anndata

Manage annotated single-cell matrices with AnnData objects for scalable analysis.

22|4|Updated May 25, 2026
One-click install
npx skills add https://github.com/crazymsn/academic-skills --skill anndata-crazymsn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: anndata
Source: https://github.com/crazymsn/academic-skills/tree/main/academic-skills/anndata
Command: npx skills add https://github.com/crazymsn/academic-skills --skill anndata-crazymsn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AnnData provides a robust data structure for handling annotated matrices used in single-cell experiments, consolidating X, obs, var, layers, and embeddings for scalable analysis.

Core Features & Use Cases

  • Data structure: X, obs, var, layers, obsm, varm, uns, raw
  • Integration with Scanpy and Muon; reading/writing; backed mode for large datasets
  • Common workflows: single-cell RNA-seq, batch integration, reproducibility

Quick Start

Create or load an AnnData object and begin exploring its X, obs, var, and embeddings using Scanpy.

Frequently Asked Questions about anndata

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

FAQPage Schema
How do I manage annotated matrices for single-cell RNA-seq data efficiently?

To manage annotated matrices for single-cell RNA-seq data, use the AnnData object to consolidate X, obs, var, layers, and embeddings, enabling scalable analysis within the scverse ecosystem.

What data structure should I use for large-scale scRNA-seq workflows like normalization and embedding?

For large-scale scRNA-seq workflows like normalization and embedding, an annotated matrix structure organizes observations and variables while supporting backed storage to handle large datasets efficiently.

Can I use Scanpy and Muon for batch integration across large datasets?

Yes, you can perform batch integration across large datasets using Scanpy and Muon, which interoperate with the AnnData object to support reproducible single-cell analysis workflows.

Does backed mode support efficient I/O for single-cell data?

Backed mode supports efficient I/O for single-cell data by enabling out-of-memory operations on large annotated matrices, allowing you to process datasets that exceed available RAM.

What is the best way to store raw data and multiple layers in a single-cell experiment?

The best way to store raw data and multiple layers is using an annotated matrix structure that specifies core components like layers, raw, and uns, consolidating all experiment metadata.

When do I need an annotated matrix structure for single-cell analysis?

You need an annotated matrix structure for single-cell analysis when handling complex experiments that require consolidating embeddings, observations, variables, and unstructured annotations for reproducible processing.