aCGH.read.Sprocs {aCGH}R Documentation

Create object of class "aCGH" from Sproc files

Description

This function reads in two-channel Array Complimentary Genomic Habrydization Sproc files, flags them for bad quality and missing data, and creates object of class aCGH.

Usage

aCGH.read.Sprocs(fnames, latest.mapping.file = NULL, maxsd = 0.2,
                 minreplic = 2, chrom.remove.threshold = 24,
                 prop.missing = 0.25, sample.names = fnames,
                 sample.quality.threshold = 0.4,
                 cols = c("Log2Rat", "Log2StdDev", "NReplic", "Bad.P"), unmapScreen=TRUE, dupRemove = TRUE)

Arguments

fnames a vector of character strings containing the file names of each Sproc data file.
latest.mapping.file The name of an optional file that contains newer clone mapping different from the clone mapping used at the time when the arrays were created.
maxsd maximum of standard deviation of log2 ratios used in pre-filtering.
minreplic minimum number of replicates per clone for a single chip used to calculate the log2 ratios.
chrom.remove.threshold Chromosomes are ordered and numbered as usual, except for X and Y chromosome, which in for Homo sapiens genome have numbers 23 and 24 repsectivelly, in for Mus musculus 20 and 21, etc.
prop.missing Clones are screened out and if the proportion missing in the samples is prop.missing they are removed.
sample.names Sample names. If they are missing, the file names are used after stripping the characters after the last dot in the filename if one exists; for example 'myfile.txt' becomes myfile.
sample.quality.threshold Mark those samples that have their proportion of missing values sample.quality.threshold.
cols character vector of length 4 containing the following Sproc file column names: log2 ratios, std. deviations of the log2 ratios, number of replicates for each clone and flags for bad clones. Defaults to c("Log2Rat", "Log2StdDev", "NReplic", "Bad.P"). Note that all the whitespace characters in the column names will be replaced with dots.
unmapScreen Indicator for whether clones with incomplete mapping information should be removed from the dataset. Note that leaving them in may cause plotting routines fail. Defaults to TRUE
dupRemove Indicator for whether clones with duplicate names should be averaged and removed from the dataset leaving only one occurence of each duplicated set.Defaults to TRUE

Value

Object of class aCGH.

Author(s)

Jane Fridlyand, Peter Dimitrov

See Also

aCGH

Examples


datadir <- system.file("examples", package = "aCGH")
latest.mapping.file <-
      file.path(datadir, "human.clones.info.Jul03.txt")
ex.acgh <-
        aCGH.read.Sprocs(dir(path = datadir,
                        pattern = paste("*", "*sproc*", sep = "\."),
                        full.names = TRUE), latest.mapping.file,
                        chrom.remove.threshold = 23)
#    aCGH.read.Sprocs(dir(path = datadir,
#                     pattern = paste("*", "txt", sep = "\."),
#                     full.names = TRUE), latest.mapping.file,
#                     chrom.remove.threshold = 23)
ex.acgh

## Testing if creating the object went right. Should all be true.

all(log2.ratios(ex.acgh)[ 1, ] == c(-0.077698 , 0.007389))
clone.name <- "RP11-88I12"
all(log2.ratios(ex.acgh)[ clone.name, ] == c(0.096387 , 0.050999))


[Package aCGH version 1.1.4 Index]