CellScoring

This function performs cell scoring on a Seurat object. It calculates scores for a given set of features and adds the scores as metadata to the Seurat object.

Usage

CellScoring(
  srt,
  features = NULL,
  slot = "data",
  assay = NULL,
  split.by = NULL,
  IDtype = "symbol",
  species = "Homo_sapiens",
  db = "GO_BP",
  termnames = NULL,
  db_update = FALSE,
  db_version = "latest",
  convert_species = TRUE,
  Ensembl_version = 103,
  mirror = NULL,
  minGSSize = 10,
  maxGSSize = 500,
  method = "Seurat",
  classification = TRUE,
  name = "",
  new_assay = FALSE,
  BPPARAM = BiocParallel::bpparam(),
  seed = 11,
  ...
)

Arguments

srt: A Seurat object
features: A named list of feature lists for scoring. If NULLL, db will be used to create features sets.
slot: The slot(layer) of the Seurat object to use for scoring. Defaults to "data".
assay: The assay of the Seurat object to use for scoring. Defaults to NULL, in which case the default assay of the object is used.
split.by: A cell metadata variable used for splitting the Seurat object into subsets and performing scoring on each subset. Defaults to NULL.
IDtype: A character vector specifying the type of gene IDs in the srt object or geneID argument. This argument is used to convert the gene IDs to a different type if IDtype is different from result_IDtype.
species: A character vector specifying the species for which the analysis is performed.
db: A character vector specifying the name of the database to be used for enrichment analysis.
termnames: A vector of term names to be used from the database. Defaults to NULL, in which case all features from the database are used.
db_update: A logical value indicating whether the gene annotation databases should be forcefully updated. If set to FALSE, the function will attempt to load the cached databases instead. Default is FALSE.
db_version: A character vector specifying the version of the database to be used. This argument is ignored if db_update is TRUE. Default is "latest".
convert_species: A logical value indicating whether to use a species-converted database when the annotation is missing for the specified species. The default value is TRUE.
Ensembl_version: Ensembl database version. If NULL, use the current release version.
mirror: Specify an Ensembl mirror to connect to. The valid options here are 'www', 'uswest', 'useast', 'asia'.
minGSSize: A numeric value specifying the minimum size of a gene set to be considered in the enrichment analysis.
maxGSSize: A numeric value specifying the maximum size of a gene set to be considered in the enrichment analysis.
method: The method to use for scoring. Can be "Seurat", "AUCell", or "UCell". Defaults to "Seurat".
classification: Whether to perform classification based on the scores. Defaults to TRUE.
name: The name of the assay to store the scores in. Only used if new_assay is TRUE. Defaults to an empty string.
new_assay: Whether to create a new assay for storing the scores. Defaults to FALSE.
BPPARAM: The BiocParallel parameter object. Defaults to BiocParallel::bpparam().
seed: The random seed for reproducibility. Defaults to 11.
...: Additional arguments to be passed to the scoring methods.

Examples

data("pancreas_sub")
ccgenes <- CC_GenePrefetch("Mus_musculus")
#> Connect to the Ensembl archives...
#> Using the 103 version of biomart...
#> Connecting to the biomart...
#> Searching the dataset hsapiens ...
#> Connecting to the dataset hsapiens_gene_ensembl ...
#> Converting the geneIDs...
#> 97 genes mapped with ensembl_symbol
#> ==============================
#> 97 genes mapped
#> 0 genes unmapped
#> ==============================
#> Error in unnest(data = geneID_collapse, cols = colnames(geneID_collapse)[sapply(geneID_collapse,     class) %in% c("list", "AsIs")], keep_empty = FALSE): could not find function "unnest"
pancreas_sub <- CellScoring(
  srt = pancreas_sub,
  features = list(S = ccgenes$S, G2M = ccgenes$G2M),
  method = "Seurat", name = "CC"
)
#> Warning: The following arguments are not used: drop
#> Warning: The following arguments are not used: drop
#> Error in as.vector(data): no method for coercing this S4 class to a vector
CellDimPlot(pancreas_sub, "CC_classification")
#> Error in CellDimPlot(pancreas_sub, "CC_classification"): CC_classification is not in the meta.data of srt object.
FeatureDimPlot(pancreas_sub, "CC_G2M")
#> Warning: CC_G2M are not in the features of srt.
#> Error in FeatureDimPlot(pancreas_sub, "CC_G2M"): There are no valid features present.

if (FALSE) { # \dontrun{
data("panc8_sub")
panc8_sub <- Integration_SCP(panc8_sub,
  batch = "tech", integration_method = "Seurat"
)
CellDimPlot(panc8_sub, group.by = c("tech", "celltype"))

panc8_sub <- CellScoring(
  srt = panc8_sub, slot = "data", assay = "RNA",
  db = "GO_BP", species = "Homo_sapiens",
  minGSSize = 10, maxGSSize = 100,
  method = "Seurat", name = "GO", new_assay = TRUE
)
panc8_sub <- Integration_SCP(panc8_sub,
  assay = "GO",
  batch = "tech", integration_method = "Seurat"
)
CellDimPlot(panc8_sub, group.by = c("tech", "celltype"))

pancreas_sub <- CellScoring(
  srt = pancreas_sub, slot = "data", assay = "RNA",
  db = "GO_BP", species = "Mus_musculus",
  termnames = panc8_sub[["GO"]]@meta.features[, "termnames"],
  method = "Seurat", name = "GO", new_assay = TRUE
)
pancreas_sub <- Standard_SCP(pancreas_sub, assay = "GO")
CellDimPlot(pancreas_sub, "SubCellType")

pancreas_sub[["tech"]] <- "Mouse"
panc_merge <- Integration_SCP(
  srtList = list(panc8_sub, pancreas_sub),
  assay = "GO",
  batch = "tech", integration_method = "Seurat"
)
CellDimPlot(panc_merge, group.by = c("tech", "celltype", "SubCellType", "Phase"))

genenames <- make.unique(capitalize(rownames(panc8_sub[["RNA"]]), force_tolower = TRUE))
panc8_sub <- RenameFeatures(panc8_sub, newnames = genenames, assay = "RNA")
head(rownames(panc8_sub))
panc_merge <- Integration_SCP(
  srtList = list(panc8_sub, pancreas_sub),
  assay = "RNA",
  batch = "tech", integration_method = "Seurat"
)
CellDimPlot(panc_merge, group.by = c("tech", "celltype", "SubCellType", "Phase"))
} # }

Usage

Arguments

See also

Examples