@@ -49,14 +49,14 @@

 #' @param logfile Character. Messages will be redirected to a file named

 #'  `logfile`. If NULL, messages will be printed to stdout.  Default NULL.

 #' @param verbose Logical. Whether to print log messages. Default TRUE.

-#' 

+#'

 #' @return A \link[SingleCellExperiment]{SingleCellExperiment} object with

 #'  'decontX_Contamination' and 'decontX_Clusters' added to the

 #'  \link[SummarizedExperiment]{colData} slot. Additionally, the

 #' decontaminated counts will be added as an assay called 'decontXCounts'.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDecontX(sce)

 #' @export

 runDecontX <- function(inSCE,

@@ -84,8 +84,8 @@ runDecontX <- function(inSCE,

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

-  inSCE <- celda::decontX(x = inSCE, 

-                          batch = sample, 

+  inSCE <- celda::decontX(x = inSCE,

+                          batch = sample,

                           assayName = useAssay,

                           z = z,

                           maxIter = maxIter,

@@ -100,7 +100,7 @@ runDecontX <- function(inSCE,

                           verbose = verbose)

   #argsList <- argsList[!names(argsList) %in% ("...")]

-  

+

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

   inSCE@metadata$runDecontX$packageVersion <- utils::packageDescription("celda")$Version

@@ -330,7 +330,7 @@

 #'  'doublet_finder_doublet_score'.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDoubletFinder(sce)

 #' @export

 #' @importFrom SummarizedExperiment colData colData<-

@@ -1,115 +1,115 @@

-#' Uniform Manifold Approximation and Projection(UMAP) algorithm for

-#' dimension reduction.

-#'

-#' @param inSCE Input \linkS4class{SingleCellExperiment} object.

-#' @param useAssay Indicate which assay to use. The default is "counts".

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

-#' @param reducedDimName a name to store the results of the dimension reduction

-#' coordinates obtained from this method. This is stored in the SingleCellExperiment

-#' object in the reducedDims slot. Default "UMAP".

-#' @param logNorm Whether the counts will need to be log-normalized prior to

-#' generating the UMAP via scater::logNormCounts. Default TRUE.

-#' @param nNeighbors The size of local neighborhood used for

-#'   manifold approximation. Larger values result in more global

-#'   views of the manifold, while smaller values result in more

-#'   local data being preserved. Default 30.

-#'    See `?uwot::umap` for more information.

-#' @param nIterations number of iterations performed during layout optimization.

-#' Default is 200.

-#' @param alpha initial value of "learning rate" of layout optimization. Default is 1.

-#' @param minDist The effective minimum distance between embedded points.

-#'    Smaller values will result in a more clustered/clumped

-#'    embedding where nearby points on the manifold are drawn

-#'    closer together, while larger values will result on a more

-#'    even dispersal of points. Default 0.01.

-#'    See `?uwot::umap` for more information.

-#' @param spread The effective scale of embedded points. In combination with

-#'    ‘min_dist’, this determines how clustered/clumped the

-#'    embedded points are. Default 1.

-#'    See `?uwot::umap` for more information.

-#' @param pca Logical. Whether to perform dimensionality reduction with PCA

-#' before UMAP.

-#' @param initialDims  Number of dimensions from PCA to use as

-#' input in UMAP. Default 50.

-#'

-#' @return A \linkS4class{SingleCellExperiment} object with the reduced

-#' dimensions updated under reducedDimName specified.

-#' @export

-#'

-#' @examples

-#' data(scExample, package = "singleCellTK")

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

-#' umap_res <- getUMAP(inSCE = sce, useAssay = "counts",

-#'                     reducedDimName = "UMAP", logNorm = TRUE,

-#'                     nNeighbors = 30, alpha = 1,

-#'                     nIterations = 200, spread = 1, pca = TRUE,

-#'                     initialDims = 50)

-#' reducedDims(umap_res)

-

-getUMAP <- function(inSCE, useAssay = "counts",

-                    sample = NULL,

-                    reducedDimName = "UMAP",

-                    logNorm = TRUE,

-                    nNeighbors = 30,

-                    nIterations = 200,

-                    alpha = 1,

-                    minDist = 0.01,

-                    spread = 1,

-                    pca = TRUE,

-                    initialDims = 50) {

-  if (!inherits(inSCE, "SingleCellExperiment")){

-    stop("Please use a SingleCellExperiment object")

-  }

-  #test for assay existing

-    if (!all(useAssay %in% names(assays(inSCE)))){

-        stop("assay '", useAssay, "' does not exist.")

-    }

-

-    if(!is.null(sample)) {

-        if(length(sample) != ncol(inSCE)) {

-            stop("'sample' must be the same length as the number of columns in 'inSCE'")

-        }

-    } else {

-        sample = rep(1, ncol(inSCE))

-    }

-    samples <- unique(sample)

-    umapDims = matrix(nrow = ncol(inSCE), ncol = 2)

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

-        useAssayTemp = useAssay

-	sceSampleInd <- sample == samples[i]

-        sceSample <- inSCE[, sceSampleInd]

-        if(logNorm){

-	    sceSample <- scater_logNormCounts(sceSample, useAssay = useAssay)

-            useAssayTemp = "ScaterLogNormCounts"

-        }

-

-        matColData <- SummarizedExperiment::assay(sceSample, useAssayTemp)

-        matColData <- as.matrix(matColData)

-

-        if (isTRUE(pca)) {

-          if(initialDims > ncol(matColData)){

-            doPCA <- ncol(matColData)

-          }else{

-            doPCA <- initialDims

-          }

-        } else {

-            doPCA <- NULL

-        }

-        if(nNeighbors > ncol(matColData)){

-          nNeighbors <- ncol(matColData)

-        }

-

-        umapRes <- uwot::umap(t(matColData), n_neighbors = nNeighbors,

-                              learning_rate = alpha,

-                              min_dist = minDist, spread = spread,

-                              n_sgd_threads = 1, pca = doPCA,

-                              n_epochs = nIterations)

-        if (is.null(rownames(sceSample))) {

-            rownames(umapRes) <- colnames(sceSample)

-        }

-        umapDims[sceSampleInd, ] = umapRes

-    }

-    colnames(umapDims) <- c("UMAP1", "UMAP2")

-    SingleCellExperiment::reducedDim(inSCE, reducedDimName) <- umapDims

-    return(inSCE)

-}

+#' Uniform Manifold Approximation and Projection(UMAP) algorithm for

+#' dimension reduction.

+#'

+#' @param inSCE Input \linkS4class{SingleCellExperiment} object.

+#' @param useAssay Indicate which assay to use. The default is "counts".

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

+#' @param reducedDimName a name to store the results of the dimension reduction

+#' coordinates obtained from this method. This is stored in the SingleCellExperiment

+#' object in the reducedDims slot. Default "UMAP".

+#' @param logNorm Whether the counts will need to be log-normalized prior to

+#' generating the UMAP via scater::logNormCounts. Default TRUE.

+#' @param nNeighbors The size of local neighborhood used for

+#'   manifold approximation. Larger values result in more global

+#'   views of the manifold, while smaller values result in more

+#'   local data being preserved. Default 30.

+#'    See `?uwot::umap` for more information.

+#' @param nIterations number of iterations performed during layout optimization.

+#' Default is 200.

+#' @param alpha initial value of "learning rate" of layout optimization. Default is 1.

+#' @param minDist The effective minimum distance between embedded points.

+#'    Smaller values will result in a more clustered/clumped

+#'    embedding where nearby points on the manifold are drawn

+#'    closer together, while larger values will result on a more

+#'    even dispersal of points. Default 0.01.

+#'    See `?uwot::umap` for more information.

+#' @param spread The effective scale of embedded points. In combination with

+#'    ‘min_dist’, this determines how clustered/clumped the

+#'    embedded points are. Default 1.

+#'    See `?uwot::umap` for more information.

+#' @param pca Logical. Whether to perform dimensionality reduction with PCA

+#' before UMAP.

+#' @param initialDims  Number of dimensions from PCA to use as

+#' input in UMAP. Default 50.

+#'

+#' @return A \linkS4class{SingleCellExperiment} object with the reduced

+#' dimensions updated under reducedDimName specified.

+#' @export

+#'

+#' @examples

+#' data(scExample, package = "singleCellTK")

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

+#' umap_res <- getUMAP(inSCE = sce, useAssay = "counts",

+#'                     reducedDimName = "UMAP", logNorm = TRUE,

+#'                     nNeighbors = 30, alpha = 1,

+#'                     nIterations = 200, spread = 1, pca = TRUE,

+#'                     initialDims = 50)

+#' reducedDims(umap_res)

+

+getUMAP <- function(inSCE, useAssay = "counts",

+                    sample = NULL,

+                    reducedDimName = "UMAP",

+                    logNorm = TRUE,

+                    nNeighbors = 30,

+                    nIterations = 200,

+                    alpha = 1,

+                    minDist = 0.01,

+                    spread = 1,

+                    pca = TRUE,

+                    initialDims = 50) {

+  if (!inherits(inSCE, "SingleCellExperiment")){

+    stop("Please use a SingleCellExperiment object")

+  }

+  #test for assay existing

+    if (!all(useAssay %in% names(assays(inSCE)))){

+        stop("assay '", useAssay, "' does not exist.")

+    }

+

+    if(!is.null(sample)) {

+        if(length(sample) != ncol(inSCE)) {

+            stop("'sample' must be the same length as the number of columns in 'inSCE'")

+        }

+    } else {

+        sample = rep(1, ncol(inSCE))

+    }

+    samples <- unique(sample)

+    umapDims = matrix(nrow = ncol(inSCE), ncol = 2)

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

+        useAssayTemp = useAssay

+	sceSampleInd <- sample == samples[i]

+        sceSample <- inSCE[, sceSampleInd]

+        if(logNorm){

+	    sceSample <- scater_logNormCounts(sceSample, useAssay = useAssay)

+            useAssayTemp = "ScaterLogNormCounts"

+        }

+

+        matColData <- SummarizedExperiment::assay(sceSample, useAssayTemp)

+        matColData <- as.matrix(matColData)

+

+        if (isTRUE(pca)) {

+          if(initialDims > ncol(matColData)){

+            doPCA <- ncol(matColData)

+          }else{

+            doPCA <- initialDims

+          }

+        } else {

+            doPCA <- NULL

+        }

+        if(nNeighbors > ncol(matColData)){

+          nNeighbors <- ncol(matColData)

+        }

+

+        umapRes <- uwot::umap(t(matColData), n_neighbors = nNeighbors,

+                              learning_rate = alpha,

+                              min_dist = minDist, spread = spread,

+                              n_sgd_threads = 1, pca = doPCA,

+                              n_epochs = nIterations)

+        if (is.null(rownames(sceSample))) {

+            rownames(umapRes) <- colnames(sceSample)

+        }

+        umapDims[sceSampleInd, ] = umapRes

+    }

+    colnames(umapDims) <- c("UMAP1", "UMAP2")

+    SingleCellExperiment::reducedDim(inSCE, reducedDimName) <- umapDims

+    return(inSCE)

+}

@@ -47,7 +47,7 @@ reportDropletQC <- function(inSCE, output_file = NULL,

 #' @return .html file

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' \dontrun{

 #' sce <- runCellQC(sce)

 #' reportCellQC(inSCE = sce)

@@ -84,7 +84,7 @@ reportCellQC <- function(inSCE, output_file = NULL,

 #' @return .html file

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' \donttest{

 #' sce <- runDecontX(sce)

 #' sce <- getUMAP(sce)

@@ -165,12 +165,6 @@ reportQCTool <- function(inSCE, algorithm=c("BarcodeRankDrops",

 #' @param output_dir name of the output directory to save the rendered file. If

 #' \code{NULL} the file is stored to the current working directory.

 #' Default \code{NULL}.

-#' @examples

-#' data(scExample, package = "singleCellTK")

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

-#' sce <- runDEAnalysis(inSCE = sce, groupName1 = "Sample1", method = "DESeq2",

-#'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40, analysisName = "DESeq2")

-#' reportDiffExp(sce, study = "DESeq2", output_file = "DESeq2_res")

 #' @return .html file

 #' @export

 reportDiffExp <- function(inSCE, study,

@@ -262,7 +262,7 @@ plotDEGRegression <- function(inSCE, useResult, threshP = FALSE, labelBy = NULL,

 #' @examples

 #' data(scExample, package = "singleCellTK")

 #' \dontrun{

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDEAnalysis(inSCE = sce, groupName1 = "Sample1", method = "DESeq2",

 #'  groupName2 = "Sample2", index1 = 1:100, index2 = 101:190, analysisName = "DESeq2")

 #' plotDEGHeatmap(sce, useResult = "DESeq2", fdrThreshold = 1)

@@ -177,7 +177,7 @@ dataAnnotationColor <- function(inSCE, axis = NULL,

 #' @param ... Other arguments passed to \code{\link[ComplexHeatmap]{Heatmap}}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

-#' plotSCEHeatmap(sce[1:3,1:3])

+#' plotSCEHeatmap(sce[1:3,1:3], useAssay = "counts")

 #' @return A \code{\link[ComplexHeatmap]{Heatmap}} object

 #' @export

 #' @author Yichen Wang

@@ -15,7 +15,7 @@

 #'

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- getUMAP(inSCE = sce, useAssay = "counts", reducedDimName = "UMAP")

 #' plotUMAP(sce, shape = "No Shape", reducedDimName = "UMAP",

 #'          runUMAP = TRUE, useAssay = "counts")

@@ -127,7 +127,7 @@

 #' \code{\link{runANOVA}}

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDEAnalysis(inSCE = sce, groupName1 = "Sample1", method = "DESeq2",

 #'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40, analysisName = "DESeq2")

 #' @return Input SCE object with \code{metadata(inSCE)} updated with name

@@ -190,7 +190,7 @@ runDEAnalysis <- function(method = 'MAST', ...){

 #' Default \code{FALSE}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDESeq2(inSCE = sce, groupName1 = "Sample1",

 #'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40, analysisName = "DESeq2")

 #'

@@ -317,7 +317,7 @@ runDESeq2 <- function(inSCE, useAssay = 'counts', index1 = NULL,

 #' Default \code{FALSE}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce@assays@data$logcounts <- log10(counts(sce) + 1)

 #' sce <- runLimmaDE(inSCE = sce, groupName1 = "Sample1",

 #'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40, analysisName = "Limma")

@@ -440,7 +440,7 @@ runLimmaDE <- function(inSCE, useAssay = 'logcounts', index1 = NULL,

 #' Default \code{FALSE}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce@assays@data$logcounts <- log10(counts(sce) + 1)

 #' sce <- runANOVA(inSCE = sce, groupName1 = "Sample1",

 #'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40,

@@ -586,7 +586,7 @@ runANOVA <- function(inSCE, useAssay = 'logcounts', index1 = NULL,

 #' Default \code{FALSE}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce@assays@data$logcounts <- log10(counts(sce) + 1)

 #' sce <- runMAST(inSCE = sce, groupName1 = "Sample1",

 #'  groupName2 = "Sample2", index1 = 1:20, index2 = 21:40, analysisName = "MAST")

@@ -20,7 +20,7 @@

 #' of \code{inSCE}.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' \donttest{

 #' sce <- runCellQC(sce)

 #' }

@@ -118,7 +118,7 @@

 #' QC stats stored in the inSCE.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sampleSummaryStats(sce, simple = TRUE)

 #' @importFrom magrittr %>%

 #' @export

@@ -23,7 +23,7 @@

 #'  Please refer to the documentation of \link[scds]{cxds} for details.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runCxds(sce)

 #' @export

 #' @importFrom SummarizedExperiment colData colData<-

@@ -31,9 +31,9 @@

 runCxds <- function(inSCE,

     sample = NULL,

     seed = 12345,

-    ntop = 500, 

+    ntop = 500,

     binThresh = 0,

-    verb = FALSE, 

+    verb = FALSE,

     retRes = FALSE,

     estNdbl = FALSE,

     useAssay = "counts") {

@@ -48,7 +48,7 @@ runCxds <- function(inSCE,

     }

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

-    

+

     ## Getting current arguments

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

     argsList <- mget(names(formals()),sys.frame(sys.nframe()))

@@ -71,19 +71,19 @@ runCxds <- function(inSCE,

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

         counts(sceSample) <- .convertToMatrix(mat)

-        

+

         result <- NULL

         nGene <- 500

         while(!inherits(result, "SingleCellExperiment") & nGene > 0) {

           try({result <- withr::with_seed(seed, scds::cxds(sce = sceSample,

-                                                           ntop = nGene, 

-                                                           binThresh = binThresh, 

-                                                           verb = verb, 

+                                                           ntop = nGene,

+                                                           binThresh = binThresh,

+                                                           verb = verb,

                                                            retRes = retRes,

                                                            estNdbl = estNdbl))}, silent = TRUE)

           nGene <- nGene - 100

-        }  

-        

+        }

+

         if (!inherits(result, "try-error") & !is.null(result)) {

           if ("cxds_call" %in% colnames(SummarizedExperiment::colData(result))) {

               output[sceSampleInd, ] <- SummarizedExperiment::colData(result)[,

@@ -95,12 +95,12 @@ runCxds <- function(inSCE,

         } else {

           output[sceSampleInd, ] <- NA

           warning(paste0("'cxds' from package 'scds' did not complete successfully for sample", samples[i]))

-        }            

+        }

     }

     colnames(output) <- paste0("scds_", colnames(output))

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

-    

+

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

     inSCE@metadata$runCxds$packageVersion <- utils::packageDescription("scds")$Version

@@ -120,7 +120,7 @@ runCxds <- function(inSCE,

 #'  separately. If NULL, then all cells will be processed together.

 #'  Default NULL.

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

-#' @param ntop See \link[scds]{bcds} for more information. Default \code{500}. 

+#' @param ntop See \link[scds]{bcds} for more information. Default \code{500}.

 #' @param srat See \link[scds]{bcds} for more information. Default \code{1}.

 #' @param verb See \link[scds]{bcds} for more information. Default \code{FALSE}.

 #' @param retRes See \link[scds]{bcds} for more information. Default \code{FALSE}.

@@ -135,7 +135,7 @@ runCxds <- function(inSCE,

 #'  Please refer to the documentation of \link[scds]{bcds} for details.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runBcds(sce)

 #' @export

 #' @importFrom SummarizedExperiment colData colData<-

@@ -143,11 +143,11 @@ runCxds <- function(inSCE,

 runBcds <- function(inSCE,

     sample = NULL,

     seed = 12345,

-    ntop = 500, 

-    srat = 1, 

+    ntop = 500,

+    srat = 1,

     verb = FALSE,

     retRes = FALSE,

-    nmax = "tune", 

+    nmax = "tune",

     varImp = FALSE,

     estNdbl = FALSE,

     useAssay = "counts"

@@ -163,7 +163,7 @@ runBcds <- function(inSCE,

     }

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

-    

+

     ## Getting current arguments

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

     argsList <- mget(names(formals()),sys.frame(sys.nframe()))

@@ -186,22 +186,22 @@ runBcds <- function(inSCE,

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

         counts(sceSample) <- .convertToMatrix(mat)

-        

+

         result <- NULL

         nGene <- 500

         while(!inherits(result, "SingleCellExperiment") & nGene > 0) {

           try({result <- withr::with_seed(seed,

             scds::bcds(sce = sceSample,

                        ntop = nGene,

-                       srat = srat, 

+                       srat = srat,

                        verb = verb,

                        retRes = retRes,

-                       nmax = nmax, 

+                       nmax = nmax,

                        varImp = varImp,

                        estNdbl = estNdbl

             ))}, silent = TRUE)

           nGene <- nGene - 100

-        }  

+        }

         if (!inherits(result, "try-error") & !is.null(result)) {

           if ("bcds_call" %in% colnames(SummarizedExperiment::colData(result))) {

@@ -214,13 +214,13 @@ runBcds <- function(inSCE,

         } else {

           output[sceSampleInd, ] <- NA

           warning(paste0("'bcds' from package 'scds' did not complete successfully for sample", samples[i]))

-        }  

- 

+        }

+

     }

     colnames(output) <- paste0("scds_", colnames(output))

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

-    

+

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

     inSCE@metadata$runBcds$packageVersion <- utils::packageDescription("scds")$Version

@@ -241,7 +241,7 @@ runBcds <- function(inSCE,

 #'  Default NULL.

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

 #' @param nTop The number of top varialbe genes to consider. Used in both \code{csds}

-#' and \code{bcds}. Default \code{500}. 

+#' and \code{bcds}. Default \code{500}.

 #' @param cxdsArgs See \link[scds]{cxds_bcds_hybrid} for more information. Default \code{NULL}.

 #' @param bcdsArgs See \link[scds]{cxds_bcds_hybrid} for more information. Default \code{NULL}.

 #' @param verb See \link[scds]{cxds_bcds_hybrid} for more information. Default \code{FALSE}.

@@ -256,7 +256,7 @@ runBcds <- function(inSCE,

 #'  details.

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runCxdsBcdsHybrid(sce)

 #' @export

 #' @importFrom SummarizedExperiment colData colData<-

@@ -266,7 +266,7 @@ runCxdsBcdsHybrid <- function(inSCE,

     seed = 12345,

     nTop = 500,

     cxdsArgs = list(),

-    bcdsArgs = list(), 

+    bcdsArgs = list(),

     verb = FALSE,

     estNdbl = FALSE,

     force = FALSE,

@@ -309,14 +309,14 @@ runCxdsBcdsHybrid <- function(inSCE,

         result <- NULL

         nGene <- 500

         while(!inherits(result, "SingleCellExperiment") & nGene > 0) {

-          try({result <- withr::with_seed(seed, scds::cxds_bcds_hybrid(sce = sceSample, 

-                                                                       cxdsArgs=c(list(ntop = nGene), cxdsArgs), 

-                                                                       bcdsArgs=c(list(ntop = nGene), bcdsArgs), 

+          try({result <- withr::with_seed(seed, scds::cxds_bcds_hybrid(sce = sceSample,

+                                                                       cxdsArgs=c(list(ntop = nGene), cxdsArgs),

+                                                                       bcdsArgs=c(list(ntop = nGene), bcdsArgs),

                                                                        verb = verb,

                                                                        estNdbl = estNdbl,

                                                                        force = force))}, silent = TRUE)

           nGene <- nGene - 100

-        }  

+        }

         if (!inherits(result, "try-error") & !is.null(result)) {

           if ("hybrid_call" %in% colnames(SummarizedExperiment::colData(result))) {

@@ -329,12 +329,12 @@ runCxdsBcdsHybrid <- function(inSCE,

         } else {

           output[sceSampleInd, ] <- NA

           warning(paste0("'cxds_bcds_hybrid' from package 'scds' did not complete successfully for sample", samples[i]))

-        }   

+        }

     }

     colnames(output) <- paste0("scds_", colnames(output))

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

-    

+

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

     inSCE@metadata$runCxdsBcdsHybrid$packageVersion <- utils::packageDescription("scds")$Version

@@ -88,7 +88,7 @@

 #' @seealso \link[scran]{doubletCells}

 #' @examples

 #' data(scExample, package = "singleCellTK")

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

+#' sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")

 #' sce <- runDoubletCells(sce)

 #' @export

 #' @importFrom SummarizedExperiment colData colData<-

@@ -1,231 +1,231 @@

-#' @title Find doublets using \code{scrublet}.

-#' @description A wrapper function that calls \code{scrub_doublets} from python

-#'  module \code{scrublet}. Simulates doublets from the observed data and uses

-#'  a k-nearest-neighbor classifier to calculate a continuous

-#'  \code{scrublet_score} (between 0 and 1) for each transcriptome. The score

-#'  is automatically thresholded to generate \code{scrublet_call}, a boolean

-#'  array that is \code{TRUE} for predicted doublets and \code{FALSE}

-#'  otherwise.

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

-#'  Needs \code{counts} in assays slot.

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

-#'  Scrublet will be run on cells from each sample separately. If NULL, then

-#'  all cells will be processed together. Default \code{NULL}.

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

-#'  'counts'.

-#' @param simDoubletRatio Numeric. Number of doublets to simulate relative to

-#'  the number of observed transcriptomes. Default 2.0.

-#' @param nNeighbors Integer. Number of neighbors used to construct the KNN

-#'  graph of observed transcriptomes and simulated doublets. If \code{NULL},

-#'  this is set to \code{round(0.5 * sqrt(n_cells))}. Default \code{NULL}.

-#' @param minDist Float Determines how tightly UMAP packs points together. If \code{NULL},

-#'  this is set to 0.1. Default \code{NULL}.

-#' @param expectedDoubletRate The estimated doublet rate for the experiment.

-#'  Default 0.1.

-#' @param stdevDoubletRate Uncertainty in the expected doublet rate.

-#'  Default 0.02.

-#' @param syntheticDoubletUmiSubsampling Numeric. Rate for sampling UMIs

-#'  when creating synthetic doublets. If 1.0, each doublet is created by simply

-#'  adding the UMIs from two randomly sampled observed transcriptomes. For

-#'  values less than 1, the UMI counts are added and then randomly sampled at

-#'  the specified rate. Defuault: 1.0.

-#' @param useApproxNeighbors Boolean. Use approximate nearest neighbor method

-#'  (\code{annoy}) for the KNN classifier. Default \code{TRUE}.

-#' @param distanceMetric Character. Distance metric used when finding nearest

-#'  neighbors.

-#'  For list of valid values, see the documentation for \code{annoy} (if

-#'  \code{useApproxNeighbors} is \code{TRUE}) or

-#'  \code{sklearn.neighbors.NearestNeighbors} (if \code{useApproxNeighbors} is

-#'  \code{FALSE}). Default "euclidean".

-#' @param getDoubletNeighborParents Boolean. If \code{TRUE}, return the

-#'  parent transcriptomes that generated the doublet neighbors of each

-#'  observed transcriptome. This information can be used to infer the cell

-#'  states that generated a given doublet state. Default \code{FALSE}.

-#' @param minCounts Numeric. Used for gene filtering prior to PCA. Genes

-#'  expressed at fewer than \code{minCounts} in fewer than \code{minCells}

-#'  (see below) are excluded. Default 3.

-#' @param minCells Integer. Used for gene filtering prior to PCA. Genes

-#'  expressed at fewer than \code{minCounts} (see above) in fewer than

-#'  \code{minCells} are excluded. Default 3.

-#' @param minGeneVariabilityPctl Numeric. Used for gene filtering prior to

-#'  PCA. Keep the most highly variable genes (in the top

-#'  minGeneVariabilityPctl percentile), as measured by the v-statistic

-#'  (\emph{Klein et al., Cell 2015}). Default 85.

-#' @param logTransform Boolean. If \code{TRUE}, log-transform the counts matrix

-#'  (log10(1+TPM)). \code{sklearn.decomposition.TruncatedSVD} will be used for

-#'  dimensionality reduction, unless \code{meanCenter} is \code{TRUE}.

-#'  Default \code{FALSE}.

-#' @param meanCenter If \code{TRUE}, center the data such that each gene has a

-#'  mean of 0. \code{sklearn.decomposition.PCA} will be used for

-#'  dimensionality reduction. Default \code{TRUE}.

-#' @param normalizeVariance Boolean. If \code{TRUE}, normalize the data such

-#'  that each gene has a variance of 1.

-#'  \code{sklearn.decomposition.TruncatedSVD} will be used for dimensionality

-#'  reduction, unless \code{meanCenter} is \code{TRUE}. Default \code{TRUE}.

-#' @param nPrinComps Integer. Number of principal components used to embed

-#'  the transcriptomes prior to k-nearest-neighbor graph construction.

-#'  Default 30.

-#' @param tsneAngle Float. Determines angular size of a distant node as measured

-#'  from a point in the t-SNE plot. If default, it is set to 0.5 Default \code{NULL}.

-#' @param tsnePerplexity Integer. The number of nearest neighbors that

-#'  is used in other manifold learning algorithms.

-#'  If default, it is set to 30. Default \code{NULL}.

-#' @param verbose Boolean. If \code{TRUE}, print progress updates. Default

-#'  \code{TRUE}.

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

-#' @return A \link[SingleCellExperiment]{SingleCellExperiment} object with

-#'  \code{scrub_doublets} output appended to the

-#'  \link[SummarizedExperiment]{colData} slot. The columns include

-#'  \emph{scrublet_score} and \emph{scrublet_call}.

-#' @examples

-#' data(scExample, package = "singleCellTK")

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

-#' sce <- runScrublet(sce)

-#' @export

-#' @importFrom reticulate py_module_available py_set_seed import

-#' @importFrom SummarizedExperiment colData colData<-

-#' @importFrom SingleCellExperiment reducedDim

-runScrublet <- function(inSCE,

-  sample = NULL,

-  useAssay = "counts",

-  simDoubletRatio = 2.0,

-  nNeighbors = NULL,

-  minDist = NULL,

-  expectedDoubletRate = 0.1,

-  stdevDoubletRate = 0.02,

-  syntheticDoubletUmiSubsampling = 1.0,

-  useApproxNeighbors = TRUE,

-  distanceMetric = "euclidean",

-  getDoubletNeighborParents = FALSE,

-  minCounts = 3,

-  minCells = 3L,

-  minGeneVariabilityPctl = 85,

-  logTransform = FALSE,

-  meanCenter = TRUE,

-  normalizeVariance = TRUE,

-  nPrinComps = 30L,

-  tsneAngle = NULL,

-  tsnePerplexity = NULL,

-  verbose = TRUE,

-  seed = 12345) {

-

-  if (!reticulate::py_module_available(module = "scrublet")) {

-    warning("Cannot find python module 'scrublet', please install Conda and",

-      " run sctkPythonInstallConda() or run sctkPythonInstallVirtualEnv().",

-      "If one of these have been previously run to install the modules,",

-      "make sure to run selectSCTKConda() or selectSCTKVirtualEnvironment(),",

-      " respectively, if R has been restarted since the module installation.",

-      " Alternatively, Scrublet can be installed on the local machine",

-      "with pip (e.g. pip install scrublet) and then the 'use_python()'",

-      " function from the 'reticulate' package can be used to select the",

-      " correct Python environment.")

-    return(inSCE)

-  }

-

-  if (!is.null(seed)) {

-    reticulate::py_set_seed(seed = seed)

-  }

-

-  if (!is.null(sample)) {

-    if (length(sample) != ncol(inSCE)) {

-      stop("'sample' must be the same length as the number of",

-        " columns in 'inSCE'")

-    }

-  } else {

-    sample = rep(1, ncol(inSCE))

-  }

-

-  message(paste0(date(), " ... Running 'scrublet'"))

-

-  ##  Getting current arguments values

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

-  argsList <- mget(names(formals()),sys.frame(sys.nframe()))

-

-  ## Define result matrix for all samples

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

-    scrublet_score = numeric(ncol(inSCE)),

-    scrublet_call = logical(ncol(inSCE)))

-

-  ## Loop through each sample and run scrublet

-  error <- try({

-    samples <- unique(sample)

-    umapDims <- matrix(ncol = 2,

-                       nrow = ncol(inSCE))

-    rownames(umapDims) = colnames(inSCE)

-    colnames(umapDims) = c("UMAP_1", "UMAP_2")

-

-    tsneDims <- matrix(ncol = 2,

-                       nrow = ncol(inSCE))

-    rownames(tsneDims) = colnames(inSCE)

-    colnames(tsneDims) = c("TSNE_1", "TSNE_2")

-

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

-      sceSampleInd <- sample == samples[i]

-      sceSample <- inSCE[, sceSampleInd]

-

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

-      mat <- .convertToMatrix(mat)

-

-      scr <- scrublet$Scrublet(counts_matrix = t(mat),

-        sim_doublet_ratio = simDoubletRatio,

-        n_neighbors = nNeighbors,

-        expected_doublet_rate = expectedDoubletRate,

-        stdev_doublet_rate = stdevDoubletRate)

-

-      result <- scr$scrub_doublets(

-        synthetic_doublet_umi_subsampling = syntheticDoubletUmiSubsampling,

-        use_approx_neighbors = useApproxNeighbors,

-        distance_metric = distanceMetric,

-        get_doublet_neighbor_parents = getDoubletNeighborParents,

-        min_counts = minCounts,

-        min_cells = as.integer(minCells),

-        min_gene_variability_pctl = minGeneVariabilityPctl,

-        log_transform = logTransform,

-        mean_center = meanCenter,

-        normalize_variance = normalizeVariance,

-        n_prin_comps = as.integer(nPrinComps),

-        verbose = verbose)

-

-      output[sceSampleInd, "scrublet_score"] <- result[[1]]

-      output[sceSampleInd, "scrublet_call"] <- result[[2]]

-

-      ## Extract UMAP and TSNE coordinates

-      if (is.null(nNeighbors) && is.null(minDist)){

-        umap_coordinates <- scrublet$get_umap(scr$manifold_obs_)

-      }else {

-        umap_coordinates <- scrublet$get_umap(scr$manifold_obs_,

-                                              n_neighbors=as.integer(nNeighbors),

-                                              min_dist=minDist)

-      }

-      umapDims[sceSampleInd, ] <- umap_coordinates

-

-    if (is.null(tsneAngle) && is.null(tsnePerplexity)){

-      tsne_coordinates <- scrublet$get_tsne(scr$manifold_obs_)

-    }else {

-      tsne_coordinates <- scrublet$get_tsne(scr$manifold_obs_,

-                                            angle=tsneAngle,

-                                            perplexity=as.integer(tsnePerplexity))

-    }

-    tsneDims[sceSampleInd, ] <- tsne_coordinates

-

-  }

-

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

-  }, silent = TRUE)

-

-  if (inherits(error, "try-error")) {

-    warning("Scrublet did not complete successfully. Returning SCE without",

-      " making any changes. Error given by Scrublet: \n\n", error)

-  }

-

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

-

-  ## add scrublet version to metadata

-  version <- pkg_resources$require("scrublet")[[1]]

-  inSCE@metadata$scrublet$packageVersion <- version

-  reducedDim(inSCE,'scrublet_TSNE') <- tsneDims

-  reducedDim(inSCE,'scrublet_UMAP') <- umapDims

-

-  return(inSCE)

-}

-

+#' @title Find doublets using \code{scrublet}.

+#' @description A wrapper function that calls \code{scrub_doublets} from python

+#'  module \code{scrublet}. Simulates doublets from the observed data and uses

+#'  a k-nearest-neighbor classifier to calculate a continuous

+#'  \code{scrublet_score} (between 0 and 1) for each transcriptome. The score

+#'  is automatically thresholded to generate \code{scrublet_call}, a boolean

+#'  array that is \code{TRUE} for predicted doublets and \code{FALSE}

+#'  otherwise.

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

+#'  Needs \code{counts} in assays slot.