leiden(adata, resolution=1, *, restrict_to=None, random_state=0, key_added='leiden', adjacency=None, directed=True, use_weights=True, n_iterations=-1, partition_type=None, copy=False)¶
Cluster cells into subgroups [Traag18].
Cluster cells using the Leiden algorithm [Traag18], an improved version of the Louvain algorithm [Blondel08]. The Louvain algorithm has been proposed for single-cell analysis by [Levine15].
This requires having ran
The annotated data matrix.
A parameter value controlling the coarseness of the clustering. Higher values lead to more clusters. Set to None if overriding partition_type to one that doesn’t accept a resolution_parameter.
Change the initialization of the optimization.
Restrict the clustering to the categories within the key for sample annotation, tuple needs to contain (obs_key, list_of_categories).
adata.obs key under which to add the cluster labels. (default: ‘leiden’)
Sparse adjacency matrix of the graph, defaults to adata.uns[‘neighbors’][‘connectivities’].
Whether to treat the graph as directed or undirected.
If True, edge weights from the graph are used in the computation (placing more emphasis on stronger edges).
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.
Whether to copy adata or modify it inplace.
Any further arguments to pass to ~leidenalg.find_partition (which in turn passes arguments to the partition_type).
Array of dim (number of samples) that stores the subgroup id (‘0’, ‘1’, …) for each cell.
A dict with the values for the parameters resolution, random_state, and n_iterations.