How does this skill handle very large datasets?

The skill provides implementation patterns for 'backed' mode, which keeps the data on disk (h5ad) and only loads specific subsets into memory as needed, and utilizes sparse matrices to minimize RAM usage.

What is an AnnData object?

An AnnData object is a specialized Python structure that stores a central data matrix (X) alongside metadata for observations (obs) and variables (var), plus additional multi-dimensional annotations (obsm, varm).

Can I use this for non-genomics data?

Yes. While developed for single-cell genomics, AnnData is a general-purpose framework for any scientific data requiring strict alignment between a data matrix and its associated metadata.

Which file formats are supported?

It supports high-performance formats like .h5ad and .zarr, as well as common formats including CSV, MTX, Loom, and 10X HDF5.

AnnData Scientific Data Management

Name: AnnData Scientific Data Management
Author: Microck

byMicrock

•

データサイエンスとML

Manages annotated data matrices and single-cell genomics datasets using the AnnData Python framework and scverse ecosystem.

概要

The AnnData skill empowers Claude to efficiently handle large-scale biological datasets by managing annotated data matrices where experimental measurements are linked with observation and variable metadata. It provides comprehensive guidance for creating, subsetting, and concatenating AnnData objects, performing single-cell RNA-seq workflows, and integrating with tools like Scanpy and scvi-tools. This skill is essential for bioinformatics and data science tasks involving high-dimensional data, sparse matrices, and memory-intensive genomics analysis.

主な機能

Seamless integration patterns for Scanpy and the scverse ecosystem
81 GitHub stars
Advanced data concatenation and batch integration strategies
Efficient I/O operations for .h5ad, .zarr, Loom, and 10X genomics formats
Comprehensive AnnData object manipulation including X, obs, var, and layers
Memory-efficient processing using backed mode and sparse matrices

ユースケース

Single-cell RNA-seq data preprocessing, filtering, and quality control
Handling multi-gigabyte genomics files using memory-efficient on-disk backed mode
Integrating and normalizing multiple experimental batches into a unified dataset

概要

主な機能

Seamless integration patterns for Scanpy and the scverse ecosystem
81 GitHub stars
Advanced data concatenation and batch integration strategies
Efficient I/O operations for .h5ad, .zarr, Loom, and 10X genomics formats
Comprehensive AnnData object manipulation including X, obs, var, and layers
Memory-efficient processing using backed mode and sparse matrices

ユースケース

Single-cell RNA-seq data preprocessing, filtering, and quality control
Handling multi-gigabyte genomics files using memory-efficient on-disk backed mode
Integrating and normalizing multiple experimental batches into a unified dataset