Extracts a tidy tibble with one row per cell per feature. Expression values are provided as one column per requested layer. Optionally includes embedding coordinates and metadata, joined by cell_id.
Usage
fetch_feature_data(
object,
features,
layer = "data",
assay = NULL,
reductions = NULL,
dims = 1:2,
metadata = TRUE
)Arguments
- object
Seurat object.
- features
character. Gene names to extract (required). Becomes a factor column preserving the input order.
- layer
character. Layer(s) to extract expression from. One of "counts", "data", "scale.data", or a combination (default: "data").
- assay
character. Assay to use (default: NULL = DefaultAssay).
- reductions
character. Reduction names to include (default: NULL = none).
- dims
integer. Dimensions to extract per reduction (default: 1:2).
- metadata
logical or character. TRUE = all metadata columns, FALSE = none, or a character vector of specific column names.
Value
A tibble with columns: cell_id, features (factor), one column per layer, optional embedding columns, optional metadata columns.
Examples
if (FALSE) { # \dontrun{
# Basic usage -- normalized expression
fetch_feature_data(seurat_obj, features = c("CD3D", "CD8A"))
# Multiple layers
fetch_feature_data(seurat_obj, features = "CD3D",
layer = c("counts", "data"))
# With embeddings and specific metadata
fetch_feature_data(seurat_obj, features = c("CD3D", "CD14"),
reductions = "umap",
metadata = c("seurat_clusters", "condition"))
# Explicit assay
fetch_feature_data(seurat_obj, features = "CD3D",
assay = "RNA", layer = "counts")
} # }