Skip to contents

Save a Seurat object to a Loom file

Source: R/SaveLoom.R
writeLoom.Rd

Export a Seurat object to Loom format (HDF5-based file format optimized for storing annotated matrices). Loom files are compatible with the loompy Python package and other tools in the bioinformatics community. This format is useful for sharing data with Python-based analysis workflows or archiving analysis results.

Usage

writeLoom(object, filename, overwrite = FALSE, verbose = TRUE, ...)

as.loom(x, ...)

# Default S3 method
writeLoom(object, filename, overwrite = FALSE, verbose = TRUE, ...)

# S3 method for class 'Seurat'
writeLoom(
  object,
  filename = paste0(Project(object = object), ".loom"),
  overwrite = FALSE,
  verbose = TRUE,
  ...
)

# Default S3 method
as.loom(x, filename, overwrite = FALSE, verbose = TRUE, ...)

# S3 method for class 'H5File'
as.loom(x, ...)

# S3 method for class 'Seurat'
as.loom(
  x,
  filename = paste0(Project(object = x), ".loom"),
  overwrite = FALSE,
  verbose = TRUE,
  ...
)

Arguments

object

A Seurat object to save

filename

Name of file to save the object to. If not provided, defaults to <ProjectName>.loom. The .loom extension is added automatically if not present.

overwrite

Logical; if TRUE, overwrite an existing file. Default is FALSE.

verbose

Logical; if TRUE (default), show progress updates

...

Arguments passed to other methods

x

A Seurat object to convert

Value

writeLoom: Invisibly returns the filename of the saved file

as.loom: A loom object

Details

The Loom format organizes data as follows:

  • /matrix: Main expression matrix (features x cells)

  • /row_attrs: Feature/gene-level annotations

  • /col_attrs: Cell/sample-level metadata (cell names, cluster assignments, etc.)

  • /layers: Additional expression layers if present

When saving a Seurat object:

  • Default assay data becomes the main matrix

  • Cell metadata and feature annotations are preserved

  • Dimensional reductions are stored in col_attrs

  • The SEURAT_ASSAY attribute stores the assay name for roundtrip loading

See also

readLoom to load Loom files back as Seurat objects writeH5Seurat to save in h5Seurat format scConvert for converting between formats Loom documentation

Examples

if (FALSE) { # \dontrun{
library(Seurat)
library(scConvert)

# Create a Seurat object (or use an existing one)
seurat_obj <- CreateSeuratObject(counts = pbmc_small$RNA@counts)

# Save to Loom format
writeLoom(seurat_obj, filename = "my_data.loom")

# Save with overwrite if needed
writeLoom(seurat_obj, filename = "my_data.loom", overwrite = TRUE)

# Load it back
loaded_obj <- readLoom("my_data.loom")

# For sharing with Python tools
writeLoom(seurat_obj, filename = "data_for_python.loom")
# Use in Python with: adata = loompy.connect("data_for_python.loom")
} # }