Why is 'is_primary_data == True' emphasized in the documentation?

This filter is critical to ensure that you are only analyzing unique cells, as the Census may contain duplicate entries from different versions of the same study.

Does this skill require a specific Census version?

While it defaults to the latest stable release, the skill includes patterns to specify a specific census_version for better reproducibility in scientific research.

What organisms are supported by the CELLxGENE Census skill?

The Census currently supports standardized single-cell genomics data for both Homo sapiens (human) and Mus musculus (mouse).

Can I integrate this data with standard bioinformatics tools?

Yes, the skill includes direct integration patterns for the Scanpy library for downstream analysis and visualization, as well as PyTorch for machine learning workflows.

How does the skill handle datasets too large for local memory?

The skill provides patterns for out-of-core processing using axis queries and iterative batching, allowing you to compute statistics or train models on millions of cells without crashing your environment.

CELLxGENE Census Analysis

Name: CELLxGENE Census Analysis
Author: K-Dense-AI

byK-Dense-AI

•

2,188

数据科学与机器学习

Accesses and analyzes over 61 million standardized single-cell genomics records from the CZ CELLxGENE Census.

关于

This skill provides programmatic access to the CZ CELLxGENE Census, enabling researchers to query and analyze one of the world's largest collections of standardized single-cell genomics data. It allows for precise filtering by cell type, tissue, disease, and donor metadata across human and mouse organisms. With specialized patterns for out-of-core processing, integration with Scanpy for bioinformatics pipelines, and experimental PyTorch support for machine learning, this skill facilitates population-scale analysis and model training directly within genomic research workflows.

主要功能

2,188 GitHub stars
Query 61M+ cells across human and mouse organisms with standardized metadata
Seamless integration with Scanpy for visualization and dimensionality reduction
Experimental PyTorch data loaders for large-scale machine learning model training
Efficient out-of-core processing for datasets that exceed available RAM
Filter by cell type, tissue, disease state, and specific donor IDs

使用场景

Performing cross-tissue gene expression analysis to identify disease-specific markers
Conducting meta-analyses of existing single-cell datasets without manual data curation
Training deep learning models for automated cell type annotation using population-scale data

关于

主要功能

2,188 GitHub stars
Query 61M+ cells across human and mouse organisms with standardized metadata
Seamless integration with Scanpy for visualization and dimensionality reduction
Experimental PyTorch data loaders for large-scale machine learning model training
Efficient out-of-core processing for datasets that exceed available RAM
Filter by cell type, tissue, disease state, and specific donor IDs

使用场景

Performing cross-tissue gene expression analysis to identify disease-specific markers
Conducting meta-analyses of existing single-cell datasets without manual data curation
Training deep learning models for automated cell type annotation using population-scale data