[Application and progress of artificial intelligence agents in drug development].

Area of Science:

• Biomedical research
• Computational chemistry
• Artificial intelligence in medicine

Background:

• Drug discovery faces challenges like high costs, risks, and long timelines.
• Emerging artificial intelligence (AI) agents offer potential solutions.
• AI agents transition from tools to active collaborators in research.

Purpose of the Study:

• To systematically review AI agents' architecture and applications in drug development.
• To highlight AI's role in improving efficiency and mitigating risks.
• To provide a reference for researchers in AI-driven drug discovery.

Main Methods:

• Review of AI agent capabilities, including autonomous reasoning and tool utilization.
• Analysis of AI applications across drug discovery stages: target identification, molecular design, synthesis planning.
• Discussion of AI integration with automated experimental platforms.

Main Results:

• AI agents can identify novel targets and mechanisms through knowledge integration.
• Automation of complex tasks like molecular design and synthesis planning is feasible.
• AI agents facilitate closing the loop from virtual design to physical experimentation.

Conclusions:

• AI agents represent a paradigm shift in drug discovery, moving towards integrated platforms and general-purpose biomedical agents.
• Further development is needed to address current limitations and unlock full potential.
• AI agents are poised to significantly transform the efficiency and success rate of drug development.