Can this skill help with batch effect correction?

Yes, it provides implementation patterns for ComBat batch correction and offers guidance on integrating external tools like Harmony or scVI.

What data formats does this skill support?

The skill primarily works with AnnData (.h5ad) files but also includes specialized loaders for 10X Genomics (MTX/H5) and standard CSV formats.

How does it handle publication-quality visualization?

The skill includes a dedicated plotting guide and templates to generate high-DPI figures, custom color palettes, and multi-panel layouts suitable for academic journals.

Does it support trajectory inference?

Yes, it includes pre-configured workflows for PAGA (Partition-based graph abstraction) and diffusion pseudotime analysis to study cellular development.

Is it suitable for large-scale datasets?

Scanpy is built for scalability; this skill leverages its efficient handling of sparse matrices and memory-mapped files via the AnnData framework.

Scanpy Single-Cell Analysis

Name: Scanpy Single-Cell Analysis
Author: Yezez9

byYezez9

0•

数据科学与机器学习

Performs comprehensive single-cell RNA-seq analysis pipelines including quality control, clustering, and visualization using the Scanpy toolkit.

This skill empowers Claude to execute end-to-end single-cell transcriptomics workflows, from raw data processing to publication-ready visualization. It provides standardized implementations for quality control, normalization, batch correction, and differential expression, ensuring best practices like Leiden clustering and UMAP dimensionality reduction are applied correctly to AnnData objects. It is an essential resource for bioinformaticians and researchers looking to streamline exploratory scRNA-seq data analysis and trajectory inference within a Python-based ecosystem.

主要功能

01Automated QC and normalization workflows for 10X, H5AD, and CSV data

02Advanced dimensionality reduction using PCA, UMAP, and t-SNE

03Leiden and Louvain clustering with automated marker gene identification

04Trajectory inference and pseudotime analysis via PAGA

05Publication-quality plotting for heatmaps, dot plots, and violin plots

060 GitHub stars

使用场景

01Identifying and annotating specific cell types within heterogeneous tissue samples

02Processing raw 10X Genomics output into filtered and normalized AnnData objects

03Conducting differential expression analysis between treated and control cell populations

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yezez9/research-agent scanpy

For use in Claude.ai and ChatGPT