@@ -105,7 +105,8 @@ Imports:

     yaml,

     rmarkdown,

     magrittr,

-    kableExtra

+    kableExtra,

+    scDblFinder

 RoxygenNote: 7.1.1

 Suggests:

     testthat,

@@ -1116,7 +1116,6 @@ plotDoubletFinderResults <- function(inSCE,

 #'  Default TRUE.

 #' @param dots Boolean. If TRUE, will plot dots for each violin plot.

 #'  Default TRUE.

-#' @param logScore Boolean. If TRUE, the log normalized doublet score will be used.

 #' @param reducedDimName Saved dimension reduction name in the

 #' \linkS4class{SingleCellExperiment} object. Required.

 #' @param xlab Character vector. Label for x-axis. Default NULL.

@@ -1173,7 +1172,6 @@ plotDoubletCellsResults <- function(inSCE,

                                     violin=TRUE,

                                     boxplot=FALSE,

                                     dots=TRUE,

-                                    logScore=TRUE,

                                     reducedDimName=NULL,

                                     xlab=NULL,

                                     ylab=NULL,

@@ -1212,13 +1210,9 @@ plotDoubletCellsResults <- function(inSCE,

   }

   sampleVector <- sample

-  if (logScore) {

-    coldata = "scran_doubletCells_score_log10"

-    titleDoubletCells <- "DoubletCells Doublet Score, log10"

-  } else {

-    coldata = "scran_doubletCells_score"

-    titleDoubletCells <- "DoubletCells Doublet Score"

-  }

+

+  coldata = "scran_doubletCells_score"

+  titleDoubletCells <- "DoubletCells Doublet Score"

   samples <- unique(sample)

   if (length(samples) > 1) {

@@ -1,81 +1,60 @@

-.runDoubletCells <- function(cell.matrix = cell.matrix,

-                              k = k,

-                              nIters = nIters,

-                              size.factors.norm = NULL,

-                              size.factors.content = NULL,

-                              subset.row = NULL,

-                              block = 10000,

-                              d = 50,

-                              force.match=FALSE,

-                              force.k=20,

-                              force.ndist=3,

-                              BNPARAM=BNPARAM,

-                              BSPARAM=BSPARAM,

-                              BPPARAM=BPPARAM

-                              ) {

-

-  cell.matrix <- .convertToMatrix(cell.matrix)

-

-  scores <- matrix(scran::doubletCells(cell.matrix, k = k,

-                                       niters = nIters,

-                                       size.factors.norm = NULL,

-                                       size.factors.content = NULL,

-                                       subset.row = NULL,

-                                       block = 10000,

-                                       d = 50,

-                                       force.match=FALSE,

-                                       force.k=20,

-                                       force.ndist=3,

-                                       BNPARAM=BNPARAM,

-                                       BSPARAM=BSPARAM,

-                                       BPPARAM=BPPARAM

-                                       ), ncol=1)

-  scores <- cbind(scores,log10(scores[,1]+1))

-  colnames(scores) <- c("scran_doubletCells_score", "scran_doubletCells_score_log10")

-

-

-  return(scores)

-}

-

-

-#' @title Detect doublet cells using \link[scran]{doubletCells}.

-#' @description A wrapper function for \link[scran]{doubletCells}. Identify

+# .runDoubletCells <- function(cell.matrix = cell.matrix,

+#                               k = k,

+#                               nIters = nIters,

+#                               size.factors.norm = NULL,

+#                               size.factors.content = NULL,

+#                               subset.row = NULL,

+#                               block = 10000,

+#                               d = 50,

+#                               force.match=FALSE,

+#                               force.k=20,

+#                               force.ndist=3,

+#                               BNPARAM=BNPARAM,

+#                               BSPARAM=BSPARAM,

+#                               BPPARAM=BPPARAM

+#                               ) {

+# 

+#   cell.matrix <- .convertToMatrix(cell.matrix)

+# 

+#   scores <- matrix(scran::doubletCells(cell.matrix, k = k,

+#                                        niters = nIters,

+#                                        size.factors.norm = NULL,

+#                                        size.factors.content = NULL,

+#                                        subset.row = NULL,

+#                                        block = 10000,

+#                                        d = 50,

+#                                        force.match=FALSE,

+#                                        force.k=20,

+#                                        force.ndist=3,

+#                                        BNPARAM=BNPARAM,

+#                                        BSPARAM=BSPARAM,

+#                                        BPPARAM=BPPARAM

+#                                        ), ncol=1)

+#   scores <- cbind(scores,log10(scores[,1]+1))

+#   colnames(scores) <- c("scran_doubletCells_score", "scran_doubletCells_score_log10")

+# 

+# 

+#   return(scores)

+# }

+# 

+

+#' @title Detect doublet cells using \link[scDblFinder]{scDblFinder}.

+#' @description A wrapper function for \link[scDblFinder]{scDblFinder}. Identify

 #'  potential doublet cells based on simulations of putative doublet expression

 #'  profiles. Generate a doublet score for each cell.

 #' @param inSCE A \link[SingleCellExperiment]{SingleCellExperiment} object.

 #' @param sample Character vector. Indicates which sample each cell belongs to.

-#'  \link[scran]{doubletCells} will be run on cells from each sample separately.

+#'  \link[scDblFinder]{scDblFinder} will be run on cells from each sample separately.

 #' @param useAssay  A string specifying which assay in the SCE to use.

 #' @param nNeighbors Number of nearest neighbors used to calculate density for

 #'  doublet detection. Default 50.

 #' @param simDoublets Number of simulated doublets created for doublet

 #'  detection. Default 10000.

 #' @param seed Seed for the random number generator. Default 12345.

-#' @param size.factors.norm A numeric vector of size factors for normalization

-#'  of \code{x} prior to PCA and distance calculations. If \code{NULL}, defaults

-#'  to size factors derived from the library sizes of \code{x}. For the SingleCellExperiment

-#'  method, the default values are taken from \code{\link{sizeFactors}(x)}, if they are available.

-#' @param size.factors.content A numeric vector of size factors for RNA content

-#'  normalization of \code{x} prior to simulating doublets. #' This is orthogonal to

-#'  the values in \code{size.factors.norm}

-#' @param subset.row See \code{?"\link{scran-gene-selection}"}.

-#' @param block An integer scalar controlling the rate of doublet generation,

-#'  to keep memory usage low.

-#' @param d An integer scalar specifying the number of components to retain after the PCA.

-#' @param force.match A logical scalar indicating whether remapping of simulated

-#'  doublets to original cells should be performed.

-#' @param force.k An integer scalar specifying the number of neighbours to use for

-#'  remapping if \code{force.match=TRUE}.

-#' @param force.ndist A numeric scalar specifying the bandwidth for remapping

-#'  if \code{force.match=TRUE}.

-#' @param BNPARAM A \code{\link[BiocNeighbors]{BiocNeighborParam}} object specifying the nearest neighbor algorithm.

-#' This should be an algorithm supported by \code{\link[BiocNeighbors]{findNeighbors}}.

-#' @param BSPARAM A \code{\link[BiocSingular]{BiocSingularParam}} object specifying the algorithm to

-#'  use for PCA, if \code{d} is not \code{NA}.

 #' @param BPPARAM A \code{\link{BiocParallelParam}} object specifying whether the

 #'  neighbour searches should be parallelized.

-#' @details This function is a wrapper function for \link[scran]{doubletCells}.

-#'  \code{runDoubletCells} runs \link[scran]{doubletCells} for each

+#' @details This function is a wrapper function for \link[scDblFinder]{scDblFinder}.

+#'  \code{runDoubletCells} runs \link[scDblFinder]{scDblFinder} for each

 #'  \code{sample} within \code{inSCE} iteratively. The

 #'  resulting doublet scores for all cells will be appended to the

 #'  \link{colData} of \code{inSCE}.

@@ -98,16 +77,16 @@ runDoubletCells <- function(inSCE,

     nNeighbors = 50,

     simDoublets = max(10000, ncol(inSCE)),

     seed = 12345,

-    size.factors.norm = NULL,

-    size.factors.content = NULL,

-    subset.row = NULL,

-    block = 10000,

-    d = 50,

-    force.match=FALSE,

-    force.k=20,

-    force.ndist=3,

-    BNPARAM=BiocNeighbors::KmknnParam(),

-    BSPARAM=BiocSingular::bsparam(),

+    # size.factors.norm = NULL,

+    # size.factors.content = NULL,

+    # subset.row = NULL,

+    # block = 10000,

+    # d = 50,

+    # force.match=FALSE,

+    # force.k=20,

+    # force.ndist=3,

+    # BNPARAM=BiocNeighbors::KmknnParam(),

+    # BSPARAM=BiocSingular::bsparam(),

     BPPARAM=BiocParallel::SerialParam()

 ) {

   #argsList <- as.list(formals(fun = sys.function(sys.parent()), envir = parent.frame()))

@@ -124,45 +103,54 @@ runDoubletCells <- function(inSCE,

   message(paste0(date(), " ... Running 'doubletCells'"))

   ## Define result matrix for all samples

-  output <- S4Vectors::DataFrame(row.names = colnames(inSCE),

-            scran_doubletCells_score = numeric(ncol(inSCE)),

-            scran_doubletCells_score_log10 = numeric(ncol(inSCE)))

+  # output <- S4Vectors::DataFrame(row.names = colnames(inSCE),

+  #           scran_doubletCells_score = numeric(ncol(inSCE)),

+  #           scran_doubletCells_score_log10 = numeric(ncol(inSCE)))

   ## Loop through each sample and run barcodeRank

-  samples <- unique(sample)

-  for (i in seq_len(length(samples))) {

-    sceSampleInd <- sample == samples[i]

-    sceSample <- inSCE[, sceSampleInd]

-

-    mat <- SummarizedExperiment::assay(sceSample, i = useAssay)

-

-    result <- withr::with_seed(seed,

-              .runDoubletCells(cell.matrix = mat,

-                               k = nNeighbors,

-                               nIters = simDoublets,

-                               size.factors.norm = NULL,

-                               size.factors.content = NULL,

-                               subset.row = NULL,

-                               block = 10000,

-                               d = 50,

-                               force.match=FALSE,

-                               force.k=20,

-                               force.ndist=3,

-                               BNPARAM=BNPARAM,

-                               BSPARAM=BSPARAM,

-                               BPPARAM=BPPARAM

-                               ))

-

-    output[sceSampleInd, ] <- result

-  }

+  #samples <- unique(sample)

+  

+  inSCE <- withr::with_seed(seed,

+                            scDblFinder::scDblFinder(sce = inSCE,

+                            samples = sample,

+                            artificialDoublets = simDoublets,

+                            k = nNeighbors,

+                            verbose = FALSE

+                            ))

+  names(SummarizedExperiment::colData(inSCE)) <- gsub(pattern = "scDblFinder\\.",

+                                                      "scran_doubletCells_",

+                                                      names(SummarizedExperiment::colData(inSCE)))

+  

+  # for (i in seq_len(length(samples))) {

+  #   sceSampleInd <- sample == samples[i]

+  #   sceSample <- inSCE[, sceSampleInd]

+  # 

+  #   mat <- SummarizedExperiment::assay(sceSample, i = useAssay)

+  # 

+  #   result <- withr::with_seed(seed,

+  #             .runDoubletCells(cell.matrix = mat,

+  #                              k = nNeighbors,

+  #                              nIters = simDoublets,

+  #                              size.factors.norm = NULL,

+  #                              size.factors.content = NULL,

+  #                              subset.row = NULL,

+  #                              block = 10000,

+  #                              d = 50,

+  #                              force.match=FALSE,

+  #                              force.k=20,

+  #                              force.ndist=3,

+  #                              BNPARAM=BNPARAM,

+  #                              BSPARAM=BSPARAM,

+  #                              BPPARAM=BPPARAM

+  #                              ))

+  # 

+  #   output[sceSampleInd, ] <- result

+  # }

+

+  argsList <- argsList[!names(argsList) %in% c("BPPARAM")]

-  argsList <- argsList[!names(argsList) %in% c("BNPARAM","BSPARAM","BPPARAM")]

-  #dotList <- list(...)

-  #dotList <- dotList[!names(dotList) %in% c("BNPARAM","BSPARAM","BPPARAM")]

-  #argsList <- c(argsList, dotList)

   inSCE@metadata$runDoubletCells <- argsList[-1]

-  inSCE@metadata$runDoubletCells$packageVersion <- utils::packageDescription("scran")$Version

-  colData(inSCE) = cbind(colData(inSCE), output)

+  inSCE@metadata$runDoubletCells$packageVersion <- utils::packageDescription("scDblFinder")$Version

   return(inSCE)

 }

@@ -359,7 +359,7 @@ description_DoubletCells<- descriptionDoubletCells()

 i="DoubletCells"

 cat(paste0('## ', i, ' \n'))

-doubletCellData <- c("scran_doubletCells_score_log10")

+doubletCellData <- c("scran_doubletCells_score")

 skipDoubletCell <- any(!doubletCellData %in% names(colData(sce.qc)))

 if (skipDoubletCell) {

@@ -13,7 +13,6 @@ plotDoubletCellsResults(

   violin = TRUE,

   boxplot = FALSE,

   dots = TRUE,

-  logScore = TRUE,

   reducedDimName = NULL,

   xlab = NULL,

   ylab = NULL,

@@ -67,8 +66,6 @@ Default TRUE.}

 \item{dots}{Boolean. If TRUE, will plot dots for each violin plot.

 Default TRUE.}

-\item{logScore}{Boolean. If TRUE, the log normalized doublet score will be used.}

-

 \item{reducedDimName}{Saved dimension reduction name in the

 \linkS4class{SingleCellExperiment} object. Required.}

@@ -2,7 +2,7 @@

 % Please edit documentation in R/scran_doubletCells.R

 \name{runDoubletCells}

 \alias{runDoubletCells}

-\title{Detect doublet cells using \link[scran]{doubletCells}.}

+\title{Detect doublet cells using \link[scDblFinder]{scDblFinder}.}

 \usage{

 runDoubletCells(

   inSCE,

@@ -11,16 +11,6 @@ runDoubletCells(

   nNeighbors = 50,

   simDoublets = max(10000, ncol(inSCE)),

   seed = 12345,

-  size.factors.norm = NULL,

-  size.factors.content = NULL,

-  subset.row = NULL,

-  block = 10000,

-  d = 50,

-  force.match = FALSE,

-  force.k = 20,

-  force.ndist = 3,

-  BNPARAM = BiocNeighbors::KmknnParam(),

-  BSPARAM = BiocSingular::bsparam(),

   BPPARAM = BiocParallel::SerialParam()

 )

 }

@@ -28,7 +18,7 @@ runDoubletCells(

 \item{inSCE}{A \link[SingleCellExperiment]{SingleCellExperiment} object.}

 \item{sample}{Character vector. Indicates which sample each cell belongs to.

-\link[scran]{doubletCells} will be run on cells from each sample separately.}

+\link[scDblFinder]{scDblFinder} will be run on cells from each sample separately.}

 \item{useAssay}{A string specifying which assay in the SCE to use.}

@@ -40,37 +30,6 @@ detection. Default 10000.}

 \item{seed}{Seed for the random number generator. Default 12345.}

-\item{size.factors.norm}{A numeric vector of size factors for normalization

-of \code{x} prior to PCA and distance calculations. If \code{NULL}, defaults

-to size factors derived from the library sizes of \code{x}. For the SingleCellExperiment

-method, the default values are taken from \code{\link{sizeFactors}(x)}, if they are available.}

-

-\item{size.factors.content}{A numeric vector of size factors for RNA content

-normalization of \code{x} prior to simulating doublets. #' This is orthogonal to

-the values in \code{size.factors.norm}}

-

-\item{subset.row}{See \code{?"\link{scran-gene-selection}"}.}

-

-\item{block}{An integer scalar controlling the rate of doublet generation,

-to keep memory usage low.}

-

-\item{d}{An integer scalar specifying the number of components to retain after the PCA.}

-

-\item{force.match}{A logical scalar indicating whether remapping of simulated

-doublets to original cells should be performed.}

-

-\item{force.k}{An integer scalar specifying the number of neighbours to use for

-remapping if \code{force.match=TRUE}.}

-

-\item{force.ndist}{A numeric scalar specifying the bandwidth for remapping

-if \code{force.match=TRUE}.}

-

-\item{BNPARAM}{A \code{\link[BiocNeighbors]{BiocNeighborParam}} object specifying the nearest neighbor algorithm.

-This should be an algorithm supported by \code{\link[BiocNeighbors]{findNeighbors}}.}

-

-\item{BSPARAM}{A \code{\link[BiocSingular]{BiocSingularParam}} object specifying the algorithm to

-use for PCA, if \code{d} is not \code{NA}.}

-

 \item{BPPARAM}{A \code{\link{BiocParallelParam}} object specifying whether the

 neighbour searches should be parallelized.}

 }

@@ -80,13 +39,13 @@ A \link[SingleCellExperiment]{SingleCellExperiment} object with the

  \link{colData} slot.

 }

 \description{

-A wrapper function for \link[scran]{doubletCells}. Identify

+A wrapper function for \link[scDblFinder]{scDblFinder}. Identify

  potential doublet cells based on simulations of putative doublet expression

  profiles. Generate a doublet score for each cell.

 }

 \details{

-This function is a wrapper function for \link[scran]{doubletCells}.

- \code{runDoubletCells} runs \link[scran]{doubletCells} for each

+This function is a wrapper function for \link[scDblFinder]{scDblFinder}.

+ \code{runDoubletCells} runs \link[scDblFinder]{scDblFinder} for each

  \code{sample} within \code{inSCE} iteratively. The

  resulting doublet scores for all cells will be appended to the

  \link{colData} of \code{inSCE}.

@@ -51,7 +51,7 @@ test_that(desc = "Testing plotSCEViolin functions", {

 sceres <- sceres[, colData(sceres)$type != 'EmptyDroplet']

 sceres <- runCellQC(sceres, algorithms = c("QCMetrics", "cxds", "bcds", "cxds_bcds_hybrid",

                                               "doubletFinder", "decontX"))

-sceres <- runDoubletCells(sceres, size.factors.norm = rep(1, ncol(sceres)))

+sceres <- runDoubletCells(sceres)

 context("Testing QC functions")