scverse/scanpy

Health v2 · maintenance only

Adoption not part of the score

2548 stars · 765 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Scanpy is a scalable Python toolkit for analyzing single-cell gene expression data, built alongside the anndata data structure. It provides preprocessing, visualization, clustering, trajectory inference, and differential expression testing, efficiently handling datasets of over a million cells with experimental Dask support for out-of-memory data.

Use cases

• analyze single-cell RNA-seq gene expression data in Python
• cluster and visualize millions of single cells
• perform trajectory inference on single-cell datasets
• run differential expression testing between cell groups
• preprocess and quality-control single-cell count matrices
• analyze single-cell datasets too large to fit in memory
• build scRNA-seq analysis pipelines with anndata

When to choose

• you need a mature, widely adopted Python toolkit for single-cell transcriptomics
• your datasets scale to hundreds of thousands or millions of cells
• you want preprocessing, clustering, visualization, and differential expression in one package
• you prefer Python over R-based single-cell tools like Seurat
• you want GPU-accelerated workflows via compatible ecosystem tools like rapids-singlecell

When to avoid

• you need bulk RNA-seq or non-single-cell genomics analysis
• you require a stable internal API - only the documented public API is supported
• you need out-of-memory processing at scale - Dask compatibility is still experimental
• your workflow is R-centric and you prefer Bioconductor ecosystems

Facets

library · maturity stable

data-science machine-learning data-visualization analytics bioinformatics data-science machine-learning data-visualization python cross-platform single-cell transcriptomics genomics anndata clustering trajectory-inference differential-expression scverse dask bioinformatics gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="scverse/scanpy")