spdata

spdata#

Functions to work with spatialdata objects.

scportrait.tools.spdata.write_image(sdata: SpatialData, image: ndarray | DataArray | DataTree | Array, image_name: str, channel_names: list[str] = None, scale_factors: list[int] = None, chunks: tuple[int, int, int] | None = None, overwrite=False, transform: None = None, rgb: bool = False)#

Add the image to the spatialdata object

Parameters:

  • image – Image to be written to the spatialdata object. Can be an array (numpy or dask) or an already validated image model.

  • image_name – Name of the image to be written to the spatialdata object.

  • channel_names – List of channel names for the image. Default is None.

  • scale_factors – List of scale factors for the image. Default is [2, 4, 8]. This will load the image at 4 different resolutions to allow for fluid visualization.

  • chunks – Chunk size for the image. Default is (1, 1000, 1000).

  • overwrite – Whether to overwrite existing data. Default is False.

scportrait.tools.spdata.add_element_sdata(sdata: SpatialData, element: Any, element_name: str, overwrite: bool = True)#

Add an element to the SpatialData object.

Parameters:

  • sdata – SpatialData object

  • element – Element to add

  • element_name – Name of the element to be added

  • overwrite – Whether to overwrite the element if it already exists

Returns:

the SpatialData object is updated on file

Return type:

None

processing (pp)#

scportrait.tools.spdata.pp.percentile_normalize_image(sdata: SpatialData, image_name: str, lower_percentile: float = 0.1, upper_percentile: float = 99.9, rescaled_image_name: str | None = None, scale_factors: list[int] | None = None, overwrite: bool = True, chunks: tuple[int, int, int] | None = None) → None#

Percentile Normalize an image in a spatialdata object.

Parameters:

  • sdata – SpatialData object containing the image to be percentile normalized.

  • image_name – Name of the image to be percentile normalized.

  • lower_percentile – Lower percentile for normalization. Default is 0.1.

  • upper_percentile – Upper percentile for normalization. Default is 99.9.

  • rescaled_image_name – Name of the rescaled image. Default is None.

  • overwrite – Whether to overwrite existing data. Default is True.

  • chunks – Chunk size for the image. Default is None.

Returns:

the SpatialData object is updated on file with a new element containing the rescaled image with the name “rescaled_image_name”. If this is not provided, the name will be “{image_name}_rescaled”. If the name is identical to the input image the input image will be overwritten.

Contents