stereo.algorithm.paste.pairwise_align#

stereo.algorithm.paste.pairwise_align(sliceA, sliceB, alpha=0.1, dissimilarity='kl', use_rep=None, G_init=None, a_distribution=None, b_distribution=None, norm=False, numItermax=200, filter_gene=True, backend=<ot.backend.NumpyBackend object>, use_gpu=False, return_obj=False, verbose=False, gpu_verbose=True, **kwargs)[source]#

Calculates and returns optimal alignment of two slices.

Parameters:

  • sliceA (StereoExpData) – Slice A to align.

  • sliceB (StereoExpData) – Slice B to align.

  • alpha (float) – Alignment tuning parameter. Note: 0 <= alpha <= 1.

  • dissimilarity (str) – Expression dissimilarity measure: 'kl' or 'euclidean'.

  • use_rep (Optional[str]) – If None, uses slice.X to calculate dissimilarity between spots, otherwise uses the representation given by slice.obsm[use_rep].

  • optional) (b_distribution (array-like,) – Initial mapping to be used in FGW-OT, otherwise default is uniform mapping.

  • optional) – Distribution of sliceA spots, otherwise default is uniform.

  • optional) – Distribution of sliceB spots, otherwise default is uniform.

  • numItermax (int) – Max number of iterations during FGW-OT.

  • norm (bool) – If True, scales spatial distances such that neighboring spots are at distance 1. Otherwise, spatial distances remain unchanged.

  • backend – Type of backend to run calculations. For list of backends available on system: ot.backend.get_backend_list().

  • use_gpu (bool) – If True, use gpu. Otherwise, use cpu. Currently we only have gpu support for Pytorch.

  • return_obj (bool) – If True, additionally returns objective function output of FGW-OT.

  • verbose (bool) – If True, FGW-OT is verbose.

  • gpu_verbose (bool) – If True, print whether gpu is being used to user.

Return type:

Tuple[ndarray, Optional[int]]

Returns:

Alignment of spots. If return_obj = True, additionally returns objective function output of FGW-OT.