GetStabilityUnion: Stability measures for gene lists

In GeneSelector: Stability and Aggregation of ranked gene lists

Description

The similarity of multiple ordered genelists is assessed by counting the size of the union ('union count') for each position in the list. The higher the union count, the less stable are the ordered lists. Similarly to the 'overlap score' of Yang et al. (2006), Lottaz et al. (2006), we compute a weighted average of the union count entitled 'union score' .

Usage

GetStabilityUnion(RR, decay = c("linear", "quadratic", "exponential"),
                    alpha = 1, noinformation = 0,...)

Arguments

RR	An object of class RepeatedRanking
decay	Argument controlling the weight decay of the weights used for the computation of the union score. If decay="linear", then we have weight 1/l for list position l, if decay="quadratic", then the weight is 1/l^2 and if decay="exponential", then the weight is exp(-alpha*l) where alpha is a tuning parameter, specified via the argument alpha. Weights are used only for the union score, not for the union count.
alpha	s. decay.
noinformation	If noinformation is a positive integer, union count and -score in the no-information case are approximated by randomly generating multiple lists and computing scores noinformation times and averaging the results. Note that this procedure can be rather slow, depending on the number and the length of the lists to be compared.
...	Currently unused argument.

Value

An object of class StabilityOverlap

Note

• Union count and union score are suitably normalized to fall into the unit interval for the sake of better interpretability, with zero corresponding to maximal instability.

• Note that this function yields exactly one stability score for multiple lists, as opposed to the pairwise approach in GetStabilityOverlap and GetStabilityDistance.

Author(s)

Martin Slawski
Anne-Laure Boulesteix

References

Stolovitzky, G. (2003). Gene selection in microarray data: the elephant, the blind men, and our algorithms. Current Opinion in Structural Biology 13, 370-376.

Jurman, G., Merler, S., Barla, A., Paoli, S., Galea, A., Furlanello, C. (2008).
Algebraic stability indicators for ranked lists in molecular profiling. Bioinformatics 24, 258-264

Yang, X., Bentink, S., Scheid, S., Spang, R. (2006)
Similarities of ordered gene lists. Journal of Bioinformatics and Computational Biology 4, 693-708

Lottaz, C., Yang, X., Scheid, S., Spang, R. (2006)
OrderedList - a Bioconductor package for detecting similarity in ordered gene lists. Bioinformatics, 22, 2315-2316

See Also

RepeatRanking

Examples

 ## Load toy gene expression data
data(toydata)
### class labels
yy <- toydata[1,]
### gene expression
xx <- toydata[-1,]
### get ranking 
ordT <- RankingTstat(xx, yy, type="unpaired")
### Generate Leave-One-Out
loo <- GenerateFoldMatrix(y = yy, k=1)
### Repeat Ranking with t-statistic
loor_ordT <- RepeatRanking(ordT, loo)
### assess stability
stab_union_ordT <- GetStabilityUnion(loor_ordT, decay="linear")
### display the result
plot(stab_union_ordT)

GeneSelector documentation built on May 1, 2019, 11:35 p.m.