Skip to contents

Fetch feature expression data, reduction coordinates, or metadata from single-cell objects

fetch_data.Rd

This function extends the behavior of SeuratObject's FetchData to other single-cell objects, allowing for expression data, metdadata, or reduction coordinates to be pulled using consistent syntax.

Usage

fetch_data(object, vars = NULL, layer = NULL, cells = NULL, ...)

# S3 method for class 'Seurat'
fetch_data(object, vars = NULL, layer = NULL, cells = NULL, slot = NULL, ...)

# S3 method for class 'SingleCellExperiment'
fetch_data(object, vars, layer = NULL, cells = NULL, ...)

# S3 method for class 'AnnDataR6'
fetch_data(object, vars, layer = NULL, cells = NULL, ...)

Arguments

object

A single-cell object. Currently, Seurat, SingleCellExperiment, and anndata objects are supported. If a Seruat object is passed to this generic, the FetchData method from the SeruatObject method will be ran. For all other objects, methods that replicate the behavior of FetchData in that object will be ran.

vars

A character vector with desired features, metadata variables, or reduction dimensions to pull from the object. By default, features are returned from the default assay (or "experiment" in SingleCellExperiment objects, or "modality" in anndata objects). To pull feature expression data, the assay to pull data from should be defined using the "key" of the assay before the feature name. To determine the key that corresponds to the assay to pull data from, run all_keys. For more information, see our user guide.

layer

For feature expression data, the layer to pull data from. Layers are referred to as "slots" in Seurat objects v4 and earlier, and "assays" in SingleCellExperiment objects.

cells

A character vector of cells to include, as they are named in the object (i.e. according to colNames(object)). If NULL, data will be returned for all cells in the object.

...

Additional parameters, beyond the ones listed above, to be passed to S3 methods. This includes the following, all of which are documented in the parameter entries below:

  • fetch_data.Seurat: slot parameter

slot

This parameter is added for backwards compatability with Seruat v4 and earlier. This was deprecated in Seurat version 5.0.0. If you have Seruat v5.0.0 or later, this should not be used. This parameter should also not be used for SingleCellExperiment objects or anndata objects and will not work at all for these object classes. Use layer instead.

Value

A data.frame with the requested vars as columns and the cells as rows.

Details

See our GitHub.io website for additional information and examples.

Methods (by class)

  • fetch_data(Seurat): Seurat objects. This will run FetchData from the SeuratObject package.

  • fetch_data(SingleCellExperiment): SingleCellExperiment objects

  • fetch_data(AnnDataR6): anndata Objects

Examples

# Feature expression data
fetch_data(AML_Seurat, vars = "rna_FLT3") |> str()
#> 'data.frame':	250 obs. of  1 variable:
#>  $ rna_FLT3: num  0 0 0 0 0 ...

# Reduction coordinates
fetch_data(AML_Seurat, vars = c("UMAP_1", "UMAP_2")) |> str()
#> 'data.frame':	250 obs. of  2 variables:
#>  $ UMAP_1: num  -1.64 -1.5 -1.45 -1.38 -1.41 ...
#>  $ UMAP_2: num  9.9 10.13 10.21 10.51 3.39 ...
fetch_data(AML_Seurat, vars = c("PC_1", "PC_2", "PC_3")) |> str()
#> 'data.frame':	250 obs. of  3 variables:
#>  $ PC_1: num  3.91 3.67 2.97 2.88 42.94 ...
#>  $ PC_2: num  1.676 -0.157 0.716 1.103 -11.258 ...
#>  $ PC_3: num  1.86 3.03 2.6 3.68 -4 ...

# Metadata
fetch_data(AML_Seurat, vars = c("condensed_cell_type", "Batch", "nCount_RNA")) |> str()
#> 'data.frame':	250 obs. of  3 variables:
#>  $ condensed_cell_type: chr  "Plasma cells" "Plasma cells" "Plasma cells" "Plasma cells" ...
#>  $ Batch              : chr  "BM_200AB" "BM_200AB" "BM_200AB" "BM_200AB" ...
#>  $ nCount_RNA         : num  10863 8403 8100 8151 8828 ...