GeneSelector (version 2.22.0)

AggregatePenalty: Aggregation of repeated rankings using a variance penalty approach

Description

The idea behind this form of aggregation is to find a compromise between quality on the one hand, represented by the list position/rank, and variability on the other hand. The latter is assessed by calling the function dispersion.

Usage

AggregatePenalty(RR, dispersion = c("sd", "mad", "iqr"), center = NULL, gamma = 0.05,...)

Arguments

RR

An object of class RepeatedRanking.

dispersion

The dispersion measure to be used (s. dispersion):

"sd"

standard deviation,

"mad"

median absolute deviation,

"iqr"

interquartile range.

center

Optional numeric vector specifying for each gene the rank serving as center/location parameter for dispersion. If center = NULL, the reference ranking RR@original@ranking is used.

gamma

As basis of the aggregated ranking, the quantity (1-gamma)*center + gamma * dispersion is used, i.e. the variability aspect dominates as gamma tends to one.

...

Further arguments passed to dispersion.

Value

AggregatedRanking.

See Also

RepeatRanking, AggregateSimple, AggregateSVD, AggregateMC

Examples

## Load toy gene expression data
data(toydata)
### class labels
yy <- toydata[1,]
### gene expression
xx <- toydata[-1,]
### run RankingTstat
ordT <- RankingTstat(xx, yy, type="unpaired")
### Generate Leave-one-out Foldmatrix
loo <- GenerateFoldMatrix(y = yy, k=1)
### Get all rankings
loor_ordT <- RepeatRanking(ordT, loo)
### aggregate rankings
agg_pen_ordT <- AggregatePenalty(loor_ordT, dispersion = "iqr", gamma = 0.3)
toplist(agg_pen_ordT)