R/load_eupath_go.R

In khughitt/EuPathDB: Provides access to pathogen annotation resources available on EuPathDB databases

#' Shortcut for loading annotation data from a eupathdb-based orgdb.
#'
#' Every time I go to load the annotation data from an orgdb for a parasite, it
#' takes me an annoyingly long time to get the darn flags right.  As a result I
#' wrote this to shortcut that process.  Ideally, one should only need to pass
#' it a species name and get out a nice big table of annotation data.
#'
#' @param query String containing a unique portion of the desired species.
#' @param webservice Which eupath webservice is desired?
#' @param eu_version Gather data from a specific eupathdb version?
#' @param wanted_fields If not provided, this will gather all columns starting
#'   with 'annot'.
#' @param gene_ids Set of gene IDs to download GO data.
#' @param columns Wanted GO columns.
#' @return Big huge data frame of annotation data.
#' @export
load_eupath_go <- function(query, webservice="tritrypdb",
                           eu_version=NULL, wanted_fields=NULL,
                           gene_ids=NULL, columns=c("go", "evidence")) {

  pkg <- NULL
  entry <- NULL
  if ("OrgDb" %in% class(query)) {
    pkg <- query
  } else if ("character" %in% class(query)) {
    entry <- get_eupath_entry(species=query, webservice=webservice)
    pkg_names <- get_eupath_pkgnames(entry=entry, eu_version=eu_version)
    pkg_installedp <- pkg_names[["orgdb_installed"]]
    if (isFALSE(pkg_installedp)) {
      stop("The required package is not installed.")
    }
    pkg <- as.character(pkg_names[["orgdb"]])
  } else if ("data.frame" %in% class(query)) {
    entry <- query
    pkg_names <- get_eupath_pkgnames(entry=entry, eu_version=eu_version)
    pkg_installedp <- pkg_names[["orgdb_installed"]]
    if (isFALSE(pkg_installedp)) {
      stop("The required package is not installed.")
    }
    pkg <- as.character(pkg_names[["orgdb"]])
  }

  org <- load_orgdb_go(orgdb=pkg)
  return(org)
}