R/getKdModels.R

In ETHZ-INS/scanMiRData: miRNA Affinity models for the scanMiR package

#' getKdModels
#'
#' Returns a collection of miRNA KdModels for the requested species.
#' \emph{Note that by default, low-confidence miRNAs are not returned}
#' (see `categories`).
#'
#' @param species Either "hsa" (human), "mmu" (mouse) or "rno" (rat)
#' @param categories The categories of miRNAs to return. Any
#' combination of "Low-confidence", "Poorly conserved",
#' "Conserved across mammals", "Conserved across vertebrates". If NULL
#' (default), all categories are returned. miRNAs with NA as
#' conservation are not returned  unless `categories==NULL`.
#'
#' @return An object of class \code{\link[scanMiR]{KdModelList}}
#' @export
#' @import scanMiR
#' @importFrom utils data
#' @examples
#' mods <- getKdModels("rno")
#' summary(mods)
getKdModels <- function( species=c("hsa","mmu","rno"), categories=NULL){
  cats <- c("Low-confidence","Poorly conserved","Conserved across mammals",
            "Conserved across vertebrates")
  if(!is.null(categories))
    categories <- match.arg(categories, cats, several.ok=TRUE)
  species <- match.arg(species)
  mods <- get(data(list=species, package="scanMiRData", envir = environment()))
  if(is.null(categories)) return(KdModelList(mods))
  con <- conservation(mods)
  if(all(is.na(con))){
    warning("The collection has no conservation information; the `categories`",
            " argument will be ignored.")
    return(mods)
  }
  KdModelList(mods[con %in% categories])
}