Extract Sample-Level Metadata from SingleCellExperiment Object

Extracts sample-wise metadata from a SingleCellExperiment object by identifying which colData columns have consistent values within each sample. Identical logic to Seurat metadata extraction but adapted for SCE's colData structure. Returns a data frame suitable for setup.tdr.obj(.meta = ...).

Usage

get.meta.SCE(.sce.obj, .sample.var, .verbose = TRUE)

Arguments

.sce.obj

SingleCellExperiment object with cell-level metadata in colData.

.sample.var

Character: column name in colData(.sce.obj) identifying samples.

.verbose

Logical: print progress messages? Default TRUE.

Value

Data frame with one row per sample, containing only sample-level metadata columns. Rownames are sample IDs from .sample.var.

Details

SingleCellExperiment stores cell-level metadata in colData. This function:

1. Converts colData to data frame

2. Groups by .sample.var

3. Identifies columns with unique values within each sample (sample-level)

4. Excludes varying columns (cell-level) with warning

5. Returns one row per sample

Common sample-level variables include experimental conditions, batches, donors. Common cell-level variables include QC metrics, clusters, cell types.

See also

get.meta for automatic format detection, get.meta.Seurat for Seurat objects, get.cells.SCE for extracting count data

Examples

if (FALSE) { # \dontrun{
# Extract sample metadata from SCE object
meta <- get.meta.SCE(.sce.obj = sce.obj,
                     .sample.var = "sample_id")

# Use with get.cells
cells <- get.cells.SCE(.sce.obj = sce.obj,
                       .meta = meta,
                       .sample.var = "sample_id")
} # }