stereo.algorithm.regulatory_network_inference.RegulatoryNetworkInference.main#
- RegulatoryNetworkInference.main(database=None, motif_anno=None, tfs=None, target_genes=None, auc_threshold=0.5, num_workers=None, res_key='regulatory_network_inference', seed=None, cache=False, cache_res_key='regulatory_network_inference', save_regulons=True, save_loom=False, fn_prefix=None, method='grnboost', ThreeD_slice=False, prune_kwargs={}, hotspot_kwargs={}, use_raw=True)[source]#
Enables researchers to infer transcription factors (TFs) and gene regulatory networks.
- Parameters:
database (
Optional
[str
]) – the sequence of databases.motif_anno (
Optional
[str
]) – the name of the file that contains the motif annotations to use.tfs (
Union
[str
,list
,None
]) – list of target transcription factors. If None or ‘all’, the list of gene_names will be used.target_genes (
Optional
[list
]) – optional list of gene names (strings). Required when a (dense or sparse) matrix is passed as ‘expression_data’ instead of a DataFrameauc_threshold (
float
) – the fraction of the ranked genome to take into account for the calculation of the Area Under the recovery Curve.num_workers (
Optional
[int
]) – if not using a cluster, the number of workers to use for the calculation. None of all available CPUs need to be used.res_key (
str
) – the key for storage of inference regulatory network result.seed (
Optional
[int
]) – optional random seed for the regressors. Default None.cache (
bool
) – whether to use cache files. Need to provide adj.csv, motifs.csv and auc.csv.save_regulons (
bool
) – whether to save regulons into a csv file.save_loom (
bool
) – whether to save the result as a loom file.fn_prefix (
Optional
[str
]) – the prefix of file name for saving regulons or loom.method (
str
) – the method to inference GRN, ‘grnboost’ or ‘hotspot’.ThreeD_slice (
bool
) – whether to use 3D slice data.prune_kwargs (
dict
) – dict, other parameters of pyscenic.prune.prune2df.hotspot_kwargs (
dict
) – dict, other parameters for ‘hotspot’ method.
- Returns:
Computation result of inference regulatory network is stored in self.result where the result key is ‘regulatory_network_inference’.