@@ -93,6 +93,7 @@ import(grid)

 import(gridExtra, except = c(combine))

 import(magrittr)

 import(stats, except = c(start, end))

+import(uwot)

 importFrom(MAST,FromMatrix)

 importFrom(MAST,summary)

 importFrom(MAST,zlm)

@@ -140,8 +141,6 @@ importFrom(scales,brewer_pal)

 importFrom(scales,dscale)

 importFrom(scales,hue_pal)

 importFrom(stringi,stri_list2matrix)

-importFrom(umap,umap)

-importFrom(umap,umap.defaults)

 importFrom(withr,with_seed)

 useDynLib(celda,"_colSumByGroup")

 useDynLib(celda,"_colSumByGroupChange")

@@ -596,33 +596,26 @@ setGeneric("celdaTsne",

 #'  requires more memory. Default 25000.

 #' @param minClusterSize Integer. Do not subsample cell clusters below this

 #'  threshold. Default 100.

-#' @param initialDims Integer. PCA will be used to reduce the dimentionality

-#'  of the dataset. The top 'initialDims' principal components will be used

-#'  for umap. Default 20.

 #' @param modules Integer vector. Determines which features modules to use for

 #'  tSNE. If NULL, all modules will be used. Default NULL.

 #' @param seed Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 #'  a default value of 12345 is used. If NULL, no calls to

 #'  \link[withr]{with_seed} are made.

-#' @param umapConfig An object of class "umapConfig" specifying parameters to

 #'  the UMAP algorithm.

-#' @return Numeric Matrix of dimension `ncol(counts)` x 2, colums representing

-#'  the "X" and "Y" coordinates in the data's t-SNE represetation.

-#' @examples

+#' @param ... Additional parameters to `uwot::umap`

+#' @return A two column matrix of UMAP coordinates#' @examples

 #' data(celdaCGSim, celdaCGMod)

-#' tsneRes <- celdaUmap(celdaCGSim$counts, celdaCGMod)

-#' @importFrom umap umap.defaults

+#' umapRes <- celdaUmap(celdaCGSim$counts, celdaCGMod)

 #' @export

 setGeneric("celdaUmap",

     signature = "celdaMod",

     function(counts,

         celdaMod,

-        maxCells = 25000,

+        maxCells = NULL,

         minClusterSize = 100,

-        initialDims = 20,

         modules = NULL,

         seed = 12345,

-        umapConfig = umap::umap.defaults) {

+	    ...) {

         standardGeneric("celdaUmap")

     })

@@ -1042,8 +1042,6 @@ setMethod("celdaHeatmap", signature(celdaMod = "celda_C"),

 #' @param initialDims Integer. PCA will be used to reduce the dimentionality

 #'  of the dataset. The top 'initialDims' principal components will be used

 #'  for tSNE. Default 20.

-#' @param modules Integer vector. Determines which features modules to use for

-#'  tSNE. If NULL, all modules will be used. Default NULL.

 #' @param perplexity Numeric. Perplexity parameter for tSNE. Default 20.

 #' @param maxIter Integer. Maximum number of iterations in tSNE generation.

 #'  Default 2500.

@@ -1063,7 +1061,6 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

         maxCells = NULL,

         minClusterSize = 100,

         initialDims = 20,

-        modules = NULL,

         perplexity = 20,

         maxIter = 2500,

         seed = 12345) {

@@ -1074,7 +1071,6 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

                 maxCells = maxCells,

                 minClusterSize = minClusterSize,

                 initialDims = initialDims,

-                modules = modules,

                 perplexity = perplexity,

                 maxIter = maxIter)

         } else {

@@ -1084,7 +1080,6 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

                     maxCells = maxCells,

                     minClusterSize = minClusterSize,

                     initialDims = initialDims,

-                    modules = modules,

                     perplexity = perplexity,

                     maxIter = maxIter))

         }

@@ -1098,15 +1093,13 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

     maxCells = NULL,

     minClusterSize = 100,

     initialDims = 20,

-    modules = NULL,

     perplexity = 20,

     maxIter = 2500) {

     preparedCountInfo <- .prepareCountsForDimReductionCeldaC(counts,

         celdaMod,

         maxCells,

-        minClusterSize,

-        modules)

+        minClusterSize)

     res <- .calculateTsne(preparedCountInfo$norm,

         perplexity = perplexity,

@@ -1117,7 +1110,7 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

     final <- matrix(NA, nrow = ncol(counts), ncol = 2)

     final[preparedCountInfo$cellIx, ] <- res

     rownames(final) <- colnames(counts)

-    colnames(final) <- c("tsne_1", "tsne_2")

+    colnames(final) <- c("tSNE1", "tSNE_2")

     return(final)

 }

@@ -1136,44 +1129,73 @@ setMethod("celdaTsne", signature(celdaMod = "celda_C"),

 #'  Default NULL.

 #' @param minClusterSize Integer. Do not subsample cell clusters below this

 #'  threshold. Default 100.

-#' @param modules Integer vector. Determines which features modules to use for

-#'  UMAP. If NULL, all modules will be used. Default NULL.

 #' @param seed Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 #'  a default value of 12345 is used. If NULL, no calls to

 #'  \link[withr]{with_seed} are made.

-#' @param umapConfig An object of class "umap.config" specifying parameters to

-#'  the UMAP algorithm.

+#' @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 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 0.2. 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 Integer. Number of dimensions from PCA to use as

+#' input in UMAP. Default 50.

+#' @param nThreads Number of threads to use. Default 1.

+#' @param ... Other parameters to pass to `uwot::umap`.

 #' @seealso `celda_C()` for clustering cells and `celdaHeatmap()` for displaying

 #'  expression.

 #' @examples

 #' data(celdaCSim, celdaCMod)

 #' umapRes <- celdaUmap(celdaCSim$counts, celdaCMod)

-#' @return A two column matrix of umap coordinates

+#' @return A two column matrix of UMAP coordinates

 #' @export

 setMethod("celdaUmap", signature(celdaMod = "celda_C"),

     function(counts,

         celdaMod,

         maxCells = NULL,

         minClusterSize = 100,

-        modules = NULL,

         seed = 12345,

-        umapConfig = umap::umap.defaults) {

+        nNeighbors = 30,

+        minDist = 0.2,

+        spread = 1,

+        pca = TRUE,

+        initialDims = 50,

+        nThreads = 1,        

+        ...) {

         if (is.null(seed)) {

             res <- .celdaUmapC(counts = counts,

                 celdaMod = celdaMod,

                 maxCells = maxCells,

                 minClusterSize = minClusterSize,

-                modules = modules,

-                umapConfig = umapConfig)

+                nNeighbors = nNeighbors,

+                minDist = minDist,

+                spread = spread,

+                pca = pca,

+                initialDims = initialDims,

+                nThreads = nThreads,

+                ...)

         } else {

             with_seed(seed,

                 res <- .celdaUmapC(counts = counts,

                     celdaMod = celdaMod,

                     maxCells = maxCells,

                     minClusterSize = minClusterSize,

-                    modules = modules,

-                    umapConfig = umapConfig))

+                    nNeighbors = nNeighbors,

+                    minDist = minDist,

+                    spread = spread,

+                    pca = pca,

+                    initialDims = initialDims,

+                    nThreads = nThreads,

+                    ...))

         }

         return(res)

@@ -1184,19 +1206,31 @@ setMethod("celdaUmap", signature(celdaMod = "celda_C"),

     celdaMod,

     maxCells = NULL,

     minClusterSize = 100,

-    modules = NULL,

-    umapConfig = umap::umap.defaults) {

+    nNeighbors = 30,

+    minDist = 0.2,

+    spread = 1,

+    pca = TRUE,

+    initialDims = 50,

+    nThreads = 1,

+    ...) {

     preparedCountInfo <- .prepareCountsForDimReductionCeldaC(counts,

         celdaMod,

         maxCells,

-        minClusterSize,

-        modules)

-    res <- .calculateUmap(preparedCountInfo$norm, umapConfig)

+        minClusterSize)

+    umapRes <- .calculateUmap(preparedCountInfo$norm,

+        nNeighbors = nNeighbors,

+        minDist = minDist,

+        spread = spread,

+        pca = pca,

+        initialDims = initialDims,        

+        nThreads = nThreads,

+        ...)

+

     final <- matrix(NA, nrow = ncol(counts), ncol = 2)

-    final[preparedCountInfo$cellIx, ] <- res

+    final[preparedCountInfo$cellIx, ] <- umapRes

     rownames(final) <- colnames(counts)

-    colnames(final) <- c("umap_1", "umap_2")

+    colnames(final) <- c("UMAP_1", "UMAP_2")

     return(final)

 }

@@ -1204,8 +1238,7 @@ setMethod("celdaUmap", signature(celdaMod = "celda_C"),

 .prepareCountsForDimReductionCeldaC <- function(counts,

     celdaMod,

     maxCells = NULL,

-    minClusterSize = 100,

-    modules = NULL) {

+    minClusterSize = 100) {

     counts <- .processCounts(counts)

     compareCountMatrix(counts, celdaMod)

@@ -1389,7 +1389,7 @@ setMethod("celdaTsne", signature(celdaMod = "celda_CG"),

     final <- matrix(NA, nrow = ncol(counts), ncol = 2)

     final[preparedCountInfo$cellIx, ] <- res

     rownames(final) <- colnames(counts)

-    colnames(final) <- c("tsne_1", "tsne_2")

+    colnames(final) <- c("tSNE_1", "tSNE_2")

     return(final)

 }

@@ -1411,12 +1411,24 @@ setMethod("celdaTsne", signature(celdaMod = "celda_CG"),

 #' @param minClusterSize Integer. Do not subsample cell clusters below this

 #'  threshold. Default 100.

 #' @param modules Integer vector. Determines which features modules to use for

-#'  tSNE. If NULL, all modules will be used. Default NULL.

+#'  UMAP. If NULL, all modules will be used. Default NULL.

 #' @param seed Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 #'  a default value of 12345 is used. If NULL, no calls to

 #'  \link[withr]{with_seed} are made.

-#' @param umapConfig Object of class `umap.config`. Configures parameters for

-#'  umap. Default `umap::umap.defaults`.

+#' @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 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 0.2. 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 nThreads Number of threads to use. Default 1.

+#' @param ... Other parameters to pass to `uwot::umap`.

 #' @seealso `celda_CG()` for clustering features and cells and `celdaHeatmap()`

 #'  for displaying expression.

 #' @examples

@@ -1431,7 +1443,11 @@ setMethod("celdaUmap",

         minClusterSize = 100,

         modules = NULL,

         seed = 12345,

-        umapConfig = umap::umap.defaults) {

+        nNeighbors = 30,

+        minDist = 0.2,

+        spread = 1,

+        nThreads = 1,

+        ...) {

         if (is.null(seed)) {

             res <- .celdaUmapCG(counts = counts,

@@ -1439,7 +1455,11 @@ setMethod("celdaUmap",

                 maxCells = maxCells,

                 minClusterSize = minClusterSize,

                 modules = modules,

-                umapConfig = umapConfig)

+                nNeighbors = nNeighbors,

+                minDist = minDist,

+                spread = spread,

+                nThreads = nThreads,

+                ...)

         } else {

             with_seed(seed,

                 res <- .celdaUmapCG(counts = counts,

@@ -1447,7 +1467,11 @@ setMethod("celdaUmap",

                     maxCells = maxCells,

                     minClusterSize = minClusterSize,

                     modules = modules,

-                    umapConfig = umapConfig))

+                    nNeighbors = nNeighbors,

+                    minDist = minDist,

+                    spread = spread,

+                    nThreads = nThreads,

+                    ...))

         }

         return(res)

@@ -1459,18 +1483,28 @@ setMethod("celdaUmap",

     maxCells = NULL,

     minClusterSize = 100,

     modules = NULL,

-    umapConfig = umap::umap.defaults) {

+    nNeighbors = nNeighbors,

+    minDist = minDist,

+    spread = spread,

+    nThreads = nThreads,

+    ...) {

     preparedCountInfo <- .prepareCountsForDimReductionCeldaCG(counts,

         celdaMod,

         maxCells,

         minClusterSize,

         modules)

-    umapRes <- .calculateUmap(preparedCountInfo$norm, umapConfig)

+    umapRes <- .calculateUmap(preparedCountInfo$norm,

+        nNeighbors = nNeighbors,

+        minDist = minDist,

+        spread = spread,

+        nThreads = nThreads,

+        ...)

+

     final <- matrix(NA, nrow = ncol(counts), ncol = 2)

     final[preparedCountInfo$cellIx, ] <- umapRes

     rownames(final) <- colnames(counts)

-    colnames(final) <- c("umap_1", "umap_2")

+    colnames(final) <- c("UMAP_1", "UMAP_2")

     return(final)

 }

@@ -1101,12 +1101,24 @@ setMethod("celdaTsne", signature(celdaMod = "celda_G"),

 #' @param minClusterSize Integer. Do not subsample cell clusters below this

 #'  threshold. Default 100.

 #' @param modules Integer vector. Determines which features modules to use for

-#'  tSNE. If NULL, all modules will be used. Default NULL.

+#'  UMAP. If NULL, all modules will be used. Default NULL.

 #' @param seed Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 #'  a default value of 12345 is used. If NULL, no calls to

 #'  \link[withr]{with_seed} are made.

-#' @param umapConfig Object of class `umap.config`. Configures parameters for

-#'  umap. Default `umap::umap.defaults`.

+#' @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 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 0.2. 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 nThreads Number of threads to use. Default 1.

+#' @param ... Other parameters to pass to `uwot::umap`.

 #' @seealso `celda_G()` for clustering features and cells  and `celdaHeatmap()`

 #'  for displaying expression

 #' @examples

@@ -1117,11 +1129,15 @@ setMethod("celdaTsne", signature(celdaMod = "celda_G"),

 setMethod("celdaUmap", signature(celdaMod = "celda_G"),

     function(counts,

         celdaMod,

-        maxCells = 25000,

+        maxCells = NULL,

         minClusterSize = 100,

         modules = NULL,

         seed = 12345,

-        umapConfig = umap::umap.defaults) {

+        nNeighbors = 30,

+        minDist = 0.2,

+        spread = 1,

+        nThreads = 1,

+        ...) {

         if (is.null(seed)) {

             res <- .celdaUmapG(counts = counts,

@@ -1129,7 +1145,11 @@ setMethod("celdaUmap", signature(celdaMod = "celda_G"),

                 maxCells = maxCells,

                 minClusterSize = minClusterSize,

                 modules = modules,

-                umapConfig = umapConfig)

+                nNeighbors = nNeighbors,

+                minDist = minDist,

+                spread = spread,

+                nThreads = nThreads,

+                ...)

         } else {

             with_seed(seed,

                 res <- .celdaUmapG(counts = counts,

@@ -1137,7 +1157,11 @@ setMethod("celdaUmap", signature(celdaMod = "celda_G"),

                     maxCells = maxCells,

                     minClusterSize = minClusterSize,

                     modules = modules,

-                    umapConfig = umapConfig))

+                    nNeighbors = nNeighbors,

+                    minDist = minDist,

+                    spread = spread,

+                    nThreads = nThreads,

+                    ...))

         }

         return(res)

@@ -1149,14 +1173,24 @@ setMethod("celdaUmap", signature(celdaMod = "celda_G"),

     maxCells = NULL,

     minClusterSize = 100,

     modules = NULL,

-    umapConfig = umap::umap.defaults) {

+    nNeighbors = nNeighbors,

+    minDist = minDist,

+    spread = spread,

+    nThreads = nThreads,

+    ...) {

     preparedCountInfo <- .prepareCountsForDimReductionCeldaCG(counts,

         celdaMod,

         maxCells,

         minClusterSize,

         modules)

-    umapRes <- .calculateUmap(preparedCountInfo$norm, umapConfig)

+    umapRes <- .calculateUmap(preparedCountInfo$norm,

+        nNeighbors = nNeighbors,

+        minDist = minDist,

+        spread = spread,

+        nThreads = nThreads,

+        ...)

+        

     final <- matrix(NA, nrow = ncol(counts), ncol = 2)

     final[preparedCountInfo$cellIx, ] <- umapRes

     rownames(final) <- colnames(counts)

@@ -445,14 +445,14 @@ plotDimReduceCluster <- function(dim1,

 # @param maxIter Numeric vector. Determines iterations for tsne. Default 1000.

 # @param doPca Logical. Whether to perform

 # dimensionality reduction with PCA before tSNE.

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

-# input in tSNE.

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

+# input in tSNE. Default 50.

 #' @importFrom Rtsne Rtsne

 .calculateTsne <- function(norm,

     perplexity = 20,

     maxIter = 2500,

     doPca = FALSE,

-    initialDims = 20) {

+    initialDims = 50) {

     res <- Rtsne::Rtsne(

         norm,

@@ -467,12 +467,35 @@ plotDimReduceCluster <- function(dim1,

 }

-# Run the umap algorithm for dimensionality reduction

+# Run the UMAP algorithm for dimensionality reduction

 # @param norm Normalized count matrix.

-# @param umapConfig An object of class umap.config,

-# containing configuration parameters to be passed to umap.

-# Default umap::umap.defualts.

-#' @importFrom umap umap

-.calculateUmap <- function(norm, umapConfig = umap::umap.defaults) {

-    return(umap::umap(norm, umapConfig)$layout)

+# @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 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 0.2. 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 Integer. Number of dimensions from PCA to use as

+# input in UMAP. Default 50.

+# @param nThreads Number of threads to use. Default 1.

+# @param ... Other parameters to pass to `uwot::umap`.

+#' @import uwot

+.calculateUmap <- function(norm, nNeighbors = 30, minDist = 0.2, spread = 1, pca=FALSE, initialDims=50, nThreads = 1, ...) {

+    if (isTRUE(pca)) {

+      doPCA <- initialDims

+    } else {

+      doPCA <- NULL

+    }

+    

+    return(uwot::umap(norm, n_neighbors=nNeighbors,

+    		min_dist = minDist, spread = spread,

+    		n_threads = nThreads, n_sgd_threads = 1, pca = doPCA, ...))

 }

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

 \title{tSNE for celda_C}

 \usage{

 \S4method{celdaTsne}{celda_C}(counts, celdaMod, maxCells = NULL,

-  minClusterSize = 100, initialDims = 20, modules = NULL,

-  perplexity = 20, maxIter = 2500, seed = 12345)

+  minClusterSize = 100, initialDims = 20, perplexity = 20,

+  maxIter = 2500, seed = 12345)

 }

 \arguments{

 \item{counts}{Integer matrix. Rows represent features and columns represent

@@ -28,9 +28,6 @@ threshold. Default 100.}

 of the dataset. The top 'initialDims' principal components will be used

 for tSNE. Default 20.}

-\item{modules}{Integer vector. Determines which features modules to use for

-tSNE. If NULL, all modules will be used. Default NULL.}

-

 \item{perplexity}{Numeric. Perplexity parameter for tSNE. Default 20.}

 \item{maxIter}{Integer. Maximum number of iterations in tSNE generation.

@@ -6,8 +6,9 @@

 \title{umap for celda_C}

 \usage{

 \S4method{celdaUmap}{celda_C}(counts, celdaMod, maxCells = NULL,

-  minClusterSize = 100, modules = NULL, seed = 12345,

-  umapConfig = umap::umap.defaults)

+  minClusterSize = 100, seed = 12345, nNeighbors = 30,

+  minDist = 0.2, spread = 1, pca = TRUE, initialDims = 50,

+  nThreads = 1, ...)

 }

 \arguments{

 \item{counts}{Integer matrix. Rows represent features and columns represent

@@ -24,18 +25,37 @@ Default NULL.}

 \item{minClusterSize}{Integer. Do not subsample cell clusters below this

 threshold. Default 100.}

-\item{modules}{Integer vector. Determines which features modules to use for

-UMAP. If NULL, all modules will be used. Default NULL.}

-

 \item{seed}{Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 a default value of 12345 is used. If NULL, no calls to

 \link[withr]{with_seed} are made.}

-\item{umapConfig}{An object of class "umap.config" specifying parameters to

-the UMAP algorithm.}

+\item{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.}

+

+\item{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 0.2. See `?uwot::umap` for more information.}

+

+\item{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.}

+

+\item{pca}{Logical. Whether to perform

+dimensionality reduction with PCA before UMAP.}

+

+\item{initialDims}{Integer. Number of dimensions from PCA to use as

+input in UMAP. Default 50.}

+

+\item{nThreads}{Number of threads to use. Default 1.}

+

+\item{...}{Other parameters to pass to `uwot::umap`.}

 }

 \value{

-A two column matrix of umap coordinates

+A two column matrix of UMAP coordinates

 }

 \description{

 Embeds cells in two dimensions using umap based on a `celda_C`

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

 \usage{

 \S4method{celdaUmap}{celda_CG}(counts, celdaMod, maxCells = NULL,

   minClusterSize = 100, modules = NULL, seed = 12345,

-  umapConfig = umap::umap.defaults)

+  nNeighbors = 30, minDist = 0.2, spread = 1, nThreads = 1, ...)

 }

 \arguments{

 \item{counts}{Integer matrix. Rows represent features and columns represent

@@ -25,14 +25,30 @@ Default NULL.}

 threshold. Default 100.}

 \item{modules}{Integer vector. Determines which features modules to use for

-tSNE. If NULL, all modules will be used. Default NULL.}

+UMAP. If NULL, all modules will be used. Default NULL.}

 \item{seed}{Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 a default value of 12345 is used. If NULL, no calls to

 \link[withr]{with_seed} are made.}

-\item{umapConfig}{Object of class `umap.config`. Configures parameters for

-umap. Default `umap::umap.defaults`.}

+\item{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.}

+

+\item{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 0.2. See `?uwot::umap` for more information.}

+

+\item{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.}

+

+\item{nThreads}{Number of threads to use. Default 1.}

+

+\item{...}{Other parameters to pass to `uwot::umap`.}

 }

 \value{

 A two column matrix of umap coordinates

@@ -5,9 +5,9 @@

 \alias{celdaUmap,celda_G-method}

 \title{umap for celda_G}

 \usage{

-\S4method{celdaUmap}{celda_G}(counts, celdaMod, maxCells = 25000,

+\S4method{celdaUmap}{celda_G}(counts, celdaMod, maxCells = NULL,

   minClusterSize = 100, modules = NULL, seed = 12345,

-  umapConfig = umap::umap.defaults)

+  nNeighbors = 30, minDist = 0.2, spread = 1, nThreads = 1, ...)

 }

 \arguments{

 \item{counts}{Integer matrix. Rows represent features and columns represent

@@ -25,14 +25,30 @@ Default NULL.}

 threshold. Default 100.}

 \item{modules}{Integer vector. Determines which features modules to use for

-tSNE. If NULL, all modules will be used. Default NULL.}

+UMAP. If NULL, all modules will be used. Default NULL.}

 \item{seed}{Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 a default value of 12345 is used. If NULL, no calls to

 \link[withr]{with_seed} are made.}

-\item{umapConfig}{Object of class `umap.config`. Configures parameters for

-umap. Default `umap::umap.defaults`.}

+\item{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.}

+

+\item{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 0.2. See `?uwot::umap` for more information.}

+

+\item{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.}

+

+\item{nThreads}{Number of threads to use. Default 1.}

+

+\item{...}{Other parameters to pass to `uwot::umap`.}

 }

 \value{

 A two column matrix of umap coordinates

@@ -4,9 +4,8 @@

 \alias{celdaUmap}

 \title{Embeds cells in two dimensions using umap.}

 \usage{

-celdaUmap(counts, celdaMod, maxCells = 25000, minClusterSize = 100,

-  initialDims = 20, modules = NULL, seed = 12345,

-  umapConfig = umap::umap.defaults)

+celdaUmap(counts, celdaMod, maxCells = NULL, minClusterSize = 100,

+  modules = NULL, seed = 12345, ...)

 }

 \arguments{

 \item{counts}{Integer matrix. Rows represent features and columns represent

@@ -22,28 +21,21 @@ requires more memory. Default 25000.}

 \item{minClusterSize}{Integer. Do not subsample cell clusters below this

 threshold. Default 100.}

-\item{initialDims}{Integer. PCA will be used to reduce the dimentionality

-of the dataset. The top 'initialDims' principal components will be used

-for umap. Default 20.}

-

 \item{modules}{Integer vector. Determines which features modules to use for

 tSNE. If NULL, all modules will be used. Default NULL.}

 \item{seed}{Integer. Passed to \link[withr]{with_seed}. For reproducibility,

 a default value of 12345 is used. If NULL, no calls to

-\link[withr]{with_seed} are made.}

-

-\item{umapConfig}{An object of class "umapConfig" specifying parameters to

+\link[withr]{with_seed} are made.

 the UMAP algorithm.}

+

+\item{...}{Additional parameters to `uwot::umap`}

 }

 \value{

-Numeric Matrix of dimension `ncol(counts)` x 2, colums representing

- the "X" and "Y" coordinates in the data's t-SNE represetation.

+A two column matrix of UMAP coordinates#' @examples

+data(celdaCGSim, celdaCGMod)

+umapRes <- celdaUmap(celdaCGSim$counts, celdaCGMod)

 }

 \description{

 Embeds cells in two dimensions using umap.

 }

-\examples{

-data(celdaCGSim, celdaCGMod)

-tsneRes <- celdaUmap(celdaCGSim$counts, celdaCGMod)

-}