scmagnify.tools.lineage_classifer

scmagnify.tools.lineage_classifer#

scmagnify.tools.lineage_classifer(data, modal='RNA', time_key='palantir_pseudotime', fate_prob_key='cellrank_fate_probabilities', q=0.01, eps=0.01, celltype_key='celltype', min_cells_per_type=10, filter_small_celltypes=True, key_added='cell_state_masks', save_tmp=True)#

Select cells along lineage branches using pseudotime and fate probabilities.

Parameters:

  • data (AnnData | MuData) – Single cell data object. Can be an anndata.AnnData, mudata.MuData, scmagnify.GRNMuData

  • modal (str (default: 'RNA')) – Modality key (e.g., ‘RNA’, ‘ATAC’) when using multi-modal data. mudata.MuData or scmagnify.GRNMuData must be provided.

  • time_key (str (default: 'palantir_pseudotime')) – Key in obs that stores pseudotime values.

  • fate_prob_key (str (default: 'cellrank_fate_probabilities')) – Key in adata.obsm for fate probabilities.

  • q (float (default: 0.01)) – Quantile to set dynamic thresholds (0–1). Default 1e-2.

  • eps (float (default: 0.01)) – Small constant subtracted from the threshold. Default 1e-2.

  • celltype_key (str (default: 'celltype')) – Key in adata.obs that stores cell type annotations. Default “celltype”.

  • min_cells_per_type (int (default: 10)) – Minimum number of cells per cell type within each lineage. Cell types with fewer cells than this threshold are removed from that lineage. Default 10.

  • filter_small_celltypes (bool (default: True)) – Whether to apply lineage-wise low-count cell type filtering. Default True.

  • key_added (str (default: 'cell_state_masks')) – Key under which boolean masks are stored in adata.obsm.

  • save_tmp (bool (default: True)) – Whether to save masks to CSV under settings.tmpfiles_dir.

Returns:

adata.obsm[key_added]

DataFrame of boolean masks per fate.

AnnData | MuData

Data with lineage masks stored in .obsm.

Contents