anndata

Load and manage annotated data matrices from h5ad and zarr files.

Updated May 24, 2026
One-click install
npx skills add https://github.com/Estrella-231/Mathematical_modeling_tongmeng --skill anndata-estrella-231
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: anndata
Source: https://github.com/Estrella-231/Mathematical_modeling_tongmeng/tree/main/.agents/skills/anndata
Command: npx skills add https://github.com/Estrella-231/Mathematical_modeling_tongmeng --skill anndata-estrella-231

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AnnData organizes large, annotated measurement matrices with consistent metadata so you can analyze single-cell style datasets without losing the link between data, observations, and features.

Core Features & Use Cases

  • Central annotated matrix container: Keeps X plus structured metadata in obs (rows) and var (columns), along with multi-dimensional annotations in obsm/varm and pairwise graphs in obsp/varp.
  • Efficient storage and I/O: Loads and saves common formats like h5ad and zarr, supports compression, and enables backed mode for datasets that don’t fit in RAM.
  • Workflow integration: Acts as the interoperable data model for Scanpy-style preprocessing, downstream probabilistic modeling (e.g., scvi-tools), and population-scale querying (e.g., cellxgene-census).

Quick Start

Load a single-cell dataset from an h5ad file into an AnnData object named adata for further preprocessing and analysis.

Frequently Asked Questions about anndata

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

FAQPage Schema
How do I load an h5ad file into an annotated matrix for single-cell analysis?

You can load an h5ad file into an AnnData object to create an annotated matrix that couples single-cell measurement data with observation and feature metadata for downstream analysis.

Can I process single-cell datasets that are too large to fit in RAM?

Yes, you can use backed mode to load large single-cell datasets from h5ad or zarr storage into memory-efficient annotated matrices without reading everything into RAM at once.

What is the best way to concatenate multiple batches of single-cell data?

Concatenating batches of single-cell data involves merging multiple annotated matrices while correctly preserving the X matrix, observation metadata in obs, and feature metadata in var.

How do I prepare annotated matrices for scvi-tools modeling?

Prepare inputs for scvi-tools by loading single-cell data into an AnnData-compatible structure with correct handling of X, obs, var, and multi-dimensional annotations like obsm.

Does this data integration approach support zarr storage?

Yes, this data integration approach supports both h5ad and zarr formats, allowing you to save and load annotated matrices with compression for efficient scientific workflows.

When do I need an annotated matrix container instead of a plain data array?

You need an annotated matrix container when working with single-cell datasets to maintain tight coupling between measurement data, observation metadata, and feature metadata during subsetting.