype or outcome of interest. This goal can naturally be cast as a variable selection regression problem, with the SNPs as the covariates in the regression. Characteristic features of genome-wide association studies include the following: (i) a focus primarily on identifying relevant variables, rather than on prediction; and (ii) many relevant covariates may have tiny effects, making it effectively impossible to confidently identify the complete “correct” subset of variables. Taken together, these factors put a premium on having interpretable measures of confidence for individual covariates being included in the model, which we argue is a strength of BVSR compared with alternatives such as penalized regression methods. Here we focus primarily on analysis of quantitative phenotypes, and on appropriate prior specification for BVSR in this setting, emphasizing the idea of considering what the priors imply about the total proportion of variance in outcome explained by relevant covariates. We also emphasize the potential for BVSR to estimate this proportion of variance explained, and hence shed light on the issue of “missing heritability” in genome-wide association studies. More generally, we demonstrate that, despite the apparent computational challenges, BVSR can provide useful inferences in these large-scale problems, and in our simulations produces better power and predictive performance compared with standard single-SNP analyses and the penalized regression method LASSO. Methods described here are implemented in a software package, pi-MASS, available from the Guan Lab website http://bcm.edu/cnrc/mcmcmc/pimass.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Regression shrinkage and selection via the lasso
Tibshirani, R. (1996) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1996.tb02080.x · Cited by 51,482
Equation of State Calculations by Fast Computing Machines
Metropolis, N., Rosenbluth, A. W., Rosenbluth, M. N., Teller, A. H., & Teller, E. (1953) · The Journal of Chemical Physics · doi:10.1063/1.1699114 · Cited by 37,192
Regularization and Variable Selection Via the Elastic Net
Zou, H., & Hastie, T. (2005) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00503.x · Cited by 20,957
Monte Carlo sampling methods using Markov chains and their applications.
Hastings, W. K. (1970) · Biometrika · doi:10.1093/biomet/57.1.97 · Cited by 15,195 · cited 3×
“In brief, it is a Metropolis–Hastings algorithm [Metropolis et al. (1953); Hastings (1970)], using a simple local proposal to jointly update h,π,γ.”
BackgroundMethodPrincipal components analysis corrects for stratification in genome-wide association studies
Price, A. L., Patterson, N. J., Plenge, R. M., Weinblatt, M. E., Shadick, N. A., & Reich, D. (2006) · Nature Genetics · doi:10.1038/ng1847 · Cited by 10,248 · cited 2×
“…been corrected for prior to analysis, for example, by letting y be the residuals from regressing the observed phenotype values against measures of population structure, obtained, for example, by model-based clustering [Pritchard et al. (2000)] or principal components analysis [Price et al. (2006)].”
BackgroundMethodInfluentialGenome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls
The Wellcome Trust Case Control Consortium, Management Committee, Burton, P. R., Clayton, D. G., Cardon, L. R., Craddock, N., Deloukas, P., Duncanson, A., Kwiatkowski, D. P., McCarthy, M. I., et al. (2007) · Nature · doi:10.1038/nature05911 · Cited by 9,442 · cited 2×
“Current typical GWAS [e.g., Wellcome Trust Case Control Consortium (2007)] measure hundreds of thousands, or millions, of genetic variants (typically Single Nucleotide Polymorphisms, or SNPs), in hundreds, thousands, or tens of thousands of individuals, with the primary goal being to identify which…”
BackgroundEfron, B., Hastie, T., Johnstone, I., & Tibshirani, R. (2004) · The Annals of Statistics · doi:10.1214/009053604000000067 · Cited by 9,357
Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
Fan, J., & Li, R. (2001) · Journal of the American Statistical Association · doi:10.1198/016214501753382273 · Cited by 8,900
Davey Smith, G., & Ebrahim, S. (2003) · International Journal of Epidemiology · doi:10.1093/ije/dyg070 · Cited by 5,882 · cited 2×
“This is because in GWAS reverse causation is generally implausible (phenotypes cannot causally affect genotype, since genotype comes first temporally), and there are few potential unmeasured confounders that could affect both genotype and y [Smith and Ebrahim (2003)].”
BackgroundInfluentialCommon SNPs explain a large proportion of the heritability for human height
Yang, J., Benyamin, B., McEvoy, B. P., Gordon, S., Henders, A. K., Nyholt, D. R., Madden, P. A., Heath, A. C., Martin, N. G., Montgomery, G. W., et al. (2010) · Nature Genetics · doi:10.1038/ng.608 · Cited by 4,446
Ridker, P. M., Rifai, N., Rose, L., Buring, J. E., & Cook, N. R. (2002) · New England Journal of Medicine · doi:10.1056/nejmoa021993 · Cited by 3,604
Bayesian Analysis of Binary and Polychotomous Response Data
Albert, J. H., & Chib, S. (1993) · Journal of the American Statistical Association · doi:10.1080/01621459.1993.10476321 · Cited by 3,167
Variable Selection via Gibbs Sampling
George, E. I., & McCulloch, R. E. (1993) · Journal of the American Statistical Association · doi:10.1080/01621459.1993.10476353 · Cited by 2,828
Sure Independence Screening for Ultrahigh Dimensional Feature Space
Fan, J., & Lv, J. (2008) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2008.00674.x · Cited by 2,711
A new multipoint method for genome-wide association studies by imputation of genotypes
Marchini, J., Howie, B., Myers, S., McVean, G., & Donnelly, P. (2007) · Nature Genetics · doi:10.1038/ng2088 · Cited by 2,609 · cited 2×
“To merge the two data sets, we used genotype imputation [Servin and Stephens (2007); Marchini et al. (2007)], using the software package BIMBAM [Guan and Stephens (2008)] to replace missing or unmeasured genotypes with their posterior mean given the observed genotype data [see Guan and Stephens…”
MethodAssociation Mapping in Structured Populations
Pritchard, J. K., Stephens, M., Rosenberg, N. A., & Donnelly, P. (2000) · The American Journal of Human Genetics · doi:10.1086/302959 · Cited by 2,051
Self-citationScheet, P., & Stephens, M. (2006) · The American Journal of Human Genetics · doi:10.1086/502802 · Cited by 1,941
Self-citationPersonal genomes: The case of the missing heritability
Maher, B. (2008) · Nature · doi:10.1038/456018a · Cited by 1,737 · cited 4×
“As a result, BVSR has the potential to shed light on the so-called “missing heritability” observed in many GWAS [Maher (2008); Yang et al. (2010)].”
BackgroundResultInfluentialBayesian Variable Selection in Linear Regression
Mitchell, T. J., & Beauchamp, J. J. (1988) · Journal of the American Statistical Association · doi:10.1080/01621459.1988.10478694 · Cited by 1,424
Are Rare Variants Responsible for Susceptibility to Complex Diseases?
Pritchard, J. K. (2001) · The American Journal of Human Genetics · doi:10.1086/321272 · Cited by 1,197 · cited 2×
““common” genetic variants (those with a population frequency exceeding a few percent), the relative contribution of these two factors is connected to the contentious topic of the relative contributions of common vs rare variants to phenotypic variation and disease risk [Pritchard (2001)].”
BackgroundMixtures of g Priors for Bayesian Variable Selection
Liang, F., Paulo, R., Molina, G., Clyde, M. A., & Berger, J. O. (2008) · Journal of the American Statistical Association · doi:10.1198/016214507000001337 · Cited by 1,089
Optimal predictive model selection
Barbieri, M. M., & Berger, J. O. (2004) · The Annals of Statistics · doi:10.1214/009053604000000238 · Cited by 812
Genome-wide association analysis by lasso penalized logistic regression
Wu, T. T., Chen, Y. F., Hastie, T., Sobel, E., & Lange, K. (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp041 · Cited by 785
A review of Bayesian variable selection methods: what, how and which
O'Hara, R. B., & Sillanpää, M. J. (2009) · Bayesian Analysis · doi:10.1214/09-ba403 · Cited by 771
Nonparametric regression using Bayesian variable selection
Smith, M., & Kohn, R. (1996) · Journal of Econometrics · doi:10.1016/0304-4076(95)01763-1 · Cited by 580
Rao-Blackwellisation of sampling schemes
Casella, G. (1996) · Biometrika · doi:10.1093/biomet/83.1.81 · Cited by 546
High-Resolution Mapping of Expression-QTLs Yields Insight into Human Gene Regulation
Veyrieras, J.-B., Kudaravalli, S., Kim, S. Y., Dermitzakis, E. T., Gilad, Y., Stephens, M., & Pritchard, J. K. (2008) · PLoS Genetics · doi:10.1371/journal.pgen.1000214 · Cited by 536 · cited 3×
“, those with a lower minor allele frequency) have larger effect sizes; see Wakefield (2009) for relevant discussion.”
BackgroundInfluentialSelf-citationClayton, D. G., Walker, N. M., Smyth, D. J., Pask, R., Cooper, J. D., Maier, L. M., Smink, L. J., Lam, A. C., Ovington, N. R., Stevens, H. E., et al. (2005) · Nature Genetics · doi:10.1038/ng1653 · Cited by 526 · cited 1×
“This can cause false positive associations if genotyping error is correlated with phenotype (which it can be, particularly in case-control studies if the DNA quality differs appreciably between cases and controls [Clayton et al. (2005)]).”
BackgroundImputation-Based Analysis of Association Studies: Candidate Regions and Quantitative Traits
Servin, B., & Stephens, M. (2007) · PLoS Genetics · doi:10.1371/journal.pgen.0030114 · Cited by 515
Self-citationBayesian statistical methods for genetic association studies
Stephens, M., & Balding, D. J. (2009) · Nature Reviews Genetics · doi:10.1038/nrg2615 · Cited by 468
Self-citationRaychaudhuri, S., Plenge, R. M., Rossin, E. J., Ng, A. C. Y., International Schizophrenia Consortium, Purcell, S. M., Sklar, P., Scolnick, E. M., Xavier, R. J., Altshuler, D., et al. (2009) · PLoS Genetics · doi:10.1371/journal.pgen.1000534 · Cited by 431
Bayes factors for genome‐wide association studies: comparison with P‐values
Wakefield, J. (2009) · Genetic Epidemiology · doi:10.1002/gepi.20359 · Cited by 404
Familial and genetic determinants of systemic markers of inflammation: the NHLBI family heart study
Pankow, J. S., Folsom, A. R., Cushman, M., Borecki, I. B., Hopkins, P. N., Eckfeldt, J. H., & Tracy, R. P. (2001) · Atherosclerosis · doi:10.1016/s0021-9150(00)00586-4 · Cited by 375
Ridker, P. M., Pare, G., Parker, A., Zee, R. Y. L., Danik, J. S., Buring, J. E., Kwiatkowski, D., Cook, N. R., Miletich, J. P., & Chasman, D. I. (2008) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2008.03.015 · Cited by 343
Simultaneous Analysis of All SNPs in Genome-Wide and Re-Sequencing Association Studies
Hoggart, C. J., Whittaker, J. C., De Iorio, M., & Balding, D. J. (2008) · PLoS Genetics · doi:10.1371/journal.pgen.1000130 · Cited by 321 · cited 1×
“We chose the LASSO as a representative of penalized regression approaches both because of its popularity and because previous papers have applied it to the specific context of GWAS [e.g., Hoggart et al. (2008); Wu et al. (2009)].”
MethodBayesian Model Averaging for Linear Regression Models
Raftery, A. E., Madigan, D., & Hoeting, J. A. (1997) · Journal of the American Statistical Association · doi:10.2307/2291462 · Cited by 294
Genome-Wide Association of Lipid-Lowering Response to Statins in Combined Study Populations
Barber, M. J., Mangravite, L. M., Hyde, C. L., Chasman, D. I., Smith, J. D., McCarty, C. A., Li, X., Wilke, R. A., Rieder, M. J., Williams, P. T., et al. (2010) · PLoS ONE · doi:10.1371/journal.pone.0009763 · Cited by 231
Self-citationReiner, A. P., Barber, M. J., Guan, Y., Ridker, P. M., Lange, L. A., Chasman, D. I., Walston, J. D., Cooper, G. M., Jenny, N. S., Rieder, M. J., et al. (2008) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2008.03.017 · Cited by 186
Self-citationPractical Issues in Imputation-Based Association Mapping
Guan, Y., & Stephens, M. (2008) · PLoS Genetics · doi:10.1371/journal.pgen.1000279 · Cited by 173
Self-citationBayes Model Averaging with Selection of Regressors
Brown, P. J., Vannucci, M., & Fearn, T. (2002) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/1467-9868.00348 · Cited by 165
Bayesian Meta-Analysis of Genetic Association Studies with Different Sets of Markers
Verzilli, C. J., Shah, T., Casas, J. P., Chapman, J. M., Sandhu, M. S., Debenham, S. L., Boekholdt, S. M., Khaw, K., Wareham, N. J., Judson, R., et al. (2008) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2008.01.016 · Cited by 63
Heritability and Expression of C‐Reactive Protein in Type 2 Diabetes in the Diabetes Heart Study
Lange, L. A., Burdon, K. P., Langefeld, C. D., Liu, Y., Beck, S. R., Rich, S. S., Freedman, B. I., Brosnihan, K. B., Herrington, D. M., Wagenknecht, L. E., et al. (2006) · Annals of Human Genetics · doi:10.1111/j.1469-1809.2006.00280.x · Cited by 39
Cited by
Polygenic prediction via Bayesian regression and continuous shrinkage priors
Ge, T., Chen, C.-Y., Ni, Y., Feng, Y.-C. A., & Smoller, J. W. (2019) · Nature Communications · doi:10.1038/s41467-019-09718-5 · Cited by 1,922
“…and Goddard, 2004; Meuwissen et al., 2001; Verbyla et al., 2009, 2010; Yi et al., 2003], BayesR [Erbe et al., 2012; Moser et al., 2015], BVSR [Guan and Stephens, 2011], BSLMM [Zhou et al., 2013], and DPR [Zeng and Zhou, 2017], have used discrete mixture priors on SNP effect sizes, we expect…”
MethodBayesian statistics and modelling
van de Schoot, R., Depaoli, S., King, R., Kramer, B., Märtens, K., Tadesse, M. G., Vannucci, M., Gelman, A., Veen, D., Willemsen, J., et al. (2021) · Nature Reviews Methods Primers · doi:10.1038/s43586-020-00001-2 · Cited by 1,176
A simple new approach to variable selection in regression, with application to genetic fine mapping
Wang, G., Sarkar, A., Carbonetto, P., & Stephens, M. (2020) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/rssb.12388 · Cited by 1,121
Self-citationFrom genome-wide associations to candidate causal variants by statistical fine-mapping
Schaid, D. J., Chen, W., & Larson, N. B. (2018) · Nature Reviews Genetics · doi:10.1038/s41576-018-0016-z · Cited by 885
FINEMAP: efficient variable selection using summary data from genome-wide association studies
Benner, C., Spencer, C. C. A., Havulinna, A. S., Salomaa, V., Ripatti, S., & Pirinen, M. (2016) · Bioinformatics · doi:10.1093/bioinformatics/btw018 · Cited by 815
Metadata
- Authors
- Guan, Y., & Stephens, M.
- Venue
- The Annals of Applied Statistics
- Publisher
- Institute of Mathematical Statistics
- Published
- 2011-09-01
- Volume
- 5
- Issue
- 3
- Pages
- 1780-1815
- ISSN
- 1932-6157 (ISSN-L)
- OpenAlex
- W2032230252
- Cited by
- 302
- References
- 42
- FWCI
- 14.19
- Type
- journal article
- Language
- en
- Open access
- bronze
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.