stlearn.spatial.morphology.adjust

stlearn.spatial.morphology.adjust(adata: AnnData, use_data: str = 'X_pca', radius: float = 50.0, rates: int = 1, method: Literal['mean', 'median', 'sum'] = 'mean', similarity_matrix: Literal['cosine', 'euclidean', 'pearson', 'spearman'] = 'cosine', copy: bool = False) → AnnData | None[source]

SME normalisation: Using spot location information and tissue morphological features to correct spot gene expression

Parameters:

  • adata (AnnData) – Annotated data matrix.

  • use_data (str, default "X_pca") – Input date to be adjusted by morphological features. choose one from [“raw”, “X_pca”, “X_umap”]

  • radius (float, default 50.0) – Radius to select neighbour spots.

  • rates (int, default 1) – Number of times to add the aggregated neighbor contribution. Higher values increase the strength of morphological adjustment.

  • method ({'mean', 'median', 'sum'}, default 'mean') – Method for aggregating neighbor contributions.

  • similarity_matrix ({'cosine', 'euclidean', 'pearson', 'spearman'}, default 'cosine') – Method to calculate morphological similarity between spots.

  • copy (bool, default False) – Return a copy instead of writing to adata.

Returns:

  • Depending on copy, returns or updates adata with the following fields.

  • **[use_data]_morphology** (adata.obsm field) – Add SME normalised gene expression matrix