read.ilmn.idat: Read Illumina gene expression iDAT files.
In drmjc/lumidat: Methods for importing Illumina gene expression iDAT files

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

This function can decrypt Illumina gene expression iDAT files (aka version 1 iDAT files). It will temporarily create GenomeStudio-compatible output files using preprocess.illumina.idat, and then run read.ilmn.

  read.ilmn.idat(files = NULL, path = NULL, zipfile = NULL,
    clmfile = NULL,
    probeID = c("ProbeID", "ArrayAddressID", "Sequence", "NuID"),
    manifestfile = NULL,
    collapseMode = c("none", "max", "median", "mean"),
    backgroundCorrect = FALSE, controls = TRUE,
    expr = "AVG_Signal", other.columns = "Detection",
    verbose = FALSE, memory = "-Xmx1024m", ...)

`controls`	logical: if `TRUE` (the default), gene-probes and the control-probes will be imported; if `FALSE`, only the gene-probes will be imported. See `read.ilmn` for more details.
`expr`	character string giving the keyword in the names of the expression intensity columns. default=`AVG_Signal`; other options are “MIN_Signal”, “MAX_Signal”. See `read.ilmn` for more details.
`other.columns`	character vector giving the keywords in the names of extra columns required, such as “Detection”, “Avg_NBEADS”, “BEAD_STDEV” etc. Each keyword corresponds to one type of column. See `read.ilmn` for more details.
`...`	any other parameters are passed on to `read.columns`.
`files`	A character vector of at least one file name
`path`	The path to the directories where the files are. This defaults to the current working directory.
`zipfile`	the path to a zip file containing idat files. This over-rides the `files` argument.
`clmfile`	`NULL`, or the path to a GenePattern-CLM file [5]. This represents a mechanism for renaming, and reordering the samples in the resulting object. Column 1 is the idat name (not path), Column 2 is the desired sample name, column 3 is the biological class of each sample. The order of the rows specifies the order of the samples in the result.
`probeID`	This controls which value to identify each probe by. Allowable values are “ArrayAddressID”, “ProbeID”, “Sequence”, “NuID”.
`manifestfile`	The full path to the Array manifest file in TXT format. See `list_illumina_manifest_files` and then `download_illumina_manifest_file` to download manifest files directly from Illumina.
`collapseMode`	Collapse probes to genes using the specified mode. Valid values are “none” (the default), “max”, “median” and “mean” (the GenomeStudio default).
`backgroundCorrect`	logical, if TRUE then peform background correction on only the gene-level probes; if FALSE, do no correction.
`verbose`	logical, if `TRUE`, print informative messages
`memory`	The maximum Java memory heapsize. eg: "-Xmx1024m", or "-Xmx4g", which reserves 1GB or 4G of RAM, respectively. default="-Xmx1024m"

limma::read.ilmn allows you control over importing the control data via the ctrlpath parameter. In read.ilmn.idat, you use the controls parameter: if controls=TRUE, then both the gene-probes, and the control-probes will be imported into an EListRaw-class object.

See preprocess.illumina.idat for more details on Illumina arrays, manifest files, probeID naming options, collapsing probes to genes, and background correcting. Note that limma provides neqc for doing background correction (so we usually set backgroundCorrect=FALSE). Also, we usually prefer to collapse from probes to genes after: normalizing, and filtering on probe quality, and expression level, thus we usually select collapseMode="none". See read.ilmn for more details on expr, other.columns parameters. We do not allow access to the annotation parameter, as lumidat doesn't add any additional columns which could be used for annotating probes (ie, not the “SYMBOL” parameter as listed in ?read.ilmn)

An EListRaw-class object with the following components:

`E`	numeric matrix of raw intensities.
`genes`	data.frame of probe annotation.
`targets`	data.frame of sample information.
`other`	list of other column data.

Mark Cowley m.cowley@garvan.org.au

read.ilmn, preprocess.illumina.idat

# iDAT files+path as input
path <- system.file("extdata", package="lumidat")
files <- c("5356583020_A_Grn.idat", "5356583020_B_Grn.idat")
manifestfile <- system.file("extdata", "HumanHT-12_V3_0_R1_99999999.txt", package="lumidat")
res <- read.ilmn.idat(files, path, manifestfile=manifestfile, probeID="NuID", controls=TRUE)
res

# zipfile as input
path <- system.file("extdata", package="lumidat")
files <- c("5356583020_A_Grn.idat", "5356583020_B_Grn.idat")
manifestfile <- system.file("extdata", "HumanHT-12_V3_0_R1_99999999.txt", package="lumidat")
zipfile <- tempfile(fileext=".zip")
zip(zipfile, file.path(path, files), flags="-r9Xq")
res <- read.ilmn.idat(zipfile=zipfile, manifestfile=manifestfile, probeID="ProbeID", controls=FALSE)
res

# clm file
clmfile <- system.file("extdata", "5356583020.clm", package="lumidat")
res <- read.ilmn.idat(zipfile=zipfile, manifestfile=manifestfile, probeID="ProbeID", controls=FALSE, clmfile=clmfile)
res$targets
#    SampleNames
# 1            A
# 2            B

## Not run: 
# Get the Human HT12 V4 manifest file:
manifestfile <- download_illumina_manifest_file("HumanHT-12_V4_0_R2_15002873_B", "txt")

## End(Not run)