Mutcomputex MCP

MutComputeX MCP

用于蛋白质建模的MutComputeX MCP服务器，从 MutComputeX官方教程.

概述

这个MutComputeX MCP服务器提供了两种蛋白质分析工具，突变记录和似然估计。在这里，我们将它们应用于蛋白质分析的MCP服务。

安装

# Create and activate virtual environment
mamba env create -p ./env python=3.10 pip -y
mamba activate ./env
pip install loguru sniffio biopython scipy numpy requests pandas tensorflow==2.15.0 black==23.1.0

pip install --ignore-installed fastmcp

从aws下载模型参数。

# Install aws if not installed.
curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
unzip awscliv2.zip
sudo ./aws/install

# Download mokdels
cd repo/MutComputeX/models
./download_models.sh

本地使用

1.计算突变概率矩阵

python scripts/run_spired.py --fasta examples/test.fasta --output examples/spired --repo repo/MutComputeX

2.估计MutComputeX的可能性

python scripts/run_spired_fitness.py --fasta examples/test.fasta --output examples/spired --repo repo/MutComputeX

MCP使用

安装MCP服务器

fastmcp install claude-code tool-mcps/mutcomputex_mcp/src/mutcomputex_mcp.py --python tool-mcps/mutcomputex_mcp/env/bin/python
fastmcp install gemini-cli tool-mcps/mutcomputex_mcp/src/mutcomputex_mcp.py --python tool-mcps/mutcomputex_mcp/env/bin/python

呼叫MCP服务

1.计算突变概率矩阵

Can you help train a ProtTrans model for data @examples/case2.1_subtilisin/ and save it to 
@examples/case2.1_subtilisin/prot-t5_fitness using the ProtTrans mcp server with ProtT5-XL model.

Please convert the relative path to absolution path before calling the MCP servers. 

2.估计MutComputeX的可能性