omicverse.single.SCENIC

omicverse.single.SCENIC#

class omicverse.single.SCENIC(adata, db_glob=None, motif_path=None, n_jobs=8, species=None, data_dir='./data/scenic', download=True, db_names=None)[source]#
__init__(adata, db_glob=None, motif_path=None, n_jobs=8, species=None, data_dir='./data/scenic', download=True, db_names=None)[source]#

Initialize a SCENIC workflow.

Parameters:

  • adata (anndata.AnnData) – Input expression object.

  • db_glob (str, sequence of str or None) – Ranking database feather path pattern or explicit feather file list. If None, species is used to prepare cisTarget resources automatically.

  • motif_path (str or None) – Motif-to-TF annotation table. If None, species is used to prepare the matching motif annotation table automatically.

  • n_jobs (int) – Number of workers used by SCENIC steps.

  • species ({'human', 'mouse', 'fly'} or None) – Species used for automatic cisTarget resource preparation. Aliases such as 'hg38', 'mm10' and 'dm6' are accepted.

  • data_dir (str) – Directory used to cache downloaded cisTarget resources.

  • download (bool) – Whether to download missing automatic resources. If False, missing files raise FileNotFoundError with the expected URLs.

  • db_names (str, sequence of str or None) – Ranking database names to use for automatic resources. Human and mouse support '500bp' and '10kb'; fly supports 'default'. If None, all available gene-based ranking databases for the species are used.

Methods

__init__(adata[, db_glob, motif_path, ...])

Initialize a SCENIC workflow.

cal_grn([method, layer, tf_names])

Infer a GRN and store the edge list on the SCENIC object.

cal_regulons([rho_mask_dropouts, seed])

Contents