R/data_preprocessing.R

In EvaYiwenWang/PLSDAbatch: PLSDA-batch

#' Prefiltering for Microbiome Data
#'
#' This function prefilters the data to remove samples or microbial variables
#' with excess zeroes.
#'
#' @param data The data to be prefiltered. The samples in rows and variables
#' in columns.
#' @param keep.spl The minimum counts of a sample to be kept.
#' @param keep.var The minimum proportion of counts of a variable to be kept.
#'
#' @return \code{PreFL} returns a list that contains the following components:
#' \item{data.filter}{The filtered data matrix.}
#' \item{sample.idx}{The indexs of samples kept.}
#' \item{var.idx}{The indexs of variables kept.}
#' \item{zero.prob}{The proportion of zeros of the input data.}
#'
#' @author Yiwen Wang, Kim-Anh Lê Cao
#'
#'
#' @references
#' \insertRef{le2016mixmc}{PLSDAbatch}
#'
#' @export
#'
#' @examples
#' library(TreeSummarizedExperiment) # for functions assays()
#' data('AD_data')
#'
#' ad.count <- assays(AD_data$FullData)$Count # microbial count data
#' ad.filter.res <- PreFL(data = ad.count)
#'
#' # The proportion of zeroes of the AD count data
#' ad.zero.prob <- ad.filter.res$zero.prob
#'
#' # The filtered AD count data
#' ad.filter <- ad.filter.res$data.filter
#'
#'
PreFL <- function(data,
                keep.spl = 10,
                keep.var = 0.01){

    # zero prob
    zero.prob <- mean(data == 0)

    # sample
    keep.sample <- which(rowSums(data) > keep.spl)
    data <- data[keep.sample, ]

    # variable
    keep.var <- which(colSums(data)*100/(sum(colSums(data))) > keep.var)
    data.out <- as.matrix(data[, keep.var])

    result <- list(data.filter = data.out,
                    sample.idx = keep.sample,
                    var.idx = keep.var,
                    zero.prob = zero.prob)

    return(invisible(result))
}