\HeaderA{PMIDAmat}{A function to compute the probe to PubMed id incidence matrix.}{PMIDAmat}
\keyword{manip}{PMIDAmat}
\begin{Description}\relax
For a given chip or a given set of genes, it computes the mapping from
probes to PubMed id.
\end{Description}
\begin{Usage}
\begin{verbatim}
PMIDAmat(pkg, gene=NULL)
\end{verbatim}
\end{Usage}
\begin{Arguments}
\begin{ldescription}
\item[\code{pkg}] The package name of the chip for which the incidence matrix should
be computed.
\item[\code{gene}] A character vector of interested probe set ids or NULL (default).
\end{ldescription}
\end{Arguments}
\begin{Details}\relax
Not much to say, just find which probes are associated with which
PubMed ids and return the incidence matrix, with PubMed ids as rows and probes as
columns. 

To specify a set of probes to use, let the argument \code{gene} to be
a vector of probe ids.   Bt this way, the calculations are not
involved with non-interested genes/PubMed ids so that the whole
process could finish soon.
\end{Details}
\begin{Value}
A matrix containing zero or one, depending on whether the probe
(column) is associated with a PubMed id (row).
\end{Value}
\begin{Author}\relax
R. Gentleman
\end{Author}
\begin{Examples}
\begin{ExampleCode}
  if(require("hgu95av2") ) {
    probe <- names(as.list(hgu95av2ACCNUM))
    Amat = PMIDAmat("hgu95av2", gene=sample(probe, 10))
 }
\end{ExampleCode}
\end{Examples}