GeneSelector (version 2.22.0)

AggregateMC: Aggregation of repeated rankings using a Markov chain approach

Description

All obtained rankings are aggregated on the basis of Markov chain model, in which each gene constitutes an element of the state space. For details, see DeConde et al. (2006).

Usage

AggregateMC(RR, maxrank, type=c("MC4", "MCT"), epsilon = 0.15)

Arguments

RR

An object of class RepeatedRanking.

maxrank

Due to time- and memory requirements, the computation is limited to a reduced set of candidate genes. A gene is selected as candidate only if at least of one its ranks is smaller than or equal to maxrank. The remainder is assigned the rank maxrank+1 as rank after aggregation.

type

Specifies the computation of the matrix of transition probabilities. If type = "MC4", the transition probabilities are forced to be binary, while they may principally range from zero to one if type = "MCT", see DeConde et al. (2006) for details.

epsilon

A second parameter concerning the computation of the transition matrix, necessary to guarantee ergodicity and hence existence of a unique stationary distribution of the Markov chain. The value epsilon = 0.15, 0 < epsilon < 1, is recommended in DeConde et al. (2006).

Value

AggregatedRanking.

References

DeConde, R. P., Hawley, S., Falcon, S., Clegg, N., Knudsen, B., Etzioni, R. (2006). Combining results of microarray experiments: a rank aggregation approach. Statistical Applications in Genetics and Molecular Biology 5, 15

See Also

RepeatRanking, AggregateSVD, AggregatePenalty, AggregateSimple

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_MC_ordT <- AggregateMC(loor_ordT, type = "MCT", maxrank = 100)
toplist(agg_MC_ordT)