omicverse-single-cell-annotation

Annotate clustered AnnData objects with OmicVerse cell-type labels.

13|2|Updated Apr 5, 2026

Search Tags:#single-cell #omicverse #annotation #anndata #celltypist #gpt4celltype #scsa

One-click install

npx skills add https://github.com/omicverse/omicverse-skills --skill omicverse-single-cell-annotation-omicverse

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: omicverse-single-cell-annotation
Source: https://github.com/omicverse/omicverse-skills/tree/main/src/omicverse_skills/skills/single-cell-annotation
Command: npx skills add https://github.com/omicverse/omicverse-skills --skill omicverse-single-cell-annotation-omicverse

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

This Skill turns a clustered single-cell AnnData object into reusable cell-type annotations, so you do not have to manually rework notebook logic each time you need labels, reference selection, or output validation.

Core Features & Use Cases

Backend Selection: Choose the right annotation path for the task, including CellTypist, gpt4celltype, or SCSA.
Reusable Annotation Workflow: Resolve references, attach models or databases, and run the correct backend on an annotation-ready AnnData object.
Validation and Outputs: Check that the expected prediction columns are written back to adata.obs before downstream plotting or analysis.
Use Case: A researcher with clustered PBMC data can map clusters to cell types, compare backend options, and confirm the resulting annotation columns in one consistent workflow.

Quick Start

Ask the agent to annotate your clustered AnnData object with the OmicVerse backend you want and verify that the matching prediction column appears in adata.obs.

Frequently Asked Questions about omicverse-single-cell-annotation

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

FAQPage Schema

How do I annotate clustered single-cell AnnData objects with cell types?▼

Single-cell annotation backends differ in reference handling: CellTypist uses trained models, gpt4celltype leverages language models, and SCSA uses database references to map clusters to cell types within AnnData.

What is the best way to compare CellTypist, gpt4celltype, and SCSA for single-cell annotation?▼

Run multiple single-cell annotation backends on the same clustered AnnData object, then compare the resulting prediction columns such as celltypist_prediction, gpt4celltype_prediction, and scsa_prediction in adata.obs.

Do I need clustered data before running single-cell annotation workflows?▼

Yes, single-cell annotation requires a preprocessed, clustering-ready AnnData object with a valid cluster_key to map clusters to cell types accurately.

Why are my single-cell annotation prediction columns missing from adata.obs?▼

Prediction columns like celltypist_prediction are missing from adata.obs if cluster_key validation fails or the selected backend does not complete its reference resolution and annotation process correctly.