GeneSelector (version 2.22.0)

RankingPermutation: Ranking based on permutation tests.

Description

The function is a wrapper for mt.sample.teststat from the package multtest (Dudoit et al., 2003). The ranking is based on permutation p-values first, followed by the absolute value of the statistic.

Usage

RankingPermutation(x, y, type = "unpaired", B = 100, gene.names = NULL, ...)

Arguments

x

A matrix of gene expression values with rows corresponding to genes and columns corresponding to observations or alternatively an object of class ExpressionSet.

y

If x is a matrix, then y may be a numeric vector or a factor with at most two levels. If x is an ExpressionSet, then y is a character specifying the phenotype variable in the output from pData.

type

Only the two sample case, type="unpaired" is possible.

B

The number of permutations to generate. Defaults to 100, but should be increased if computing power admits. Taking B too high, however, can lead to long computation time, especially if the function is called from RepeatRanking

gene.names

An optional vector of gene names.

...

Further arguments passed to mt.sample.teststat from the package multtest. Can be used, for example, to select the statistic to be computed. By default this is "t.equalvar" (t-test with equal variances assumed).

Value

GeneRanking

References

Dudoit, S., Shaffer, J.P., Boldrick, J.C. (2003). Multiple Hypothesis Testing in Microarray Experiments Statistical Science, 18, 71-103

See Also

RepeatRanking, RankingTstat, RankingFC, RankingWelchT, RankingWilcoxon, RankingBaldiLong, RankingFoxDimmic, RankingLimma, RankingEbam, RankingWilcEbam, RankingSam, RankingShrinkageT, RankingSoftthresholdT

Examples

### Load toy gene expression data
data(toydata)
### class labels
yy <- toydata[1,]
### gene expression
xx <- toydata[-1,]
### run RankingPermutation (100 permutations)
perm <- RankingPermutation(xx, yy, B=100, type="unpaired")