c2v.seurat.integrate_data#
- c2v.seurat.integrate_data(adata, batch_key, method='cca', layer_name='integrated', scale_integrated=True, scale_max_value=10.0, l2_norm=True, n_comps=30, k_anchor=5, k_filter=200, k_score=30, max_features=200, n_trees=50, eps=0, k_weight=100, progress_bar=True, only_anchors=False, mnn_graph_key='seurat_mnn', uns_key='seurat_integration')#
Performs single-cell data integration via Seurat’s CCA or RPCA approach.
- Parameters:
- adata sc.AnnData
Annotated data matrix.
- batch_key str
Key in adata.obs that contains batch information.
- method Literal[‘cca’, ‘rpca’] (default:
'cca') Integration method to use. Either “cca” or “rpca”.
- layer_name str (default:
'integrated') Name of the layer to store the integrated expression matrix.
- scale_integrated bool (default:
True) Whether to scale the integrated expression matrix.
- scale_max_value float (default:
10.0) Maximum value to scale the integrated expression matrix.
- l2_norm bool (default:
True) Perform L2 normalization on the CCA cell embeddings after dimensional reduction.
- n_comps int (default:
30) Which dimensions to use from the CCA to specify the neighbor search space.
- k_anchor int (default:
5) How many neighbors (k) to use when picking anchors.
- k_filter int (default:
200) How many neighbors (k) to use when filtering anchors.
- k_score int (default:
30) How many neighbors (k) to use when scoring anchors.
- max_features int (default:
200) The maximum number of features to use when specifying the neighborhood search space in the anchor filtering.
- n_trees int (default:
50) More trees gives higher precision when using annoy approximate nearest neighbor search.
- eps float (default:
0) Error bound on the neighbor finding algorithm.
- k_weight int (default:
100) Number of neighbors to consider when weighting anchors.
- progress_bar bool (default:
True) Whether to show a progress bar.
- return_anchors
Whether to return the anchors.
- only_anchors bool
- mnn_graph_key str
- uns_key str
- Return type:
ro.rinterface.Sexp
- Returns:
anchors Seurat anchors object.