Skip to contents

This function processes the metadata and pairwise matrices in a `SingleCellExperiment` object. It organizes metadata into distinct categories (`uns`, `obsp`, `obsm`, `varp`, `varm`) based on their dimensions and relationship to the main data matrix (`X`). Metadata matrices matching observation or variable dimensions are moved into the corresponding slots, while remaining metadata is retained in `uns`. Pairwise matrices for observations and variables are identified and organized into `obsp` and `varp`, respectively.

Usage

process_metadata_and_pairwise(sce, alt_exps_data, X, obsm = list())

Arguments

sce

A `SingleCellExperiment` object containing the data to be processed.

alt_exps_data

A list of alternative experiment data to be added to the metadata.

X

The main data matrix (`anndata.X`) for the AnnData object. This is used to validate dimensions of metadata and pairwise matrices.

obsm

A list of observation matrices, which can be extended during processing.

Value

A list containing:

uns

A list of remaining metadata that was not categorized into other slots.

obsp

A list of pairwise observation matrices.

varp

A list of pairwise variable matrices.

obsm

A list of observation matrices.

varm

A list of variable matrices.