ProcessMultiBlock: ProcessMultiBlock

In f-puig/R.ComDim: ComDim analysis under R environment

ProcessMultiBlock

Description

Apply a custom function to Process the MultiBlocks and/or to select some blocks. If multiple functions are used, the order followed is FUN, FUN.SelectVars, and FUN.SelecBlocks.

Usage

ProcessMultiBlock(
  MB = MB,
  blocks = NULL,
  vars = NULL,
  FUN = NULL,
  FUN.SelectVars = NULL,
  FUN.SelectBlocks = NULL
)

Arguments

MB	The MultiBlocks structure.
blocks	The blocks to apply the processing. It can be a vector of integers or a vector with the block names. Facultative.
vars	The variables to kept. It is a list object with the same length as 'blocks'. Each element in the list contains the position of the variables in the nth block.
FUN	The function used for data processing.
FUN.SelectVars	The function used to select variables. If applied to a numeric data matrix, it should return a vector of TRUE/FALSE with the same length as ncol.
FUN.SelectBlocks	The function used to select blocks. If applied to a numeric data matrix, it should return TRUE/FALSE.

Value

The MultiBlock

Examples

b1 = matrix(rnorm(500),10,50)
b2 = matrix(rnorm(800),10,80)
# Build MultiBlock by adding one data block at a time:0))
mb <- BuildMultiBlock(b1, newSamples = paste0('sample_',1:1
mb <- BuildMultiBlock(b2, growingMB = mb)
# Prepare custom functions
BY100 <- function(x) { 100*(x/max(x)) } # Normalize each block to 100%
norm100000 <- function(x){norm(x)>100000} # Keep blocks with a norm higher than 100000.
thr5 <- function(x) {x > max(x, na.rm = TRUE) * 0.05} # Keep variables larger than 5% in the block.
mb <- ProcessMultiBlock(MB, FUN = BY100, FUN.SelectBlocks = norm100000, FUN.SelectVars = thr5)

f-puig/R.ComDim documentation built on Feb. 20, 2024, 6:49 a.m.