cospar.pp.remove_cell_cycle_correlated_genes

cospar.pp.remove_cell_cycle_correlated_genes(adata, cycling_gene_list=['Ube2c', 'Hmgb2', 'Hmgn2', 'Tuba1b', 'Ccnb1', 'Tubb5', 'Top2a', 'Tubb4b'], corr_threshold=0.1, confirm_change=False)

Remove cell-cycle correlated genes.

Take pre-selected highly variable genes, and compute their correlation with the set of cell cycle genes. If confirm_change=True, remove those having absolute correlation score are above given correlation threshold. It is a prerequisite to run get_highly_variable_genes() first.

Warning: the default cell cycle gene sets are from mouse genes. Please convert them to upper case if you want to apply it to human data. Also, consider using your own gene sets.

Parameters

adata : AnnData object

cycling_gene_list : list, optional

A list of cell cycle correlated genes to compute correlation with.

corr_threshold : float, optional (default: 0.1)

Highly variable genes with absolute correlation score about this threshold will be removed from the highly variable gene list.

confirm_change : bool, optional (default: False)

If set True, adata.var[‘highly_variable’] will be updated to exclude cell cycle correlated genes.