The effect of phenotypic outliers and non-normality on rare-variant association testing

Auer, Paul L.; Reiner, Alex P.; Leal, Suzanne M.

doi:10.1038/ejhg.2015.270

Cited by 39 publications

(45 citation statements)

References 30 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently Auer et al (2016) also reported that single SNV-based and SNV-set-based RV tests can be nonrobust to phenotypic outliers and nonnormality, which is in agreement with the main theme of this paper and highlights the importance of the topic studied here. They recommended the INV transformation for nonnormally distributed traits.…”

Section: Discussionsupporting

confidence: 90%

On Robust Association Testing for Quantitative Traits and Rare Variants

Wei

Cao

Zhang

et al. 2016

G3 Genes|Genomes|Genetics

View full text Add to dashboard Cite

With the advance of sequencing technologies, it has become a routine practice to test for association between a quantitative trait and a set of rare variants (RVs). While a number of RV association tests have been proposed, there is a dearth of studies on the robustness of RV association testing for nonnormal distributed traits, e.g., due to skewness, which is ubiquitous in cohort studies. By extensive simulations, we demonstrate that commonly used RV tests, including sequence kernel association test (SKAT) and optimal unified SKAT (SKAT-O), are not robust to heavy-tailed or right-skewed trait distributions with inflated type I error rates; in contrast, the adaptive sum of powered score (aSPU) test is much more robust. Here we further propose a robust version of the aSPU test, called aSPUr. We conduct extensive simulations to evaluate the power of the tests, finding that for a larger number of RVs, aSPU is often more powerful than SKAT and SKAT-O, owing to its high data-adaptivity. We also compare different tests by conducting association analysis of triglyceride levels using the NHLBI ESP whole-exome sequencing data. The QQ plots for SKAT and SKAT-O were severely inflated (λ = 1.89 and 1.78, respectively), while those for aSPU and aSPUr behaved normally. Due to its relatively high robustness to outliers and high power of the aSPU test, we recommend its use complementary to SKAT and SKAT-O. If there is evidence of inflated type I error rate from the aSPU test, we would recommend the use of the more robust, but less powerful, aSPUr test.

show abstract

Section: Discussionsupporting

confidence: 90%

On Robust Association Testing for Quantitative Traits and Rare Variants

Wei

Cao

Zhang

et al. 2016

G3 Genes|Genomes|Genetics

View full text Add to dashboard Cite

show abstract

“…Residuals from this model were then transformed using the rank-based inverse normal transformation to control type I error. 16 Autosomal and X chromosome variants were then tested for association with each WBC trait using either Rvtests or RAREMETALWORKER software packages. Both packages generate single variant association score summary statistics, variance-covariance matrices containing LD relationships between …”

Section: Discussionmentioning

confidence: 99%

Large-Scale Exome-wide Association Analysis Identifies Loci for White Blood Cell Traits and Pleiotropy with Immune-Mediated Diseases

Tajuddin

Schick

Eicher

et al. 2016

The American Journal of Human Genetics

View full text Add to dashboard Cite

White blood cells play diverse roles in innate and adaptive immunity. Genetic association analyses of phenotypic variation in circulating white blood cell (WBC) counts from large samples of otherwise healthy individuals can provide insights into genes and biologic pathways involved in production, differentiation, or clearance of particular WBC lineages (myeloid, lymphoid) and also potentially inform the genetic basis of autoimmune, allergic, and blood diseases. We performed an exome array-based meta-analysis of total WBC and subtype counts (neutrophils, monocytes, lymphocytes, basophils, and eosinophils) in a multi-ancestry discovery and replication sample of ∼157,622 individuals from 25 studies. We identified 16 common variants (8 of which were coding variants) associated with one or more WBC traits, the majority of which are pleiotropically associated with autoimmune diseases. Based on functional annotation, these loci included genes encoding surface markers of myeloid, lymphoid, or hematopoietic stem cell differentiation (CD69, CD33, CD87), transcription factors regulating lineage specification during hematopoiesis (ASXL1, IRF8, IKZF1, JMJD1C, ETS2-PSMG1), and molecules involved in neutrophil clearance/apoptosis (C10orf54, LTA), adhesion (TNXB), or centrosome and microtubule structure/function (KIF9, TUBD1). Together with recent reports of somatic ASXL1 mutations among individuals with idiopathic cytopenias or clonal hematopoiesis of undetermined significance, the identification of a common regulatory 3' UTR variant of ASXL1 suggests that both germline and somatic ASXL1 mutations contribute to lower blood counts in otherwise asymptomatic individuals. These association results shed light on genetic mechanisms that regulate circulating WBC counts and suggest a prominent shared genetic architecture with inflammatory and autoimmune diseases.

show abstract

“…Non-normality can inflate type I error rates (Auer et al 2016), however if only a few datasets analyzed are not normally distributed then no further statistical analyses will be performed as ANOVA is reasonably robust with respect to the normality assumption (Schmider et al 2010). Normality was performed with a confidence level of 0.95…”

Section: Statistical Analysesmentioning

confidence: 99%