ap_mads2: MAD normalization
In cekehe/rappp: QC, data transformation and visualization for PAPP

View source: R/data_transformation_2_diff_input.R

ap_mads2

R Documentation

MAD normalization

Description

Sample based normalization to number of Median Absolute Deviations (MADs) from the median (or other quantile probability) for Autoimmunity profiling data.

Usage

ap_mads2(
  x,
  center_prob = 0.5,
  constant = 1,
  na.rm = TRUE,
  low = FALSE,
  high = FALSE,
  check.names = FALSE
)

Arguments

`x`	list with at least two elements, see Details for naming and content.
`center_prob`	value in [0,1] passed to prob in `quantile()`, defaults to the median.
`constant`	constant for `mad()` function, default is 1 (compared to 1.4826 in base function).
`na.rm`	logical, indicating whether NA values should be stripped before the computation proceeds. Altered default from `quantile()` and `mad()`.
`low`	if TRUE, compute the ‘lo-median’, i.e., for even sample size, do not average the two middle values, but take the smaller one.(From `mad()`).
`high`	if TRUE, compute the ‘hi-median’, i.e., take the larger of the two middle values for even sample size.(From `mad()`).
`check.names`	logical, altered default from `data.frame()`.

Details

The input values will be normalized per sample to the number of MADs from the median using the algorithm MADs = (MFI - median )/MAD, where MAD is calculated using mad(constant=1).

The x list needs to include at least the elements:

MFI = assay mfi,

BEADS = Beads info, if any should be excluded then these should be annotated in a column called "Filtered". Any beads with no text (ie. "" or NA) or "NegControl" in such column will be included in the transformation.