scanpy.tl.louvain

Contents

scanpy.tl.louvain#

scanpy.tl.louvain(adata, resolution=None, *, random_state=0, restrict_to=None, key_added='louvain', adjacency=None, flavor='vtraag', directed=True, use_weights=False, partition_type=None, partition_kwargs=mappingproxy({}), neighbors_key=None, obsp=None, copy=False)[source]#

Cluster cells into subgroups [Blondel et al., 2008, Levine et al., 2015, Traag, 2015].

Deprecated since version 1.12.0: Use scanpy.tl.leiden() instead.

Cluster cells using the Louvain algorithm [Blondel et al., 2008] in the implementation of Traag [2015]. The Louvain algorithm was proposed for single-cell analysis by Levine et al. [2015].

This requires having run neighbors() or bbknn() first, or explicitly passing a adjacency matrix.

Parameters:
adata AnnData

The annotated data matrix.

resolution float | None (default: None)

For the default flavor ('vtraag') or for `RAPIDS`, you can provide a resolution (higher resolution means finding more and smaller clusters), which defaults to 1.0. See “Time as a resolution parameter” in Lambiotte et al. [2014].

random_state int | RandomState | None (default: 0)

Change the initialization of the optimization.

restrict_to tuple[str, Sequence[str]] | None (default: None)

Restrict the clustering to the categories within the key for sample annotation, tuple needs to contain (obs_key, list_of_categories).

key_added str (default: 'louvain')

Key under which to add the cluster labels. (default: 'louvain')

adjacency csr_array | csc_array | csr_matrix | csc_matrix | None (default: None)

Sparse adjacency matrix of the graph, defaults to neighbors connectivities.

flavor Literal['vtraag', 'igraph', 'rapids'] (default: 'vtraag')

Choose between to packages for computing the clustering.

'vtraag'

Much more powerful than 'igraph', and the default.

'igraph'

Built in igraph method.

'rapids'

GPU accelerated implementation.

Deprecated since version 1.10.0: Use rapids_singlecell.tl.louvain() instead.

directed bool (default: True)

Interpret the adjacency matrix as directed graph?

use_weights bool (default: False)

Use weights from knn graph.

partition_type type[MutableVertexPartition] | None (default: None)

Type of partition to use. Only a valid argument if flavor is 'vtraag'.

partition_kwargs Mapping[str, Any] (default: mappingproxy({}))

Key word arguments to pass to partitioning, if vtraag method is being used.

neighbors_key str | None (default: None)

Use neighbors connectivities as adjacency. If not specified, louvain looks .obsp[‘connectivities’] for connectivities (default storage place for pp.neighbors). If specified, louvain looks .obsp[.uns[neighbors_key][‘connectivities_key’]] for connectivities.

obsp str | None (default: None)

Use .obsp[obsp] as adjacency. You can’t specify both obsp and neighbors_key at the same time.

copy bool (default: False)

Copy adata or modify it inplace.

Return type:

AnnData | None

Returns:

Returns None if copy=False, else returns an AnnData object. Sets the following fields:

adata.obs['louvain' | key_added]pandas.Series (dtype category)

Array of dim (number of samples) that stores the subgroup id ('0', '1', …) for each cell.

adata.uns['louvain' | key_added]['params']dict

A dict with the values for the parameters resolution, random_state, and n_iterations.

Contents