globalIntensityNormalization: The normalization divides the intensity of each variable in a...

View source: R/normMethods.R

globalIntensityNormalizationR Documentation

The normalization divides the intensity of each variable in a sample with the sum of intensities of all variables in the sample and multiplies with the median of sum of intensities of all variables in all samples. The normalized data is then log2-transformed.

Description

The normalization divides the intensity of each variable in a sample with the sum of intensities of all variables in the sample and multiplies with the median of sum of intensities of all variables in all samples. The normalized data is then log2-transformed.

Usage

globalIntensityNormalization(rawMatrix, noLogTransform = FALSE)

Arguments

rawMatrix

Target matrix to be normalized

noLogTransform

Assumes no need for log transformation

Value

Normalized and log-transformed matrix

Examples

data(example_data_only_values_small)
normMatrix <- globalIntensityNormalization(example_data_only_values)

ComputationalProteomics/NormalyzerDE documentation built on Oct. 18, 2024, 9:57 p.m.