> ## Documentation Index
> Fetch the complete documentation index at: https://docs.scmcphub.org/llms.txt
> Use this file to discover all available pages before exploring further.

# scanpy MCP Server Tools

Here is doc for scanpy-mcp, but you can refer to scanpy docs at [https://scanpy.readthedocs.io/](https://scanpy.readthedocs.io/)

## IO Module

* `read`: Read data from various sources (10X directory, h5ad files, 10x files, text files)
* `write`: Save AnnData object to a file

## Preprocessing Module

* `subset_cells`: Filter or subset cells based on total genes expressed counts, number of cells, or values in adata.obs
* `subset_genes`: Filter or subset genes based on number of cells, counts, or values in adata.var
* `calculate_qc_metrics`: Calculate quality control metrics (total counts, gene number, mitochondrial genes)
* `log1p`: Logarithmize the data matrix
* `normalize_total`: Normalize counts per cell to the same total count
* `highly_variable_genes`: Annotate highly variable genes in the dataset
* `regress_out`: Regress out unwanted sources of variation
* `scale`: Scale the data to unit variance and zero mean
* `combat`: Perform batch effect correction using ComBat
* `scrublet`: Detect and remove doublets using Scrublet
* `neighbors`: Compute a neighborhood graph of observations

## Tools Module

* `tsne`: Perform t-distributed stochastic neighborhood embedding (t-SNE) for visualization
* `umap`: Perform Uniform Manifold Approximation and Projection (UMAP) for visualization
* `draw_graph`: Perform force-directed graph drawing
* `diffmap`: Compute Diffusion Maps for dimensionality reduction
* `embedding_density`: Calculate the density of cells in an embedding
* `leiden`: Perform Leiden clustering algorithm for community detection
* `louvain`: Perform Louvain clustering algorithm for community detection
* `dendrogram`: Compute hierarchical clustering dendrogram
* `dpt`: Perform Diffusion Pseudotime (DPT) analysis
* `paga`: Perform Partition-based graph abstraction
* `ingest`: Map labels and embeddings from reference to query data
* `rank_genes_groups`: Rank genes for characterizing groups
* `filter_rank_genes_groups`: Filter ranked genes groups
* `marker_gene_overlap`: Compute overlap between marker genes
* `score_genes`: Score genes based on their expression
* `score_genes_cell_cycle`: Score genes based on cell cycle phase
* `pca`: Perform Principal Component Analysis

## Plotting Module

* `pca`: Create a scatter plot in PCA coordinates
* `diffmap`: Plot diffusion map embedding of cells
* `violin`: Create violin plots of one or more variables
* `stacked_violin`: Create stacked violin plots for compact visualization
* `heatmap`: Create a heatmap of gene expression values
* `dotplot`: Create a dot plot of expression values per gene for each group
* `matrixplot`: Create a heatmap of mean expression values per group
* `tracksplot`: Create a compact plot of gene expression
* `scatter`: Create a scatter plot of two variables
* `embedding`: Create a scatter plot for user-specified embedding basis (e.g., UMAP, t-SNE)
* `embedding_density`: Plot the density of cells in an embedding
* `rank_genes_groups`: Plot ranking of genes based on differential expression
* `rank_genes_groups_dotplot`: Create a dot plot of ranked genes groups
* `clustermap`: Create a hierarchical clustering heatmap
* `highly_variable_genes`: Plot highly variable genes
* `pca_variance_ratio`: Plot PCA variance ratio