opendvp.tl.scimap_spatial_lda

opendvp.tl.scimap_spatial_lda#

scimap_spatial_lda(adata, x_coordinate='X_centroid', y_coordinate='Y_centroid', z_coordinate=None, phenotype='phenotype', method='radius', radius=30, knn=10, imageid='imageid', num_motifs=10, random_state=0, subset=None, verbose=True, label='spatial_lda', **kwargs)#
Parameters:
adata (anndata.AnnData):

AnnData object, containing spatial gene expression data.

x_coordinate (str, required):

Column name in adata denoting the x-coordinates.

y_coordinate (str, required):

Column name in adata denoting the y-coordinates.

z_coordinate (str, optional):

Column name in adata for z-coordinates, for 3D spatial data.

phenotype (str, required):

Column name in adata indicating cell phenotype or classification.

method (str, optional):

Neighborhood definition method: ‘radius’ for fixed distance, ‘knn’ for K nearest neighbors.

radius (int, optional):

Radius defining local neighborhoods (when method=’radius’).

knn (int, optional):

Number of nearest neighbors for neighborhood definition (when method=’knn’).

imageid (str, optional):

Column name in adata specifying image identifiers, for analyses within specific images.

num_motifs (int, optional):

Number of latent motifs to identify.

random_state (int, optional):

Seed for random number generator, ensuring reproducibility.

subset (str, optional):

Specific image identifier for targeted analysis.

verbose (bool, optional):

If True, enables progress and informational messages.

label (str, optional):

Custom label for storing results in adata.uns.

Returns:

The input adata object, updated with spatial LDA results in adata.uns[label].

Return type:

anndata.AnnData

Examples

# Analyze spatial motifs using the radius method
adata = dvp.tl.scimap_spatial_lda(
    adata,
    x_coordinate="X_centroid",
    y_coordinate="Y_centroid",
    method="radius",
    radius=50,
    num_motifs=10,
    label="lda_radius_50",
)

# KNN method with specific image subset
adata = dvp.tl.scimap_spatial_lda(
    adata,
    x_coordinate="X_centroid",
    y_coordinate="Y_centroid",
    method="knn",
    knn=15,
    num_motifs=15,
    subset="image_01",
    label="lda_knn_15_image_01",
)

# 3D spatial data analysis using the radius method
adata = dvp.tl.scimap_spatial_lda(
    adata,
    x_coordinate="X_centroid",
    y_coordinate="Y_centroid",
    z_coordinate="Z_centroid",
    method="radius",
    radius=100,
    num_motifs=20,
    label="lda_3D_radius_100",
)
Contents