FBN.valueCenter: The SNP normalization function

In FBN: FISH Based Normalization and Copy Number Inference of SNP Microarray Data

The SNP normalization function

Description

Normalization of the raw SNP microarray values, by multiplication (on linear scale) or addition (in log scale) of all the raw SNP values with the normalization factor. The normalization factor is estimated such that it brings the normalizingValue of the raw SNP values onto the nominalValueCN.

Usage

FBN.valueCenter(inputData, normalizingValue, nominalValueCN, 
	logScale)
FBN.valueCenter(inputData = NULL, normalizingValue = NULL, 
	nominalValueCN = 2, logScale = FALSE)

Arguments

inputData	The vector of raw SNP values, as they come out from, e.g. Circular Binary Segmentation in DNAcopy package from Bioconductor
normalizingValue	The value representing the center of the cluster identified as having a certain CN
nominalValueCN	The nominal value representing a certain CN on which the normalizingValue has to be brought.
logScale	A logical value, specifying wether the data is on linear (FALSE) or logarithmic scale (TRUE).

Details

The nominalValueCN is a real value representing the CN, e.g. CN=2 has a nominalValueCN of 2, but all other CN=n (n != 2) will have a nominalValueCN different from n. Such nominalValueCN is identified by the FBN.kmeans function.

Value

Returns a vector containing the normalized values of the inputData

Author(s)

Adrian Andronache [e-mail]
Luca Agnelli [e-mail]

See Also

FBN.kmeans, FBNormalization

Examples

require(stats)
require(graphics)
x = c(rnorm(1000, 1, .1), rnorm(1000, 1.5, .1))
y = FBN.valueCenter(x, normalizingValue = 1, nominalValueCN = 2, 
	logScale = FALSE)
par(mfrow = c(2, 1), new = FALSE)
h = hist(x)
par(new = TRUE)
plot(1, 0, col = 'red', xlim = c(min(h$breaks), max(h$breaks)), 
	ylim = c(0,max(h$counts)), xlab = NA, ylab = NA)
par(new = FALSE)
h = hist(y)
par(new = TRUE)
plot(2, 0, col = 'red', xlim = c(min(h$breaks), max(h$breaks)), 
	ylim = c(0,max(h$counts)), xlab = NA, ylab = NA)

FBN documentation built on July 9, 2023, 5:18 p.m.