Can it help with manual cell type annotation?

Yes, it provides workflows for visualizing marker genes across clusters and mapping cluster IDs to biological cell types using metadata dictionaries.

Which single-cell data formats are supported?

The skill supports all standard formats compatible with Scanpy, including .h5ad (AnnData), 10X Genomics (MTX and H5), and CSV expression matrices.

Can I perform batch correction with this skill?

Yes, the skill includes guidance for batch correction techniques like ComBat and provides patterns for integrating with other scverse tools.

Does it support trajectory and pseudotime analysis?

Absolutely. It provides implementation patterns for PAGA (Partition-based graph abstraction) and diffusion pseudotime (DPT) to model cellular development.

Scanpy Single-Cell Analysis

Name: Scanpy Single-Cell Analysis
Author: K-Dense-AI

byK-Dense-AI

•

2,188

Data Science & ML

Performs comprehensive single-cell RNA-seq analysis workflows including quality control, normalization, clustering, and cell-type annotation.

About

The Scanpy skill empowers researchers and bioinformaticians to execute professional-grade single-cell RNA-seq analysis within the Claude environment. Built on the scalable Scanpy Python toolkit and AnnData structure, it provides standardized patterns for loading 10X or .h5ad data, performing rigorous quality control, and executing complex dimensionality reduction like UMAP and t-SNE. Whether you are identifying novel cell clusters, finding marker genes, or conducting trajectory inference, this skill provides the domain-specific logic and implementation patterns required to move from raw transcriptomic data to publication-ready biological insights.

Key Features

End-to-end scRNA-seq processing from raw data loading to cell type annotation
High-performance clustering and marker gene identification using Leiden algorithms
Automated quality control scripts for mitochondrial filtering and count normalization
Customizable plotting for publication-quality heatmaps, dot plots, and violin plots
Advanced dimensionality reduction support for PCA, UMAP, t-SNE, and PAGA
2,188 GitHub stars

Use Cases

Conducting differential expression analysis between treated and control experimental groups
Mapping cellular differentiation pathways through trajectory inference and pseudotime analysis
Analyzing raw 10X Genomics or .h5ad datasets to identify discrete cellular subpopulations

About

Key Features

End-to-end scRNA-seq processing from raw data loading to cell type annotation
High-performance clustering and marker gene identification using Leiden algorithms
Automated quality control scripts for mitochondrial filtering and count normalization
Customizable plotting for publication-quality heatmaps, dot plots, and violin plots
Advanced dimensionality reduction support for PCA, UMAP, t-SNE, and PAGA
2,188 GitHub stars

Use Cases

Conducting differential expression analysis between treated and control experimental groups
Mapping cellular differentiation pathways through trajectory inference and pseudotime analysis
Analyzing raw 10X Genomics or .h5ad datasets to identify discrete cellular subpopulations