Preprocess Illumina arrays into a usable form. A very fancy wrapper.

ProcessIlluminaArray(idatFiles, contrast, sampleTable = NULL,
  idatDir = NULL, geneset.filter = NULL, min.fold.change = 1,
  min.intensity = 5.3, db.probequality = NULL)

Arguments

idatFiles

character vector of .idat files with full path

contrast

limma contrast to be taken of sample conditions with three args: c(COLUMN, CONDITION1, CONDITION2) such that the contrast will be "CONDITION1 - CONDITION2" selected from the given COLUMN

sampleTable

data.frame with info about samples (rows) with a header row and columns as follows: Filename : filename (basename) of sample (A.idat, B.idat, etc.) SampleID : full name of sample [Condition][1..N] : 1 through N columns of conditions (subset, time, etc.) to be used with contrast argument these can be named in the header row

idatDir

directory containing .idat files; if this is specified and sampleTable is not specified then a .txt of sampletable describing the samples must be in the same directory

Value

Returns a list with data that is read in, fitted data, a matrix of expression values, and the analysis output from limma::toptable

Examples

idatFiles <- list.files(d1, pattern = ".idat", full.names = TRUE)
contrast <- "ZEB2KO - WT"
sampleTable <- data.frame(Filename = c("A.idat", "B.idat"),
                          SampleID = c("001", "002"),
                          Condition = c("KO", "WT"))
ProcessIlluminaArray(idatFiles, sampleTable = sampleTable, contrast = c("Condition", "KO", "WT"),
                     geneset.filter = c("Tbx21", "Bcl6", other.genes),
                     db.probequality = illuminaMousev2PROBEQUALITY)