Sample clustering¶

cluster runs K-means on the sample embedding obsm['X_DR_sample'] and returns a {sample_id: cluster_label} dict. The assignments are the standard input to proportion_test and raisinfit via their sample_to_clade arguments.

Call¶

from sampledisco.sample_clustering.cluster import cluster

sample_clusters, _ = cluster(
    pseudobulk_adata=pseudo_adata,
    output_dir="sampledisco_demo_output/rna",
    number_of_clusters=4,
    random_state=0,
)

Output¶

Writes → sampledisco_demo_output/rna/sample_cluster/:

kmeans_clusters_sample.csv — sample ↔ cluster table.
kmeans_sample_embedding.png — 2D scatter of the first two embedding dimensions, colored by cluster.

The pseudobulk AnnData also gets a cluster_sample_kmeans column in .obs.

Result¶

K-means on sample embedding

Sample clusters on the first two components of the sample embedding.

See the API page for the full parameter list.