Scvelo Tools

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

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

Available Tools

run_scvelo_velocity (Type: RemoteTool)

Run scVelo (RNA velocity; Bergen et al., Nature Biotechnology 2020) on a single-cell AnnData and …

run_scvelo_velocity tool specification

Tool Information:

  • Name: run_scvelo_velocity

  • Type: RemoteTool

  • Description: Run scVelo (RNA velocity; Bergen et al., Nature Biotechnology 2020) on a single-cell AnnData and return a velocity-derived pseudotime ordering of cells plus a per-cell velocity confidence. The input .h5ad MUST contain spliced and unspliced count layers (e.g. from velocyto, kb-python / kallisto|bustools, or STARsolo); without them RNA velocity cannot be estimated and the tool returns an error. Pipeline: filter_and_normalize -> moments -> velocity -> velocity_graph -> velocity_pseudotime + velocity_confidence. When a cluster_key is supplied, also returns the mean pseudotime per cluster (a coarse trajectory ordering).

Parameters:

  • adata_path (string) (required) Server-accessible path or URL to an .h5ad AnnData that contains spliced and unspliced layers.

  • cluster_key (string) (optional) obs column to summarize pseudotime by (optional); returns mean velocity_pseudotime per cluster as a coarse ordering.

  • mode (string) (optional) Velocity estimation mode: ‘deterministic’, ‘stochastic’ (default), or ‘dynamical’.

Example Usage:

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