fitMultiscanAffine.RGData: Fits an affine model to signals from a multi-scanned slide
In HenrikBengtsson/aroma: An R Object-oriented Microarray Analysis package

Description Usage Arguments Details Value Author(s) References See Also

Fits an affine model to signals from a multi-scanned slide.

This function is used internally by *calibrateMultiscan() and is not intended to be used by the end-user. Developers of new algorithms may use it, but there is no guarantee that it will exist in the future.

1
2

## S3 method for class 'RGData'
fitMultiscanAffine(this, slides=NULL, channels=c("R", "G"), aShift=NULL, method="L1", maxIter=30, reps=0.02, acc=1e-04, satSignal=2^16 - 1, bootstrap=FALSE, R=NULL, verbose=FALSE, ...)

`slides`	Slides to be used in the fit and that are to be calibrated. If `NULL`, all slides are considered.
`channels`	Specifies which channels to be fitted. If `NULL`, all channels are considered.
`method`	`character` specifying method for IWPCA fitting. If `"L1"`, the distances are minimised in L_1. No other methods are available.
`maxIter`	The maximum number of re-weigthed iterations.
`acc`	The accuracy for the stopping criterion.
`satSignal`	Signals equal to or above this threshold will not be used in the fitting.
`bootstrap`	If `TRUE`, bootstrap is applied to estimate the mean and standard deviation of the bootstraped estimates.
`R`	The number of bootstrap samples. Only effective if `bootstrap==TRUE`.
`reps`	Small value, which is added to the absolute value of the residuals to avoid "1/0" weights, i.e. "(1/0+reps)".
`verbose`	If `TRUE`, extra information is printed while running.
`...`	Additional arguments accepted by `iwpca`. For instance, a N `vector` of weights for each observation may be given, otherwise they get the same weight.

Fitting is done by iterated re-weighted principal component analysis (IWPCA).

Returns a list of fitted AffineModelFit objects for each channel.

Henrik Bengtsson (http://www.braju.com/R/)

[1] H. Bengtsson, J. Vallon-Christersson and G. JÃ¶nsson, Calibration and assessment of channel-specific biases in microarray data with extended dynamical range, BMC Bioinformatics, 5:177, 2004.

To calibrate the data using these fits see *calibrateMultiscan(). For a similar strategy for between channel normalization see *normalizeAffine(). For more information see RGData.

HenrikBengtsson/aroma documentation built on May 7, 2019, 12:56 a.m.