ehrapy.tools.louvain#

ehrapy.tools.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 observations into subgroups [Blondel08] [Levine15] [Traag17].

Cluster observations using the Louvain algorithm [Blondel08] in the implementation of [Traag17]. The Louvain algorithm has been proposed for single-cell analysis by [Levine15]. This requires having ran neighbors() or bbknn() first, or explicitly passing a adjacency matrix.

Parameters:

  • adata (AnnData) – AnnData object object containing all observations.

  • resolution (Optional[float]) – For the default flavor ('vtraag'), 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 [Lambiotte09].

  • random_state (Union[None, int, RandomState]) – Random seed of the initialization of the optimization.

  • restrict_to (Optional[Tuple[str, Sequence[str]]]) – Restrict the clustering to the categories within the key for sample annotation, tuple needs to contain (obs_key, list_of_categories).

  • key_added (str) – Key under which to add the cluster labels. (default: 'louvain')

  • adjacency (Optional[spmatrix]) – Sparse adjacency matrix of the graph, defaults to neighbors connectivities.

  • flavor (Literal['vtraag', 'igraph', 'rapids']) – Choose between to packages for computing the clustering. 'vtraag' is much more powerful, and the default.

  • directed (bool) – Interpret the adjacency matrix as directed graph?

  • use_weights (bool) – Use weights from knn graph.

  • partition_type (Optional[Type[MutableVertexPartition]]) – Type of partition to use. Only a valid argument if flavor is 'vtraag'.

  • partition_kwargs (Mapping[str, Any]) – Key word arguments to pass to partitioning, if vtraag method is being used.

  • neighbors_key (Optional[str]) – 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 (Optional[str]) – Use .obsp[obsp] as adjacency. You can’t specify both obsp and neighbors_key at the same time.

  • copy (bool) – Whether to copy adata or modify it inplace.

Returns:

adata.obs['louvain'] (pandas.Series, dtype category) Array of dim (number of samples) that stores the subgroup id ('0', '1', …) for each observation.

AnnData When copy=True is set, a copy of adata with those fields is returned.

Return type:

By default (copy=False), updates adata with the following fields