Why should I always filter for primary data?

Filtering for 'is_primary_data == True' ensures you are not counting duplicate cells that may appear in multiple datasets within the census.

How does it handle large datasets that don't fit in RAM?

It utilizes out-of-core processing via axis queries and iterative batch processing to handle datasets that exceed available system memory.

Can I use this for machine learning?

Yes, the skill includes specific patterns for using experimental PyTorch dataloaders to train models directly on millions of cells from the Census.

Is scanpy supported by this skill?

Yes, the skill provides specific implementation patterns for loading Census data directly into AnnData objects for seamless use with scanpy workflows.

CZ CELLxGENE Census Analysis

Name: CZ CELLxGENE Census Analysis
Author: pur3v4d3r

bypur3v4d3r

•

データサイエンスとML

Facilitates programmatic access to over 61 million standardized single-cell genomics data points for advanced querying and analysis.

概要

The cellxgene-census skill provides a specialized interface for exploring and analyzing the world's largest versioned collection of standardized single-cell genomics data from CZ CELLxGENE Discover. It enables researchers to efficiently query millions of cells by cell type, tissue, or disease, retrieve raw gene expression matrices, and seamlessly integrate this data into modern machine learning pipelines using PyTorch or traditional analysis frameworks like scanpy. This skill is essential for population-scale single-cell studies, training large-scale genomics models, and performing cross-dataset integrations without the need for manual data downloading and preprocessing.

主な機能

Query 61M+ standardized single-cell data points from human and mouse
Retrieve raw gene expression matrices and pre-calculated embeddings
Native compatibility with scanpy for downstream bioinformatics analysis
Filter data by cell type, tissue, disease, and donor metadata
Integrated PyTorch data loading for machine learning workflows
1 GitHub stars

ユースケース

Training machine learning models on massive single-cell genomics datasets
Performing cross-tissue or cross-disease expression analysis at scale
Identifying specific cell type markers across thousands of different datasets

概要

主な機能

Query 61M+ standardized single-cell data points from human and mouse
Retrieve raw gene expression matrices and pre-calculated embeddings
Native compatibility with scanpy for downstream bioinformatics analysis
Filter data by cell type, tissue, disease, and donor metadata
Integrated PyTorch data loading for machine learning workflows
1 GitHub stars

ユースケース

Training machine learning models on massive single-cell genomics datasets
Performing cross-tissue or cross-disease expression analysis at scale
Identifying specific cell type markers across thousands of different datasets