R/32-dyntoy.R

In duohongrui/simmethods: A collection of 40+ simulation methods for singlce-cell RNA seqencing data

#' Estimate Parameters From Real Datasets by dyntoy
#'
#' This function is used to estimate useful parameters from a real dataset by
#' using \code{infer_trajectory} function in dynwrap package.
#'
#' @param ref_data A count matrix. Each row represents a gene and each column
#' represents a cell.
#' @param verbose Logical.
#' @param seed An integer of a random seed.
#' @param other_prior A list with names of certain parameters. Some methods need
#' extra parameters to execute the estimation step, so you must input them. In
#' simulation step, the number of cells, genes, groups, batches, the percent of
#' DEGs are usually customed, so before simulating a dataset you must point it out.
#' See `Details` below for more information.
#' @importFrom dynwrap infer_trajectory wrap_expression add_grouping
#' @return A list contains the estimated parameters and the results of execution
#' detection.
#' @export
#' @details
#' In dyntoy, users can input cell group information if it is available. If cell
#' group information is not provided, the procedure will detect cell groups by
#' kmeans automatically.
#' See `Examples` for more instructions.
#'
#' @references
#' Github URL: <https://github.com/dynverse/dyntoy>
#' @examples
#' \dontrun{
#' ref_data <- simmethods::data
#'
#' estimate_result <- simmethods::dyntoy_estimation(
#'   ref_data = ref_data,
#'   other_prior = NULL,
#'   verbose = TRUE,
#'   seed = 111
#' )
#'
#' ## estimation with cell group information
#' group_condition <- paste0("Group", as.numeric(simmethods::group_condition))
#' estimate_result <- simmethods::dyntoy_estimation(
#'   ref_data = ref_data,
#'   other_prior = list(group.condition = group_condition),
#'   verbose = TRUE,
#'   seed = 111
#' )
#' }
#'
dyntoy_estimation <- function(ref_data,
                              verbose = FALSE,
                              other_prior = NULL,
                              seed){
  ##############################################################################
  ####                               Check                                   ###
  ##############################################################################
  if(!requireNamespace("tislingshot", quietly = TRUE)){
    message("tislingshot is not installed on your device...")
    message("Installing tislingshot...")
    devtools::install_github("dynverse/ti_slingshot/package/")
  }
  if(!requireNamespace("NbClust", quietly = TRUE)){
    message("NbClust is not installed on your device...")
    message("Installing NbClust...")
    utils::install.packages("NbClust")
  }
  if(!is.matrix(ref_data)){
    ref_data <- as.matrix(ref_data)
  }
  if(is.null(other_prior[["group.condition"]])){
    message("Performing k-means and determin the best number of clusters...")
    clust <- NbClust::NbClust(data = t(ref_data),
                              distance = 'euclidean',
                              min.nc = 2,
                              max.nc = sqrt(nrow(t(ref_data))),
                              method = "kmeans",
                              index = "dunn")
    other_prior[["group.condition"]] <- paste0("Group", clust[["Best.partition"]])
  }
  ref_data <- dynwrap::wrap_expression(counts = t(ref_data),
                                       expression = log2(t(ref_data) + 1))
  ## Add group
  if(verbose){
    message("Add grouping to data...")
  }
  ref_data <- dynwrap::add_grouping(dataset = ref_data,
                                    grouping = other_prior[["group.condition"]])
  ##############################################################################
  ####                            Estimation                                 ###
  ##############################################################################
  if(verbose){
    message("Estimating parameters using dyntoy")
  }
  # Estimation
  estimate_detection <- peakRAM::peakRAM(
    estimate_result <- dynwrap::infer_trajectory(dataset = ref_data,
                                                 method = tislingshot::ti_slingshot(),
                                                 parameters = NULL,
                                                 give_priors = NULL,
                                                 seed = seed,
                                                 verbose = verbose)
  )
  estimate_result <- list(estimate_result = estimate_result,
                          data_dim = c(length(ref_data[["feature_ids"]]),
                                       length(ref_data[["cell_ids"]])))
  ##############################################################################
  ####                           Ouput                                       ###
  ##############################################################################
  estimate_output <- list(estimate_result = estimate_result,
                          estimate_detection = estimate_detection)
  return(estimate_output)
}


#' Simulate Datasets by dyntoy
#'
#' @param parameters A object generated by [dynwrap::infer_trajectory()]
#' @param other_prior A list with names of certain parameters. Some methods need
#' extra parameters to execute the estimation step, so you must input them. In
#' simulation step, the number of cells, genes, groups, batches, the percent of
#' DEGs are usually customed, so before simulating a dataset you must point it out.
#' See `Details` below for more information.
#' @param return_format A character. Alternative choices: list, SingleCellExperiment,
#' Seurat, h5ad. If you select `h5ad`, you will get a path where the .h5ad file saves to.
#' @param verbose Logical. Whether to return messages or not.
#' @param seed A random seed.
#' @export
#' @details
#' In dyntoy, users can only set `nCells` and `nGenes` to specify the number of cells and genes in the
#' simulated dataset. See `Examples` for instructions.
#'
#' @references
#' Github URL: <https://github.com/dynverse/dyntoy>
#' @examples
#' \dontrun{
#' ref_data <- simmethods::data
#'
#' ## estimation with cell group information
#' group_condition <- paste0("Group", as.numeric(simmethods::group_condition))
#' estimate_result <- simmethods::dyntoy_estimation(
#'   ref_data = ref_data,
#'   other_prior = list(group.condition = group_condition),
#'   verbose = TRUE,
#'   seed = 111
#' )
#'
#' # 1) Simulate with default parameters
#' simulate_result <- simmethods::dyntoy_simulation(
#'   parameters = estimate_result[["estimate_result"]],
#'   other_prior = NULL,
#'   return_format = "list",
#'   verbose = TRUE,
#'   seed = 111
#' )
#' ## counts
#' counts <- simulate_result[["simulate_result"]][["count_data"]]
#' dim(counts)
#'
#' # 2) 2000 cells and 5000 genes
#' simulate_result <- simmethods::dyntoy_simulation(
#'   parameters = estimate_result[["estimate_result"]],
#'   other_prior = list(nCells = 2000,
#'                      nGenes = 5000),
#'   return_format = "list",
#'   verbose = TRUE,
#'   seed = 111
#' )
#'
#' ## counts
#' counts <- simulate_result[["simulate_result"]][["count_data"]]
#' dim(counts)
#' }
#'
dyntoy_simulation <- function(parameters,
                              other_prior = NULL,
                              return_format,
                              verbose = FALSE,
                              seed
){

  ##############################################################################
  ####                            Environment                                ###
  ##############################################################################
  if(!requireNamespace("dyntoy", quietly = TRUE)){
    message("dyntoy is not installed on your device")
    message("Installing dyntoy...")
    devtools::install_github("dynverse/dyntoy")
  }
  ##############################################################################
  ####                               Check                                   ###
  ##############################################################################
  model <- parameters[["estimate_result"]][["milestone_network"]]
  # nCells
  if(!is.null(other_prior[["nCells"]])){
    nCells <- other_prior[["nCells"]]
  }else{
    nCells <- parameters[["data_dim"]][2]
  }
  # nGenes
  if(!is.null(other_prior[["nGenes"]])){
    nGenes <- other_prior[["nGenes"]]
  }else{
    nGenes <- parameters[["data_dim"]][1]
  }
  # de.prob
  if(!is.null(other_prior[["de.prob"]])){
    de.prob <- other_prior[["de.prob"]]
  }else{
    de.prob <- 0.1
  }

  # Return to users
  message(paste0("nCells: ", nCells))
  message(paste0("nGenes: ", nGenes))
  message(paste0("de.prob: ", de.prob))
  ##############################################################################
  ####                            Simulation                                 ###
  ##############################################################################
  if(verbose){
    message("Simulating datasets using dyntoy")
  }
  # Seed
  set.seed(seed)
  # Estimation
  simulate_detection <- peakRAM::peakRAM(
    simulate_result <- dyntoy::generate_dataset(model = model,
                                                num_cells = nCells,
                                                num_features = nGenes,
                                                differentially_expressed_rate = de.prob)
  )
  ##############################################################################
  ####                        Format Conversion                              ###
  ##############################################################################
  de_gene <- simulate_result[["tde_overall"]][["differentially_expressed"]]
  simulate_result <- t(as.matrix(simulate_result[["counts"]]))
  colnames(simulate_result) <- paste0("Cell", 1:ncol(simulate_result))
  rownames(simulate_result) <- paste0("Gene", 1:nrow(simulate_result))
  ## col_data
  col_data <- data.frame("cell_name" = colnames(simulate_result))
  ## row_data
  row_data <- data.frame("gene_name" = rownames(simulate_result),
                         "de_gene" = ifelse(de_gene, "yes", "no"))
  # Establish SingleCellExperiment
  simulate_result <- SingleCellExperiment::SingleCellExperiment(list(counts = simulate_result),
                                                                colData = col_data,
                                                                rowData = row_data)
  simulate_result <- simutils::data_conversion(SCE_object = simulate_result,
                                               return_format = return_format)

  ##############################################################################
  ####                           Ouput                                       ###
  ##############################################################################
  simulate_output <- list(simulate_result = simulate_result,
                          simulate_detection = simulate_detection)
  return(simulate_output)
}