OLIN (version 1.50.0)

ino: Intensity-dependent normalisation of two-colour microarrays

Description

This functions performs intensity-dependent normalisation based on local regression by locfit.

Usage

ino(object,alpha=0.3,weights=NA,bg.corr="subtract",...)

Arguments

object

object of class “marrayRaw” or “marrayNorm”

alpha

smoothing parameter

weights

matrix of weights for local regression. Rows correspond to the spotted probe sequences, columns to arrays in the batch. These may be derived from the matrix of spot quality weights as defined for “maRaw” objects.

bg.corr

backcorrection method (for “marrayRaw” objects) : “none” or “subtract”(default).

...

Further arguments for locfit function.

Value

“marrayNorm” with normalised logged ratios

Details

The function ino regresses the average logged fold changes (M) with respect to the average logged spot intensity (A). The residuals of this fit are the normalised logged fold changes. The parameter alpha specifies the fraction of points that are included in the neighbourhood and thus has a value between 0 and 1. Larger alpha values lead to smoother fits.

See Also

maNorm, locfit.raw,olin, oin, lin

Examples

# LOADING DATA
  data(sw)

# INTENSITY-DEPENDENT NORMALISATION
  norm.ino <- ino(sw)

# MA-PLOT OF NORMALISATION RESULTS OF FIRST ARRAY
  plot(maA(norm.ino)[,1],maM(norm.ino)[,1],main="INO")
 
# CORRESPONDING MXY-PLOT
   mxy.plot(maM(norm.ino)[,1],Ngc=maNgc(norm.ino),Ngr=maNgr(norm.ino),
                Nsc=maNsc(norm.ino),Nsr=maNsr(norm.ino),main="INO")