GSEA-MSigDB / GSEA_R

Updated implementation of the GSEA-P R application for modern R distributions
Other
94 stars 36 forks source link

GSEA R Implementation v1.2

Revisions to the GSEA-P R Application (Subramanian and Tamayo et al 2005)

Preliminary revisions to the original GSEA implementation in R to enable it to run on modern R distributions. Recommended R minimum version: 3.6

Additionally, a rudimentary Collapse dataset function has been backported from the Java GSEA application, however differences in the implementation result in inconsistencies with the desktop collapse function.

GSEA-R v1.2 contains a preliminary implementation of GSEA-preranked. This implementation has not been thoroughly tested and is reliant on undocumented gene set permutation code.

The collapse dataset feature requires the dplyr package from tidyverse

You may install this requirement from CRAN with install.packages("dplyr")

The original R-GSEA documentation indicates that it supports "phenotype permutation" mode only, however, code was present in the application to perform gene set permutation testing. This code has been enabled, but not tested.

R-GSEA remains unsupported by the GSEA-MSigDB Team.

See the original R-GSEA readme here: https://software.broadinstitute.org/cancer/software/gsea/wiki/index.php/R-GSEA_Readme

Running this application

Install the GSEA package with: library("devtools") then install_github("GSEA-MSigDB/GSEA_R")

Optionally, a helper script has been provided to simplify use of this package. Initialize the helper script with the source() command by calling source(system.file('extdata', 'Run.GSEA.R', package = 'GSEA'))

The helper script will import the GSEA functions from the GSEA package, and begin a series of prompts for loading files required for GSEA and setting experiment parameters.