Generic Wrapper function for running dimensionality reduction

runDimReduce(
  inSCE,
  method = c("scaterPCA", "seuratPCA", "seuratICA", "rTSNE", "seuratTSNE",
    "scaterUMAP", "seuratUMAP"),
  useAssay = NULL,
  useReducedDim = NULL,
  useAltExp = NULL,
  reducedDimName,
  nComponents = 20,
  seed = NULL,
  ...
)

Arguments

inSCE	Input SingleCellExperiment object.
method	One from `"scaterPCA"`, `"seuratPCA"`, `"seuratICA"`, `"rTSNE"`, `"seuratTSNE"`, `"scaterUMAP"` and `"seuratUMAP"`.
useAssay	Assay to use for computation. If `useAltExp` is specified, `useAssay` has to exist in `assays(altExp(inSCE, useAltExp))`. Default `"counts"`.
useReducedDim	The low dimension representation to use for embedding computation. Default `NULL`.
useAltExp	The subset to use for computation, usually for the selected variable features. Default `NULL`.
reducedDimName	The name of the result matrix. Required.
nComponents	Specify the number of dimensions to compute with the selected method in case of PCA/ICA and the number of components to use in the case of TSNE/UMAP methods.
seed	Random seed for reproducibility of results. Default `NULL` will use global seed in use by the R environment.
...	The other arguments for running a specific algorithm. Please refer to the one you use.

Value

The input SingleCellExperiment object with reducedDim updated with the result.

Details

Wrapper function to run one of the available dimensionality reduction algorithms integrated within SCTK from scaterPCA, seuratPCA, seuratICA, getTSNE, seuratRunTSNE, getUMAP and seuratRunUMAP. Users can use an assay by specifying useAssay, use the assay in an altExp by specifying both useAltExp and useAssay, or use a low-dimensionality representation by specifying useReducedDim.

Examples

data(scExample, package = "singleCellTK")
sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'")
sce <- runNormalization(sce, useAssay = "counts",
                        outAssayName = "logcounts_scaled",
                        normalizationMethod = "logNormCounts",
                        scale = TRUE)
#> Normalization performed using logNormCounts method.
#> Z.Score scaling performed on the input data.
sce <- runDimReduce(inSCE = sce, method = "scaterPCA",
                    useAssay = "logcounts_scaled",
                    reducedDimName = "PCA")
#> Thu Dec 23 11:36:13 2021 ... Computing Scater PCA.