ADaCGH2 (version 2.12.0)

cutFile: Cut a file and return individual columns

Description

A text file is split by individual columns, creating as many files as individual columns. Splitting is used using multiple cores, if available. Specified columns are renamed as ID.txt, Chrom.txt, Pos.txt, and other columns can be deleted. The individual files can then all be read from a given directory with inputToADaCGH.

Usage

cutFile(filename, id.col, chrom.col, pos.col, sep = "\t", cols = NULL,  mc.cores = detectCores(), delete.columns = NULL, fork = FALSE)

Arguments

filename

Name of the text file with the data. The file contains at least a column for probe ID, a column for Chromosome, and a column for position, and one or more for data. The column is expected to have a first row with an identifier. All columns are to be separated by a single separating ccharacter colsep.

id.col

The number of the column (starting from 1) that should be used as ID. This is the name before any columns are possibly deleted (see argument delete.columns).

chrom.col

The number of the column (starting from 1) that should be used as the identifier for Chromosome. This is the name before any columns are possibly deleted (see argument delete.columns).

pos.col

The number of the column (starting from 1) that should be used as the position (the coordinates). This is the name before any columns are possibly deleted (see argument delete.columns).

cols

The number of columns of the file. If not specified, we try to guess it. But guessing can be dangerous.

sep

The field or column separator, similar to sep in read.table, etc. The default is a tab. A space can be specified as 'sep = " "'. If you use the default in, say, read.table, which is 'sep = ""', then multiple consecutive tab or space field separators are taken as one ---this is also the behavior in read.table or awk, which is what we use.

mc.cores

The number of processes to launch simultaneously.

delete.columns

The number of the columns (starting from 1) that you do not want to preserve. You probably do not want to have too many of these, and if you do you should note that the file is cut into pieces BEFORE dealing with the unwanted columns so many unwanted columns will mean that we are doing many unwanted calls to cut.

fork

Should we fork R processes, via mclapply, or just send several system commands from this R process. The default (FALSE) is probably the most reasonable option for large files.

Value

This function is used for it main effect: cutting a file into individual one-column files. These files are names "col_1.txt", "col_2.txt", etc, and there are three called "ID.txt", "Chrom.txt", "Pos.txt". The files are created in the current working directory.As we move the files corresponding to "ID", "Chrom", and "Position", the stdout output is shown to the user (to check that things worked).After calling this function, you can call inputToADaCGH.

Details

This function is unlikely to work under Windows unless MinGW or similar are installed (and even then it might not work). This function should work under Mac OS, and it does in the machines we've tried it, but it seems not to work on the BioC testing machine.

This function basically calls "head" and "awk" using system, and trying to divide all the jobs into as many cores as you specify (argument cores).

Examples

## Read a tab separated file, and assign the first,
## second, and third positions to ID, Chrom, and Position

if( (.Platform$OS.type == "unix") && (Sys.info()['sysname'] != "Darwin") ) {
## This will not work in Windows, and might, or might not, work under Mac

fnametxt <- list.files(path = system.file("data", package = "ADaCGH2"),
                         full.names = TRUE, pattern = "inputEx.txt")
cutFile(fnametxt, 1, 2, 3, sep = "\t")

## Verify we have ID, Chrom, Pos
c("ID.txt", "Chrom.txt", "Pos.txt") %in% list.files(getwd())

## verify some other column

c("col_5.txt") %in% list.files(getwd())

## Read a white space separated file, and assign the first, second, and
## third positions to ID, Chrom, and Position, but remove the fifth
## column

fnametxt2 <- list.files(path = system.file("data", package = "ADaCGH2"),
                         full.names = TRUE, pattern = "inputEx-sp.txt")
cutFile(fnametxt2, 1, 2, 3, sep = " ", delete.columns = 5)

}