R/DataDesc.R

In LXQin/PRECISION.seq: Comparison of Depth Normalization Methods

#' MiRNA Sequencing Benchmark Data
#'
#' Myxofibrosarcoma (MXF) and pleomorphic malignant fibrous histiocytoma (PMFH) are the two most common and aggressive subtypes of genetically complex soft tissue sarcoma.
#' This dataset includes three libraries used for sequencing 54 individual tumor samples. Library preparation and read capture were each processed by a single experienced technician in one run.
#'
#'
#' @format A data frame with 1033 rows and 54 columns. Here are the examples of the column and row naming rule:
#' \describe{
#'   \item{MXF2516_D13}{One sample belonging to MXF and library D with sample ID MXF2516 and library ID D13.}
#'   \item{hsa-let-7a-2*}{Gene ID.}
#' }
"data.benchmark"

#' MiRNA Sequencing Test Data
#'
#' Myxofibrosarcoma (MXF) and pleomorphic malignant fibrous histiocytoma (PMFH) are the two most common and aggressive subtypes of genetically complex soft tissue sarcoma.
#' MiRNAs for the same 54 tumors used for the benchmark data were re-sequenced using neither uniform handling nor balanced library assignment.
#' In this study these samples were sequenced in the order of sample collection and processed in multiple runs.
#'
#'
#' @format A dataframe with 1033 rows and 54 columns. Here are the examples of the column and row naming rule:
#' \describe{
#'   \item{MXF2516}{One sample belonging to MXF sample ID MXF2516.}
#'   \item{hsa-let-7a-2*}{Gene ID.}
#' }
"data.test"

#' MiRNA Information
#'
#' @format Information of the microRNAs in test and benchmark dataset, including their names, exact sequence, length and GC content.
"data.miR.info"

#' Sample Labels of the test and benchmark data
#'
#' @format A set of labels for the test and benchmark samples, either "MXF" (myxofibrosarcoma) or "PMFH" (pleomorphic malignant fibrous histiocytoma).
"data.group"

#' Simulation Plan
#'
#' @format A dataframe including information of allocation about 20,000 simulation datasets, which contains the proportion of differential expression and median of mean difference for each dataset.
"data.simulation"


#' Simulated Data
#'
#' Function for providing simulated datasets according to different proportions of DE and medians of mean difference.
#' The range of DE proportion available is between 0 and 0.387, and the range of median of mean difference is between -2.52 and 4.59.
#'
#' @param proportion the range of proportion of DE for simulated dataset filtering
#' @param median the range of median of mean difference for simulated dataset filtering
#' @param numsets number of simulated datasets requested for returning (randomly drawn from the available sets). If it exceeds the maximum number of available datasets, all the available sets will be returned.
#'
#' @return list containing list of simulated benchmark data and test data
#' @import magrittr
#' @import dplyr
#' @export
#'
#' @examples
#' simulated <- simulated.data(c(0.0175, 0.0225), c(-0.5, 0.5), 10)
simulated.data <- function(proportion, median, numsets){

  proportion_L <- proportion[1]
  proportion_R <- proportion[2]
  median_L <- median[1]
  median_R <- median[2]

  if (proportion_L <=0 | proportion_R >= 0.387 | median_L <= -2.52 | median_R >= 4.59) {
    stop("The range exceeds the availability.")
  }

  benchmark_simu <- data.benchmark
  test_simu <- data.test
  colnames(benchmark_simu) <- sub(".*_", "", colnames(benchmark_simu))
  colnames(test_simu) <- colnames(benchmark_simu)
  s <- data.simulation %>%
    filter(proportion > proportion_L & proportion < proportion_R) %>%
    filter(median > median_L & median < median_R)
  rowselect <- if(nrow(s) > numsets){
    sample(nrow(s), numsets)
  }else{
    1:nrow(s)
    warning("Number of datasets that can be provided is less than the demand.")
  }

  s <- as.matrix(s[rowselect, 1:54])
  data.simulated <- list()
  for (i in 1:nrow(s)) {
    simulated_benchmark <-  benchmark_simu[, s[i,]]
    simulated_test <-  test_simu[, s[i,]]
    data.simulated[[i]] <- list(simulated_benchmark = simulated_benchmark,
                                simulated_test = simulated_test,
                                simulated_group = c(rep('MXF',27),rep('PMFH',27)))
  }
  return(data.simulated)
}