integrate_pca: Integration of multiple Assays into linear combinations by...

In subhadeep1024/OMICsPCA: An R package for quantitative integration and analysis of multiple omics assays from heterogeneous samples

Description

This function integrates multiple assays read by InputOMICs() into linear combinations using PCA. Following earlier analyses, user may want to exclude some variables/columns/cell lines which should be supplied through the "exclude" argument (see vignettes for more explanation)

PCA is done by the function PCA() from package "FactoMineR". Additional arguments may be supplied to PCA() through "..." .

Usage

integrate_pca(Assays, name, exclude, mergetype = 1,
groupinfo = NULL, ...)

Arguments

Assays	A vector of names of an Assays read by "InputOMICs()"
name	Name of the "MultiAssayExperiment" object containing the assay data
exclude	A list that indicates which columns to exclude. See vignettes for details
mergetype	2 options : 1 = subset of groups should be entered by user and PCA is run on this subset to integrate Assays into PCs 2 = PCA is run on all groups
groupinfo	same as plot_density()
...	aditional arguments supplied to function "PCA()" from package "FactoMineR"

Details

The "mergetype" argument lets user decide whether to combine all groups or a subset of groups created by "create_group". If mergetype is set to 1, "integrateAssays" asks for the name of groups which should be a charatcer or charagter vector (see vignettes for more details).

Value

"integrateAssays()" returns a list containing the start and end column of each Assay as ordered in the "Assays" argument.

Author(s)

Subhadeep Das <[e-mail]>

Examples

int_PCA <- integrate_pca(Assays = c("H2az",
"H3k9ac"), name = multi_assay, mergetype = 2,
exclude = list(0,c(1,9)), graph = FALSE,
groupinfo = groupinfo_ext)

subhadeep1024/OMICsPCA documentation built on March 17, 2020, 5:23 p.m.