rnb.execute.dreduction

Description

Performs principal component analysis (PCA) and multi-dimensional scaling (MDS) of the samples in the given methylation dataset.

Usage

1
rnb.execute.dreduction(rnb.set, target = "sites")

Arguments

rnb.set

Methylation dataset as an object of type inheriting RnBSet. This dataset must contain at least four samples.

target

character singleton specifying the level of DNA methylation infromation. If this is "sites", the DNA methylation information for the individual sites or probes is analyzed. Otherwise, this should be one of the supported region types, as returned by rnb.region.types.

Details

Row names in the returned matrices are sample identifiers, determined based on the package option "identifiers.column". See RnBeads Options for more information on this option.

Value

Results of the dimension reduction in the form of a list with the following elements:

pca

Results of the PCA as returned by the function prcomp.

mds

List of two elements - "manhattan" and "euclidean", each of which is a two-column matrix storing the coordinates of the samples in a two-dimensional space. The matrices are computed using the function isoMDS.

Author(s)

Yassen Assenov

See Also

rnb.run.exploratory for running the whole exploratory analysis module

Examples

1
2
3
4
5
6
7
library(RnBeads.hg19)
data(small.example.object)
regs <- c("sites", summarized.regions(rnb.set.example))
dreduction <- function(x) rnb.execute.dreduction(rnb.set.example, x)
pcoordinates <- lapply(regs, dreduction)
names(pcoordinates) <- regs
str(pcoordinates)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.