Powerful and robust cross‐phenotype association test for case‐parent trios

Abstract

There has been increasing interest in identifying genes within the human genome that influence multiple diverse phenotypes. In the presence of pleiotropy, joint testing of these phenotypes is not only biologically meaningful but also statistically more powerful than univariate analysis of each separate phenotype accounting for multiple testing. Although many cross-phenotype association tests exist, the majority of such methods assume samples composed of unrelated subjects and therefore are not applicable to family-based designs, including the valuable case-parent trio design. In this paper, we describe a robust gene-based association test of multiple phenotypes collected in a case-parent trio study. Our method is based on the kernel distance covariance (KDC) method, where we first construct a similarity matrix for multiple phenotypes and a similarity matrix for genetic variants in a gene; we then test the dependency between the two similarity matrices. The method is applicable to either common variants or rare variants in a gene, and resulting tests from the method are by design robust to confounding due to population stratification. We evaluated our method through simulation studies and observed that the method is substantially more powerful than standard univariate testing of each separate phenotype. We also applied our method to phenotypic and genotypic data collected in case-parent trios as part of the Genetics of Kidneys in Diabetes (GoKinD) study and identified a genome-wide significant gene demonstrating cross-phenotype effects that was not identified using standard univariate approaches.

MF‐TOWmuT: Testing an optimally weighted combination of common and rare variants with multiple traits using family data

“…MF‐KM is computationally intensive and their software cannot be easily adapted to an arbitrary number of traits, so we excluded it from our comparison. Fischer et al (2018) proposed a two‐stage method for gene association with multiple traits from case‐parent trios. Specifically, in the first stage, GAMuT is performed for each gene using the phenotypes and genotypes of the parents.…”

“…Family samples have greater power than unrelated samples to detect the association between rare variants and traits due to the enriched rare variants in family samples. A number of methods have been developed to detect association between multiple genetic variants and multiple traits using family as well as unrelated samples (Chen, Manichaikul, & Rich, 2009; Chen, Meigs, & Dupuis, 2013; Feng, Elston, & Zhu, 2011; Fischer, Jiang, Broadaway, Conneely, & Epstein, 2018; Jiang, Conneely, & Epstein, 2014; Jiang & McPeek, 2014; Lasky‐Su, Murphy, McQueen, Weiss, & Lange, 2010; Schifano et al, 2012; Wang et al, 2016; Won et al, 2015; Yan et al, 2015; Zhu & Xiong, 2012). These include methods based on linear and generalized linear mixed models that can incorporate the relatedness of family samples (Jiang et al, 2014; Lee et al, 2012; Lee et al, 2017b; Schifano et al, 2012; Wu & Pankow, 2016; Wu et al, 2010, 2011; Yan et al, 2015) and methods based on quasi‐likelihood (Wang et al, 2016; Won et al, 2015).…”

“…As of now, a large number of methods have been developed only for rare variants association studies (Broadaway et al, 2016; Jiang & McPeek, 2014; Lee et al, 2012, 2014; Li & Leal, 2008; Madsen & Browning, 2009; Wang et al, 2016; Wu & Pankow, 2016; Wu et al, 2011; Yan et al, 2015). Methods that can combine common variants and rare variants in association studies have also been developed (Feng et al, 2011; Fischer et al, 2018; He, Avery, & Lin, 2013; Ionita‐Laza, Lee, Makarov, Buxbaum, & Lin, 2013; Kim, Zhang, & Pan, 2016; Lasky‐Su et al, 2010; Maity, Sullivan, & Tzeng, 2012; Zhu & Xiong, 2012). However, due to the proportion and composition of causal variants, those methods are not uniformly the most powerful.…”

“…Additionally, as mentioned before, these methods cannot incorporate covariates in their analysis, which will be undoubtedly less powerful. The method proposed in Fischer et al (2018) is actually a two‐step scheme and only applicable to trios, and their method cannot allow for arbitrary pedigree structures. Similarly, the method proposed in Feng et al (2011) is only applicable to sib‐pairs and cannot be applied to family data with arbitrary pedigree structures.…”

A review of kernel methods for genetic association studies

“…Although these types of transmission disequilibrium tests were developed years ago for single genetic markers, only recently were they extended to kernel methods that test the association of a set of genetic markers with a single trait using trios of child-parents or nuclear families (Jiang, Conneely, & Epstein, 2014). Further extensions to test the association of multiple traits with multiple genetic variants for childparent trios have recently been developed (Fischer, Jiang, Broadaway, Conneely, & Epstein, 2018). This approach offers promise for testing rare variants in this type of data.…”