read.cel.complete {affxparser}R Documentation

Parsing an Affymetrix CEL file

Description

Parsing an Affymetrix CEL file using the Fusion SDK. The intention is to let users obtain all (but possibly a subset of) information in the CEL file.

Usage

read.cel.complete(fname, indices = NULL, read.intensities = TRUE,
                  read.stdvs = TRUE, read.pixels = TRUE,
                  read.xy = FALSE, read.outliers = TRUE,
                  read.masked = TRUE, verbose = 0)

Arguments

fname the name of the CEL file.
indices a vector of indices indicating which features to read. If the argument is NULL all features will be returned.
read.intensities a logical: will the intensities be returned.
read.stdvs a logical: will the standard deviations be returned.
read.pixels a logical: will the number of pixels be returned.
read.xy a logical: will the (x,y) coordinates be returned.
read.outliers a logical: will the outliers be return.
read.masked a logical: will the masked features be returned.
verbose how verbose do we want to be. 0 is no verbosity, higher numbers mean more verbose output. At the moment the values 1 and 2 are supported.

Details

This function reads everything from a CEL file. It is considered a bug if the file contains information not accessible by this function, please report it.

The (input and output) interface makes heavily use of the concept of indices. An (x,y) coordinate is converted to an integer by the formula (y + 1) times numberOfColsInCelFile + (x + 1), assuming that the (x,y) coordinates start at (0,0).

The file returns a list with components (possible empty depending on the input flags) corresponding to the various flags. The outliers and masked components are special in that they are vectors of indices of features flagged as respectively outlier/masked by the Affymetrix image software.

Value

A named list with components described below. The order of the intensities, stdvs and pixels corresponds to the ordering of the indices argument.

header The header of the CEL file. Equivalent to the output from read.cel.header, see the documentation for that function.
intensities A vector of mode nummeric containing the intensity associated with each feature, unless read.intensities = FALSE in which case it will be NULL. Restricted to features within the indices range.
stdvs A vector of mode nummeric containing the standard deviation associated with each feature, unless read.stdvs = FALSE in which case it will be NULL. Restricted to features within the indices range.
pixels A vector of mode integer containing the number of pixels associated with each feature, unless read.pixels = FALSE in which case it will be NULL. Restricted to features within the indices range.
x A vector of mode integer containing the x coordinate associated with each feature, unless read.xy = FALSE in which case it will be NULL. Restricted to features with the indices range.
y A vector of mode integer containing the y coordinate associated with each feature, unless read.xy = FALSE in which case it will be NULL. Restricted to features with the indices range.
outliers A vector of mode integer which the indices of the features flagged as outliers, unless read.outliers = NULL in which case it will be NULL. Note that there is a difference between outliers = NULL and outliers = integer(0) - the last case happens when read.outliers = TRUE but there are no outliers. In case an indices argument is given, only indices within the indices argument are returned.
masked A vector of mode integer which the indices of the features flagged as masked, unless read.masked = NULL in which case it will be NULL. Note that there is a difference between masked = NULL and masked = integer(0) - the last case happens when read.masked = TRUE but there are no masked features. In case an indices argument is given, only indices within the indices argument are returned.

Note

Memory usage: the Fusion SDK allocates memory for the entire CEL file, when the file is accessed (but does not actually read the file into memory). Using the indices argument will therefore only affect the memory use of the final object (as well as speed), not the memory allocated in the C function used to parse the file. This should be a minor problem however.

The outlier/masked flag are computed by the Affymetrix image analysis software. The current community view seems to be that this should be done based on statistical modelling of the actual probe intensities and should be based on the choice of preprocessing algorithm. Mopst algorithms are only using the intensities from the CEL file.

Author(s)

James Bullard, bullard@stat.berkeley.edu and Kasper Daniel Hansen, khansen@stat.berkeley.edu

See Also

affxparserInfo for a package overview and read.cel.header for a description of the header output. Often a user only wants to read the intensities, look at read.cel.intensities for a function specialized for that use. See xy2indices from affy and xy2i from makePlatformDesign for a discussion of the index approach.

Examples

## Not run: 
read.cel.complete("~/testFile.CEL", read.npixels = FALSE)
## End(Not run)

[Package affxparser version 1.2.0 Index]