Slingshot Tools

Configuration File: remote_tools/slingshot_tools.json Tool Type: Remote Tools Count: 1

This page contains all tools defined in the slingshot_tools.json configuration file.

Available Tools

run_slingshot_trajectory (Type: RemoteTool)

Infer single-cell lineages and pseudotime with Slingshot (Street 2018): from a low-dimensional em…

run_slingshot_trajectory tool specification

Tool Information:

  • Name: run_slingshot_trajectory

  • Type: RemoteTool

  • Description: Infer single-cell lineages and pseudotime with Slingshot (Street 2018): from a low-dimensional embedding (obsm) + cluster labels (obs), order clusters into smooth lineages via a minimum spanning tree + simultaneous principal curves, and assign each cell a pseudotime along every lineage it belongs to. Robust for tree-shaped differentiation. Returns the lineage structure (ordered cluster sequences) and per-cell pseudotime; optionally anchor with a known start and/or terminal clusters. Input is a server-accessible .h5ad.

Parameters:

  • adata_path (string) (required) Server-accessible path to an .h5ad AnnData with a reduced embedding in obsm and cluster labels in obs.

  • embedding_key (string) (optional) obsm key of the embedding to build the trajectory on (default ‘X_pca’; e.g. ‘X_umap’).

  • cluster_key (string) (required) obs column with cluster labels whose centroids are ordered into lineages.

  • start_cluster (string) (optional) Known root cluster to start every lineage from (optional but recommended for directionality).

  • end_clusters (array) (optional) Known terminal cluster(s) to force as lineage endpoints (optional).

  • n_dims (integer) (optional) Use only the first n_dims columns of the embedding (default: all; for X_pca, 10-20 is typical).

Example Usage:

query = {
    "name": "run_slingshot_trajectory",
    "arguments": {
        "adata_path": "example_value",
        "cluster_key": "example_value"
    }
}
result = tu.run(query)