winsorize: Winsorization of copy number data

In luciansmith/copynumber: Segmentation of single- and multi-track copy number data by penalized least squares regression.

Description

Outliers in copy number data are detected and modified using MAD or PCF Winsorization.

Usage

winsorize(data, pos.unit = "bp", arms = NULL, method = "mad", tau = 2.5, 
          k = 25, gamma = 40, iter = 1, assembly = "hg19", digits = 4, 
          return.outliers = FALSE, save.res = FALSE, file.names = NULL, 
          verbose = TRUE)

Arguments

data	either a data frame or the name of a tab-separated file from which copy number data can be read. The rows of the data frame or file should represent the probes. Column 1 must hold numeric or character chromosome numbers, column 2 the numeric local probe positions, and subsequent column(s) the numeric copy number measurements for one or more samples. The header of copy number columns should give sample IDs.
pos.unit	the unit used to represent the probe positions. Allowed options are "mbp" (mega base pairs), "kbp" (kilo base pairs) or "bp" (base pairs). By default assumed to be "bp".
arms	optional character vector containing chromosome arms (denoted 'p' and 'q') corresponding to the chromosomes and positions found in data. If not specified chromosome arms are found using the built-in genome assembly version determined by assembly.
method	the Winsorization method to be applied, must be one of "mad" (default) or "pcf".
tau	Winsorization threshold, default is 2.5.
k	the half window size to be applied in median filtering, default is 25.
gamma	penalty for each discontinuity in the pcf curve, default is 40. Only applicable when method="pcf".
iter	number of iterations in PCF Winsorization, default is 1.
assembly	a string specifying which genome assembly version should be applied to determine chromosome arms. Allowed options are "hg19", "hg18", "hg17" and "hg16" (corresponding to the four latest human genome annotations in the UCSC genome browser).
digits	the number of decimals to be applied when reporting results. Default is 4.
return.outliers	logical value indicating whether a data frame identifying outliers should be returned, default is FALSE.
save.res	logical value indicating whether results should be saved in text files, default is FALSE.
file.names	optional character vector of length two giving the name of the files where the Winsorized data and outlier statuses, respectively, should be saved if save.res=TRUE.
verbose	logical value indicating whether or not to print a progress message each time Winsorization is finished for a new chromosome arm.

Details

The copy number data are either MAD Winsorized or PCF Winsorized as described in Nilsen and Liestoel et al. (2012). Winsorization is done separately on each chromosome arm in each sample.

Value

If return.outliers = TRUE a list with the following components:

wins.data	a data frame with chromosome numbers in the first column, probe positions in the second and the Winsorized copy number values for the sample(s) in subsequent column(s).
wins.outliers	a data frame with chromosome numbers in the first column, probe positions in the second and outlier statuses for each sample in the subsequent column(s). The values +/- 1 indicate that the observation is an outlier, whereas the value 0 indicates that it is not.

If return.outliers = FALSE only the data frame containing the winsorized data is returned.

If save.res=TRUE the results are saved in text files with names as specified in file.names. If file.names=NULL, a folder named "Wins_res" is created in the working directory and Winsorized data and outlier statuses are saved in this directory in tab-separated files named wins.data.txt and wins.outliers.txt, respectively.

Note

Any missing values in data imply that the Winsorized value and outlier status for this probe will be missing as well. Also, if the number of probes within a chromosome arm is less than 2*k, Winsorization cannot be done and the data values are thus left unchanged.

Author(s)

Gro Nilsen, Knut Liestoel, Ole Christian Lingjaerde

References

Nilsen and Liestoel et al., "Copynumber: Efficient algorithms for single- and multi-track copy number segmentation", BMC Genomics 13:591 (2012), doi:10.1186/1471-2164-13-59

Examples

#Lymphoma data
data(lymphoma)
#Take out a smaller subset of 3 samples (using subsetData):
sub.lymphoma <- subsetData(lymphoma,sample=1:3)

#Do MAD Winsorization:
wins.data <- winsorize(data=sub.lymphoma)
         

luciansmith/copynumber documentation built on May 6, 2019, 2:32 p.m.