Skip to content
GridLMM package
Branch: master
Clone or download
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
R fix to GridLMMnet and adding Running_GridLMMnet Jun 14, 2019
man
misc fix to GridLMMnet and adding Running_GridLMMnet Jun 14, 2019
src cleanup Jan 15, 2019
vignettes Delete Running_GridLMM_GWAS.R Feb 19, 2019
.DS_Store update docs May 9, 2019
.Rbuildignore fix prior on GridLMM_GWAS_fast Aug 15, 2018
.gitattributes Initial commit Jun 30, 2018
.gitignore update docs May 9, 2019
DESCRIPTION update docs May 9, 2019
GridLMM.Rproj adding cv.glmnet functionality Jul 31, 2018
LICENSE Initial commit Jun 30, 2018
NAMESPACE import as from methods package Feb 20, 2019
README.md Update README.md Feb 12, 2019
README.nb.html adding X_map info, re-running vignette Aug 22, 2018

README.md

GridLMM is a package for fitting linear mixed models (LMMs) with multiple random effects.

The fitting process is optimized for repeated evaluation of the random effect model with different sets of fixed effects, ex. for GWAS analyses. The approximation is due to the use of a discrete grid of possible values for the random effect variance component proportions. We include functions for both frequentist and Bayesian GWAS, (Restricted) Maximum Likelihood evaluation, Bayesian Posterior inference of variance components, and Lasso/Elastic Net fitting of high-dimensional models with random effects.

Please treat this as a Beta version and let me know of issues running the functions.

Installation:

devtools::install_github('deruncie/GridLMM')

Main functions:

  • GridLMM_ML: estimates parameters of a LMM by (restricted) Maximum Likelihood
  • GridLMM_posterior: Approximates the posterior distribution of the variance component proportions of a LMM
  • GridLMM_GWAS: Runs a GWAS with the error structure a LMM. By default, uses heuristics to efficiently sample the grid. Can run Wald tests (method = 'REML'), Likelihood ratio tests (method = 'ML'), or calculate Bayes Factors (method = 'BF')
  • GridLMMnet: Fit a multiple regression with LASSO / elastic net penalty and mixed model error term.

Examples:

There is a vignette walking through the data format necessary for GridLMM and a few analyses using GridLMM_GWAS.

If you would like to build the vignette (see below), do:

devtools::install_github('deruncie/GridLMM', build_opts = c("--no-resave-data", "--no-manual"),force = TRUE)
vignette(topic = 'Running_GridLMM_GWAS',package='GridLMM')
You can’t perform that action at this time.