R/accessClass_pcOut.R

In gabrielodom/pathwayPCA: Integrative Pathway Analysis with Modern PCA Methodology and Gene Selection

#' Extract PCs and Loadings from a \code{superpcOut}- or \code{aespcOut}-class
#'    Object.
#'
#' @description Given an object of class \code{aespcOut} or \code{superpcOut},
#'    as returned by the functions \code{\link{AESPCA_pVals}} or
#'    \code{\link{SuperPCA_pVals}}, respectively, and the name or unique ID of
#'    a pathway, return a data frame of the principal components and a data
#'    frame of the loading vectors corresponding to that pathway.
#'
#' @details Match the supplied pathway character string to either the
#'    \code{pathways} or \code{terms} columns of the \code{pVals_df} data frame
#'    within the \code{pcOut} object. Then, subset the \code{loadings_ls} and
#'    \code{PCs_ls} lists for their entries which match the supplied pathway.
#'    Finally, return a list of the PCs, loadings, and the pathway ID and name.
#'
#' @param pcOut An object of classes \code{superpcOut} or \code{aespcOut} as
#'    returned by the \code{\link{SuperPCA_pVals}} or \code{\link{AESPCA_pVals}}
#'    functions, respectively.
#' @param pathway_char A character string of the name or unique identifier of a
#'    pathway
#' @param ... Dots for additional arguments (currently unused).
#'
#' @return A list of four elements:
#'    \itemize{
#'      \item{\code{PCs} : }{A data frame of the principal components}
#'      \item{\code{Loadings} : }{A matrix of the loading vectors with features
#'        in the row names}
#'      \item{\code{pathway} : }{The unique pathway identifier for the
#'        \code{pcOut} object}
#'      \item{\code{term} : }{The name of the pathway}
#'    }
#'
#' @examples
#'
#'   ###  Load Data  ###
#'   data("colonSurv_df")
#'   data("colon_pathwayCollection")
#'
#'   ###  Create -Omics Container  ###
#'   colon_Omics <- CreateOmics(
#'     assayData_df = colonSurv_df[, -(2:3)],
#'     pathwayCollection_ls = colon_pathwayCollection,
#'     response = colonSurv_df[, 1:3],
#'     respType = "survival"
#'   )
#'
#'   ###  Calculate Supervised PCA Pathway p-Values  ###
#'   colon_superpc <- SuperPCA_pVals(
#'     colon_Omics,
#'     numPCs = 2,
#'     parallel = TRUE,
#'     numCores = 2,
#'     adjustment = "BH"
#'   )
#'
#'   ###  Extract PCs and Loadings  ###
#'   getPathPCLs(
#'     colon_superpc,
#'     "KEGG_PENTOSE_PHOSPHATE_PATHWAY"
#'   )
#'
#'
#' @rdname getPathPCLs
#' @export getPathPCLs
getPathPCLs <- function(pcOut, pathway_char, ...){
  UseMethod("getPathPCLs")
}

#' @return \code{NULL}
#'
#' @rdname getPathPCLs
#' @method getPathPCLs superpcOut
#' @export
getPathPCLs.superpcOut <- function(pcOut, pathway_char, ...){
  # browser()

  pVals_df <- pcOut$pVals_df
  ###  Check for Matches  ###
  pathID_idx <- which(pVals_df$pathways %in% pathway_char)
  term_idx   <- which(pVals_df$terms %in% pathway_char)
  existDescr <- "description" %in% colnames(pVals_df)
  if(length(pathID_idx) + length(term_idx) == 0){
    stop("Supplied pathway does not match any pathway in the supplied object.")
  }

  ###  Match pathway ID to pathway name  ###
  if(length(term_idx) == 1){

    pathID   <- as.character(pVals_df[term_idx, "pathways"])
    pathway  <- pathway_char
    descript <- ifelse(
      existDescr,
      as.character(pVals_df[term_idx, "description"]),
      NA_character_
    )

  } else if(length(pathID_idx) == 1) {

    pathID   <- pathway_char
    pathway  <- as.character(pVals_df[pathID_idx, "terms"])
    descript <- ifelse(
      existDescr,
      as.character(pVals_df[pathID_idx, "description"]),
      NA_character_
    )

  } else {
    stop("Multiple pathways are not allowed.")
  }

  ###  Select the PCs and Loadings that Match this Pathway ID  ###
  tidyClasses <- c("tbl_df", "tbl", "data.frame")

  PCs_df <- data.frame(
    sampleID = attr(pcOut$PCs_ls, "sampleIDs"),
    pcOut$PCs_ls[[pathID]],
    stringsAsFactors = FALSE
  )

  Load_mat <- pcOut$loadings_ls[[pathID]]
  Load_df  <- data.frame(
    featureID = rownames(Load_mat),
    Load_mat,
    row.names = NULL,
    stringsAsFactors = FALSE
  )

  class(PCs_df) <- class(Load_df) <- tidyClasses

  ###  Return  ###
  list(
    PCs = PCs_df,
    Loadings = Load_df,
    pathway = pathID,
    term = pathway,
    description = descript
  )

}

#' @return \code{NULL}
#'
#' @rdname getPathPCLs
#' @method getPathPCLs aespcOut
#' @export
getPathPCLs.aespcOut <- getPathPCLs.superpcOut