OmicsMarkeR (version 1.4.2)

pairwise.stability: Pairwise Stability Metrics

Description

Conducts all pairwise comparisons of features selected following bootstrapping. Also known as the data perturbation ensemble approach.

Usage

pairwise.stability(features, stability.metric, nc)

Arguments

features

A matrix of selected features

stability.metric

string indicating the type of stability metric. Available options are "jaccard" (Jaccard Index/Tanimoto Distance), "sorensen" (Dice-Sorensen's Index), "ochiai" (Ochiai's Index), "pof" (Percent of Overlapping Features), "kuncheva" (Kuncheva's Stability Measures), "spearman" (Spearman Rank Correlation), and "canberra" (Canberra Distance)

nc

Number of variables in original dataset

Value

A list is returned containing:

comparisons

Matrix of pairwise comparisons

overall

The average of all pairwise comparisons

References

He. Z. & Weichuan Y. (2010) Stable feature selection for biomarker discovery. Computational Biology and Chemistry 34 215-225.

Examples

# pairwise.stability demo

# For demonstration purposes only!!!
some.numbers <- seq(20)

# matrix of Metabolites identified (e.g. 5 trials)
features <- 
    replicate(5, paste("Metabolite", sample(some.numbers, 10), sep="_"))

# nc may be omitted unless using kuncheva
pairwise.stability(features, "jaccard")