sopa.segmentation.shapes.solve_conflicts(cells, threshold=0.5, patch_indices=None, return_indices=False)
+
+Resolve segmentation conflicts (i.e. overlap) after running segmentation on patches
+ + + +Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
cells |
+
+ list[Polygon]
+ |
+
+
+
+ List of cell polygons + |
+ + required + | +
threshold |
+
+ float
+ |
+
+
+
+ When two cells are overlapping, we look at the area of intersection over the area of the smallest cell. If this value is higher than the |
+
+ 0.5
+ |
+
patch_indices |
+
+ ndarray | None
+ |
+
+
+
+ Patch from which each cell belongs. + |
+
+ None
+ |
+
return_indices |
+
+ bool
+ |
+
+
+
+ If |
+
+ False
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ ndarray[Polygon]
+ |
+
+
+
+ Array of resolved cells polygons + |
+
sopa/segmentation/shapes.pysopa.stats.mean_distance(adata, group_key, target_group_key=None, ignore_zeros=False)
+
+Mean distance between two groups (typically, between cell-types, or between cell-types and domains)
+ +The distance is a number of hops, i.e. a distance of 10 between a pDC and a T cell means that there are 10 cells on the closest path from one to the other cell.
+Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
adata |
+
+ AnnData
+ |
+
+
+
+ An |
+ + required + | +
group_key |
+
+ str
+ |
+
+
+
+ Key of |
+ + required + | +
target_group_key |
+
+ str | None
+ |
+
+
+
+ Key of |
+
+ None
+ |
+
ignore_zeros |
+
+ bool
+ |
+
+
+
+ If |
+
+ False
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ DataFrame
+ |
+
+
+
+
|
+
sopa/stats/distance.pysopa.stats.geometrize_niches(adata, niche_key, buffer='auto', perc_area_th=0.05)
+
+Converts the niches to shapely polygons, and put into a GeoDataFrame. Note that each niche can appear multiple times, as they can be separated by other niches ; in this case, we call them different "components" of the same niche ID.
You can show niches components with GeoPandas +
+Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
adata |
+
+ AnnData | SpatialData
+ |
+
+
+
+ An |
+ + required + | +
niche_key |
+
+ str
+ |
+
+
+
+ Key of |
+ + required + | +
buffer |
+
+ int | str
+ |
+
+
+
+ Expansion radius applied on components. By default, |
+
+ 'auto'
+ |
+
perc_area_th |
+
+ float
+ |
+
+
+
+ For each niche, components whose area is less than |
+
+ 0.05
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ GeoDataFrame
+ |
+
+
+
+ A |
+
sopa/stats/morpho.pysopa.stats.niches_geometry_stats(adata, niche_key, aggregation='mean', key_added_suffix='_distance_to_niche_', **geometrize_niches_kwargs)
+
+Computes statistics over niches geometries
+ +n_components: Number of connected component of a niche (a component is a group of neighbor cells with the same niche attribute)length: Mean distance of the exterior/boundary of the components of a nichearea: Mean area of the components of a nicheroundness: Float value between 0 and 1. The higher the value, the closer to a circle. Computed via 4 * pi * area / length**2mean_distance_to_niche_X: mean distance to the niche (between the two closest points of the niches)Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
adata |
+
+ AnnData | SpatialData
+ |
+
+
+
+ An |
+ + required + | +
niche_key |
+
+ str
+ |
+
+
+
+ Key of |
+ + required + | +
aggregation |
+
+ str | list[str]
+ |
+
+
+
+ Aggregation mode. Either one string such as |
+
+ 'mean'
+ |
+
key_added_suffix |
+
+ str
+ |
+
+
+
+ Suffix added in the DataFrame columns. Defaults to "distance_to_niche". + |
+
+ '_distance_to_niche_'
+ |
+
geometrize_niches_kwargs |
+
+ str
+ |
+
+
+
+ Kwargs to the |
+
+ {}
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ GeoDataFrame
+ |
+
+
+
+ A |
+
sopa/stats/morpho.pysopa.stats.cells_to_groups(adata, group_key, key_added_prefix=None, ignore_zeros=False)
+
+Compute the hop-distance between each cell and a cell category/group.
+ + + +Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
adata |
+
+ AnnData
+ |
+
+
+
+ An |
+ + required + | +
group_key |
+
+ str
+ |
+
+
+
+ Key of |
+ + required + | +
key_added_prefix |
+
+ str | None
+ |
+
+
+
+ Prefix to the key added in |
+
+ None
+ |
+
ignore_zeros |
+
+ bool
+ |
+
+
+
+ If |
+
+ False
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ DataFrame | None
+ |
+
+
+
+ A |
+
sopa/stats/distance.pysopa.stats.spatial_neighbors(adata, radius, library_key=None, percentile=None, set_diag=False)
+
+Create a Delaunay graph from spatial coordinates. This function comes from squidpy.
+ + + +Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
adata |
+
+ AnnData | SpatialData
+ |
+
+
+
+ AnnData object + |
+ + required + | +
radius |
+
+ tuple[float, float] | None
+ |
+
+
+
+ tuple that prunes the final graph to only contain edges in interval |
+ + required + | +
library_key |
+
+ str | None
+ |
+
+
+
+ Optional batch key in adata.obs + |
+
+ None
+ |
+
percentile |
+
+ float | None
+ |
+
+
+
+ Percentile of the distances to use as threshold. + |
+
+ None
+ |
+
set_diag |
+
+ bool
+ |
+
+
+
+ Whether to set the diagonal of the spatial connectivities to |
+
+ False
+ |
+
sopa/stats/_build.pysopa.utils.data.uniform(*_, length=2048, cell_density=0.01, n_points_per_cell=50, n_genes=5, c_coords=['DAPI', 'CK', 'CD3', 'CD20'], sigma_factor=0.4, seed=0, save_vertices=False, apply_blur=True)
+
+Generate a dummy dataset composed of cells generated uniformly in a square. It also has transcripts.
+ + + +Parameters:
+| Name | +Type | +Description | +Default | +
|---|---|---|---|
length |
+
+ int
+ |
+
+
+
+ Size of the square, in pixels + |
+
+ 2048
+ |
+
cell_density |
+
+ float
+ |
+
+
+
+ Density of cells per pixel^2 + |
+
+ 0.01
+ |
+
n_points_per_cell |
+
+ int
+ |
+
+
+
+ Mean number of transcripts per cell + |
+
+ 50
+ |
+
n_genes |
+
+ int
+ |
+
+
+
+ Number of gene names + |
+
+ 5
+ |
+
c_coords |
+
+ list[str]
+ |
+
+
+
+ Channel names + |
+
+ ['DAPI', 'CK', 'CD3', 'CD20']
+ |
+
sigma_factor |
+
+ float
+ |
+
+
+
+ Factor used to determine |
+
+ 0.4
+ |
+
seed |
+
+ int
+ |
+
+
+
+ Numpy random seed + |
+
+ 0
+ |
+
save_vertices |
+
+ bool
+ |
+
+
+
+ Whether to save the vertices of the cells (as points) + |
+
+ False
+ |
+
apply_blur |
+
+ bool
+ |
+
+
+
+ Whether to apply gaussian blur on the cells (without blur, cells are just one pixel) + |
+
+ True
+ |
+
Returns:
+| Type | +Description | +
|---|---|
+ SpatialData
+ |
+
+
+
+ A SpatialData object with a 2D image ( |
+
sopa/utils/data.py16 +17 +18 +19 +20 +21 +22 +23 +24 +25 +26 +27 +28 +29 +30 +31 +32 +33 +34 +35 +36 +37 +38 +39 +40 +41 +42 +43 +44 +45 +46 +47 +48 +49 +50 +51 +52 +53 +54 +55 +56 +57 +58 +59 +60 +61 +62 +63 +64 +65 +66 +67 +68 +69 +70 +71 +72 +73 +74 +75 +76 +77 +78 +79 +80 +81 +82 +83 +84 +85 +86 +87 +88 +89 +90 +91 +92 +93 +94 +95 +96 +97 +98 +99 | |
sopa.utils.data.blobs(*_, length=1024, n_points=10000, c_coords=['DAPI', 'CK', 'CD3', 'CD20'], **kwargs)
+
+Adapts the blobs dataset from SpatialData for sopa. Please refer to the SpatialData documentation
+ +sopa/utils/data.py