twocols2tworows: Function to reshape a data frame or file

In PBR/PediHaplotyper: PediHaplotyper assigns haploblock alleles based on phased marker data in pedigrees

Description

This is a function to help in formatting marker data produced by PediHaplotyper for use by other software. It takes a data frame or file with two consecutive columns per marker and one row per individual and converts this to two rows per individual and one column per marker. It is more user-friendly than reshape() for this particular case.

Usage

twocols2tworows(source, target = "", sep = "", skip = 0, 
  na.strings = "NA", sep.out = "\t", na.out = "", ...)

Arguments

source	Either a data frame or a file name: the data to be transposed. If a file name, the file should be readable by read.table() with header=TRUE; all other parameters for read.table have the default values of read.table but can be specified.
target	Ignored if source is a data frame, else the name of the transposed file
sep	Ignored if source is a data frame, else the separator (passed to read.table)
skip	Ignored if source is a data frame, else the number of lines to skip (passed to read.table)
na.strings	Ignored if source is a data frame, else the strings used to represent NA values (passed to read.table)
sep.out	Ignored if source is a data frame, else the separator to be used in the target file
na.out	Ignored if source is a data frame, else the representation of NA values to be used in the target file
...	Ignored if source is a data frame, else further parameters to be passed to read.table

Details

The source data consist of one column with names of individuals, followed by two consecutive columns per marker that contain the marker alleles inherited from parent 1 and parent 2. The name of the first column for each marker is the marker name, the name of the second column is ignored, as are the row names. This is the format used by PediHaplotyper for its input and output marker data. This function reformats the data such that there is only one column per marker and two consecutive rows per individual.

If source is a file name the file is read using read.table; it must contain a header line, but in contrast to read.table the headers are not modified if they are not valid identifiers or if duplicate headers occur. The column names of the original and rehaped data frame may be invalid as identifiers; that does not affect any of the PediHaplotyper functions.

Value

The return value is always the reshaped data frame. If source and target are file names also the target file is created, overwriting any previous file.

Examples

data(mrkdat) # a data frame with data for two markers and two individuals
s <- transpose(mrkdat)
s # a data frame arranged as the source of twocols2tworows
twocols2tworows(s)

PBR/PediHaplotyper documentation built on Feb. 3, 2021, 12:03 a.m.