anndata

Creates, reads, writes, subsets, concatenates and transforms AnnData matrices for workflows.

74|5|Updated Dec 10, 2025
One-click install
npx skills add https://github.com/dralkh/seerai --skill anndata-dralkh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: anndata
Source: https://github.com/dralkh/seerai/tree/main/skills/anndata
Command: npx skills add https://github.com/dralkh/seerai --skill anndata-dralkh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you organize, read, transform, and save annotated matrix data when your analysis needs both measurements and rich metadata in one reliable structure. It removes the friction of juggling dense, sparse, backed, and on-disk formats while keeping observation and variable annotations aligned.

Core Features & Use Cases

  • Data structure management: Work with the full AnnData model, including X, obs, var, layers, obsm, varm, obsp, varp, uns, and raw.
  • File I/O and conversion: Read and write h5ad, zarr, csv, mtx, loom, and other supported inputs with efficient storage options.
  • Data manipulation at scale: Subset, transpose, rename, concatenate, and filter datasets while preserving metadata integrity and memory efficiency.
  • Research workflows: Ideal for single-cell RNA-seq, multimodal scverse analysis, batch integration, and large-dataset processing with backed mode.
  • Use case: A researcher can load multiple h5ad batches, merge them with batch labels, store raw counts, filter to highly variable genes, and export a processed dataset for downstream analysis.

Quick Start

Ask the Skill to help you create, inspect, transform, concatenate, and save an AnnData dataset for a single-cell analysis workflow.

Frequently Asked Questions about anndata

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

FAQPage Schema
How do I merge multiple h5ad batches while preserving metadata in an AnnData object?

You can merge multiple h5ad batches by concatenating AnnData objects while preserving observation and variable annotations. This process maintains metadata alignment across obs, var, layers, and uns entries during batch integration.

What is the best way to manage large single-cell datasets with memory-efficient storage?

Managing large single-cell datasets uses AnnData backed mode for memory-efficient storage. This approach keeps measurement matrices on disk while allowing subsetting, filtering, and transformation without loading the entire dataset into memory.

Can I read and write single-cell data across h5ad, zarr, and sparse matrix formats?

Yes, annotated matrix data supports reading and writing across h5ad, zarr, csv, mtx, and loom formats. This I/O capability handles dense, sparse, and on-disk inputs while keeping observation and variable metadata aligned.

Does backed mode in AnnData support subsetting and transposing high-dimensional workflows?

Backed mode supports subsetting and transposing high-dimensional workflows in AnnData objects. You can filter datasets and manipulate the data structure while maintaining memory efficiency and metadata integrity across obs and var annotations.

Why does metadata misalignment occur when concatenating single-cell RNA-seq datasets?

Metadata misalignment occurs when observation and variable annotations do not match during concatenation. AnnData objects preserve metadata integrity by aligning obs, var, layers, and uns entries during single-cell RNA-seq batch merging.

What is an annotated matrix for single-cell analysis and when do I need it?

An annotated matrix for single-cell analysis is a data structure holding measurements and rich metadata in one reliable object. You need it when organizing high-dimensional workflows that require aligned observation and variable annotations.