\name{triples.posterior}
\alias{triples.posterior}
\alias{print.triples}

\title{Infers a phenotypic hierarchy from triples}
\description{
  Function \code{triples.posterior} estimates the hierarchy triple-wise. In each step only a triple of nodes
  is involved and no exhaustive enumeration of model space is needed as in function \code{score}.
}
\usage{
triples.posterior(D, type="mLL",para=NULL, hyperpara=NULL,Pe=NULL,Pmlocal=NULL,Pm=NULL,lambda=0,delta=1, triples.thrsh=.5,verbose=TRUE)

\method{print}{triples}(x,...) 
}
%- maybe also 'usage' for other objects documented here.
\arguments{
  \item{D}{data matrix. Columns correspond to the nodes in the silencing scheme. Rows are phenotypes.}   
\item{type}{see \code{nem}}
  \item{para}{vector with parameters a and b for "mLL", if count matrices are used}
  \item{hyperpara}{vector with hyperparameters a0, b0, a1, b1 for "FULLmLL"}
  \item{Pe}{prior position of effect reporters. Default: uniform over nodes in hierarchy}
  \item{Pmlocal}{local model prior for the four models tested at each node: a vector of length 4 with positive entries summing to one}
  \item{triples.thrsh}{threshold used when combining tripel models for each edge. Default: only edges appearing in more than half of triples are included in the final graph.}
  \item{Pm}{prior on model graph (n x n matrix) with entries 0 <= priorPhi[i,j] <= 1 describing the probability of an edge between gene i and gene j.}
  \item{lambda}{regularization parameter to incorporate prior assumptions.}  
  \item{delta}{regularization parameter for automated E-gene subset selection (CONTmLLRatio only)}
  \item{verbose}{do you want to see progress statements printed or not? Default: TRUE}  

  \item{x}{nem object}
  \item{...}{other arguments to pass}
}

\details{
  \code{triples.posterior} is an alternative to exhaustive search 
  by the function \code{score} and more accurate than \code{pairwise.posterior}. 
  For each triple of perturbed genes 
  it chooses between the 29 possible models. It then uses model averaging to combine the triple-models into a final graph.
  
  
  \code{print.triples} gives an overview over the 'triples' object.
}

\value{
nem object
}

\references{
  Markowetz F, Kostka D, Troyanskaya OG, Spang R:
  Nested effects models for high-dimensional phenotyping screens.
  Bioinformatics. 2007; 23(13):i305-12.
}

\author{Florian Markowetz <URL: http://genomics.princeton.edu/~florian>}


\seealso{\code{\link{score}}, \code{\link{nem}}}
\examples{
   data("BoutrosRNAi2002") 
   res <- nem(BoutrosRNAiDiscrete[,9:16],para=c(.13,.05),inference="triples")
   
   # plot graph
   plot(res,what="graph")
   
   # plot posterior over effect positions
   plot(res,what="pos")
   
   # estimate of effect positions
   res$mappos
   
}
\keyword{graphs}