SCAN.UPC (version 2.14.0)

SCAN: Single-Channel Array Normalization (SCAN) and Universal exPression Codes (UPC) for Affymetrix microarrays

Description

This function is used to normalize Affymetrix .CEL files via the SCAN and UPC methods.

Usage

SCAN(celFilePattern, outFilePath = NA, convThreshold = 0.01, annotationPackageName = NA, probeSummaryPackage = NA, probeLevelOutDirPath = NA, exonArrayTarget=NA, batchFilePath=NA, verbose = TRUE)
SCANfast(celFilePattern, outFilePath = NA, convThreshold = 0.50, annotationPackageName = NA, probeSummaryPackage = NA, probeLevelOutDirPath = NA, exonArrayTarget=NA, batchFilePath=NA, verbose = TRUE)
UPC(celFilePattern, outFilePath = NA, convThreshold = 0.01, annotationPackageName = NA, probeSummaryPackage = NA, probeLevelOutDirPath = NA, exonArrayTarget = NA, modelType="nn", batchFilePath=NA, verbose = TRUE)
UPCfast(celFilePattern, outFilePath = NA, convThreshold = 0.50, annotationPackageName = NA, probeSummaryPackage = NA, probeLevelOutDirPath = NA, exonArrayTarget = NA, modelType="nn", batchFilePath=NA, verbose = TRUE)

Arguments

celFilePattern

Absolute or relative path to the input file to be processed. This is the only required parameter. To process multiple files, wildcard characters can be used (e.g., "*.CEL"). Alternatively, a Gene Expression Omnibus identifier (e.g., GSE22309 or GSM555237) can be specified.

outFilePath

Absolute or relative path where the output file will be saved. This is optional.

convThreshold

Convergence threshold that determines at what point the mixture-model parameters have stabilized. The default value should be suitable in most cases. However, if the model fails to converge, it may be useful to adjust this value. (This parameter is optional.)

annotationPackageName

The name of an annotation package that specifies the layout and sequences of the probes. This is optional. By default, the correct annotation package should be identified in most cases. However, with this option allows the user to specify the package explicitly if needed.

probeSummaryPackage

An R package that specifies alternative probe/gene mappings. This is optional. See note below for more details.

probeLevelOutDirPath

Absolute or relative path to a directory where probe-level normalized values can be saved. This is optional. By default, the probe-level values will be discarded after they have been summarized. However, if the user has a need to repeatedly process the same file (perhaps to try various probe/gene mappings), this option can be useful because SCAN will retrieve previously normalized values if a probe-level file exists, rather than renormalize the raw data. The user should be aware that probe-level files may consume a considerable amount of disk space.

exonArrayTarget

The type of probes to be used. This parameter is optional and should only be specified when Affymetrix Exon 1.0 ST arrays are being processed. This parameter allows the user to specify the subset of probes that should be used and how the probes should be grouped. Available options are NA, "core", "extended", "full", or "probeset". When "probeset" is used, all probes will be used, and the probes will be grouped according to the Affymetrix probeset definitions. When "core", "extended", or "full" are used, the probes that Affymetrix has defined to fall within each classification will be used, and probes will be grouped by Entrez Gene IDs (as defined in the corresponding annotation package). It is recommended to specify "probeset" when the probeSummaryPackage parameter is being used so that all probes will be considered.

modelType

The type of mixture model to be used. This value can be either "nn" (default) or "nn_bayes."

batchFilePath

Absolute or relative path to a tab-separated text file that indicates batch (and optionally, covariate information) for each sample. Optional.

verbose

Whether to output more detailed status information as files are normalized. Default is TRUE.

Value

An ExpressionSet object that contains a row for each probeset/gene/transcript and a column for each input file. SCAN values will be on a log2 scale, centered at zero. UPC values will range between zero and one (lower values indicate that the gene is inactive and higher values indicating that the gene is active).

References

Piccolo SR, Sun Y, Campbell JD, Lenburg ME, Bild AH, and Johnson WE. A single-sample microarray normalization method to facilitate personalized-medicine workflows. Genomics, 2012, 100:6, pp. 337-344. Piccolo SR, Withers MR, Francis OE, Bild AH and Johnson WE. Multi-platform single-sample estimates of transcriptional activation. Proceedings of the National Academy of Sciences of the United States of America, 2013, 110(44):11778-17783.

Examples

## Not run: 
# # SCAN normalize a CEL file from GEO
# normalized = SCAN("GSM555237")
# 
# # UPC normalize a CEL file from GEO
# normalized = UPC("GSM555237")
# 
# # Normalize a CEL file and save output to a file
# normalized = SCAN("GSM555237", "output_file.txt")
# 
# # Normalize a CEL file and summarize at the gene level using BrainArray
# # mappings for Entrez Gene. First it is necessary to install the package
# # and obtain the package name. For demonstration purposes, this file
# # will be downloaded manually from GEO.
# tmpDir = tempdir()
# getGEOSuppFiles("GSM555237", makeDirectory=FALSE, baseDir=tmpDir)
# celFilePath = file.path(tmpDir, "GSM555237.CEL.gz")
# pkgName = InstallBrainArrayPackage(celFilePath, "17.0.1", "hs", "entrezg")
# normalized = SCAN(celFilePath, probeSummaryPackage=pkgName)
# 
# # Normalize multiple files in parallel on multiple cores within a given
# # computer. It is also possible using the doParallel package to spread
# # the workload across multiple computers on a cluster.
# library(doParallel)
# registerDoParallel(cores=2)
# result = SCAN("GSE22309")
# ## End(Not run)