Meta-analysis of quantitative pleiotropic traits for next-generation sequencing with multivariate functional linear models

Abstract

To analyze next-generation sequencing data, multivariate functional linear models are developed for a meta-analysis of multiple studies to connect genetic variant data to multiple quantitative traits adjusting for covariates. The goal is to take the advantage of both meta-analysis and pleiotropic analysis in order to improve power and to carry out a unified association analysis of multiple studies and multiple traits of complex disorders. Three types of approximate F -distributions based on Pillai-Bartlett trace, HotellingLawley trace, and Wilks's Lambda are introduced to test for association between multiple quantitative traits and multiple genetic variants. Simulation analysis is performed to evaluate false-positive rates and power of the proposed tests. The proposed methods are applied to analyze lipid traits in eight European cohorts. It is shown that it is more advantageous to perform multivariate analysis than univariate analysis in general, and it is more advantageous to perform meta-analysis of multiple studies instead of analyzing the individual studies separately. The proposed models require individual observations. The value of the current paper can be seen at least for two reasons: (a) the proposed methods can be applied to studies that have individual genotype data; (b) the proposed methods can be used as a criterion for future work that uses summary statistics to build test statistics to meta-analyze the data. INTRODUCTION Meta-analysis of multiple studies and pleiotropy analysis of multiple traits are two areas in association studies that recently have received extensive attention in the literature. 1-10 To our knowledge, metaanalysis and pleiotropy analysis have been performed separately so far, and there are no gene-based meta-analysis methods for combining multiple studies together and for carrying out a unified pleiotropy analysis. Here, multivariate functional linear models (MFLM) are developed to connect genetic variant data to multiple quantitative traits adjusting for covariates in a meta-analysis context. The goal is to take the advantage of both meta-analysis and pleiotropy analysis in order to improve power and to carry out a unified analysis of multiple studies and multiple quantitative traits of complex disorders.A noticeable feature of next-generation sequencing data is that dense panels of genetic variants are available via high-throughput sequencing technology, and so we face high-dimension genetic data. [11][12][13][14] The genetic data can consist of rare variants, or common variants, or a combination of the two, where the rare variants' minor allele frequencies (MAFs) are less than 0.01 ∼ 0.05. The high dimensionality of genetic data and the presence of dense rare variants raise

Efficacy and safety of COVID-19 vaccines for patients with spinal tumors receiving denosumab treatment: An initial real−clinical experience study

“…χ2 test and Fisher Exact Test were used to compare between the groups. Due to the sample size and concerns regarding false positives in the multivariate analysis, the subgroup analyses were not conducted on different types of cancer ( 18 ). All statistical tests were two-tailed.…”

Multivariate variable selection by means of null-beamforming

“…The concept of power, defined as the variance of a signal, is borrowed from the research field of signal processing, where sensor observations y j , 1 ≤ j ≤ J are often assumed weakly stationary [20,23]. In genetics, the above concept describes the so-called pleiotropic genetic effect of a single gene on multiple phenotypic traits, where multivariate linear models have been developed to connect genetic variant data to multiple quantitative traits [5]. In the multivariate randomeffects regression setting, the power is the variance component of a covariate in conditional covariance matrix C = cov(y j |X) given X, where we model regression coefficients of the multiple responses to each covariate as realizations from a random variable with a finite second moment.…”

“…Statistical connectivity patterns in the selected covariates are a hallmark feature for connecting pleiotropic traits such as drug inhibitory concentrations to genetic variants in genetics and for studying functional networks in neuroscience [5,14]. Here, to quantify such patterns, we compute the regression coefficientbased Pearson correlation coefficient for each pair of the selected covariates.…”

OpenMendel: a cooperative programming project for statistical genetics

“…Owing to page and time constraints, we do not offer encyclopedic coverage of recent advances in GWAS or sequence analysis. Many promising methods are left unmentioned, for example meta-analysis based on summary statistics [15,19,26,41,52,83] or estimation of fine-scale population structure [58]. Instead we focus on topics related to projects already underway in OPENMENDEL.…”