The Use of Permutation Tests for Variance Components in Linear Mixed Models

Permutation Tests for Assessing Potential Non-Linear Associations between Treatment Use and Multivariate Clinical Outcomes

In many psychometric applications, the relationship between the mean of an outcome and a quantitative covariate is too complex to be described by simple parametric functions; instead, flexible nonlinear relationships can be incorporated using penalized splines. Penalized splines can be conveniently represented as a linear mixed effects model (LMM), where the coefficients of the spline basis functions are random effects. The LMM representation of penalized splines makes the extension to multivariate outcomes relatively straightforward. In the LMM, no effect of the quantitative covariate on the outcome corresponds to the null hypothesis that a fixed effect and a variance component are both zero. Under the null, the usual asymptotic chi-square distribution of the likelihood ratio test for the variance component does not hold. Therefore, we propose three permutation tests for the likelihood ratio test statistic: one based on permuting the quantitative covariate, the other two based on permuting residuals. We compare via simulation the Type I error rate and power of the three permutation tests obtained from joint models for multiple outcomes, as well as a commonly used parametric test. The tests are illustrated using data from a stimulant use disorder psychosocial clinical trial.

Early protein energy malnutrition impacts life-long developmental trajectories of the sources of EEG rhythmic activity

Protein Energy Malnutrition (PEM) has lifelong consequences on brain development and cognitive function. We studied the lifelong developmental trajectories of resting-state EEG source activity in 66 individuals with histories of Protein Energy Malnutrition (PEM) limited to the first year of life and in 83 matched classmate controls (CON) who are all participants of the 49 years longitudinal Barbados Nutrition Study (BNS). qEEGt source z-spectra measured deviation from normative values of EEG rhythmic activity sources at 5-11 years of age and 40 years later at 45-51 years of age. The PEM group showed qEEGt abnormalities in childhood, including a developmental delay in alpha rhythm maturation and an insufficient decrease in beta activity. These profiles may be correlated with accelerated cognitive decline.

Detecting heritable phenotypes without a model using fast permutation testing for heritability and set-tests

Testing for association between a set of genetic markers and a phenotype is a fundamental task in genetic studies. Standard approaches for heritability and set testing strongly rely on parametric models that make specific assumptions regarding phenotypic variability. Here, we show that resulting p-values may be inflated by up to 15 orders of magnitude, in a heritability study of methylation measurements, and in a heritability and expression quantitative trait loci analysis of gene expression profiles. We propose FEATHER, a method for fast permutation-based testing of marker sets and of heritability, which properly controls for false-positive results. FEATHER eliminated 47% of methylation sites found to be heritable by the parametric test, suggesting a substantial inflation of false-positive findings by alternative methods. Our approach can rapidly identify heritable phenotypes out of millions of phenotypes acquired via high-throughput technologies, does not suffer from model misspecification and is highly efficient.

Standard approaches for heritability and set testing in statistical genetics rely on parametric models that might not hold in reality and give inflated p-values. Here, the authors develop a fast method for permutation-based testing of marker sets and of heritability that does not suffer from model misspecification.

Fingerprint resampling: A generic method for efficient resampling

In resampling methods, such as bootstrapping or cross validation, a very similar computational problem (usually an optimization procedure) is solved over and over again for a set of very similar data sets. If it is computationally burdensome to solve this computational problem once, the whole resampling method can become unfeasible. However, because the computational problems and data sets are so similar, the speed of the resampling method may be increased by taking advantage of these similarities in method and data. As a generic solution, we propose to learn the relation between the resampled data sets and their corresponding optima. Using this learned knowledge, we are then able to predict the optima associated with new resampled data sets. First, these predicted optima are used as starting values for the optimization process. Once the predictions become accurate enough, the optimization process may even be omitted completely, thereby greatly decreasing the computational burden. The suggested method is validated using two simple problems (where the results can be verified analytically) and two real-life problems (i.e., the bootstrap of a mixed model and a generalized extreme value distribution). The proposed method led on average to a tenfold increase in speed of the resampling method.

Permutation-based variance component test in generalized linear mixed model with application to multilocus genetic association study

Background

In many medical studies the likelihood ratio test (LRT) has been widely applied to examine whether the random effects variance component is zero within the mixed effects models framework; whereas little work about likelihood-ratio based variance component test has been done in the generalized linear mixed models (GLMM), where the response is discrete and the log-likelihood cannot be computed exactly. Before applying the LRT for variance component in GLMM, several difficulties need to be overcome, including the computation of the log-likelihood, the parameter estimation and the derivation of the null distribution for the LRT statistic.

Methods

To overcome these problems, in this paper we make use of the penalized quasi-likelihood algorithm and calculate the LRT statistic based on the resulting working response and the quasi-likelihood. The permutation procedure is used to obtain the null distribution of the LRT statistic. We evaluate the permutation-based LRT via simulations and compare it with the score-based variance component test and the tests based on the mixture of chi-square distributions. Finally we apply the permutation-based LRT to multilocus association analysis in the case–control study, where the problem can be investigated under the framework of logistic mixed effects model.

Results

The simulations show that the permutation-based LRT can effectively control the type I error rate, while the score test is sometimes slightly conservative and the tests based on mixtures cannot maintain the type I error rate. Our studies also show that the permutation-based LRT has higher power than these existing tests and still maintains a reasonably high power even when the random effects do not follow a normal distribution. The application to GAW17 data also demonstrates that the proposed LRT has a higher probability to identify the association signals than the score test and the tests based on mixtures.

Conclusions

In the present paper the permutation-based LRT was developed for variance component in GLMM. The LRT outperforms existing tests and has a reasonably higher power under various scenarios; additionally, it is conceptually simple and easy to implement.

Fast and powerful heritability inference for family-based neuroimaging studies

Heritability estimation has become an important tool for imaging genetics studies. The large number of voxel- and vertex-wise measurements in imaging genetics studies presents a challenge both in terms of computational intensity and the need to account for elevated false positive risk because of the multiple testing problem. There is a gap in existing tools, as standard neuroimaging software cannot estimate heritability, and yet standard quantitative genetics tools cannot provide essential neuroimaging inferences, like family-wise error corrected voxel-wise or cluster-wise P-values. Moreover, available heritability tools rely on P-values that can be inaccurate with usual parametric inference methods. In this work we develop fast estimation and inference procedures for voxel-wise heritability, drawing on recent methodological results that simplify heritability likelihood computations (Blangero et al., 2013). We review the family of score and Wald tests and propose novel inference methods based on explained sum of squares of an auxiliary linear model. To address problems with inaccuracies with the standard results used to find P-values, we propose four different permutation schemes to allow semi-parametric inference (parametric likelihood-based estimation, non-parametric sampling distribution). In total, we evaluate 5 different significance tests for heritability, with either asymptotic parametric or permutation-based P-value computations. We identify a number of tests that are both computationally efficient and powerful, making them ideal candidates for heritability studies in the massive data setting. We illustrate our method on fractional anisotropy measures in 859 subjects from the Genetics of Brain Structure study.

A cautionary note on ignoring polygenic background when mapping quantitative trait loci via recombinant congenic strains

In gene mapping, it is common to test for association between the phenotype and the genotype at a large number of loci, i.e., the same response variable is used repeatedly to test a large number of non-independent and non-nested hypotheses. In many of these genetic problems, the underlying model is a mixed model consistent of one or very few major genes concurrently with a genetic background effect, usually thought as of polygenic nature and, consequently, modeled through a random effects term with a well-defined covariance structure dependent upon the kinship between individuals. Either because the interest lies only on the major genes or to simplify the analysis, it is habitual to drop the random effects term and use a simple linear regression model, sometimes complemented with testing via resampling as an attempt to minimize the consequences of this practice. Here, it is shown that dropping the random effects term has not only extreme negative effects on the control of the type I error rate, but it is also unlikely to be fixed by resampling because, whenever the mixed model is correct, this practice does not allow to meet some basic requirements of resampling in a gene mapping context. Furthermore, simulations show that the type I error rates when the random term is ignored can be unacceptably high. As an alternative, this paper introduces a new bootstrap procedure to handle the specific case of mapping by using recombinant congenic strains under a linear mixed model. A simulation study showed that the type I error rates of the proposed procedure are very close to the nominal ones, although they tend to be slightly inflated for larger values of the random effects variance. Overall, this paper illustrates the extent of the adverse consequences of ignoring random effects term due to polygenic factors while testing for genetic linkage and warns us of potential modeling issues whenever simple linear regression for a major gene yields multiple significant linkage peaks.