scanpy

Build nearest neighbor graph from PCA space and compute UMAP embedding for visualization. Input is PCA-processed AnnData object and output is neighbor graph, UMAP embedding, and visualization.

Calculate quality control metrics, visualize QC distributions, and filter low-quality cells and genes. Input is single-cell count data in AnnData format and output is QC plots, filtered data, and doublet scores.

Perform Leiden clustering on the neighborhood graph and visualize results. Input is AnnData with neighborhood graph and output is clustered data with UMAP visualization.

Normalize count data using median total counts scaling followed by log1p transformation. Input is quality-controlled AnnData object and output is normalized expression data.

Identify highly variable genes for feature selection using specified method. Input is normalized AnnData object and output is feature selection plot and filtered data.

Perform multi-resolution clustering, marker gene analysis, and differential expression for cell type annotation. Input is clustered AnnData object and output is multi-resolution plots, marker analysis, and differential expression results.

Perform principal component analysis for dimensionality reduction and visualization. Input is feature-selected AnnData object and output is PCA embeddings and variance plots.