multiGSEA

The multiGSEA package was built to facilitate the use of gene sets in the analysis of high throughput genomics data (primarily RNA-seq). Analysts can orchestrate any number of GSEA methods across a specific contrast using the unified interface provided by the multiGSEA function, and a shiny application is provided that facilitates the exploration and interpretation of GSEA results.

• The multiGSEA function is a wrapper that orchestrates the execution of any number of user-specified gene set enrichment analyses (GSEA) over a particular experimental contrast of interest. This will create a MultiGSEAResult object which stores the results of each GSEA method internally, allowing for easy query and retrieval.
• A sister multiGSEA.shiny package provides an explore function, which is invoked on MultiGSEAREsult objects returned from a call to multiGSEA. The shiny application facilitates interactive exploration of these GSEA results. This application can also be deployed to a shiny server and can be initialized by uploading a serialized MultiGSEAResult *.rds file.

Full details that outline the use of this software package is provided in the package’s vignette, however a brief description is outlined below.

Usage

A thorough vignette is provided with this package that illustrates the multitudes of its gene-set-based analysis functionality. For convenience, a small taster of the package’s functionality is included here.

A subset of the RNA-seq data tumor/normal samples in the BRCA indication from the TCGA are provided in this package. We will use that data to perform a “camera” and “fry” gene set enrichment analysis of tumor vs normal samples using the MSigDB hallmark and c2 gene set collections with multiGSEA.

library(multiGSEA)
library(dplyr)
gdb <- getMSigGeneSetDb(c('h', 'c2'), 'human')
vm <- exampleExpressionSet(dataset='tumor-vs-normal', do.voom=TRUE)
mg <- multiGSEA(gdb, vm, vm$design, "tumor", methods=c("camera", "fry"))

We can view the top “camera” results with the smallest pvalues like so:

results(mg, "camera") %>% 
  arrange(pval) %>% 
  select(collection, name, padj) %>% 
  head
#>   collection                                        name         padj
#> 1         c2      SOTIRIOU_BREAST_CANCER_GRADE_1_VS_3_UP 2.604378e-38
#> 2         c2 ROSTY_CERVICAL_CANCER_PROLIFERATION_CLUSTER 3.601729e-37
#> 3         c2         NAKAYAMA_SOFT_TISSUE_TUMORS_PCA2_DN 1.004916e-28
#> 4         c2              KANG_DOXORUBICIN_RESISTANCE_UP 1.478337e-22
#> 5         c2                     BENPORATH_PROLIFERATION 2.437669e-22
#> 6         c2               CROONQUIST_IL6_DEPRIVATION_DN 1.934138e-21

The shift in expression of the genes within the top gene set can be visualized with the iplot function below. This plot produces interactive graphics, but rasterized versions are saved for use with this README file:

iplot(mg, 'c2', 'SOTIRIOU_BREAST_CANCER_GRADE_1_VS_3_UP', type="density")

iplot(mg, 'c2', 'SOTIRIOU_BREAST_CANCER_GRADE_1_VS_3_UP', type="boxplot")

When these plots are rendered in your workspace or an Rmarkdown document, the user can hover of the genes (dots) to see their name and differential expression statistics.

Installation

The multiGSEA suite of packages will soon be submitted to bioconductor and installable via the recommended biocLite mechanism. In the meantime, these packages can be installed like so:

# install.packages("devtools")
devtools::install_github("lianos/multiGSEA")
devtools::install_github("lianos/multiGSEA.shiny")
devtools::install_github("lianos/GeneSetDb.MSigDB.Hsapiens.v61")
devtools::install_github("lianos/GeneSetDb.MSigDB.Mmusculus.v61")