omicverse.space.calculate_gene_signature

omicverse.space.calculate_gene_signature(adata_sc, clustertype, rank=True, key='rank_genes_groups', foldchange=2, topgenenumber=20)[source]

Build a marker-gene signature table for each cell type in a reference scRNA-seq dataset.

Parameters:

  • adata_sc (AnnData) – Single-cell reference AnnData.

  • clustertype (str) – Cell-type label key in adata_sc.obs.

  • rank (bool) – Whether to use ranked markers from differential expression analysis.

  • key (str) – adata.uns key for ranked genes (used when rank=True).

  • foldchange (float) – Fold-change threshold for marker selection.

  • topgenenumber (int) – Number of top marker genes retained per cell type.

Returns:

Gene-signature matrix where each column corresponds to one cell type and each cell stores a marker gene (padded with NA when needed).

Return type:

pandas.DataFrame

Examples

>>> gene_sig = ov.space.calculate_gene_signature(adata_sc, clustertype='celltype', topgenenumber=50)