\name{pca}
\alias{pca}
\title{Perform principal component analysis}
\usage{pca(object, method=listPcaMethods(), nPcs=2, scale=c("none",
    "pareto", "vector", "uv"), center=TRUE, completeObs=TRUE,
    subset, cv=c("none", "q2"), ...)}
\description{Can be used for computing PCA on a numeric matrix for
visualisation, information extraction and missing value
imputation.}
\details{This method is wrapper function for the following set of pca
methods:

\describe{\item{svd:}{Uses classical \code{prcomp}.See
documentation for \code{\link{svdPca}}.}

\item{nipals:}{An iterative method capable of handling small
amounts of missing values. See documentation for
\code{\link{nipalsPca}}.}

\item{rnipals:}{Same as nipals but implemented in R.}

\item{bpca:}{An iterative method using a Bayesian model to handle
missing values. See documentation for \code{\link{bpca}}.}

\item{ppca:}{An iterative method using a probabilistic model to
handle missing values. See documentation for \code{\link{ppca}}.}

\item{svdImpute:}{Uses expectation maximation to perform SVD PCA
on incomplete data. See documentation for
\code{\link{svdImpute}}.}}

Scaling and centering is part of the PCA model and handled by
\code{\link{prep}}.}
\value{A \code{pcaRes} object.}
\references{Wold, H. (1966) Estimation of principal components and
related models by iterative least squares. In Multivariate
Analysis (Ed., P.R. Krishnaiah), Academic Press, NY, 391-420.

Shigeyuki Oba, Masa-aki Sato, Ichiro Takemasa, Morito Monden,
Ken-ichi Matsubara and Shin Ishii.  A Bayesian missing value
estimation method for gene expression profile
data. \emph{Bioinformatics, 19(16):2088-2096, Nov 2003}.

Troyanskaya O. and Cantor M. and Sherlock G. and Brown P. and
Hastie T. and Tibshirani R. and Botstein D. and Altman RB.  -
Missing value estimation methods for DNA microarrays.
\emph{Bioinformatics. 2001 Jun;17(6):520-5}.}
\seealso{\code{\link{prcomp}}, \code{\link{princomp}},
\code{\link{nipalsPca}}, \code{\link{svdPca}}}
\keyword{multivariate}
\author{Wolfram Stacklies, Henning Redestig}
\arguments{\item{object}{Numerical matrix with (or an object coercible to
such) with samples in rows and variables as columns. Also takes
\code{ExpressionSet} in which case the transposed expression
matrix is used.}
\item{method}{One of the methods reported by \code{pcaMethods()}}
\item{nPcs}{Number of principal components to calculate.}
\item{scale}{Scaling, see \code{\link{prep}}.}
\item{center}{Centering, see \code{\link{prep}}.}
\item{completeObs}{Sets the \code{completeObs} slot on the
resulting \code{pcaRes} object containing the original data with
but with all NAs replaced with the estimates.}
\item{subset}{A subset of variables to use for calculating the
model. Can be column names or indices.}
\item{cv}{character naming a the type of cross-validation
to be performed.}
\item{...}{Arguments to \code{\link{prep}}, the chosen pca
method and \code{\link{Q2}}.}}
\examples{data(iris)
##  Usually some kind of scaling is appropriate
pcIr <- pca(iris[,1:4], method="svd", nPcs=2)
pcIr <- pca(iris[,1:4], method="nipals", nPcs=3, cv="q2")
## Get a short summary on the calculated model
summary(pcIr)
plot(pcIr)
## Scores and loadings plot
slplot(pcIr, sl=as.character(iris[,5]))}