subtractFromLocation: Subtract Numeric Feature Measurements from a Location

In ClassifyR: A framework for cross-validated classification problems, with applications to differential variability and differential distribution testing

Description

For each numeric feature, calculates the location, and subtracts all measurements from that location.

Usage

  ## S4 method for signature 'matrix'
subtractFromLocation(measurements, training, location = c("mean", "median"),
                     absolute = TRUE, verbose = 3)
  ## S4 method for signature 'DataFrame'
subtractFromLocation(measurements, training, location = c("mean", "median"),
                     absolute = TRUE, verbose = 3)
  ## S4 method for signature 'MultiAssayExperiment'
subtractFromLocation(measurements, training, targets = names(measurements),
                           location = c("mean", "median"), absolute = TRUE, verbose = 3)

Arguments

measurements	A matrix, DataFrame or a MultiAssayExperiment object containing the data. For a matrix, the rows are for features and the columns are for samples.
training	A vector specifying which samples are in the training set.
location	Character. Either "mean" or "median".
absolute	Logical. Default: TRUE. If TRUE, then absolute values of the differences are returned. Otherwise, they are signed.
targets	If measurements is a MultiAssayExperiment, the names of the data tables to be used. "clinical" is also a valid value and specifies that numeric variables from the clinical data table will be used.
verbose	Default: 3. A progress message is shown if this value is 3.

Details

Only the samples specified by training are used in the calculation of the location. To use all samples for calculation of the location, simply provide indices of all the samples.

Value

The same class of variable as the input variable measurements is, with the numeric features subtracted from the calculated location.

Author(s)

Dario Strbenac

Examples

  aMatrix <- matrix(1:100, ncol = 10)
  subtractFromLocation(aMatrix, training = 1:5, "median")

ClassifyR documentation built on Nov. 8, 2020, 6:53 p.m.