Skip to contents

Run Gene Set Enrichment Analysis (GSEA)

Source: R/run_gsea.R
run_gsea.Rd

A wrapper around clusterProfiler::GSEA() that accepts a pre-ranked gene vector and a gene-set data frame, runs GSEA, and returns the result together with the inputs needed by downstream plotting functions (plot_pathways() and plot_pathway_correlation_network()).

Usage

run_gsea(
  gene_vec,
  term2gene,
  sort = TRUE,
  p_adjust_method = "BH",
  p_cutoff = 1,
  min_gs_size = 10,
  max_gs_size = 500,
  seed = 1,
  verbose = FALSE,
  ...
)

Arguments

gene_vec

named numeric vector Gene-level statistics (e.g. log2 fold change, t-statistic) named by gene symbol. The vector must be sorted in decreasing order before being passed in (or set sort = TRUE).

term2gene

data.frame Two-column data frame mapping gene-set names to gene symbols. Column 1 must be the gene-set / term name; column 2 must be the gene symbol. Column names are coerced to "term" and "gene" internally.

sort

logical(1) If TRUE (default), gene_vec is sorted in decreasing order before being passed to clusterProfiler::GSEA().

p_adjust_method

character(1) Method passed to clusterProfiler::GSEA() for multiple-testing correction (default "BH").

p_cutoff

numeric(1) pvalueCutoff passed to clusterProfiler::GSEA() (default 1, i.e. return all gene sets). Must be a single finite value in [0, 1].

min_gs_size

integer(1) Minimum gene-set size (default 10). Must be a single positive integer and no greater than max_gs_size.

max_gs_size

integer(1) Maximum gene-set size (default 500). Must be a single positive integer and no less than min_gs_size.

seed

integer(1) Random seed set before calling clusterProfiler::GSEA() for reproducibility (default 1). The caller's RNG state is saved before the seed is applied and fully restored on exit, so calling this function does not affect subsequent random operations in the session.

verbose

logical(1) Passed to clusterProfiler::GSEA() (default FALSE).

...

Additional arguments forwarded to clusterProfiler::GSEA().

Value

A list with three elements:

gsea_result

A gseaResult object returned by clusterProfiler::GSEA().

gene_vec

The (possibly sorted) named numeric vector that was passed to GSEA — use this as the foldChange argument of plot_pathways().

term2gene

The gene-set data frame (columns term and gene) that was passed to GSEA — use this as the gene_sets argument of plot_pathway_correlation_network().

Examples

if (FALSE) { # \dontrun{
# hallmark_t2g is bundled with the package (columns: term, gene)
data(hallmark_t2g)
set.seed(1)
all_genes <- unique(hallmark_t2g$gene)
gene_vec  <- setNames(rnorm(length(all_genes)), all_genes)

res <- run_gsea(gene_vec, term2gene = hallmark_t2g)
head(as.data.frame(res$gsea_result))
} # }