getBiomarker {singleCellTK}R Documentation

Get and plot PCA and t-SCE components for a SCtkE object

Description

Selects the 500 most variable genes in the SCE, performs PCA or t-SNE based on them and stores the values in the reducedDims slot of the SCE object.

Usage

getBiomarker(inSCE, gene, binary = "Binary", useAssay = "counts")

getPCA(inSCE, useAssay = "logcounts", reducedDimName = "PCA")

getTSNE(inSCE, useAssay = "logcounts", reducedDimName = "TSNE")

plotBiomarker(inSCE, gene, binary = "Binary", visual = "PCA",
  shape = "No Shape", x = "PC1", y = "PC2", useAssay = "counts",
  reducedDimName = "PCA")

plotPCA(inSCE, colorBy = "No Color", shape = "No Shape", pcX = "PC1",
  pcY = "PC2", reducedDimName = "PCA", runPCA = FALSE,
  useAssay = "logcounts")

plotTSNE(inSCE, colorBy = "No Color", shape = "No Shape",
  reducedDimName = "TSNE", runTSNE = FALSE, useAssay = "logcounts")

Arguments

inSCE

Input SCtkExperiment object. Required

gene

gene list

binary

"Binary" for binary expression or "Continuous" for a gradient. Default: "Binary"

useAssay

Indicate which assay to use for PCA. Default is "counts"

reducedDimName

Store the PCA data with this name. The default is PCA. The toolkit will store data with the pattern <ASSAY>_<ALGORITHM>.

visual

Type of visualization (PCA or tSNE). Default: "PCA"

shape

Shape of the points

x

x coordinate for PCA

y

y coordinate for PCA

colorBy

The variable to color clusters by

pcX

User choice for the first principal component

pcY

User choice for the second principal component

runPCA

Run PCA if the reducedDimName does not exist. the Default is FALSE.

runTSNE

Run t-SNE if the reducedDimName does not exist. the Default is FALSE.

Value

getBiomarker(): A data.frame of expression values

getPCA(): A SCtkE object with the specified reduecedDim and pcaVariances updated

getTSNE(): A SCtkE object with the specified reduecedDim and pcaVariances updated

plotBiomarker(): A Biomarker plot

plotPCA(): A PCA plot

plotTSNE(): A t-SNE plot

Functions

Examples

getBiomarker(mouseBrainSubsetSCE, gene="C1qa")

data("mouseBrainSubsetSCE")
#add a CPM assay
assay(mouseBrainSubsetSCE, "cpm") <- apply(assay(mouseBrainSubsetSCE,
                                                 "counts"),
                                           2, function(x) {
                                             x / (sum(x) / 1000000)
                                           })
mouseBrainSubsetSCE <- getPCA(mouseBrainSubsetSCE,
                              useAssay = "cpm",
                              reducedDimName = "PCA_cpm")
reducedDims(mouseBrainSubsetSCE)

data("mouseBrainSubsetSCE")
#add a CPM assay
assay(mouseBrainSubsetSCE, "cpm") <- apply(
  assay(mouseBrainSubsetSCE, "counts"), 2, function(x) {
    x / (sum(x) / 1000000)
  })
mouseBrainSubsetSCE <- getTSNE(mouseBrainSubsetSCE, useAssay = "cpm",
                               reducedDimName = "TSNE_cpm")
reducedDims(mouseBrainSubsetSCE)

data("mouseBrainSubsetSCE")
plotBiomarker(mouseBrainSubsetSCE, gene="C1qa", shape="level1class")

data("mouseBrainSubsetSCE")
plotPCA(mouseBrainSubsetSCE, colorBy = "level1class",
        reducedDimName = "PCA_counts")

data("mouseBrainSubsetSCE")
plotTSNE(mouseBrainSubsetSCE, colorBy = "level1class",
         reducedDimName = "TSNE_counts")


[Package singleCellTK version 1.0.3 Index]