explVar {MatrixQCvis}R Documentation

Retrieve the explained variance for each principal component (PCA) or axis (PCoA)

Description

The function 'explVar' calculates the proportion of explained variance for each principal component (PC, 'type = "PCA"') and axis ('type = "PCoA"').

Usage

explVar(x, params, type = c("PCA", "PCoA"))

Arguments

x

'matrix', containing no missing values ('NA'), samples in columns and features in rows

params

'list', containing the parameters for PCA and PCoA. For 'type = "PCA"' these are 'center' of type 'logical' (indicating whether the variables should be shifted to be zero centered) and 'scale' of type 'logical'(indicating whether the variables should be scaled that they have a standard variation of 1). For 'type = "PCoA"', this is 'method' of type 'character' (indicating the method for distance calculation).

type

'character', one of '"PCA"' or '"PCoA"'

Details

'explVar' uses the function 'prcomp' from the 'stats' package to retrieve the explained standard deviation per PC ('type = "PCA"') and the function 'cmdscale' from the 'stats' package to retrieve the explained variation based on eigenvalues per Axis ('type = "PCoA"').

Value

'numeric' vector with the proportion of explained variance for each PC or Axis

Author(s)

Thomas Naake

Examples

x <- matrix(1:100, nrow = 10, ncol = 10, 
    dimnames = list(1:10, paste("sample", 1:10)))
set.seed(1)
x <- x + rnorm(100)
explVar(x = x, params = list(center = TRUE, scale = TRUE), type = "PCA")
explVar(x = x, params = list(method = "euclidean"), type = "PCoA")


[Package MatrixQCvis version 1.1.2 Index]