omicverse.space.CellLoc

omicverse.space.CellLoc#

class omicverse.space.CellLoc(adata_sc, adata_sp, use_rep_sc='X_pca', use_rep_sp='X_pca')[source]#

SpatRio CellLoc class for probabilistic cell localization.

This class extends CellMap with probabilistic filtering based on cell type proportions for more accurate spatial localization. It provides methods for mapping, saving/loading results, and probabilistic assignment.

Parameters:

  • adata_sc (anndata.AnnData) – Single-cell reference AnnData.

  • adata_sp (anndata.AnnData) – Spatial AnnData to receive probabilistic localization.

  • use_rep_sc (str, default="X_pca") – Representation key in adata_sc.obsm used for transport.

  • use_rep_sp (str, default="X_pca") – Representation key in adata_sp.obsm used for transport.

  • Attributes –

    adata_sc: AnnData

    Single-cell RNA sequencing data.

    adata_sp: AnnData

    Spatial transcriptomics data.

    use_rep_sc: str

    Representation key for single-cell data.

    use_rep_sp: str

    Representation key for spatial data.

    spatrio_decon: pandas.DataFrame

    Deconvolution results after mapping.

    spatrio_map: pandas.DataFrame

    Coordinate assignment results.

  • Examples – 

    >>> import scanpy as sc
>>> import omicverse as ov
>>> # Load data
>>> adata_sc = sc.read_h5ad('single_cell.h5ad')
>>> adata_sp = sc.read_h5ad('spatial.h5ad')
>>> # Initialize CellLoc
>>> cl = ov.space.CellLoc(
...     adata_sc=adata_sc,
...     adata_sp=adata_sp
... )

__init__(adata_sc, adata_sp, use_rep_sc='X_pca', use_rep_sp='X_pca')[source]#

Initialize CellLoc for probabilistic cell localization.

Parameters:

  • adata_sc (anndata.AnnData) – Single-cell reference object to be localized.

  • adata_sp (anndata.AnnData) – Spatial transcriptomics object containing spot coordinates.

  • use_rep_sc (str, default="X_pca") – Embedding key in adata_sc.obsm used for mapping.

  • use_rep_sp (str, default="X_pca") – Embedding key in adata_sp.obsm used for mapping.

Methods

__init__(adata_sc, adata_sp[, use_rep_sc, ...])

Initialize CellLoc for probabilistic cell localization.

load_map(map_info[, sc_type, sp_type, ...])

Load precomputed transport map into the CellLoc instance.

loc_assign(**kwargs)

Assign coordinates to filtered CellLoc mappings.

loc_map([sc_type, sp_type, alpha, ...])

Run SpatRio mapping for probabilistic localization workflow.

loc_prob(spot_cell_prob[, sc_type, sc_prop, ...])

Filter transport assignments by spot-level cell-type probabilities.

save_map(path)

Save current mapping table as CSV.
Contents