tni.gsea2: Compute regulon activity by calling GSEA2 (two-tailed Gene...

In RTN: RTN: Reconstruction of Transcriptional regulatory Networks and analysis of regulons

Description

Uses GSEA2 algorithm to compute regulon activity for TNI-class objects.

Usage

tni.gsea2(object, minRegulonSize=15, sizeFilterMethod="posORneg", scale=FALSE, 
exponent=1, tnet="dpi", regulatoryElements=NULL, samples=NULL, features=NULL, refsamp=NULL, 
log=TRUE, alternative=c("two.sided", "less", "greater"), targetContribution=FALSE, 
additionalData=FALSE, verbose=TRUE, doSizeFilter=NULL)

Arguments

object	a preprocessed object of class 'TNI' TNI-class.
minRegulonSize	a single integer or numeric value specifying the minimum number of elements in a regulon. Regulons smaller than this number are removed from the analysis.
sizeFilterMethod	a single character value specifying the use of the 'minRegulonSize' argument, which is applyed to regulon's positive and negative targets. Options: "posANDneg", "posORneg", "posPLUSneg". For "posANDneg", the number of both positive and negative targets should be > 'minRegulonSize'; for "posORneg", the number of either positive or negative targets should be > 'minRegulonSize'; and for "posPLUSneg", the number of all targets should be > 'minRegulonSize'.
scale	A logical value specifying if expression values should be centered and scaled across samples (when verbose=TRUE) or not (when verbose=FALSE).
exponent	a single integer or numeric value used in weighting phenotypes in GSEA.
tnet	can take values of 'refnet', 'dpi' or 'cdt'. It refers to the version of the regulatory network that will be used for GSEA analysis.
regulatoryElements	an optional vector with transcription factor identifiers.
samples	an optional string vector containing the sample names for which will be computed the GSEA2.
features	a string vector containing features for feature selection.
refsamp	an optional string vector containing the names of the reference samples for differential expression calculations. If not provided, then the average of all samples will be used as reference.
log	a logical value. If TRUE, it will check whether the expression values are provided as logged data; if not, it will performe a log2 transformation on expression values before the differential expression calculations.
alternative	a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less".
targetContribution	This argument is used for internal calls. A single logical value specifying to return the contribution of each target in enrichment scores (when verbose=TRUE) or not (when verbose=FALSE).
additionalData	This argument is used for internal calls. A single logical value specifying to return the additional data objects (when verbose=TRUE) or not (when verbose=FALSE).
verbose	a single logical value specifying to display detailed messages (when verbose=TRUE) or not (when verbose=FALSE).
doSizeFilter	'doSizeFilter' is deprecated, please use the 'sizeFilterMethod' parameter.

Value

a list with enrichment scores for all samples in the TNI. The list contains the following elements:

differential:

A numeric "matrix" with differential enrichment scores (dES).

positive:

A numeric "matrix" with enrichment scores from positive targets.

negative:

A numeric "matrix" with enrichment scores from negative targets.

status:

A numeric "matrix" with discretized scores derived from the dES values.

regulatoryElements:

A character vector listing the regulatory elements assessed by the GSEA-2T algorithm.

sections:

A single numeric value used in internal plots.

Author(s)

Mauro Castro

See Also

TNI-class tna.gsea2 tna.plot.gsea2

Examples

data(tniData)

## Not run: 

rtni <- tni.constructor(expData=tniData$expData, 
        regulatoryElements=c("PTTG1","E2F2","FOXM1","E2F3","RUNX2"), 
        rowAnnotation=tniData$rowAnnotation)
rtni <- tni.permutation(rtni)
rtni <- tni.bootstrap(rtni)
rtni <- tni.dpi.filter(rtni)

#run GSEA2 analysis pipeline
rtni <- tni.gsea2(rtni)

#get results
regulonActivity <- tni.get(rtni, what = "regulonActivity")

#parallel version with SNOW package!
library(snow)
options(cluster=snow::makeCluster(3, "SOCK"))
rtni <- tni.gsea2(rtni)
stopCluster(getOption("cluster"))


## End(Not run)

RTN documentation built on Nov. 12, 2020, 2:02 a.m.