ListHyperGResult-class: Classes for quick GO/KEGG/CHR/miRNA target or other...

In eisa: Expression data analysis via the Iterative Signature Algorithm

Description

These classes extend the HyperGResult class from the Category package to perform enrichment calculation quickly for multiple gene sets.

Usage

## S4 method for signature 'ListHyperGResult'
summary(object, pvalue = pvalueCutoff(object), 
  categorySize = NULL)
## S4 method for signature 'ListHyperGResult'
htmlReport(r, file = "", append = FALSE,
  label = "", digits = 3, summary.args = NULL)
## S4 method for signature 'ListHyperGResult'
pvalues(r)
## S4 method for signature 'ListHyperGResult'
sigCategories(r, p)

## S4 method for signature 'ListHyperGResult'
geneCounts(r)
## S4 method for signature 'ListHyperGResult'
expectedCounts(r)
## S4 method for signature 'ListHyperGResult'
oddsRatios(r)
## S4 method for signature 'ListHyperGResult'
universeCounts(r)
## S4 method for signature 'ListHyperGResult'
geneMappedCount(r)
## S4 method for signature 'ListHyperGResult'
universeMappedCount(r)
## S4 method for signature 'ListHyperGResult'
geneIdsByCategory(r, catids = NULL)

## S4 method for signature 'ListHyperGResult'
geneIdUniverse(r, cond = FALSE)

Arguments

object,r	A ListHyperGResult object.
pvalue,p	Numeric vector of length one, the p-value limit, up to which the terms are listed.
categorySize	A numeric vector of length one, or NULL. If not NULL, then it gives the minimum number of annotated genes in the universe, in order to list the term.
file	A file name, or a connection object. The result is written here. If it is "", then the result is written to the standard output. If it is NULL, then the result is not written anywhere. (But it is always returned, invisibly, see below.)
append	Logical scalar, whether to append the HTML code to the given file, or remove its previous contents if it already exists.
label	An HTML label (<A LABEL=""> tag) to add.
digits	The number of digits to use for the numeric columns.
summary.args	A list of arguments to pass to the summary method.
catids	The categories for which the genes are listed. All categories will be listed if this argument is NULL.
cond	Currently not used.

Details

A ListHyperGResult object can store the results of hypergeometric tests, several gene sets against the same universe. ListHyperGRresult is an extension of HyperGResult, as defined in the Category package.

More precisely, ListHyperGResult is an abstract class, it is not possible to instantiate objects from it. Its extensions are be used instead: GOListHyperGResult, KEGGListHyperGResult, CHRListHyperGResult and miRNAListHyperGResult.

Value

pvalues, geneCounts, expectedCounts, oddsRatios and universeCounts return a list of named numeric vectors.

geneMappedCount returns a numeric vector, universeMappedCount returns a numeric vector of length one.

sigCategories returns a list of character vectors.

geneIdsByCategory returns a list of lists of character vectors.

geneIdUniverse returns a list of character vectors.

summary returns a list of data frames with columns: ‘Pvalue’, ‘OddsRatio’, ‘ExpCount’, ‘Count’, ‘Size’ and optionally ‘drive’.

htmlReport returns a list of chracter vectors, invisibly.

conditional returns a logical vector of length one. ontology returns a character vector of length one.

Member functions

Most of the member functions are analogous to the ones defined for HyperGResult in the Category package. Usually the only difference is that they return a list of vectors, with one entry for each gene set, instead of just a single vector.

pvalues returns the p-values of the hypergeomatric tests. A list is returned, with one numeric vector entry for each input gene set. The p-values for each gene set are ordered according to decreasing significance.

geneCounts returns the number of genes from the gene set that are annotated with the given term. This is returned for all input gene sets, in a list.

expectedCounts returns the number of genes that are expected to be annotated with the given term, just by chance. This is calculated for all input gene sets, and returned as a list.

oddsRatios returns the odds ratios for each term tested, for all gene sets, in a list of numeric vectors.

universeCounts returns the number of genes from the universe that are annotated with the given term, for all gene sets, in a list.

geneMappedCount gives the size of the gene sets, as used in the algorithm. This can be different than the size of the input gene sets, because of the elimination of duplicates and genes that are not in the universe, before the actual computation.

universeMappedCount gives the size of the gene universe, as used in the computation. This can be different than the size given by the user, because duplicates are eliminated before the computation.

sigCategories returns the significant terms, at the given p-value threshold, for all gene sets, as a list.

geneIdsByCategory returns a list of lists, one entry for each input gene set. Every entry is a list itself and for each tested term it gives the gene ids from the gene set that are annotated with the given term.

geneIdUniverse returns a list of character vectors, one for each term that was tested, giving the ids of the genes from the universe that are annotated with that term.

summary returns a list of data frames, one for each input gene set. Each data frame has columns: ‘Pvalue’, ‘OddsRatio’, ‘ExpCount’, ‘Count’, ‘Size’ and optionally ‘drive’. Each row of the data frame corresponds to a tested term.

htmlReport creates a HTML summary from a ListHyperGParams object. This consists of one table for each input gene get. The summary can be written to a file, but it is also returned in a list of character vectors. There is one list entry for each input gene set, and each element of the character vector corresponds to one line of HTML code. You need the xtable package to use this function.

The following functions are defined for GOListHyperGResult objects only.

conditional returns a logical vector of length one, whether the test was conditional or not. Conditional testing is currently not implemented, please see the GOstats package for a working implementation.

ontology returns a character vector of length one, the name of the ontology for the GO test.

Author(s)

Gabor Csardi [e-mail]

See Also

Functions for enrichment calculation of ISA modules: ISAGO, ISAKEGG, ISACHR, ISAmiRNA, ISAEnrichment.

Perhaps see also the vignette in the GOstats package.

Examples

data(ALLModulesSmall)
GO <- ISAGO(ALLModulesSmall)
GO$CC
sigCategories(GO$CC)[[1]]
summary(GO$CC)[[1]][,1:5]

eisa documentation built on Nov. 8, 2020, 6:47 p.m.