stepNorm (version 1.44.0)

seqWithinNorm: Sequential within-slide normalization function

Description

This function conducts cDNA microarray normalization in a sequential fashion. In a two-color cDNA array setting, within-slide normalization calibrates signals from the two channels to remove non-biological variation introduced by various processing steps.

Usage

seqWithinNorm(marraySet, y = "maM", subset = TRUE, loss.fun = square, A = c("loess", "rlm", "median", "none"), PT = c("median", "rlm", "loess", "none"), PL = c("median", "rlm", "loess", "none"), Spatial2D = c("none", "aov2D", "rlm2D", "loess2D", "spatialMedian"), criterion = c("BIC", "AIC"))

Arguments

marraySet

Object of class marrayRaw or class marrayNorm, containing intensity data for the batch of arrays to be normalized.

y

Name of accessor method for spot statistics, usually the log-ratio maM.

subset

A "logical" or "numeric" vector indicating the subset of points used to compute the normalization values.

loss.fun

The loss function used in calculating deviance, the default uses squared sum of residuals; for absolute sum of residuals, use abs

A

A character string specifying the normalization model for the adjustment of intensity or $A$ bias:

loess:

global intensity or $A$-dependent robust nonlinear normalization using the loess function

rlm:

global intensity or $A$-dependent robust linear normalization using the rlm function

median:

global median location normalization

none:

no normalization for the $A$ bias

If not specified, loess normalization will be applied.

PT

A character string specifying the normalization model for the adjustment of print-tip or $PT$ bias:

median:

within-print-tip-group median normalization

rlm:

within-print-tip-group robust linear normalization using the rlm function

loess:

within-print-tip-group robust nonlinear normalization using the loess function

none:

no normalization for the $PT$ bias

If not specified, median normalization within print-tip will be applied.

PL

A character string specifying the normalization model for the adjustment of well-plate or $PL$ bias:

median:

within-well-plate median normalization

rlm:

within-well-plate robust linear normalization using the rlm function

loess:

within-well-plate robust nonlinear normalization using the loess function

none:

no normalization for the $PL$ bias

If not specified, median normalization within well-plate will be applied.

Spatial2D

A character string specifying the normalization model for the adjustment of spatial 2D bias:

none:

no normalization for the spatial 2D bias

aov2D:

spatial bivariate location normalization using ANOVA

rlm2D:

spatial bivariate location normalization using the rlm function

loess2D:

spatial bivariate location normalization using the loess function

spatialMedian:

spatial location normalization using a spatial median approach (see Wilson et al. (2003) in reference)

If not specified, no normalization will be carried out in this step.

criterion

Character string specifying the criterion:

AIC:

the AIC criterion is used; see calcAIC.

BIC:

the BIC criterion is used; see calcBIC.

If no specification, BIC is used. Note that here we don't use the criterion to choose normalization model in each step. Criterion is calculated solely for informaion purpose.

Value

An object of class "list":

normdata

an object of class marrayNorm, containing the normalized intensity data.

res

a list of the sequential normalization result for each slide within the marray dataset. Each list component is also a list containing the name of the biases, deviance, equivalent number of parameters, AIC/BIC value for a certain slide.

Details

Typical systematic non-biological variations of a two-color cDNA microarray include the dependence of ratio measurements (M) on intensity ($A$), print-tip IDs ($PT$), plate IDs ($PL$) and spatial heterogeneity of the slide (Spatial 2D). The sequential normalization procedure in seqWithinNorm normalizes a slide in a sequential fashion: $A$ -> $PT$ -> $PL$ -> Spatial2D. In each step one kind of variation is targeted for correction, and the user chooses the normalization method as desired. We calculate the AIC/BIC criterion along the normalization steps, but they are not used for selection of models.

References

Y. H. Yang, S. Dudoit, P. Luu, and T. P. Speed (2001). Normalization for cDNA microarray data. In M. L. Bittner, Y. Chen, A. N. Dorsel, and E. R. Dougherty (eds), Microarrays: Optical Technologies and Informatics, Vol. 4266 of Proceedings of SPIE. D. L. Wilson, M. J. Buckley, C. A. Helliwell and I. W. Wilson (2003). New normalization methods for cDNA microarray data. Bioinformatics, Vol. 19, pp. 1325-1332.

See Also

stepWithinNorm, withinNorm, fitWithin, fit2DWithin, calcAIC, calcBIC.

Examples

# Examples use swirl dataset, for description type ? swirl
data(swirl)

# Apply sequential normalization for the first slide
# default: loess(A)->median(PT)->median(PL)-> none (Spatial2D)
## Not run: 
# res.swirl1 <- seqWithinNorm(swirl[,1])
# 
# # normalized data
# norm.swirl <- res.swirl1[[1]]
# 
# # sequential normalization information
# step.swirl <- res.swirl1[[2]]
# 
# ## End(Not run)
# median(A)->median(PT)->median(PL)->none(Spatial2D)
res.swirl <- seqWithinNorm(swirl[,1], A="median",PT="median",PL="median",Spatial2D="none")