Added more processing functions and gender prediction methods.

DESCRIPTION

@@ -10,5 +10,10 @@ License: The Unlicense (see LICENSE)
 Encoding: UTF-8
 LazyData: true
 Imports:
-    utils,affy,Biobase,arrayQualityMetrics,SCAN.UPC,doParallel
+    Biobase,
+    utils,
+    affy,
+    arrayQualityMetrics,
+    SCAN.UPC,
+    massiR
 RoxygenNote: 6.1.1

R/preprocess_data_rma.R

+#' @title Preprocess a dataset with RMA.
+#'
+#' @description This function preprocess a dataset using RMA and saves the results in
+#' a given TSV file. In addition, it returns the ESET object.
+#'
+#' The function assumes that a folder containing the raw data exists (as cel files).
+#'
+#' Note: the function does not check for the existence of folders or files.
+#'
+#' @param input_data_dir A string representing the folder that contains the input data.
+#' @param output_data_file A string representing the file that should contain the
+#'  preprocessed data.
+#' @param verbose A boolean representing whether the function should display log information. This
+#'  is TRUE by default.
+#' @param compressed A boolean representing whether the cel files are compressed. This
+#'  is TRUE by default.
+#' @return The expression data as an ESET object.
+preprocess_data_rma <- function(input_data_dir, output_data_file,
+                                 compressed = TRUE,
+                                 verbose    = TRUE) {
+
+  # We define the I/Os.
+  raw_data_input_dir <- paste0(input_data_dir, "RAW/")
+
+  # We run the RMA pre-processing method on the data.
+  input_data_files <- affy::list.celfiles(raw_data_input_dir, full.names = TRUE)
+  remove(raw_data_input_dir)
+  batch <- affy::ReadAffy(filenames = input_data_files, compress = compressed, verbose = verbose)
+  eset <- affy::rma(batch)
+
+  # We save the eset data as TSV file.
+  utils::write.table(Biobase::exprs(eset), file = output_data_file, sep = "\t", quote = FALSE)
+
+  # We clean up and log information.
+  remove(input_data_files, batch)
+  if (verbose == TRUE) {
+    message(paste0("[", Sys.time(), "] Expression data pre-processed with RMA."))
+  }
+
+  # We return the created ESET.
+  return(eset)
+}

R/read_eset.R

+#' @title Read an expression set from a TSV file.
+#'
+#' @description This function loads an expression matrix and returns an eset. The
+#' function assumes a folder with the data exists and does not
+#' check for its existence.
+#'
+#' @param data_file A string representing the file that contains the expression data.
+#' @param header A boolean indicating whether the file contains a header (default to TRUE).
+#' @param sep A string that is used as a field separator to read the data (default to
+#'  tab for TSV files).
+#' @param row_names The index of the row names (default to 1).
+#' @param as_is A boolean indicating whether R should keep the data as they are in the
+#'  file (default to TRUE).
+#' @param verbose A boolean representing whether the function should display log information. This
+#'  is TRUE by default.
+#' @return An ExpressionSet containing the preprocessed expression data.
+read_eset <- function(data_file,
+                      header    = TRUE,
+                      sep       = "\t",
+                      row_names = 1,
+                      as_is     = TRUE,
+                      verbose   = TRUE) {
+
+  # We load the matrix and creates the associated eset object.
+  exprs_mat <- as.matrix(utils::read.table(data_file,
+                                           header    = header,
+                                           sep       = sep,
+                                           row.names = row_names,
+                                           as.is     = as_is))
+  exprs_eset <- Biobase::ExpressionSet(assayData = exprs_mat)
+
+  # We clean up and log information.
+  rm(exprs_mat)
+  if (verbose == TRUE) {
+    message(paste0("[", Sys.time(), "] Expression data read."))
+  }
+
+  # We return the eset.
+  return(exprs_eset)
+}

R/read_probe_lists.R

+#' @title Reads probe lists.
+#'
+#' @description This function loads the lists of probes from TSV files (for instance,
+#' from Biomart). It then returns a list of data-frames as to be able to use them with massiR.
+#'
+#' Note: the function does not check for the existence of folders and files.
+#'
+#' @param probe_list_names A list of strings representing the names of the files to be loaded (without
+#'  their eventual prefixes / suffixes).
+#' @param filename_prefix A prefix used by all files (e.g., "mart_export_").
+#' @param filename_suffix A suffix used by all files (e.g., ".tsv").
+#' @param data_dir A string representing the folder that contains the probe data (as TSV files).
+#' @param header A boolean indicating whether the file contains a header. This is FALSE by default.
+#' @param verbose A boolean representing whether the function should display log information. This
+#'  is TRUE by default.
+#' @return A list data-frames containing the probes (as row.names only).
+read_probe_lists <- function(probe_list_names, filename_prefix = "", filename_suffix = "",
+                             data_dir, header = FALSE, verbose = TRUE) {
+
+  # We define the I/Os.
+  probe_list_fullnames <- paste0(data_dir, filename_prefix, probe_list_names, filename_suffix)
+
+  # We then load the files one by one.
+  to_return <- vector("list", length = length(probe_list_fullnames))
+  for (i in seq_len(length(probe_list_fullnames))) {
+
+    # Regular file reading.
+    probes <- utils::read.csv(probe_list_fullnames[i], sep = "\t", header = header)
+
+    # We reformat the set of probes.
+    to_return[[i]] <- data.frame(row.names = probes[[colnames(probes)[1]]])
+    remove(probes)
+  }
+
+  # We clean up and log information.
+  rm(probe_list_fullnames)
+  if (verbose == TRUE) {
+    message(paste0("[", Sys.time(), "] Probe lists retrieved."))
+  }
+
+  # We return the probe lists (as a list of data-frames).
+  return(to_return)
+}

R/run_massir.R

+#' @title Predicts the gender of all samples in a given expression set.
+#'
+#' @description This function runs massiR on an eset object. The objective is to predict the gender
+#' associated with the samples from the expression of a subset of probes (i.e., typically
+#' on probes for the Y chomosome).
+#'
+#' @param eset The ExpressionSet object to analyze.
+#' @param probes The set of probes to use to discriminate males and females (most probably
+#'  these would be Y-chromosome probes).
+#' @param pheno_data The clinical data of the dataset (if known) as to estimate the accuracy of
+#'  the massiR predictions.
+#' @param massir_threshold Am integer representing the massiR threshold (between 1 and 4). This
+#'  controls the number of probes that are considered as suitable (based on the quartiles) The
+#'  default is 4 (most conservative).
+#' @param plot  A boolean representing whether control plots need to be created. This is set to
+#'  FALSE by default.
+#' @param verbose A boolean representing whether the function should display log information. This
+#'  is TRUE by default.
+#' @return A matrix with a subset of the original signal based on the optimnal probes selected
+#'   by massiR.
+run_massir <- function(eset, probes, pheno_data,
+                       massir_threshold = 4,
+                       plot             = FALSE,
+                       verbose          = TRUE) {
+
+  # We start with some control plots.
+  if (plot == TRUE) {
+    # We subset the data with the given probes (using massiR).
+    eset_subset <- massiR::massi_y(eset, probes)
+
+    # If necessary, we plot the signal (for visual control).
+    massiR::massi_y_plot(eset_subset)
+
+    # Cleaning
+    rm(eset_subset)
+  }
+
+  # We select the most suitable probes (i.e., most varying) and subset the data
+  # again (for latter use).
+  selected_probes <- massiR::massi_select(eset, probes, threshold = massir_threshold)
+  eset_subset_best <- eset[rownames(selected_probes), ]
+
+  # We perform the clutering on the 'best' probe subset.
+  cluster_results <- massiR::massi_cluster(selected_probes)
+
+  # If necessary, we check the clustering quality and predictibility.
+  if (plot == TRUE) {
+    massiR::massi_cluster_plot(selected_probes, cluster_results)
+  }
+
+  # We also log the results of the dip analysis.
+  if (verbose == TRUE) {
+    dip_result <- massiR::massi_dip(selected_probes)
+    rm(dip_result)
+  }
+
+  # We compute simple stats.
+  cases <- table(data.frame(
+    Predictions <- cluster_results[[2]]$sex,
+    Observed    <- Biobase::pData(pheno_data)$Gender
+  ))
+  acc <- (cases[1, 1] + cases[2, 2]) / sum(cases)
+
+  # We log the simple stats.
+  if (verbose == TRUE) {
+    cases
+    acc
+  }
+
+  # We clean up and log information.
+  rm(selected_probes, cluster_results, cases, acc)
+  if (verbose == TRUE) {
+    message(paste0("[", Sys.time(), "] massiR analysis performed."))
+  }
+
+  # We return the subset of the data for later use.
+  return(eset_subset_best)
+}