new file mode 100644

@@ -0,0 +1,7 @@

+R/importOptimus.R

+R/scrublet_doubletDetection.R

+R/sce2adata.R

+R/exportSCEtoAnndata.R

+R/reticulate_setup.R

+R/importAnnData.R

+

@@ -1,12 +1,12 @@

 #' Run t-SNE embedding with Rtsne method

-#' @description T-Stochastic Neighbour Embedding (t-SNE) algorithm is commonly 

-#' for 2D visualization of single-cell data. This function wraps the 

-#' Rtsne \code{\link[Rtsne]{Rtsne}} function. 

-#' 

+#' @description T-Stochastic Neighbour Embedding (t-SNE) algorithm is commonly

+#' for 2D visualization of single-cell data. This function wraps the

+#' Rtsne \code{\link[Rtsne]{Rtsne}} function.

+#'

 #' With this funciton, users can create tSNE embedding directly from raw count

-#' matrix, with necessary preprocessing including normalization, scaling, 

-#' dimension reduction all automated. Yet we still recommend having the PCA as 

-#' input, so that the result can match with the clustering based on the same 

+#' matrix, with necessary preprocessing including normalization, scaling,

+#' dimension reduction all automated. Yet we still recommend having the PCA as

+#' input, so that the result can match with the clustering based on the same

 #' input PCA, and will be much faster.

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

 #' @param useAssay Assay to use for tSNE computation. If \code{useAltExp} is

@@ -21,33 +21,33 @@

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

 #' generating the tSNE via \code{\link{scaterlogNormCounts}}. Ignored when using

 #' \code{useReducedDim}. Default \code{FALSE}.

-#' @param useFeatureSubset Subset of feature to use for dimension reduction. A 

+#' @param useFeatureSubset Subset of feature to use for dimension reduction. A

 #' character string indicating a \code{rowData} variable that stores the logical

-#' vector of HVG selection, or a vector that can subset the rows of 

+#' vector of HVG selection, or a vector that can subset the rows of

 #' \code{inSCE}. Default \code{NULL}.

 #' @param nTop Automatically detect this number of variable features to use for

-#' dimension reduction. Ignored when using \code{useReducedDim} or using 

+#' dimension reduction. Ignored when using \code{useReducedDim} or using

 #' \code{useFeatureSubset}. Default \code{2000}.

-#' @param center Whether data should be centered before PCA is applied. Ignored 

+#' @param center Whether data should be centered before PCA is applied. Ignored

 #' when using \code{useReducedDim}. Default \code{TRUE}.

-#' @param scale Whether data should be scaled before PCA is applied. Ignored 

+#' @param scale Whether data should be scaled before PCA is applied. Ignored

 #' when using \code{useReducedDim}. Default \code{TRUE}.

-#' @param pca Whether an initial PCA step should be performed. Ignored when 

+#' @param pca Whether an initial PCA step should be performed. Ignored when

 #' using \code{useReducedDim}. Default \code{TRUE}.

 #' @param partialPCA Whether truncated PCA should be used to calculate principal

-#' components (requires the irlba package). This is faster for large input 

+#' components (requires the irlba package). This is faster for large input

 #' matrices. Ignored when using \code{useReducedDim}. Default \code{FALSE}.

 #' @param initialDims Number of dimensions from PCA to use as input in tSNE.

 #' Default \code{25}.

-#' @param theta Numeric value for speed/accuracy trade-off (increase for less 

+#' @param theta Numeric value for speed/accuracy trade-off (increase for less

 #' accuracy), set to \code{0.0} for exact TSNE. Default \code{0.5}.

-#' @param perplexity perplexity parameter. Should not be bigger than 

-#' \code{3 * perplexity < ncol(inSCE) - 1}. Default \code{30}. See 

-#' \code{\link[Rtsne]{Rtsne}} details for interpretation. 

+#' @param perplexity perplexity parameter. Should not be bigger than

+#' \code{3 * perplexity < ncol(inSCE) - 1}. Default \code{30}. See

+#' \code{\link[Rtsne]{Rtsne}} details for interpretation.

 #' @param nIterations maximum iterations. Default \code{1000}.

-#' @param numThreads Integer, number of threads to use using OpenMP, Default 

+#' @param numThreads Integer, number of threads to use using OpenMP, Default

 #' \code{1}. \code{0} corresponds to using all available cores.

-#' @param seed Random seed for reproducibility of tSNE results. 

+#' @param seed Random seed for reproducibility of tSNE results.

 #' Default \code{NULL} will use global seed in use by the R environment.

 #' @return A \linkS4class{SingleCellExperiment} object with tSNE computation

 #' updated in \code{reducedDim(inSCE, reducedDimName)}.

@@ -62,16 +62,16 @@

 #' # Run from PCA

 #' sce <- scaterlogNormCounts(sce, "logcounts")

 #' sce <- runModelGeneVar(sce)

-#' sce <- scaterPCA(sce, useAssay = "logcounts", 

+#' sce <- scaterPCA(sce, useAssay = "logcounts",

 #'                  useFeatureSubset = "HVG_modelGeneVar2000", scale = TRUE)

 #' sce <- getTSNE(sce, useReducedDim = "PCA")

 #' }

 #' @importFrom S4Vectors metadata<-

-getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL, 

-                    useAltExp = NULL, reducedDimName = "TSNE", logNorm = FALSE, 

-                    useFeatureSubset = NULL, nTop = 2000, center = TRUE, 

-                    scale = TRUE, pca = TRUE, partialPCA = FALSE, 

-                    initialDims = 25, theta = 0.5, perplexity = 30, 

+getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

+                    useAltExp = NULL, reducedDimName = "TSNE", logNorm = FALSE,

+                    useFeatureSubset = NULL, nTop = 2000, center = TRUE,

+                    scale = TRUE, pca = TRUE, partialPCA = FALSE,

+                    initialDims = 25, theta = 0.5, perplexity = 30,

                     nIterations = 1000, numThreads = 1, seed = NULL){

   params <- as.list(environment())

   params$inSCE <- NULL

@@ -80,11 +80,11 @@ getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

     stop("Please use a SingleCellExperiment object")

   }

   # matrix specification check

-  # When useReducedDim, never useAssay; 

+  # When useReducedDim, never useAssay;

   # when useAltExp, useAssay/reducedDim from there

   if (is.null(useAssay) && is.null(useReducedDim)) {

     stop("`useAssay` and `useReducedDim` cannot be NULL at the same time.")

-  } 

+  }

   if (!is.null(useAltExp)) {

     if (!useAltExp %in% SingleCellExperiment::altExpNames(inSCE)) {

       stop("Specified `useAltExp` not found.")

@@ -104,15 +104,15 @@ getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

       stop("Specified `useAssay` not found.")

     }

   }

-  

+

   if (!is.null(useAssay)) {

     if (isTRUE(logNorm)) {

-      sce <- scaterlogNormCounts(sce, "logcounts")

+      sce <- scaterlogNormCounts(sce, assayName = "logcounts", useAssay = useAssay)

       useAssay <- "logcounts"

     }

     if (!is.null(useFeatureSubset)) {

-      subset_row <- .parseUseFeatureSubset(inSCE, useFeatureSubset, 

-                                           altExpObj = sce, 

+      subset_row <- .parseUseFeatureSubset(inSCE, useFeatureSubset,

+                                           altExpObj = sce,

                                            returnType = "logical")

       sce <- sce[subset_row,]

     } else {

@@ -120,7 +120,7 @@ getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

         suppressMessages({

           sce <- runFeatureSelection(sce, useAssay, method = "modelGeneVar")

         })

-        sce <- setTopHVG(sce, method = "modelGeneVar", hvgNumber = nTop, 

+        sce <- setTopHVG(sce, method = "modelGeneVar", hvgNumber = nTop,

                          featureSubsetName = "tsneHVG")

         sce <- subsetSCERows(sce, rowData = "tsneHVG", returnAsAltExp = FALSE)

       }

@@ -139,7 +139,7 @@ getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

       mat <- mat[,seq(initialDims)]

     }

   }

-  

+

   if (is.null(perplexity)){

     perplexity <- floor(ncol(inSCE) / 5)

   }

@@ -148,11 +148,11 @@ getTSNE <- function(inSCE, useAssay = "logcounts", useReducedDim = NULL,

   message(paste0(date(), " ... Computing Rtsne."))

   .withSeed(seed, {

     tsneOut <- Rtsne::Rtsne(mat, pca_scale = scale, pca_center = center,

-                            pca = pca, partial_pca = partialPCA, 

+                            pca = pca, partial_pca = partialPCA,

                             perplexity = perplexity,

                             initial_dims = initialDims,

-                            max_iter = nIterations, theta = theta, 

-                            num_threads = numThreads) 

+                            max_iter = nIterations, theta = theta,

+                            num_threads = numThreads)

   })

   tsneOut <- tsneOut$Y

   rownames(tsneOut) <- colnames(inSCE)

@@ -44,7 +44,7 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

     } else if (entry$type == "starSolo") {

       newSce <- importSTARsolo(

         STARsoloDirs = entry$params$STARsoloDirs,

-        samples = entry$params$amples,

+        samples = entry$params$samples,

         delayedArray = delayedArray

       )

     } else if (entry$type == "busTools") {

@@ -96,7 +96,7 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

       }

     }

-    

+

     # Begin Set Tags

     if(entry$type %in% c("cellRanger2", "cellRanger3", "starSolo", "busTools", "seqc", "optimus", "example")){

       newSce <- expSetDataTag(

@@ -114,7 +114,7 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

             assayType = "raw",

             assays = "counts")

         }, silent = TRUE)

-        

+

         try({

           logcounts(newSce)

           newSce <- expSetDataTag(

@@ -122,7 +122,7 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

             assayType = "transformed",

             assays = "logcounts")

         }, silent = TRUE)

-        

+

         try({

           normcounts(newSce)

           newSce <- expSetDataTag(

@@ -130,7 +130,7 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

             assayType = "transformed",

             assays = "normcounts")

         }, silent = TRUE)

-        

+

         try({

           celda::decontXcounts(newSce)

           newSce <- expSetDataTag(

@@ -138,21 +138,21 @@ importMultipleSources <- function(allImportEntries, delayedArray = FALSE) {

             assayType = "raw",

             assays = "decontXcounts")

         }, silent = TRUE)

-        

+

         untaggedAssays <- SummarizedExperiment::assayNames(newSce)

         untaggedAssays <- untaggedAssays[! untaggedAssays %in% c('counts', 'logcounts', 'normcounts', 'decontX')]

-        

+

         newSce <- expSetDataTag(

           inSCE = newSce,

           assayType = "uncategorized",

-          assays = untaggedAssays) 

+          assays = untaggedAssays)

       }

       # End Set Tags

     }

-    

+

     sceObjs = c(sceObjs, list(newSce))

   }

-  

+

   return(combineSCE(sceList = sceObjs,

                     by.r = Reduce(base::intersect, lapply(sceObjs, function(x) { colnames(rowData(x))})),

                     by.c = Reduce(base::intersect, lapply(sceObjs, function(x) { colnames(colData(x))})),

@@ -69,7 +69,7 @@

     }

     sce <- do.call(SingleCellExperiment::cbind, res)

-    

+

     if (isTRUE(rowNamesDedup)) {

         if (any(duplicated(rownames(sce)))) {

             message("Duplicated gene names found, adding '-1', '-2', ",

@@ -77,7 +77,7 @@

         }

         sce <- dedupRowNames(sce)

     }

-    

+

     # Load metrics summary and store in sce

     metrics_summary <- .importMetricsStarSolo(STARsoloDirs, samples, "Gene", "Summary.csv")

     # sce <- setSampleSummaryStatsTable(sce, "starsolo", metrics_summary)

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

 #'  \link[base]{matrix} function). Default "Matrix".

 #' @param delayedArray Boolean. Whether to read the expression matrix as

 #'  \link{DelayedArray} object or not. Default \code{FALSE}.

-#' @param rowNamesDedup Boolean. Whether to deduplicate rownames. Default 

+#' @param rowNamesDedup Boolean. Whether to deduplicate rownames. Default

 #'  \code{TRUE}.

 #' @return A \code{SingleCellExperiment} object containing the count

 #'  matrix, the gene annotation, and the cell annotation.

@@ -195,7 +195,7 @@ importSTARsolo <- function(

   if(!identical(length(samplePaths), length(sampleNames))){

     stop("Vectors samplePaths and sampleNames must be equal in length.")

   }

-  

+

   # Processing

   metrics_summary <- list()

   for(i in seq(samplePaths)){

@@ -210,7 +210,7 @@ importSTARsolo <- function(

       colnames(metrics_summary[[i]]) <- ms_colnames_union

     }

   }

-  

+

   # Merge StarSolo summary csv files from all/multiple samples into a single data.frame

   for(i in seq_along(metrics_summary)){

     metrics_summary[[i]] <- as.data.frame(t(metrics_summary[[i]]))

@@ -19,7 +19,7 @@

 #' plotPCA(mouseBrainSubsetSCE, colorBy = "level1class",

 #'         reducedDimName = "PCA_counts")

 #'

-plotPCA <- function(inSCE, colorBy="No Color", shape="No Shape", pcX="PC1",

+plotPCA <- function(inSCE, colorBy=NULL, shape=NULL, pcX="PC1",

                     pcY="PC2", reducedDimName="PCA", runPCA=FALSE,

                     useAssay="logcounts"){

   if(!(reducedDimName %in% names(SingleCellExperiment::reducedDims(inSCE)))){

@@ -40,17 +40,11 @@ plotPCA <- function(inSCE, colorBy="No Color", shape="No Shape", pcX="PC1",

     stop("pcY dimension ", pcY, " is not in the reducedDim data")

   }

-  

+

   # Need to add back in variances in the plot axis labels

   pcXlab <- pcX

   pcYlab <- pcY

-  if (colorBy == "No Color"){

-    colorBy <- NULL

-  }

-  if (shape == "No Shape"){

-    shape <- NULL

-  }

   if (!is.null(colorBy)){

     pcaDf$color <- SingleCellExperiment::colData(inSCE)[, colorBy]

   }

@@ -17,7 +17,7 @@

 #' plotTSNE(mouseBrainSubsetSCE, colorBy = "level1class",

 #'          reducedDimName = "TSNE_counts")

 #'

-plotTSNE <- function(inSCE, colorBy="No Color", shape="No Shape",

+plotTSNE <- function(inSCE, colorBy=NULL, shape=NULL,

                      reducedDimName="TSNE", runTSNE=FALSE,

                      useAssay="logcounts"){

   if(!(reducedDimName %in% names(SingleCellExperiment::reducedDims(inSCE)))){

@@ -38,12 +38,6 @@ plotTSNE <- function(inSCE, colorBy="No Color", shape="No Shape",

   colnames(tsneDf)[2] <- "tSNE2"

   xdim <- colnames(tsneDf)[1]

   ydim <- colnames(tsneDf)[2]

-  if (colorBy == "No Color"){

-    colorBy <- NULL

-  }

-  if (shape == "No Shape"){

-    shape <- NULL

-  }

   if (!is.null(colorBy)){

     tsneDf$color <- SingleCellExperiment::colData(inSCE)[, colorBy]

   }

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

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

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

 #'

-plotUMAP <- function(inSCE, colorBy = "No Color", shape = "No Shape",

+plotUMAP <- function(inSCE, colorBy = NULL, shape = NULL,

                      reducedDimName = "UMAP", runUMAP = FALSE,

                      useAssay = "logcounts"){

   if(!(reducedDimName %in% names(SingleCellExperiment::reducedDims(inSCE)))){

@@ -41,12 +41,7 @@ plotUMAP <- function(inSCE, colorBy = "No Color", shape = "No Shape",

   colnames(UMAPDf)[2] <- "UMAP2"

   xdim <- colnames(UMAPDf)[1]

   ydim <- colnames(UMAPDf)[2]

-  if (colorBy == "No Color"){

-    colorBy <- NULL

-  }

-  if (shape == "No Shape"){

-    shape <- NULL

-  }

+

   if (!is.null(colorBy)){

     UMAPDf$color <- SingleCellExperiment::colData(inSCE)[, colorBy]

   }

@@ -45,44 +45,44 @@ reductions. Default \code{"TSNE"}.}

 generating the tSNE via \code{\link{scaterlogNormCounts}}. Ignored when using

 \code{useReducedDim}. Default \code{FALSE}.}

-\item{useFeatureSubset}{Subset of feature to use for dimension reduction. A 

+\item{useFeatureSubset}{Subset of feature to use for dimension reduction. A

 character string indicating a \code{rowData} variable that stores the logical

-vector of HVG selection, or a vector that can subset the rows of 

+vector of HVG selection, or a vector that can subset the rows of

 \code{inSCE}. Default \code{NULL}.}

 \item{nTop}{Automatically detect this number of variable features to use for

-dimension reduction. Ignored when using \code{useReducedDim} or using 

+dimension reduction. Ignored when using \code{useReducedDim} or using

 \code{useFeatureSubset}. Default \code{2000}.}

-\item{center}{Whether data should be centered before PCA is applied. Ignored 

+\item{center}{Whether data should be centered before PCA is applied. Ignored

 when using \code{useReducedDim}. Default \code{TRUE}.}

-\item{scale}{Whether data should be scaled before PCA is applied. Ignored 

+\item{scale}{Whether data should be scaled before PCA is applied. Ignored

 when using \code{useReducedDim}. Default \code{TRUE}.}

-\item{pca}{Whether an initial PCA step should be performed. Ignored when 

+\item{pca}{Whether an initial PCA step should be performed. Ignored when

 using \code{useReducedDim}. Default \code{TRUE}.}

 \item{partialPCA}{Whether truncated PCA should be used to calculate principal

-components (requires the irlba package). This is faster for large input 

+components (requires the irlba package). This is faster for large input

 matrices. Ignored when using \code{useReducedDim}. Default \code{FALSE}.}

 \item{initialDims}{Number of dimensions from PCA to use as input in tSNE.

 Default \code{25}.}

-\item{theta}{Numeric value for speed/accuracy trade-off (increase for less 

+\item{theta}{Numeric value for speed/accuracy trade-off (increase for less

 accuracy), set to \code{0.0} for exact TSNE. Default \code{0.5}.}

-\item{perplexity}{perplexity parameter. Should not be bigger than 

-\code{3 * perplexity < ncol(inSCE) - 1}. Default \code{30}. See 

+\item{perplexity}{perplexity parameter. Should not be bigger than

+\code{3 * perplexity < ncol(inSCE) - 1}. Default \code{30}. See

 \code{\link[Rtsne]{Rtsne}} details for interpretation.}

 \item{nIterations}{maximum iterations. Default \code{1000}.}

-\item{numThreads}{Integer, number of threads to use using OpenMP, Default 

+\item{numThreads}{Integer, number of threads to use using OpenMP, Default

 \code{1}. \code{0} corresponds to using all available cores.}

-\item{seed}{Random seed for reproducibility of tSNE results. 

+\item{seed}{Random seed for reproducibility of tSNE results.

 Default \code{NULL} will use global seed in use by the R environment.}

 }

 \value{

@@ -90,14 +90,14 @@ A \linkS4class{SingleCellExperiment} object with tSNE computation

 updated in \code{reducedDim(inSCE, reducedDimName)}.

 }

 \description{

-T-Stochastic Neighbour Embedding (t-SNE) algorithm is commonly 

-for 2D visualization of single-cell data. This function wraps the 

-Rtsne \code{\link[Rtsne]{Rtsne}} function. 

+T-Stochastic Neighbour Embedding (t-SNE) algorithm is commonly

+for 2D visualization of single-cell data. This function wraps the

+Rtsne \code{\link[Rtsne]{Rtsne}} function.

 With this funciton, users can create tSNE embedding directly from raw count

-matrix, with necessary preprocessing including normalization, scaling, 

-dimension reduction all automated. Yet we still recommend having the PCA as 

-input, so that the result can match with the clustering based on the same 

+matrix, with necessary preprocessing including normalization, scaling,

+dimension reduction all automated. Yet we still recommend having the PCA as

+input, so that the result can match with the clustering based on the same

 input PCA, and will be much faster.

 }

 \examples{

@@ -110,7 +110,7 @@ sce <- getTSNE(inSCE = sce, useAssay = "counts", logNorm = TRUE, nTop = 2000,

 # Run from PCA

 sce <- scaterlogNormCounts(sce, "logcounts")

 sce <- runModelGeneVar(sce)

-sce <- scaterPCA(sce, useAssay = "logcounts", 

+sce <- scaterPCA(sce, useAssay = "logcounts",

                  useFeatureSubset = "HVG_modelGeneVar2000", scale = TRUE)

 sce <- getTSNE(sce, useReducedDim = "PCA")

 }

@@ -60,7 +60,7 @@ object. Can be one of "Matrix" (as returned by

 \item{delayedArray}{Boolean. Whether to read the expression matrix as

 \link{DelayedArray} object or not. Default \code{FALSE}.}

-\item{rowNamesDedup}{Boolean. Whether to deduplicate rownames. Default 

+\item{rowNamesDedup}{Boolean. Whether to deduplicate rownames. Default

 \code{TRUE}.}

 }

 \value{

@@ -6,8 +6,8 @@

 \usage{

 plotPCA(

   inSCE,

-  colorBy = "No Color",

-  shape = "No Shape",

+  colorBy = NULL,

+  shape = NULL,

   pcX = "PC1",

   pcY = "PC2",

   reducedDimName = "PCA",

@@ -6,8 +6,8 @@

 \usage{

 plotTSNE(

   inSCE,

-  colorBy = "No Color",

-  shape = "No Shape",

+  colorBy = NULL,

+  shape = NULL,

   reducedDimName = "TSNE",

   runTSNE = FALSE,

   useAssay = "logcounts"

@@ -6,8 +6,8 @@

 \usage{

 plotUMAP(

   inSCE,

-  colorBy = "No Color",

-  shape = "No Shape",

+  colorBy = NULL,

+  shape = NULL,

   reducedDimName = "UMAP",

   runUMAP = FALSE,

   useAssay = "logcounts"

new file mode 100644

@@ -0,0 +1,32 @@

+# description

+library(singleCellTK)

+test_that(desc = "Testing description", {

+    a <- singleCellTK:::description_subsetSCECols()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionBarcodeRank()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionBCDS()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionCXDS()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionDecontX()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionDoubletFinder()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionEmptyDrops()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionRunCellQC()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionRunDropletQC()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionRunPerCellQC()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionScDblFinder()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionScdsHybrid()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionScrublet()

+    testthat::expect_type(a, "list")

+    a <- singleCellTK:::descriptionSoupX()

+    testthat::expect_type(a, "list")

+})

new file mode 100644

@@ -0,0 +1,129 @@

+# dimension reduction

+library(singleCellTK)

+library(testthat)

+context("Testing dimension reduction")

+data(scExample, package = "singleCellTK")

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

+sce <- scaterlogNormCounts(sce, "logcounts")

+sce <- runFeatureSelection(sce, useAssay = "counts")

+sce <- setTopHVG(sce, featureSubsetName = "hvg")

+sce <- setTopHVG(sce, featureSubsetName = "hvgAltExp", altExp = TRUE)

+

+test_that(desc = "Testing scaterPCA", {

+    sce <- scaterPCA(sce, useAssay = "logcounts", useFeatureSubset = "hvg",

+                     reducedDimName = "PCA1")

+    testthat::expect_true("PCA1" %in% reducedDimNames(sce))

+

+    sce <- scaterPCA(sce, useAssay = "hvgAltExplogcounts", useAltExp = "hvgAltExp",

+                     reducedDimName = "PCA2")

+    testthat::expect_true("PCA2" %in% reducedDimNames(sce))

+

+    expect_error({

+        scaterPCA(sce, useAssay = "null", useFeatureSubset = "hvg")

+    }, "Specified assay")

+    expect_error({

+        scaterPCA(sce, useAssay = "null", useAltExp = "null", useFeatureSubset = "hvg")

+    }, "Specified altExp ")

+    expect_error({

+        scaterPCA(sce, useAssay = "null", useAltExp = "hvgAltExp", useFeatureSubset = "hvg")

+    }, "not found in the specified altExp")

+

+    p1 <- plotPCA(sce, reducedDimName = "PCA1", colorBy = "type", shape = "type")

+    p2 <- plotPCA(sce, reducedDimName = "PCA3", runPCA = TRUE)

+    expect_is(p1, "ggplot")

+    expect_is(p2, "ggplot")

+    expect_error({

+        plotPCA(sce, pcX = "foo", reducedDimName = "PCA1")

+    }, regexp = "pcX dimension")

+    expect_error({

+        plotPCA(sce, pcY = "bar", reducedDimName = "PCA1")

+    }, regexp = "pcY dimension")

+    expect_error({

+        plotPCA(sce, reducedDimName = "UMAP")

+    }, regexp = "dimension not found")

+})

+

+test_that(desc = "Testing scater UMAP", {

+    sce <- scaterPCA(sce, useFeatureSubset = "hvg", seed = 12345, reducedDimName = "PCA1")

+    sce <- getUMAP(sce, useReducedDim = "PCA1", reducedDimName = "UMAP1")

+    testthat::expect_true("UMAP1" %in% reducedDimNames(sce))

+    sce <- getUMAP(sce, useAssay = "hvgAltExplogcounts", useAltExp = "hvgAltExp",

+                   reducedDimName = "UMAP2")

+    testthat::expect_true("UMAP2" %in% reducedDimNames(sce))

+    # TODO: Still some runable conditions

+    expect_error({

+        getUMAP(inSCE = 1)

+    }, "Please use a SingleCellExperiment object")

+    expect_error({

+        getUMAP(sce, useAssay = NULL, useReducedDim = NULL)

+    }, "`useAssay` and `useReducedDim` cannot be NULL")

+    expect_error({

+        getUMAP(sce, useAltExp = "altexp")

+    }, "Specified `useAltExp` not found.")

+    expect_error({

+        getUMAP(sce, useReducedDim = "TSNE")

+    }, "Specified `useReducedDim` not found.")

+    expect_error({

+        getUMAP(sce, useAssay = "TSNE")

+    }, regexp = "Specified `useAssay` not found.")

+    expect_error({

+        getUMAP(sce, sample = "batch")

+    }, regexp = "Given sample annotation ")

+    expect_error({

+        getUMAP(sce, sample = letters)

+    }, regexp = "'sample' must be the same length")

+

+    p1 <- plotUMAP(sce, reducedDimName = "UMAP1", colorBy = "type", shape = "type")

+    p2 <- plotUMAP(sce, reducedDimName = "UMAP3", runUMAP = TRUE)

+    expect_is(p1, "ggplot")

+    expect_is(p2, "ggplot")

+    expect_error({

+        plotUMAP(sce, reducedDimName = "TSNE")

+    }, regexp = "dimension not found")

+    reducedDim(sce, "UMAP4") <- cbind(reducedDim(sce, "UMAP1"), reducedDim(sce, "UMAP1"))

+    expect_warning({

+        plotUMAP(sce, reducedDimName = "UMAP4")

+    }, "More than two UMAP dimensions")

+})

+

+test_that(desc = "Testing Rtsne TSNE", {

+    sce <- scaterPCA(sce, useFeatureSubset = "hvg", seed = 12345, reducedDimName = "PCA1")

+    sce <- getTSNE(sce, useReducedDim = "PCA1", reducedDimName = "TSNE1")

+    testthat::expect_true("TSNE1" %in% reducedDimNames(sce))

+    sce <- getTSNE(sce, useAssay = "hvgAltExpcounts", useAltExp = "hvgAltExp",

+                   reducedDimName = "TSNE2", logNorm = TRUE, nTop = 50)

+    testthat::expect_true("TSNE2" %in% reducedDimNames(sce))

+    # TODO: Still some runable conditions

+    expect_error({

+        getTSNE(inSCE = 1)

+    }, "Please use a SingleCellExperiment object")

+    expect_error({

+        getTSNE(sce, useAssay = NULL, useReducedDim = NULL)

+    }, "`useAssay` and `useReducedDim` cannot be NULL")

+    expect_error({

+        getTSNE(sce, useAltExp = "altexp")

+    }, "Specified `useAltExp` not found.")

+    expect_error({

+        getUMAP(sce, useReducedDim = "TSNE")

+    }, "Specified `useReducedDim` not found.")

+    expect_error({

+        getUMAP(sce, useAssay = "TSNE")

+    }, regexp = "Specified `useAssay` not found.")

+

+    p1 <- plotTSNE(sce, reducedDimName = "TSNE1", colorBy = "type", shape = "type")

+    p2 <- plotTSNE(sce, reducedDimName = "TSNE3", runTSNE = TRUE)

+    expect_is(p1, "ggplot")

+    expect_is(p2, "ggplot")

+    expect_error({

+        plotTSNE(sce, reducedDimName = "UMAP")

+    }, regexp = "dimension not found")

+    reducedDim(sce, "TSNE4") <- cbind(reducedDim(sce, "TSNE1"), reducedDim(sce, "TSNE1"))

+    expect_warning({

+        plotTSNE(sce, reducedDimName = "TSNE4")

+    }, "More than two t-SNE dimensions")

+

+    p3 <- plotDimRed(sce, useReduction = "PCA1")

+    p4 <- plotDimRed(sce, useReduction = "PCA1", xAxisLabel = "PC_1", yAxisLabel = "PC_2")

+    expect_is(p3, "ggplot")

+    expect_is(p4, "ggplot")

+})

@@ -9,39 +9,37 @@ test_that(desc = "Testing FindHVG", {

     sce <- runModelGeneVar(sce, "seuratNormData")

     metricNames.mgv <- metadata(sce)$sctk$runFeatureSelection$modelGeneVar$rowData

     testthat::expect_true(all(metricNames.mgv %in% names(rowData(sce))))

-    

-    sce <- runSeuratFindHVG(sce, useAssay = "counts", method = "vst")

+

+    sce <- runFeatureSelection(sce, "counts", "vst")

     metricNames.vst <- metadata(sce)$sctk$runFeatureSelection$vst$rowData

     testthat::expect_true(all(metricNames.vst %in% names(rowData(sce))))

-    

-    sce <- runSeuratFindHVG(sce, useAssay = "seuratNormData", 

-                            method = "dispersion")

+

+    sce <- runFeatureSelection(sce, "seuratNormData", "dispersion")

     metricNames.disp <- metadata(sce)$sctk$runFeatureSelection$dispersion$rowData

     testthat::expect_true(all(metricNames.disp %in% names(rowData(sce))))

-    

-    sce <- runSeuratFindHVG(sce, useAssay = "seuratNormData", 

-                            method = "mean.var.plot")

+

+    sce <- runFeatureSelection(sce, "seuratNormData", "mean.var.plot")

     metricNames.mvp <- metadata(sce)$sctk$runFeatureSelection$mean.var.plot$rowData

     testthat::expect_true(all(metricNames.mvp %in% names(rowData(sce))))

-    

+

     # Test accessor functions

     sce <- setTopHVG(sce, "modelGeneVar", hvgNumber = 2000, altExp = TRUE)

     nHVG <- length(getTopHVG(sce, useFeatureSubset = "HVG_modelGeneVar2000"))

     testthat::expect_true(is.logical(rowData(sce)$HVG_modelGeneVar2000))

     testthat::expect_equal(nrow(altExp(sce, "HVG_modelGeneVar2000")), nHVG)

-    testthat::expect_equal(metadata(sce)$sctk$featureSubsets$HVG_modelGeneVar2000$useAssay, 

+    testthat::expect_equal(metadata(sce)$sctk$featureSubsets$HVG_modelGeneVar2000$useAssay,

                            "seuratNormData")

-    

-    hvgs <- getTopHVG(sce, "mean.var.plot", hvgNumber = 2000, 

+

+    hvgs <- getTopHVG(sce, "mean.var.plot", hvgNumber = 2000,

                       featureDisplay = "feature_name")

     testthat::expect_false(all(startsWith(hvgs, "ENSG00000")))

     hvgs <- getTopHVG(sce, "vst", hvgNumber = 2000)

     vm1 <- plotTopHVG(sce, "dispersion")

-    vm2 <- plotTopHVG(sce, "modelGeneVar", hvgNumber = 30, labelsCount = 10, 

+    vm2 <- plotTopHVG(sce, "modelGeneVar", hvgNumber = 30, labelsCount = 10,

                      featureDisplay = "feature_name")

-    vm3 <- plotTopHVG(sce, method = "mean.var.plot", 

+    vm3 <- plotTopHVG(sce, method = "mean.var.plot",

                       useFeatureSubset = "HVG_modelGeneVar2000")

     testthat::expect_true(inherits(vm1, "ggplot"))

     testthat::expect_true(inherits(vm2, "ggplot"))

     testthat::expect_true(inherits(vm3, "ggplot"))

-})

\ No newline at end of file

+})

@@ -11,20 +11,48 @@ context("Testing import functions")

 ## Importing scRNA-seq Data Functions

 #####################################

-test_that(desc = "Testing importBUStools", {

-  

-  sce <- importBUStools(BUStoolsDirs = system.file("extdata/BUStools_PBMC_1k_v3_20x20/genecount/", package = "singleCellTK"),

-                        samples = "PBMC_1k_v3_20x20")

-  expect_true(validObject(sce))

-})

-

-

-test_that(desc = "Testing importCellRanger", {

-  sce <- importCellRanger(cellRangerDirs = system.file("extdata",package = "singleCellTK"),

-                          sampleDirs = "hgmm_1k_v3_20x20",

-                          sampleNames = "hgmm1kv3",

-                          dataType = "filtered")

+test_that(desc = "Testing import single-cell data", {

+  allImportEntries <- list(

+    samples = list()

+  )

+  allImportEntries$samples$a <- list(

+    type = "busTools",

+    params = list(BUStoolsDirs = system.file("extdata/BUStools_PBMC_1k_v3_20x20/genecount/", package = "singleCellTK"),

+                  samples = "BUStools_20x20")

+  )

+  allImportEntries$samples$b <- list(

+    type = "cellRanger3",

+    params = list(cellRangerDirs = system.file("extdata",package = "singleCellTK"),

+                  sampleDirs = "hgmm_1k_v3_20x20",

+                  sampleNames = "cellRanger3_20x20",

+                  dataType = "filtered")

+  )

+  allImportEntries$samples$c <- list(

+    type = "seqc",

+    params = list(seqcDirs = system.file("extdata/pbmc_1k_50x50",package = "singleCellTK"),

+                  samples = "seqc_50x50",

+                  prefix = "pbmc_1k",

+                  combinedSample = FALSE)

+  )

+  allImportEntries$samples$d <- list(

+    type = "starSolo",

+    params = list(STARsoloDirs = system.file("extdata/STARsolo_PBMC_1k_v3_20x20",package = "singleCellTK"),

+                  samples = "STARsolo_20x20")

+  )

+  allImportEntries$samples$e <- list(

+    type = "files",

+    params = list(assayFile = system.file("extdata/pbmc_1k_50x50/pbmc_1k_sparse_molecule_counts.mtx",package = "singleCellTK"),

+                  annotFile = system.file("extdata/pbmc_1k_50x50/pbmc_1k_sparse_counts_barcodes.csv",package = "singleCellTK"),

+                  featureFile = system.file("extdata/pbmc_1k_50x50/pbmc_1k_sparse_counts_genes.csv",package = "singleCellTK"),

+                  assayName = "counts")

+  )

+  expect_warning({

+    sce <- importMultipleSources(allImportEntries)

+  }, "column names 'cell_barcode.x'")

   expect_true(validObject(sce))

+  summary <- as.data.frame(table(sce$sample))

+  expect_true(all(summary$Var1 %in% c("BUStools_20x20", "cellRanger3_20x20", "seqc_50x50", "STARsolo_20x20", "sample")))

+  expect_equal(summary$Freq, c(20, 20, 50, 50, 20))

 })

 test_that(desc = "Testing importDropEst", {

@@ -33,22 +61,7 @@ test_that(desc = "Testing importDropEst", {

   expect_true(validObject(sce))

 })

-

-test_that(desc = "Testing importSeqc", {

-  sce <- importSEQC(seqcDirs = system.file("extdata/pbmc_1k_50x50",package = "singleCellTK"),

-                    samples = "pbmc_1k_50x50",

-                    prefix = "pbmc_1k",

-                    combinedSample = FALSE)

-  expect_true(validObject(sce))

-})

-

-test_that(desc = "Testing importSTARSolo", {

-  sce <- importSTARsolo(STARsoloDirs = system.file("extdata/STARsolo_PBMC_1k_v3_20x20",package = "singleCellTK"),

-                        samples = "PBMC_1k_v3_20x20")

-  

-  expect_true(validObject(sce))

-})

-# 

+#

 # test_that(desc = "Testing importOptimus", {

 #   if (!reticulate::py_module_available("scipy.sparse") || (!reticulate::py_module_available("numpy"))){

 #     skip("scipy.sparse or numpy not available. Skipping testing importOptimus")

@@ -56,7 +69,7 @@ test_that(desc = "Testing importSTARSolo", {

 #   sce <- importOptimus(OptimusDirs = system.file("extdata/Optimus_20x1000",package = "singleCellTK"),

 #                        samples = "Optimus_20x1000")

 #   expect_true(validObject(sce))

-# }) 

+# })

 # test_that(desc = "Testing importAnnData", {

 #   if (!reticulate::py_module_available("anndata")){

@@ -69,17 +82,15 @@ test_that(desc = "Testing importSTARSolo", {

 ##################################

 ## Importing Gene Set Functions

 ##################################

-

+data(scExample)

 test_that(desc = "Testing importGeneSetFromGMT", {

-  data(scExample)

   gmt <- system.file("extdata/mito_subset.gmt", package = "singleCellTK")

   sce <- importGeneSetsFromGMT(inSCE = sce, file = gmt, by = NULL,

                                collectionName = "test")

   expect_true(inherits(sce@metadata$sctk$genesets$test[[1]], "GeneSet"))

-}) 

+})

 test_that(desc = "Testing importGeneSetFromList", {

-  data(scExample)

   gs1 <- rownames(sce)[1:10]

   gs2 <- rownames(sce)[11:20]

   gs <- list("geneset1" = gs1, "geneset2" = gs2)

@@ -88,10 +99,9 @@ test_that(desc = "Testing importGeneSetFromList", {

                                 collectionName = "test",

                                 by = "rownames")

   expect_true(inherits(sce@metadata$sctk$genesets$test[[1]], "GeneSet"))

-}) 

+})

 test_that(desc = "Testing importGeneSetFromCollection", {

-  data(scExample)

   gs1 <- GSEABase::GeneSet(setName = "geneset1", geneIds = rownames(sce)[1:10])

   gs2 <- GSEABase::GeneSet(setName = "geneset2", geneIds = rownames(sce)[11:20])

   gsc <- GSEABase::GeneSetCollection(list(gs1, gs2))

@@ -100,17 +110,26 @@ test_that(desc = "Testing importGeneSetFromCollection", {

                                       collectionName = "test",

                                       by = "rownames")

   expect_true(inherits(sce@metadata$sctk$genesets$test[[1]], "GeneSet"))

-}) 

+  gscList <- sctkListGeneSetCollections(sce)

+  expect_equal(gscList, "test")

+  gsList <- getGenesetNamesFromCollection(sce, "test")

+  expect_equal(gsList, c("geneset1", "geneset2"))

+})

 test_that(desc = "Testing importGeneSetFromMSigDB", {

-  data(scExample)

   sce <- importGeneSetsFromMSigDB(inSCE = sce,

                                   categoryIDs = "C2-CP",

                                   species = "Homo sapiens",

                                   mapping = "gene_symbol",

                                   by = "feature_name")

   expect_true(inherits(sce@metadata$sctk$genesets$"C2-CP"[[1]], "GeneSet"))

-}) 

-

+})

+data(scExample)

+test_that(desc = "Testing export", {

+  exportSCEtoFlatFile(sce)

+  expect_true(file.exists("./assays/SCE_counts.mtx.gz"))

+  expect_true(file.exists("./SCE_cellData.txt.gz"))

+  expect_true(file.exists("./SCE_featureData.txt.gz"))

+})