FROGENT: An End-to-End Full-process Drug Design Multi-Agent System

2025/08/14 by Qihua Pan, Dong Xu, Pan, Qihua +11 · 2 voices · 2 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Artificial Intelligence (cs.AI) #Biomolecules (q-bio.BM) #FOS: Biological sciences #FOS: Computer and information sciences #cs.AI #q-bio.BM

paper · pdf · doi:10.48550/arxiv.2508.10760

arxiv published 2025/08/14 · arxiv updated 2026/03/01

Abstract

Drug discovery is a complex, multi-step pipeline that remains heavily dependent on manual, experience-driven operations; meanwhile, existing customized artificial intelligence tools are fragmented across web applications, desktop software, and code libraries, resulting in incompatible interfaces and inefficient, burdensome workflows. To overcome these challenges, we propose FROGENT, a full-process drug design multi-agent system that leverages the planning, reasoning, and tool-use capabilities of large language models (LLMs) to unify drug discovery within a closed-loop and autonomous framework. FROGENT is a collaborative multi-agent system comprising a central Orchestrate Agent for strategic workflow coordination and three distributed agents, Retrieve, Forge, and Gauge, that employ dynamic biochemical databases, extensible tool libraries, and task-specific computational models via the Model Context Protocol. This architecture enables end-to-end execution of complex drug discovery pipelines, covering target identification, small-molecule generation, peptide optimization, and retrosynthetic planning. Across eight benchmarks spanning core drug discovery tasks, FROGENT consistently outperforms six increasingly advanced ReAct-style agents. Case studies further demonstrate its practicality and generalization across real-world small-molecule and peptide design scenarios. Overall, FROGENT not only achieves substantial gains in efficiency and accuracy, but also demonstrates the potential of LLM-based agentic systems to autonomously orchestrate drug development pipelines, reducing, or even replacing, reliance on manual, experience-driven human intervention.

Citations

• LLM-based Agentic Reasoning Frameworks: A Survey from Methods to Scenarios
• The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies
• DrugMCTS: a drug repurposing framework combining multi-agent, RAG and Monte Carlo Tree Search
• DrugPilot: LLM-based Parameterized Reasoning Agent for Drug Discovery
• Qwen3 Technical Report
• Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions
• PharmAgents: Building a Virtual Pharma with Large Language Model Agents
• DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
• DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaboration
• UniProt: the Universal Protein Knowledgebase in 2025
• Deep Lead Optimization: Leveraging Generative AI for Structural Modification
• MolCRAFT: Structure-Based Drug Design in Continuous Parameter Space
• ADMETlab 3.0: an updated comprehensive online ADMET prediction platform enhanced with broader coverage, improved performance, API functionality and decision support
• Empowering Biomedical Discovery with AI Agents
• DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design
• Gotta be SAFE: A New Framework for Molecular Design
• AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors
• MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
• AI-powered therapeutic target discovery
• Gorilla: Large Language Model Connected with Massive APIs
• ChemCrow: Augmenting large-language models with chemistry tools
• GPT-4 Technical Report
• 3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction
• Toolformer: Language Models Can Teach Themselves to Use Tools
• DiffBP: Generative Diffusion of 3D Molecules for Target Protein Binding
• Structure-based Drug Design with Equivariant Diffusion Models
• Distributing Accountability, Not Capability: Phase Separation and the LLM Workflow Quadrant in Autonomous AI Agent Architectures
• Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets
• Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
• Language Models are Few-Shot Learners

Cited by

• Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics
• MolClaw: An Autonomous Agent with Hierarchical Skills for Drug Molecule Evaluation, Screening, and Optimization

Discussions

• FROGENT (nothing to do with frogs) Is set to streamline drug discovery (Faster discovery & mfr, faster to save lives), as Cornell University reports. arxiv.org/abs/2508.10760 and this tweet summarizes [bsky, 0 points, 0 comments]
• Qihua Pan, Dong Xu, Jenna Xinyi Yao, Lijia Ma, Zexuan Zhu, Junkai Ji: FROGENT: An End-to-End Full-process Drug Design Agent https://arxiv.org/abs/2508.10760 https://arxiv.org/pdf/2508.10760 https://ar [bsky, 0 points, 0 comments]

Related

• Can AI Agents Design and Implement Drug Discovery Pipelines?
• DrugAgent: Automating AI-aided Drug Discovery Programming through LLM Multi-Agent Collaboration
• AI Agents in Drug Discovery
• Large Language Model Agent for Modular Task Execution in Drug Discovery
• Large Language Model Agent for Modular Task Execution in Drug Discovery