ini.call: Informative/Non-Informative Call
In xps: Processing and Analysis of Affymetrix Oligonucleotide Arrays including Exon Arrays, Whole Genome Arrays and Plate Arrays

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

Computes the Informative/Non-Informative Call for the exclusion of non-informative probe sets.

ini.call(xps.data,
         filename   = character(0),
         filedir    = getwd(),
         tmpdir     = "",
         weight     = 0.5,
         mu         = 0.0,
         scale      = 1.0,
         tol        = 0.00001,
         cyc        = 100,
         alpha1     = 0.4,
         alpha2     = 0.6,
         version    = "1.3.1",
         option     = "transcript",
         exonlevel  = "",
         xps.scheme = NULL,
         add.data   = TRUE,
         verbose    = TRUE)

xpsINICall(object, ...)

`xps.data`	object of class `DataTreeSet`.
`filename`	file name of ROOT data file.
`filedir`	system directory where ROOT data file should be stored.
`tmpdir`	optional temporary directory where temporary ROOT files should be stored.
`weight`	hyperparameter, usually set to 0.5 for `version="1.3.1"` and to 8.0 for `version="1.3.0"`.
`mu`	hyperparameter allowing to correct for potential bias.
`scale`	scaling parameter, usually set to 1.0 for `version="1.3.1"` and to 2.0 for `version="1.3.0"`.
`tol`	termination tolerance for EM algorithm.
`cyc`	maximum number of cycles of EM algorithm.
`alpha1`	a significance threshold in (0,alpha2).
`alpha2`	a significance threshold in (alpha1,1.0).
`version`	version of original farms package. Currently, `version="1.3.1"` and `version="1.3.0"` are implemented. Default is `version="1.3.1"`.
`option`	option determining the grouping of probes for summarization, one of ‘transcript’, ‘exon’, ‘probeset’; exon arrays only.
`exonlevel`	exon annotation level determining which probes should be used for summarization; exon/genome arrays only.
`xps.scheme`	optional alternative `SchemeTreeSet`.
`add.data`	logical. If `TRUE` call data will be added to slots `data` and `detcall`.
`verbose`	logical, if `TRUE` print status information.
`object`	object of class `DataTreeSet`.
`...`	the arguments described above.

In contrast to mas5.call this function quantifies the signal-to-noise ratio for each probe set, as described in Talloen et al. Thus, the returned p-values and detection calls have a different meaning:

The p-value that is returned estimates the signal-to-noise ratio (SNR):
P-values (SNR) of less than 0.5 indicate that there is more signal than noise and the corresponding genes are considered to be ‘informative’ for further analysis. In contrast, values greater than 0.5 indicate ‘non-informative’ genes.

The informative call is computed by thresholding the p-value as in:

call “P” if p-value < alpha1
call “M” if alpha1 <= p-value < alpha2
call “A” if alpha2 <= p-value

Here “P” should be considered as informative “I”, “M” as marginally informative, and “A” as non-informative “NI”.

The defaults for alpha1=0.4 and alpha2=0.6 are set to allow “M” calls. In order to get the same results as package ‘farms\_1.3.1’, you need to set alpha1=0.5 and alpha2=0.5.

For exon/genome arrays it is necessary to supply option and exonlevel.

Following options are valid for exon arrays only:

`transcript`:	expression levels are computed for transcript clusters, i.e. probe sets containing the same 'transcript_cluster_id'.
`exon`:	expression levels are computed for exon clusters, i.e. probe sets containing the same 'exon_id', where each exon cluster consists of one or more `probeset`s.
`probeset`:	expression levels are computed for individual probe sets, i.e. for each 'probeset_id'.

Following exonlevel annotations are valid for exon arrays:

	`core`:	probesets supported by RefSeq and full-length GenBank transcripts.
	`metacore`:	core meta-probesets.
	`extended`:	probesets with other cDNA support.
	`metaextended`:	extended meta-probesets.
	`full`:	probesets supported by gene predictions only.
	`metafull`:	full meta-probesets.
	`ambiguous`:	ambiguous probesets only.
	`affx`:	standard AFFX controls.
	`all`:	combination of above.

Following exonlevel annotations are valid for whole genome arrays:

	`core`:	probesets with category 'unique' and 'mixed'.
	`metacore`:	probesets with category 'unique' only.
	`affx`:	standard AFFX controls.
	`all`:	combination of above.

Exon levels can also be combined, with following combinations being most useful:

`exonlevel="metacore+affy"`:	core meta-probesets plus AFFX controls
`exonlevel="core+extended"`:	probesets with cDNA support
`exonlevel="core+extended+full"`:	supported plus predicted probesets

Exon level annotations are described in the Affymetrix whitepaper 'exon_probeset_trans_clust_whitepaper.pdf'.

In order to use an alternative SchemeTreeSet set the corresponding SchemeTreeSet xps.scheme.

xpsINICall is the DataTreeSet method called by function ini.call, containing the same parameters.

A CallTreeSet

Since I/NI-calls distinguish only between informative and non-informative genes, the calls are identical for all samples.

Christian Stratowa

Talloen, W., Clevert D.-A., Hochreiter, S., Amaratunga, D., Bijnens, J., Kass, S., and Gohlmann, H.W.H. (2006), I/NI-calls for the exclusion of non-informative genes: a highly effective filtering tool for microarray data. Bioinformatics 23(21):2897-2902

farms, mas5.call

## first, load ROOT scheme file and ROOT data file
scheme.test3 <- root.scheme(paste(path.package("xps"),"schemes/SchemeTest3.root",sep="/"))
data.test3 <- root.data(scheme.test3, paste(path.package("xps"),"rootdata/DataTest3_cel.root",sep="/"))

## I/NI call
call.ini <- ini.call(data.test3,"tmp_Test3INI",verbose=FALSE)

## get data.frames
snr.ini <- pvalData(call.ini)
inf.ini <- presCall(call.ini)
head(snr.ini)
head(inf.ini)

## plot results
if (interactive()) {
callplot(call.ini)
}

rm(scheme.test3, data.test3)
gc()