Preprocess and analyze single-cell RNA-seq data with Scanpy on AnnData datasets.
npx skills add https://github.com/shushuzn/Rairos --skill scanpy-shushuzn
Please help me install this Agent Skill. Skill: scanpy Source: https://github.com/shushuzn/Rairos/tree/main/skills/scanpy Command: npx skills add https://github.com/shushuzn/Rairos --skill scanpy-shushuzn
Scanpy provides an end-to-end workflow for turning raw single-cell RNA-seq data into interpretable biology through standardized QC, preprocessing, embeddings, clustering, markers, and visualizations.
Run the provided QC script on your input file named data/raw_counts.h5ad to generate a filtered dataset you can continue analyzing in Scanpy.
High-intent search queries and answers about installing and using this skill.
A complete single-cell RNA-seq workflow runs standardized QC, normalization, PCA/UMAP embeddings, Leiden clustering, and marker gene ranking on an AnnData dataset to yield annotated cell types. You load a raw counts .h5ad file and execute bundled scripts for automated preprocessing and visualization.
ScRNA-seq preprocessing requires computing QC metrics, filtering poor-quality cells and genes, and visualizing mitochondrial content to improve dataset reliability. Normalization, log-transformation, and highly variable gene identification follow to prepare the filtered AnnData object for downstream dimensionality reduction and clustering.
Yes, AnnData .h5ad files serve as the standard input and output format for scRNA-seq pipelines. The workflow loads the .h5ad dataset, builds neighborhood graphs, computes PCA/UMAP/t-SNE embeddings, runs Leiden clustering, and writes the processed results back to .h5ad files for further analysis.
Differential expression ranking identifies marker genes by applying statistical tests to compare gene expression across Leiden clusters in single-cell data. It ranks genes by significance, enabling data-driven cell-type annotation by mapping cluster-specific markers to known biological cell identities.
Yes, the scRNA-seq analysis approach supports trajectory inference by applying PAGA and diffusion pseudotime (DPT) to the processed AnnData dataset. It also includes gene set scoring to provide deeper biological interpretation of developmental trajectories alongside standard clustering and embedding visualizations.
Generating publication-ready visualizations for single-cell embeddings and markers uses sc.pl plotting functions on the processed AnnData object. It produces standardized figures for QC metrics, UMAP/t-SNE embeddings, ranked marker genes, and trajectories to interpret and communicate scRNA-seq analysis results.
ya-wayFilter scRNA-seq cells by mito and gene thresholds and remove doublets.
anthropicsFilter low-quality cells from single-cell RNA-seq data using MAD-based QC.
chenyhvvvvAnnotate cell types in spatial transcriptomics data using clustering, marker analysis, and LLM labeling.
ya-wayProcess 10X Cell Ranger scRNA-seq data into annotated cell types.
