Infercnv

Infercnv-MCP provides a natural language interface to perform Copy Number Variation (CNV) inference from scRNA-Seq data using infercnvpy, accessible through MCP clients. It combines data reading, preprocessing, CNV inference, and visualization in a single, scriptable server workflow, enabling researchers to integrate CNV analysis into chatbots, plugins, or agent systems.

How to use

You can run Infercnv-MCP either locally or on a remote server, then connect your MCP client to the server endpoint. Locally, start the stdio server and point your client to the local process. Remotely, run the HTTP server and configure your client to the provided URL.

How to install

Prerequisites: Python 3.x and a working internet connection.

Install the MCP package from PyPI.

Test running the MCP server locally or remotely.

Additional notes

The server exposes modules for IO (reading/writing scRNA-Seq data and gene positions), preprocessing (neighbors computation and data preparation), CNV inference (cnv scoring), and plotting (chromosome heatmaps, UMAP, and t-SNE visualizations). You can use these capabilities through MCP client configurations.

Available tools

IO

Reading and writing scRNA-Seq data and loading gene position information.

Preprocessing

Computing neighbors and preparing data for CNV analysis.

CNV Inference

Running CNV inference and computing CNV scores.

Plotting

Generating chromosome heatmaps and visualizations such as UMAP and t-SNE.