ProteinMCP: An Agentic AI Framework for Autonomous Protein Engineering

ABSTRACT

This paper introduces ProteinMCP, a novel framework that utilizes the Model Context Protocol (MCP) to enable autonomous AI agents to conduct protein engineering. The system coordinates between sequence generation models and physics-based simulators to iteratively optimize protein designs without human intervention. The methodology demonstrates how agentic reasoning can navigate the vast chemical space more efficiently than traditional high-throughput screening.