AutoBinder Agent: An MCP-Based Agent for End-to-End Protein Binder Design

TL;DR

The paper presents AutoBinder Agent, an MCP-based framework for end-to-end protein binder design that coordinates MaSIF, Rosetta, ProteinMPNN, and AlphaFold3 through a unified, protocol-driven workflow. By replacing fragmented scripts with a structured MCP-driven orchestration, the system achieves reproducible, auditable design processes from target structures to predicted binders. Experimental evaluation across standardized prompts demonstrates high task understanding, robust tool planning, and reliable end-to-end execution, with a composite score of $S_{\text{prompt}} = 93.86$ and strong individual metrics ($S_t$, $S_{\text{tool}}$). The work highlights the practicality and scalability of MCP-based agentic design in drug discovery, enabling automated binder generation with improved reproducibility and extensibility. It also notes areas for future work, including quantitative measures of reasoning and broader applicability to molecular design workflows.

Abstract

Modern AI technologies for drug discovery are distributed across heterogeneous platforms-including web applications, desktop environments, and code libraries-leading to fragmented workflows, inconsistent interfaces, and high integration overhead. We present an agentic end-to-end drug design framework that leverages a Large Language Model (LLM) in conjunction with the Model Context Protocol (MCP) to dynamically coordinate access to biochemical databases, modular toolchains, and task-specific AI models. The system integrates four state-of-the-art components: MaSIF (MaSIF-site and MaSIF-seed-search) for geometric deep learning-based identification of protein-protein interaction (PPI) sites, Rosetta for grafting protein fragments onto protein backbones to form mini proteins, ProteinMPNN for amino acid sequences redesign, and AlphaFold3 for near-experimental accuracy in complex structure prediction. Starting from a target structure, the framework supports de novo binder generation via surface analysis, scaffold grafting and pose construction, sequence optimization, and structure prediction. Additionally, by replacing rigid, script-based workflows with a protocol-driven, LLM-coordinated architecture, the framework improves reproducibility, reduces manual overhead, and ensures extensibility, portability, and auditability across the entire drug design process.

Figures (3)

• Figure 1: Framework of our proposed AutoBinder Agent
• Figure 2: Workflow of Binder Design by Autobinder Agent
• Figure 3: Results of All Experiments