scvi.external.stereoscope.SpatialDeconv

class scvi.external.stereoscope.SpatialDeconv(n_spots, sc_params, prior_weight='n_obs')

Bases: BaseModuleClass

Model of single-cell RNA-sequencing data for deconvolution of spatial transriptomics.

Reimplementation of the STModel module of Stereoscope [Andersson et al., 2020]: https://github.com/almaan/stereoscope/blob/master/stsc/models.py.

Parameters:

• n_spots (int) – Number of input spots

• sc_params (Tuple[ndarray]) – Tuple of ndarray of shapes [(n_genes, n_labels), (n_genes)] containing the dictionnary and log dispersion parameters

• prior_weight (Literal['n_obs', 'minibatch']) – Whether to sample the minibatch by the number of total observations or the monibatch size

Attributes table

Methods table

generative(x, ind_x)	Build the deconvolution model for every cell in the minibatch.
get_ct_specific_expression(y)	Returns cell type specific gene expression at the queried spots.
get_proportions([keep_noise])	Returns the loadings.
inference()	Inference.
loss(tensors, inference_outputs, ...[, ...])	Loss computation.
sample(tensors[, n_samples, library_size])	Sample from the model.

Attributes

training

SpatialDeconv.training: bool

Methods

generative

SpatialDeconv.generative(x, ind_x)

Build the deconvolution model for every cell in the minibatch.

get_ct_specific_expression

SpatialDeconv.get_ct_specific_expression(y)

Returns cell type specific gene expression at the queried spots.

Parameters:

y – cell types

get_proportions

SpatialDeconv.get_proportions(keep_noise=False)

Returns the loadings.

Return type:

ndarray

inference

SpatialDeconv.inference()

Inference.

loss

SpatialDeconv.loss(tensors, inference_outputs, generative_outputs, kl_weight=1.0, n_obs=1.0)

Loss computation.

Parameters:

• kl_weight (float) –

• n_obs (int) –

sample

SpatialDeconv.sample(tensors, n_samples=1, library_size=1)

Sample from the model.