dmt (version 0.8.20)

sharedVar: Shared variation retained in the combined drCCA representation

Description

A function for estimating the amount of shared variation (i.e. variation that is common to more than one data set) retained in the combined data set of given dimensionality.

Usage

sharedVar(datasets,regcca,dimension,pca=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.

dimension

The number of dimensions of projected data to be used

pca

A logical variable with default value FALSE. If the value is TRUE, the pairwise variation will also be calculated for the PCA projected data, where PCA is performed on the columnwise concatenation of the given data sets.

Value

A list of following elements is returned

oo

A matrix containing the pairwise shared variations for original data sets

cc

A matrix containing the pairwise shared variations for a drCCA projection of given dimensions

pc

A matrix containing the pairwise shared variations for a PCA projection of given dimensions, if pca = TRUE is given

mcca

Mean of shared variation between all pairs for drCCA

mpca

Mean of shared variation between all pairs for PCA, if pca = TRUE is given

Details

The function estimates the amount of shared information retained in a previously calculated drCCA solution. It calculates the shared variation between all pairs of the data sets returned from drCCA combined data for a particular dimensionality. The function also calculates the same quantities for the original data and for the simple PCA projection of the concatenation of data sets. This can be used as a comparison value. If the full dimensionality of drCCA projection or the PCA projection is used, the sum of all pairs of shared variations will be the same. The mean of shared variations for drCCA and PCA is estimated, normalized in a way that the value for original data sets will be 1. A good result will have value greater than 1. For details please refer to the reference below.

References

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

See Also

specificVar

Examples

  #     data(expdata1)
  #     data(expdata2)
  #     r <- regCCA(list(expdata1,expdata2))

  #     sharedVar(list(expdata1,expdata2),r,4)