evalTypeIError {GSEABenchmarkeR} | R Documentation |
This function evaluates the type I error rate of selected methods for enrichment analysis when applied to one or more expression datasets.
evalTypeIError(methods, exp.list, gs, alpha = 0.05, ea.perm = 1000, tI.perm = 1000, perm.block.size = -1, parallel = NULL, save2file = FALSE, out.dir = NULL, ...)
methods |
Methods for enrichment analysis. A character vector with
method names chosen from |
exp.list |
Experiment list. A |
gs |
Gene sets, i.e. a list of character vectors of gene IDs. |
alpha |
Numeric. Statistical significance level. Defaults to 0.05. |
ea.perm |
Integer. Number of permutations of the sample group assignments during enrichment analysis. Defaults to 1000. Can also be an integer vector matching the length of 'methods' to assign different numbers of permutations for different methods. |
tI.perm |
Integer. Number of permutations of the sample group assignments
during type I error rate evaluation. Defaults to 1000. Can also be an integer
vector matching the length of |
perm.block.size |
Integer. When running in parallel, splits |
parallel |
Parallel computation mode. An instance of class
|
save2file |
Logical. Should results be saved to file for subsequent
benchmarking? Defaults to |
out.dir |
Character. Determines the output directory where results are
saved to. Defaults to |
... |
Additional arguments passed to the selected enrichment methods. |
A list with an entry for each method applied. Each method entry is a list with an entry for each dataset analyzed. Each dataset entry is a list of length 2, with the first element being the runtime and the second element being the gene set ranking, as obtained from applying the respective method to the respective dataset.
Ludwig Geistlinger <Ludwig.Geistlinger@sph.cuny.edu>
sbea
and nbea
for carrying out set- and network-based enrichment analysis.
BiocParallelParam
and register
for
configuration of parallel computation.
# loading three datasets from the GEO2KEGG compendium geo2kegg <- loadEData("geo2kegg", nr.datasets=3) # only considering the first 1000 probes for demonstration geo2kegg <- lapply(geo2kegg, function(d) d[1:1000,]) # preprocessing and DE analysis for two of the datasets geo2kegg <- maPreproc(geo2kegg[2:3]) geo2kegg <- runDE(geo2kegg) # getting a subset of human KEGG gene sets gs.file <- system.file("extdata", package="EnrichmentBrowser") gs.file <- file.path(gs.file, "hsa_kegg_gs.gmt") kegg.gs <- EnrichmentBrowser::getGenesets(gs.file) # evaluating type I error rate of two methods on two datasets # NOTE: using a small number of permutations for demonstration; # for a meaningful evaluation tI.perm should be >= 1000 res <- evalTypeIError(geo2kegg, methods=c("ora", "camera"), gs=kegg.gs, ea.perm=0, tI.perm=3)