pint (version 1.22.0)

z.effects: The model parameters z and W

Description

Contribution of each sample to a dependency model, and contribution of each variable.

Usage

z.effects(model, X, Y = NULL)
W.effects(model, X, Y = NULL)

Arguments

model

The fitted dependency model.

X, Y

Data sets used in fitting the dependency modeling functions (screen.cgh.mrna or link{fit.dependency.model}). Note: Arguments must be given in the same order as in fit.dependency.model or screen.cgh.mrna. Only X is needed for dependency model for one data set.

Value

z.effects gives a projection vector over the samples and W.effects gives a projection vector over the variables.

Details

z.effects gives the contribution of each sample to the dependency score. This is approximated by projecting original data to first principal component of Wz. This is possible only when the data window is smaller than half the number of samples.

W.effects gives the contribution of each variable to the observed dependency. This is approximated with the loadings of the first principal component of Wz

Original data can be retrieved by locating the row in X (or Y) which has the same variable (gene) name than model.

References

Dependency Detection with Similarity Constraints, Lahti et al., 2009 Proc. MLSP'09 IEEE International Workshop on Machine Learning for Signal Processing, See http://www.cis.hut.fi/lmlahti/publications/mlsp09_preprint.pdf

A Probabilistic Interpretation of Canonical Correlation Analysis, Bach Francis R. and Jordan Michael I. 2005 Technical Report 688. Department of Statistics, University of California, Berkley. http://www.di.ens.fr/~fbach/probacca.pdf

Probabilistic Principal Component Analysis, Tipping Michael E. and Bishop Christopher M. 1999. Journal of the Royal Statistical Society, Series B, 61, Part 3, pp. 611--622. http://research.microsoft.com/en-us/um/people/cmbishop/downloads/Bishop-PPCA-JRSS.pdf

See Also

DependencyModel-class, screen.cgh.mrna

Examples

data(chromosome17)
window <- fixed.window(geneExp, geneCopyNum, 150, 10)

## pSimCCA model around one gene
depmodel <- fit.dependency.model(window$X, window$Y)
# Conversion from DependencyModel to GeneDependencyModel so that gene name and location can be stored
depmodel <- as(depmodel,"GeneDependencyModel")
setGeneName(depmodel) <- window$geneName
setLoc(depmodel) <- window$loc
barplot(z.effects(depmodel, geneExp, geneCopyNum))

## Plot the contribution of each genes to the model. Only the X component is plotted
## here since Wx = Wy (in SimCCA) 
barplot(W.effects(depmodel, geneExp, geneCopyNum)$X)

## plot.DpenendencyModel shows also sample and variable effects
plot(depmodel,geneExp,geneCopyNum)