ehrapy.tools.leiden

Contents

ehrapy.tools.leiden#

ehrapy.tools.leiden(edata, resolution=1, *, restrict_to=None, random_state=0, key_added='leiden', adjacency=None, use_weights=True, n_iterations=-1, neighbors_key=None, obsp=None, copy=False, **clustering_args)[source]#

Cluster observations into subgroups [TWvE19].

Cluster observations using the Leiden algorithm [TWvE19], an improved version of the Louvain algorithm [BGLL08]. It has been proposed for single-cell analysis by [LSB+15]. This requires having run neighbors(). Uses the igraph implementation (flavor="igraph" in scanpy); leidenalg is not supported.

Parameters:

  • edata (EHRData) – Central data object.

  • resolution (float, default: 1) – A parameter value controlling the coarseness of the clustering. Higher values lead to more clusters.

  • 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).

  • random_state (int | RandomState | None, default: 0) – Random seed of the initialization of the optimization.

  • key_added (str, default: 'leiden') – edata.obs key under which to add the cluster labels.

  • adjacency (spmatrix | None, default: None) – Sparse adjacency matrix of the graph, defaults to neighbors connectivities.

  • use_weights (bool, default: True) – If True, edge weights from the graph are used in the computation (placing more emphasis on stronger edges).

  • n_iterations (int, default: -1) – How many iterations of the Leiden clustering algorithm to perform. Positive values above 2 define the total number of iterations to perform. -1 has the algorithm run until it reaches its optimal clustering.

  • neighbors_key (str | None, default: None) – Use neighbors connectivities as adjacency. If not specified, leiden looks .obsp[‘connectivities’] for connectivities (default storage place for pp.neighbors). If specified, leiden 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) – Whether to copy edata or modify it inplace.

  • **clustering_args – Any further arguments passed to igraph.Graph.community_leiden.

Return type:

EHRData | None

Returns:

edata.obs[key_added] Array of dim (number of samples) that stores the subgroup id (‘0’, ‘1’, …) for each cell.

edata.uns[‘leiden’][‘params’] A dict with the values for the parameters resolution, random_state, and n_iterations.

Contents