doFIRMA: Finding Isoforms using Robust Multichip Analysis (FIRMA)

doFIRMAR Documentation

Finding Isoforms using Robust Multichip Analysis (FIRMA)

Description

Finding Isoforms using Robust Multichip Analysis (FIRMA) based on [1].

Usage

  ## S3 method for class 'AffymetrixCelSet'
doFIRMA(csR, ..., flavor=c("v1b", "v1a"), drop=TRUE, verbose=FALSE)
  ## Default S3 method:
doFIRMA(dataSet, ..., verbose=FALSE)

Arguments

csR, dataSet

An AffymetrixCelSet (or the name of an AffymetrixCelSet).

...

Additional arguments passed to FirmaModel, and to set up AffymetrixCelSet (when argument dataSet is specified).

flavor

A character string specifying the flavor of FIRMA to use.

drop

If TRUE, the FIRMA scores are returned, otherwise a named list of all intermediate and final results.

verbose

See Verbose.

Value

Returns a named list, iff drop == FALSE, otherwise only FirmaSet object (containing the FIRMA scores).

Using a custom exon-by-transcript CDF

It is strongly recommended to use a custom CDF, e.g. "core", "extended" or "full" [1]. To use a custom CDF, set it before calling this method, i.e. setCdf(csR, cdf). Do not set the standard "non-supported" Affymetrix CDF (see also Section 'Flavors').

Flavors

If flavor == "v1b" (default), then the standard "non-supported" Affymetrix CDF is used for background correction and the quantile normalization steps, and the custom CDF is used for the probe summarization and everything that follows. The advantage of this flavor is that those two first preprocessing steps will remain the same if one later changes to a different custom CDF.

If flavor == "v1a", then the custom CDF is used throughout all steps of FIRMA, which means that if one changes the custom CDF all steps will be redone.

Author(s)

Henrik Bengtsson

References

[1] E. Purdom, K. Simpson, M. Robinson, J. Conboy, A. Lapuk & T.P. Speed, FIRMA: a method for detection of alternative splicing from exon array data, Bioinformatics, 2008.


HenrikBengtsson/aroma.affymetrix documentation built on Feb. 20, 2024, 9:07 p.m.