Does it support scanpy integration?

Absolutely. The skill provides methods to load Census data directly into AnnData objects, making it compatible with standard scanpy analysis workflows.

Why should I filter for 'is_primary_data == True'?

Filtering for primary data ensures you are analyzing unique cells and avoiding duplicates that may exist across different datasets within the Census repository.

How do I ensure my research is reproducible?

You can specify a specific 'census_version' (e.g., '2023-07-25') when opening the SOMA object to ensure your analysis runs against a fixed data snapshot.

What is the CZ CELLxGENE Census?

It is a versioned collection of standardized single-cell genomics data containing over 61 million cells from human and mouse studies, accessible programmatically.

Can I use this for machine learning?

Yes, the skill includes specific patterns for PyTorch integration using experimental data loaders designed for training on massive datasets.

CELLxGENE Census Genomics

Name: CELLxGENE Census Genomics
Author: x-cmd

byx-cmd

•

데이터 과학 및 ML

Provides programmatic access to standardized single-cell genomics data for large-scale analysis and machine learning.

소개

The cellxgene-census skill empowers researchers and developers to query the Chan Zuckerberg Initiative's massive repository of over 61 million single-cell genomics data points directly within Claude. It streamlines the process of filtering datasets by metadata such as cell type, tissue, and disease status, while providing optimized workflows for data retrieval using scanpy and PyTorch. This skill is essential for performing population-scale single-cell analysis, training specialized ML models, and integrating high-quality biological data into computational pipelines without the overhead of manual data management.

주요 기능

Out-of-core processing for datasets exceeding local RAM
Filter by standardized metadata (cell type, tissue, disease, donor)
8 GitHub stars
Versioned data access for reproducible scientific research
Integrated workflows for scanpy and PyTorch training
Query 61M+ human and mouse single-cell expression data points

사용 사례

Exploring gene expression patterns across thousands of diverse cell types
Training machine learning models on massive single-cell transcriptomics datasets
Performing cross-tissue and cross-disease population-scale analysis

소개

주요 기능

Out-of-core processing for datasets exceeding local RAM
Filter by standardized metadata (cell type, tissue, disease, donor)
8 GitHub stars
Versioned data access for reproducible scientific research
Integrated workflows for scanpy and PyTorch training
Query 61M+ human and mouse single-cell expression data points

사용 사례

Exploring gene expression patterns across thousands of diverse cell types
Training machine learning models on massive single-cell transcriptomics datasets
Performing cross-tissue and cross-disease population-scale analysis