Data-specific and shared variance for several dimensionalities

Share:

Description

A function for calculating the captured variations for several different number of retained dimensions. This is a wrapper over specificVar and sharedVar to help computing both for a range of dimensionalities.

Usage

1
plotVar(datasets,regcca,dimVector,plot=FALSE)

Arguments

datasets

A list containing the data matrices to be combined. Each matrix needs to have the same number of rows (samples), but the number of columns (features) can differ. Each row needs to correspond to the same sample in every matrix.

regcca

Output of regCCA function, containing the solution of the generalized CCA.

dimVector

A list of dimensions for which the retained variations are to be computed.

plot

A logical variable with default value FALSE. If the value is TRUE, the functions creates a plot of the output.

Details

The function uses specificVar and sharedVar to do all the computation. The purpose of this function is to provide an easy way to visualize the properties of the reduced-dimensional representation created by drCCA. The function also estimates the same quantities for PCA of concatenated feature vectors to illustrate the difference to optimal linear model based on preserving the total variation in the whole collection of data sets.

Value

The function returns the data-specific and shared variance for the given values of dimensions in a list. The list has four components.

pw_cca

A vector with values as shared variances captured by drCCA for the given dimensions

pw_pca

A vector with values as shared variances captured by PCA for the given dimensions

within_cca

A vector with values as data-specific variances captured by drCCA for the given dimensions

within_pca

A vector with values as data-specific variances captured by PCA for the given dimensions

Author(s)

Abhishek Tripathi, Arto Klami

References

Tripathi A., Klami A., Kaski S. (2007), Simple integrative preprocessing preserves what is shared in data sources, submitted for publication.

See Also

sharedVar,specificVar

Examples

1
2
3
4
5
#       data(expdata1)
#       data(expdata2)
#       r <- regCCA(list(expdata1,expdata2))#

#       plotVar(list(expdata1,expdata2),r,c(1:2),4)