Meta-analysis of Multivariate Genome Wide Association Studies

Allows running a meta-analysis of multivariate Genome Wide Association Studies (GWAS) and easily visualizing results through custom plotting functions. The multivariate setting implies that results for each single nucleotide polymorphism (SNP) include several effect sizes (also known as "beta coefficients", one for each trait), as well as related variance values, but also covariance between the betas. The main goal of the package is to provide combined beta coefficients across different cohorts, together with the combined variance/covariance matrix. The method is inverse-variance based, thus each beta is weighted by the inverse of its variance-covariance matrix, before taking the average across all betas. The default options of the main function \code{multi_meta} will work with files obtained from GEMMA multivariate option for GWAS (Zhou & Stephens, 2014). It will work with any other output, as soon as columns are formatted to have the according names. The package also provides several plotting functions for QQ-plots, Manhattan Plots and custom summary plots.


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.1 by Dragana Vuckovic, 3 years ago

Browse source code at

Authors: Dragana Vuckovic

Documentation:   PDF Manual  

Task views: Meta-Analysis

GPL (>= 2) license

Imports mvtnorm, expm, ggplot2, reshape2

Depends on gtable, grid

See at CRAN