HyperGResult
class from the Category
package to perform enrichment
calculation quickly for multiple gene sets.
"summary"(object, pvalue = pvalueCutoff(object), categorySize = NULL)
"htmlReport"(r, file = "", append = FALSE, label = "", digits = 3, summary.args = NULL)
"pvalues"(r)
"sigCategories"(r, p)
"geneCounts"(r)
"expectedCounts"(r)
"oddsRatios"(r)
"universeCounts"(r)
"geneMappedCount"(r)
"universeMappedCount"(r)
"geneIdsByCategory"(r, catids = NULL)
"geneIdUniverse"(r, cond = FALSE)
ListHyperGResult
object.NULL
. If
not NULL
, then it gives the minimum number of annotated genes
in the universe, in order to list the term.""
, 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.)summary
method.NULL
.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.
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.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
.
ISAGO
, ISAKEGG
, ISACHR
,
ISAmiRNA
, ISAEnrichment
.
Perhaps see also the vignette in the GOstats
package.
data(ALLModulesSmall)
GO <- ISAGO(ALLModulesSmall)
GO$CC
sigCategories(GO$CC)[[1]]
summary(GO$CC)[[1]][,1:5]
Run the code above in your browser using DataLab