scrna-orchestrator

Coordinate end-to-end single-cell RNA-seq analysis from raw counts or 10x inputs.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/ya-way/cytoclaw-skills --skill scrna-orchestrator-ya-way
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: scrna-orchestrator
Source: https://github.com/ya-way/cytoclaw-skills/tree/main/workspace/skills/scrna-orchestrator
Command: npx skills add https://github.com/ya-way/cytoclaw-skills --skill scrna-orchestrator-ya-way

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy, pandas, scanpy, anndata, matplotlib.

What problem does it solve?

This Skill provides a local, reproducible end-to-end pipeline for single-cell RNA-seq analysis using Scanpy, including QC, optional doublet detection, clustering, marker discovery, and optional CellTypist annotation, with support for latent downstream modes and contrastive marker analysis from raw counts or 10x inputs.

Core Features & Use Cases

  • End-to-end single-cell RNA-seq workflow: QC, normalization, clustering, and marker analysis.
  • Optional advanced steps: doublet detection with Scrublet, local CellTypist annotation, and two-group contrastive marker analysis.
  • Flexible inputs: raw-count .h5ad, integrated latent representations (e.g., X_scvi), or 10x Matrix Market formats; outputs include a reproducible report, figures, and tables.

Quick Start

Load an input file (h5ad or 10x) and run the orchestrator to generate a reproducible analysis bundle.

Frequently Asked Questions about scrna-orchestrator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an end-to-end scRNA-seq analysis workflow locally from raw counts?

You can run a local scRNA-seq analysis workflow by loading raw-count h5ad or 10x Matrix Market inputs into an orchestrator that coordinates QC, normalization, clustering, and marker discovery to generate a reproducible report bundle.

Can I use CellTypist for cell type annotation within a Scanpy clustering workflow?

Yes, CellTypist can be used for local cell type annotation within a Scanpy workflow. The orchestrator supports optional CellTypist annotation alongside clustering and marker discovery for comprehensive single-cell analysis.

What is the best way to perform two-group contrastive marker analysis on single-cell data?

Two-group contrastive marker analysis can be performed by running an end-to-end scRNA-seq pipeline that supports direct comparison between specified groups, yielding contrastive marker tables alongside standard clustering and QC reports.

Does the scRNA-seq pipeline support integrated latent representations like scVI as input?

Yes, the pipeline supports integrated latent representations, such as X_scvi, as inputs. It processes these latent downstream modes directly, bypassing early preprocessing to generate clustering and marker analysis outputs.

Do I need Scrublet for doublet detection in my single-cell RNA-seq analysis?

Scrublet is an optional dependency for doublet detection in single-cell RNA-seq analysis. The pipeline enforces reproducibility by producing reports and tables with or without this advanced step, though Scanpy and Anndata are required.

What limitations exist when using Scanpy for reproducible single-cell clustering and marker discovery?

Limitations include the requirement for Scanpy and Anndata environments, while Scrublet and CellTypist remain optional. Inputs must be raw-count h5ad, 10x Matrix Market formats, or integrated latent representations to ensure reproducible analysis bundle generation.