Skip to content

jknightlab/mePipe

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

239 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

The mePipe package Build Status

The mePipe R package provides a convenient way to carry out several steps that are often required as part of an eQTL analysis. Matrix-eQTL is used to carry out the association testing. This very fast and supports the use of many covariates without a substantial increase in run time. In addition mePipe provides facilities to compute principle components of the gene expression data and include a subset of these as covariates to account for confounding variation and maximise the number significant associations. Some functionality to account for LD structure is also available. All computations can either be carried out on a single processor or run in parallel (as far as possible) on an SGE compute cluster.

Installation

Dependencies

The mePipe package requires R version 2.15 or newer to run and the following R packages have to be available. If they are not already present they need to be installed before mePipe:

The first four of these are available from CRAN and Bioconductor and can be installed from within R with

source("http://bioconductor.org/biocLite.R")
biocLite(c("MatrixEQTL", "trio", "optparse", "XML", "snow"))

Unfortunately Rsge has recently been removed from CRAN. The latest version is available from here. To install the latest version directly from GitHub, use the following command from within R (requires the devtools package)

devtools::install_github("Rsge", "humburg")

Installing mePipe from source

The latest version of the R source package is available here. Download this into your home directory and (on the machine on which you would like to add the R package) run

R CMD INSTALL mePipe_latest.tar.gz

Installing the command-line script

The R package includes the runMatrixEQTL.R script that allows use of the MatrixEQTL package from the command line. By default this will be installed into <R_library>/mePipe/exec/. When installing mePipe for the first time it is recommended to create a link to this script in a more convenient place, e.g.

mkdir ~/bin
cp -s <R_library>/mePipe/exec/runMatrixEQTL.R ~/bin/

About

Functions to support running of Matrix-eQTL analysis pipeline. This includes support for adding principle components as covariates and determining the optimal number of principal components to include.

Resources

License

Stars

2 stars

Watchers

3 watching

Forks

Packages

 
 
 

Contributors