Rapid and Robust Resampling-Based Multiple-Testing Correction with Application in a Genome-Wide Expression Quantitative Trait Loci Study

Abstract

Genome-wide expression quantitative trait loci (eQTL) studies have emerged as a powerful tool to understand the genetic basis of gene expression and complex traits. In a typical eQTL study, the huge number of genetic markers and expression traits and their complicated correlations present a challenging multiple-testing correction problem. The resampling-based test using permutation or bootstrap procedures is a standard approach to address the multiple-testing problem in eQTL studies. A brute force application of the resampling-based test to large-scale eQTL data sets is often computationally infeasible. Several computationally efficient methods have been proposed to calculate approximate resampling-based P-values. However, these methods rely on certain assumptions about the correlation structure of the genetic markers, which may not be valid for certain studies. We propose a novel algorithm, rapid and exact multiple testing correction by resampling (REM), to address this challenge. REM calculates the exact resampling-based P-values in a computationally efficient manner. The computational advantage of REM lies in its strategy of pruning the search space by skipping genetic markers whose upper bounds on test statistics are small. REM does not rely on any assumption about the correlation structure of the genetic markers. It can be applied to a variety of resampling-based multiple-testing correction methods including permutation and bootstrap methods. We evaluate REM on three eQTL data sets (yeast, inbred mouse, and human rare variants) and show that it achieves accurate resampling-based P-value estimation with much less computational cost than existing methods. The software is available at http://csbio.unc.edu/eQTL. G ENOME-WIDE studies of expression quantitative trait loci (eQTL) have been widely used to dissect the genetic basis of gene expression and molecular mechanisms underlying complex traits (Bochner 2003;Rockman and Kruglyak 2006;Michaelson et al. 2009). In a typical eQTL study, the association between each expression trait and each genetic marker [e.g., single-nucleotide polymorphism (SNP)] is assessed separately, which leads to a huge number of correlated tests. Appropriate multiple-testing correction is crucial for eQTL studies. The resampling-based test using permutation or bootstrap (Good 2005) has been widely used for multipletesting correction across multiple genetic markers for each phenotype (Barrett et al. 2005;Purcell et al. 2007) by simulating the null distribution using permuted or bootstrapped phenotype values (Westfall and Young 1993;Churchill and Doerge 1994;). More specifically, the phenotype values are randomly shuffled and reassigned to individuals with or without replacement (i.e., bootstrap and permutation, respectively). For each resampled phenotype a whole-genome scan is performed to find the maximum test statistic among all SNPs. The corrected P-value is the proportion of the resampled phenotypes where the maximum test statistics are greater than the maximum test statistic ...

RNA Sequencing and Analysis

“…To mitigate this computational burden, software has been developed such as Matrix eQTL to efficiently test the associations by modeling the effect of genotype as either additive linear (least squares model) or categorical (ANOVA model) (Shabalin 2012). Because of the large number of tests performed, it is important to correct for multiple-testing by calculating the false discovery rate (Benjamini and Hochberg 1995; Yekutieli and Benjamini 1999) or resampling using bootstrap or permutation procedures (Karlsson 2006; Zhang et al 2012). …”

The emerging landscape of health research based on biobanks linked to electronic health records: Existing resources, statistical challenges, and potential opportunities

“…In the context of correlated SNPs, some methods correct for multiple testing via analysis of the underlying linkage disequilibrium structure of the genetic data . Johnson et al, Zhang et al, and Li et al provide some simulations comparing the performance of different methods …”

“…[178][179][180] In the context of correlated SNPs, some methods correct for multiple testing via analysis of the underlying linkage disequilibrium structure of the genetic data. 181 174,182,183 An emerging challenge is the correction of multiple testing across the medical phenome x genome twodimensional landscape. With recent work regarding phenotype risk scores, there is increasing interest in studying phenotype-phenotype associations across the phenome.…”

Functional Genomic Architecture of Predisposition to Voluntary Exercise in Mice: Expression QTL in the Brain

“…We acknowledge that our approach is only one of potentially many and that the threshold estimates may be imperfect in the context of available techniques. As new approaches that are more computationally efficient (e.g., Zhang et al 2012) and applicable to the current data set become available, thresholds computed here may be revisited.…”

“…We acknowledge that our approach is only one of potentially many and that the threshold estimates may be imperfect in the context of available techniques. As new approaches that are more computationally efficient (e.g., Zhang et al 2012) and applicable to the current data set become available, thresholds computed here may be revisited.Due to our more complex breeding history (G 4 as opposed to F 2 ), some adjustment of the LOD threshold is warranted (Valdar et al 2009). The subpopulation utilized here (n = 248) was composed of 54 unique families with an average of four individuals representing each family (median = 4; minimum = 1; maximum = 11).…”