Cellrank Tools¶
Configuration File: remote_tools/cellrank_tools.json
Tool Type: Remote
Tools Count: 1
This page contains all tools defined in the cellrank_tools.json configuration file.
Available Tools¶
run_cellrank_fate (Type: RemoteTool)¶
Map single-cell fate with CellRank 2 (Weiler 2024 / Lange 2022): build a Markov transition matrix…
run_cellrank_fate tool specification
Tool Information:
Name:
run_cellrank_fateType:
RemoteToolDescription: Map single-cell fate with CellRank 2 (Weiler 2024 / Lange 2022): build a Markov transition matrix over cells with a selectable kernel (connectivity = kNN graph only, pseudotime, or RNA-velocity), coarse-grain it with a GPCCA estimator to identify terminal macrostates, and compute each cell’s probability of reaching each terminal state — the standard differentiation/lineage readout. Returns the terminal states and the per-cell fate-probability matrix (plus per-cluster mean fate probabilities when a cluster_key is supplied). Input is a server-accessible .h5ad.
Parameters:
adata_path(string) (required) Server-accessible path or URL to an .h5ad AnnData (log-normalized; raw scVelo velocity layers required only for kernel=’velocity’).kernel(string) (optional) Transition kernel: ‘connectivity’ (default; needs only a kNN graph), ‘pseudotime’ (needs pseudotime_key), or ‘velocity’ (needs scVelo velocity layers).n_states(integer) (optional) Number of macrostates to compute (default 3); terminal states are auto-selected from these.pseudotime_key(string) (optional) obs column with a precomputed pseudotime (required when kernel=’pseudotime’, e.g. ‘dpt_pseudotime’).cluster_key(string) (optional) obs column with cluster/cell-type labels; used to name macrostates and report per-cluster mean fate probabilities (optional).
Example Usage:
query = {
"name": "run_cellrank_fate",
"arguments": {
"adata_path": "example_value"
}
}
result = tu.run(query)