Label cell types with SingleR

SingleR works with a reference dataset where the cell type labeling is given. Given a reference dataset of samples (single-cell or bulk) with known labels, it assigns those labels to new cells from a test dataset based on similarities in their expression profiles.

runSingleR(
  inSCE,
  useAssay = "logcounts",
  useSCERef = NULL,
  labelColName = NULL,
  useBltinRef = c("hpca", "bpe", "mp", "dice", "immgen", "mouse", "zeisel"),
  level = c("main", "fine", "ont"),
  featureType = c("symbol", "ensembl"),
  labelByCluster = NULL
)

Arguments

inSCE

SingleCellExperiment inherited object. Required.

useAssay

character. A string specifying which assay to use for expression profile identification. Required.

useSCERef

SingleCellExperiment inherited object. An optional customized reference dataset. Default NULL.

labelColName

A single character. A string specifying the column in colData(useSCERef) that stores the cell type labeling. Default NULL.

useBltinRef

A single character. A string that specifies a reference provided by SingleR. Choose from "hpca", "bpe", "mp", "dice", "immgen", "mouse", "zeisel". See detail. Default "hpca".

level

A string for cell type labeling level. Used only when using some of the SingleR built-in references. Choose from "main", "fine", "ont". Default "main".

featureType

A string for whether to use gene symbols or Ensembl IDs when using a SingleR built-in reference. Should be set based on the type of rownames of inSCE. Choose from "symbol", "ensembl". Default "symbol".

labelByCluster

A single character. A string specifying the column name in colData(inSCE) that stores clustering labels. Use this when users want to only label cells on cluster level, instead of performing calculation on each cell. Default NULL.

Value

Input SCE object with cell type labeling updated in colData(inSCE), together with scoring metrics.

Examples

data("sceBatches")
logcounts(sceBatches) <- log1p(counts(sceBatches))
#sceBatches <- runSingleR(sceBatches, useBltinRef = "mp")