A novel algorithm for simultaneous SNP selection in high-dimensional genome-wide association studies

Abstract

BackgroundIdentification of causal SNPs in most genome wide association studies relies on approaches that consider each SNP individually. However, there is a strong correlation structure among SNPs that needs to be taken into account. Hence, increasingly modern computationally expensive regression methods are employed for SNP selection that consider all markers simultaneously and thus incorporate dependencies among SNPs.ResultsWe develop a novel multivariate algorithm for large scale SNP selection using CAR score regression, a promising new approach for prioritizing biomarkers. Specifically, we propose a computationally efficient procedure for shrinkage estimation of CAR scores from high-dimensional data. Subsequently, we conduct a comprehensive comparison study including five advanced regression approaches (boosting, lasso, NEG, MCP, and CAR score) and a univariate approach (marginal correlation) to determine the effectiveness in finding true causal SNPs.ConclusionsSimultaneous SNP selection is a challenging task. We demonstrate that our CAR score-based algorithm consistently outperforms all competing approaches, both uni- and multivariate, in terms of correctly recovered causal SNPs and SNP ranking. An R package implementing the approach as well as R code to reproduce the complete study presented here is available from http://strimmerlab.org/software/care/.

A 2,000-year reconstruction of the rain-fed maize agricultural niche in the US Southwest

“…There are many ways one could select variables once they are ranked, including removing null variables using an adaptive threshold 48 , optimized selection using compute-intensive cross-validation or classic model selection using information criteria such as Akaike's Information Criterion 49 . Here we follow the suggestion of Zuber and Strimmer 13 …”

Optimal Whitening and Decorrelation

“…Cross- and Strimmer, 2010;Zuber and Strimmer, 2011;Zuber et al, 2012). ZCAcor whitening employs W ZCA-cor = P −1/2 V −1/2 (11) as its sphering matrix.…”

An approximate Bayesian significance test for genomic evaluations

“…As an option, a GWAS approach based on a single‐marker regression model that considers the empirical correlation among SNPs (Zuber et al., ) was applied to the simulated data; it is available as R package care version 1.1.9. The top‐ranked SNPs corresponding to a false discovery rate of 5 % were selected as significant.…”

“…Ideally, the number of independent tests accounts for the LD between SNPs as was proposed by Gao, Starmer, and Martin (2008) in their simple M-method. Alternatively, a score-based test statistic accounting for the correlation between SNPs yields a ranked order of SNPs (Zuber, Duarte Silva, & Strimmer, 2012). Then, the top-ranked SNPs can be taken for further genetic investigation.…”