Celltypist Tools

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

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

Available Tools

run_celltypist_annotate (Type: RemoteTool)

Annotate single-cell RNA-seq data with CellTypist (Domínguez Conde 2022 / Xu 2023): predict a cel…

run_celltypist_annotate tool specification

Tool Information:

  • Name: run_celltypist_annotate

  • Type: RemoteTool

  • Description: Annotate single-cell RNA-seq data with CellTypist (Domínguez Conde 2022 / Xu 2023): predict a cell-type label for every cell using a pre-trained logistic-regression model and return the per-cell predicted labels plus a cell-type count summary. Optional majority-voting over-clustering harmonises predictions within transcriptionally similar groups. Input is a server-accessible .h5ad; the matrix is log1p-normalised to 10,000 counts per cell before annotation.

Parameters:

  • adata_path (string) (required) Server-accessible path or URL to an .h5ad AnnData of single-cell expression (raw or normalized counts).

  • model (string) (optional) CellTypist pre-trained model name (default ‘Immune_All_Low.pkl’). Other examples: ‘Immune_All_High.pkl’, ‘Human_Lung_Atlas.pkl’, ‘Adult_Mouse_Gut.pkl’.

  • majority_voting (boolean) (optional) Refine predictions by over-clustering so cells in the same group share a label (default true).

Example Usage:

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