gsva(expr, geneSet,
method = gsva,
kcdf = none,
min.sz = 1,
max.sz = 2147483647,
mxdiff = TRUE,
tau = 1,
kernel = TRUE,
rnaseq = FALSE,
abs.ranking = FALSE,
alpha = 0.25,
normalize = TRUE,
name.suffix = FALSE,
verbose = FALSE);
expr
A raw gene expression data matrix. Acceptable inputs are an Matrix object or an R# dataframe whose rows are features (genes) and whose columns are samples.
geneSet
A gene-set collection supplied either as a GSEA Background model or as anything that BackgroundHelper.CastBackground can coerce into one (for example a GMT file path or a list of gene-set objects).
method
The enrichment scoring method: Methods.gsva (default, kernel based ES), Methods.ssgsea, Methods.zscore or Methods.plage. [as Methods]
kcdf
The kernel density estimation type used to estimate the cumulative distribution function of each expression profile: KCDFs.Gaussian for continuous data, KCDFs.Poisson for count data, or KCDFs.none for the empirical cumulative distribution function. Defaults to KCDFs.none, i.e. the same behavior as the underlying function when no kernel is requested. [as KCDFs]
min.sz
Minimum gene-set size (number of mapped features). Gene sets with fewer mapped features are removed from the analysis. Default 1. [as integer]
max.sz
Maximum gene-set size (number of mapped features). Gene sets with more mapped features are removed from the analysis. Default unlimited. [as integer]
mxdiff
When TRUE (default) the ES range is transformed to the interval [-1, 1] by using the difference between the largest and the smallest ES values; otherwise the original ES range is kept. [as boolean]
tau
Exponent (τ, default 1) applied to the weight of the cumulative distribution function when method is Methods.gsva; higher values give more weight to the tails of the distribution. [as double]
kernel
Logical flag to use the Gaussian kernel to estimate the cumulative distribution function when method is Methods.gsva. Ignored when a non-kernel kcdf is requested. [as boolean]
rnaseq
Logical flag indicating that expr is read count data from an RNA-seq experiment; this switches the kernel to a Poisson density when a kernel is used. [as boolean]
abs.ranking
When TRUE (default FALSE) the enrichment scores are computed using the absolute values of the ranking, which is appropriate for single-sample (ssGSEA-style) scoring. [as boolean]
alpha
The α (alpha) parameter (default 0.25) of the ssGSEA scoring method, controlling the weight of the random-walk deviation. [as double]
normalize
Logical flag (default TRUE) to normalize the enrichment scores so that each gene set has a mean enrichment of zero across samples. [as boolean]
name.suffix
Append the gene-set description (names) to the pathway identifier as a suffix in the returned matrix, e.g. "hsa00010 - Glycolysis". [as boolean]
verbose
Logical flag (default FALSE) to emit progress and diagnostic messages during the calculation. [as boolean]
env
The R# runtime environment (used for error reporting). [as Environment]
GSVA is a non-parametric, unsupervised method for estimating variation of gene set enrichment through the samples of an expression data set. It performs a change in coordinate systems, transforming the data from a gene-by-sample matrix to a gene-set-by-sample matrix, thereby allowing the evaluation of pathway enrichment for each sample. This new matrix of GSVA enrichment scores facilitates applying standard analytical methods such as functional enrichment, survival analysis, clustering, CNV-pathway analysis or cross-tissue pathway analysis, in a pathway-centric manner. Hänzelmann S., Castelo R. and Guinney J. GSVA: gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics, 14:7, 2013.