\name{quantileNormalize} \alias{quantileNormalize} %- Also NEED an '\alias' for EACH other topic documented here. \title{ Quantile normalization } \description{ Apply quantile normalization to multiple bins of data, divided by a sliding window approach } \usage{ quantileNormalize(x, y, ...) } %- maybe also 'usage' for other objects documented here. \arguments{ \item{x}{ the vector of numerical data to be normalized. If \code{x} is a matrix it is interpreted as a vector. \code{x} can also be of class \code{"ExpressionSet"}. } \item{y}{ an additional vector of numerical data to be used for binning. If \code{y} is a matrix it is interpreted as a vector. \code{y} can also be of class \code{"ExpressionSet"}. } \item{\dots}{ Arguments to be passed to methods (see \code{\link{quantileNormalize-methods}}): \describe{ \item{\option{element}}{ which element of \code{AssayData} to use for a given \code{ExpressionSet} input (default is \option{"exprs"}) } \item{\option{sample}}{ which element of \code{sampleNames} to use as data (default is 1). Can be a character matching a sample name or simply an integer indicating which sample to choose. See \code{\link{getSamples}}. } \item{\option{feature}}{ which element of \code{featureData} to use as binning variable (default is 1). Can be a character matching \code{varLabel} or simply an integer indicating which feature to choose. See \code{\link{getFeatures}}. } \item{\option{num.bins}}{ number of bins (default is 10) used to divide the data } \item{\option{num.steps}}{ number of steps (default is 3) used to create bin offsets, resulting in bins of sliding windows } \item{\option{mode}}{ the binning mode to be used. This must be either \option{"continuous"} (default) or \option{"discrete"}. \option{"continuous"} mode will divide the data into density-dependent bins. \option{"discrete"} mode will divide the data uniformly by binning data values. } \item{\option{type}}{ an integer between 1 and 9 (default is 7) selecting one of the nine quantile algorithms: see \code{\link{quantile}}. } \item{\option{na.rm}}{ logical; if \option{TRUE}, missing values are removed from x and y. If \option{FALSE} any missing values cause an error. } \item{\option{\dots}}{ other arguments to be passed to \code{quantile}. See \code{\link{quantile}}.} } } } \value{ Returns a vector of normalized numerical data according to input parameters. } \author{ Reid F. Thompson (\email{rthompso@aecom.yu.edu}) } \seealso{ \code{\link{quantileNormalize-methods}}, \code{\link{quantile}} } \examples{ #demo(pipeline,package="HELP") x <- rep(1:100,10)+10*rep(1:10,each=100) y <- rep(1:20,each=50) d <- density(quantileNormalize(x,y,num.bins=20,num.steps=1,mode="discrete")) plot(density(x)) lines(d$x,d$y/3,col="red") #rm(x,y,d) } % Add one or more standard keywords, see file 'KEYWORDS' in the % R documentation directory. \keyword{ array }