results: Interrogate the results of a multiGSEA analysis stored in a...

In lianos/multiGSEA: A unified interface to a plethora of gene set enrichment analysis methods

Description

The resultNames, result, and results functions enable you to explore the results of the analysis run with multiGSEA.

The results that are stored within a MultiGSEAResult object have a more or less 1:1 mapping with the values passed as methods, parameter of the multiGSEA() call.

Generates a table to indicate the number of genesets per collection that pass a given FDR. The table provides separate groups of rows for each of the methods run in the multiGSEA() call that generated that generated x.

Usage

resultNames(x)

result(x, ...)

## S3 method for class 'MultiGSEAResult'
result(
  x,
  name = NULL,
  stats.only = FALSE,
  rank.by = c("pval", "t", "logFC"),
  add.suffix = FALSE,
  as.dt = FALSE,
  ...
)

results(
  x,
  names = resultNames(x),
  stats.only = TRUE,
  rank.by = c("pval", "logFC", "t"),
  add.suffix = length(names) > 1L,
  as.dt = FALSE
)

tabulateResults(
  x,
  names = resultNames(x),
  max.p = 0.2,
  p.col = c("padj", "padj.by.collection", "pval"),
  as.dt = FALSE
)

Arguments

x	A MultiGSEAResult() object.
name	the names of the results desired
stats.only	logical, set to FALSE if you want to return all (column-wise) data for each result. By default only the pvalues, adjusted pvalues, and rank are returned.
rank.by	the statistic to use to append a rank column for the geneset result. By default we rank by pvalue calculated by the GSEA method. You can rank the results based on the trimmed mean of the logFC's calculated for all of the features in the geneset ("logFC"), or the trimmed t-statistics of the these features ("t").
add.suffix	If TRUE, adds .name as a suffix to the columns of the method-specific statistics returned, ie. the pval column from the "camera" result will be turned to pval.camera.
names	the names of the GSEA methods to be reported. By default, this function will display results for all methods.
max.p	The maximum padj value to consider a result significant
p.col	use padj or padj.by.collection?

Details

The product of an indivdual GSEA is consumed by the corresponding do.<METHOD> function and converted into a data.table of results that is internally stored.

Use the resultNames() function to identify which results are available for interrogation. The result() function returns the statistics of one individual result, and the results() function combines the results from the specified methods into an arbitrarily wide data.table with method-suffixed column names.

Use the tabulateResults() function to create a summary table that tallies the number of significant genesets per collection, per method at the specified FDR thresholds.

Value

a data.table with the results from the requested method.

a data.table that summarizes the significant results per method per collection for the GSEA that was run

Examples

## Refer to the examples in ?multiGSEA

lianos/multiGSEA documentation built on Nov. 17, 2020, 1:26 p.m.