\name{buildEntrezGeneDb}
\alias{getGeneStuff}
\alias{buildEntrezGeneDb}
\alias{getEntrezGenesFromTaxId}


\title{builds an Entrez Gene centric organism database}

\description{
  These functions contact the NCBI eutils web services and then
  construct an organism (org.Xx.eg.db) database.  
}

\usage{
  getGeneStuff(entrezGenes, dir="files")
  buildEntrezGeneDb(entrezGenes, file="test.sqlite")
  getEntrezGenesFromTaxId(taxId)
}

\arguments{
  \item{entrezGenes}{
    character vector of entrez Gene IDs that you want annotated.  We
    presently check to make sure these are from the same species.  You
    will get an error if they aren't.
  }
  \item{file}{
    file to write the database to.  Traditionally these end with the
    .sqlite extension.
  }
  \item{dir}{
    directory for storing parsed XML.
  }  
  \item{taxId}{
    A taxonomy ID from NCBI.  You can browse these to select the species
    of interest from their website.
  }
}

\value{
  In the case of \code{getEntrezGenesFromTaxId}, this will be a
  character vector of entrez gene IDs.  In the case of
  \code{buildEntrezGeneDb}, then you will produce a SQLite database.  In
  the case of \code{getGeneStuff}, then you get back a list of named lists
  containing parsed XML values for each 800 long chunk of entrez Gene IDs.
}

\examples{
  ## seems that some of the EGs are not from the taxid specified? (more tests)
  ## EGs <- getEntrezGenesFromTaxId(taxId)
  ## temporarily, lets use known EGs to run tests
  #library(org.Hs.eg.db)
  #EGs <- Lkeys(org.Hs.egCHR)
  #EGs <- EGs[c(1:10)] ## Obviously NOT a typical step.
  ## XML is broken, so I need a workaround for this.
  #system.time(buildEntrezGeneDb(EGs,file=tempfile()))

  #debug(getGeneStuff)
  #buildEntrezGeneDb(EGs,file=tempfile())
}

\keyword{interface}