computePCs: Topological PCA

Description Usage Arguments Details Value Examples

Description

For internal use only. Performs Principal Componenent analysis.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
topoCompPCsWithResampling(resampling, exp, shrink, cliques, k)

sparseCompPCsWithResampling(resampling, exp, shrink, k)

compPCsWithResampling(resampling, exp, shrink, k)

computePCs(exp, shrink = FALSE, method = c("regular", "topological",
  "sparse"), cliques = NULL, maxPCs = 3)

topoCompPCs(exp, shrink, cliques, k)

sparseCompPCs(exp, shrink, k)

compPCs(exp, shrink, k)

Arguments

resampling

list of resampled columns

exp

a matrix

shrink

logical, whether to shrink or not.

cliques

the pathway topology summarized in a list of cliques

k

the number of components to use

method

one of 'regular', 'topological' and 'sparse'

maxPCs

the maximum number of PCs to consider

Details

Three methods are implemented: * regular: a regular PCA ('prcomp') * topological: PCA using a pathway topology. * sparse: sparse PCA analysis implemented by 'elasticnet'

Value

a list with the following elements:

x

the computed PCs

sdev

the standard deviation captured by the PCs

loadings

the loadings

a list with the following elements:

x

the computed PCs

sdev

the standard deviation captured by the PCs

loadings

the loadings

a list with the following elements:

x

the computed PCs

sdev

the standard deviation captured by the PCs

loadings

the loadings

a list with the following elements:

x

the computed PCs

sdev

the standard deviation captured by the PCs

loadings

the loadings

Examples

1
2
  fakeExp <- randomExpression(4)
  computePCs(t(fakeExp))

cavei/houseOfClipUtility documentation built on May 12, 2019, 5:23 p.m.