estimateParameter: Estimate model parameter from spikes

In CALIB: Calibration model for estimating absolute expression levels from microarray data

Description

Estimate the calibration model parameters according to the known concentration and the measured intensities of external control spikes on each array.

Usage

 estimateParameter(spike, RG, bc = FALSE, area = TRUE, errormodel = "M")

Arguments

spike	a SpikeList object.
RG	a RGList_CALIB object.
bc	a logical value. TRUE means background corrected measured intensities are used. Default is FALSE.
area	a logical value. TRUE means spot area is used to calculate measured intensities. Namly, measured intensities are calculated by foreground intensities(or background corrected intensities, if bc is TRUE ) multiply spot area. FALSE means spot area is not used. Default is TRUE .
errormodel	a character to indicate the distribution of spot capacity. "A" means spot capacity is additive. "M" means spot capacity is multiplicative. Default is "M".

Details

This function estimates calibration model parameters. In this function, the model parameters are estimated separately for each microarray, based on the measured intensities of the external control spikes and their known concentration in the hybridization solution. It accepts spike measured intensities and concentration from spike argument, which is an object of SpikeList class.

It supports different ways to calculate the measured intensities. Arguments bc and area are logical and their combinations is used for specifying four differents ways. bc indicates using background correction or not. area indicates multiplying spot area or not. The default value of these two arguments are bc = FALSE and area = TRUE.

The argument errormodel is to specify the distribution of spot capacity of each array. The spot capacity is either additive or multiplicative. Whichever distribution is more appropriate will depend largely on the type of microarray slide and spotting procedure used. The spot parameters mus and sigmas can be considered equal for all measurements of a single array.

The argument RG is for calculating the maximum intensity of each array. These maximum intensities are used to estimate the upper saturation level of each array.

More details please refer to the reference literature.

Value

An ParameterList object containing the components:

MuS	matrix containing MuS for each array.
Ka	matrix containing Ka for each array.
P1	matrix containing P1 of each dye for each array.
P2	matrix containing P2 of each dye for each array.
SigmaA	matrix containing sigma additive for each array.
SigmaM	matrix containing sigma multiplicative for each array.
SigmaS	matrix containing sigma spoterror for each array.
SpotError	matrix containing the spot error of each spot for each array.
Method	boolean values indicating the way to calculate the measured intensities.
ErrorModel	character "M" or "A" to indicate the type of spot capacity distribution.

Author(s)

Hui Zhao

References

Engelen, K., Naudts, B., DeMoor, B., Marchal, K. (2006) A calibration method for estimating absolute expression levels from microarray data. Bioinformatics 22: 1251-1258.

Examples

# load data: RG and spike
data(RG)
data(spike)

# for the measured itensities, take the default bc=FALSE and area=TRUE.
# use multiplicative spot error model
parameter <- estimateParameter(spike,RG)

CALIB documentation built on Oct. 31, 2019, 3:45 a.m.