Related Experiment Video
Updated: Jan 20, 2026

Monitoring GPCR-β-arrestin1/2 Interactions in Real Time Living Systems to Accelerate Drug Discovery
Published on: June 28, 2019
Democratising real-world drug discovery through agentic AI
Jiazhen He1, Helen Lai2, Lakshidaa Saigiridharan1
1Molecular AI, Discovery Sciences, BioPharmaceuticals R&D, AstraZeneca, Pepparedsleden 1, 431 83 Mölndal, Sweden.
Agentic systems powered by large language models (LLMs) are now aiding drug discovery. ChatInvent, an LLM-based system, is integrated into AstraZeneca's pipeline for molecular design and synthesis planning.
Area of Science:
- Artificial Intelligence in Chemistry
- Drug Discovery Technologies
- Computational Chemistry
Background:
- Agentic systems leveraging large language models (LLMs) show potential in chemical research.
- Existing systems like CoScientist, Chemcrow, and LLM-RDF demonstrate utility in cheminformatics and reaction development.
- Real-world adoption of these systems in drug discovery pipelines remains underexplored.
Purpose of the Study:
- To present a real-world example of an agentic system's adoption in drug discovery.
- To detail the development and integration of the ChatInvent system at AstraZeneca.
- To discuss the evolution, challenges, and future perspectives of agentic systems in pharmaceutical research.
Main Methods:
- Development of an agentic system named ChatInvent, based on large language models.
- Integration of ChatInvent into AstraZeneca's drug discovery pipeline.
- Evolution from a single-agent proof-of-concept to a multi-agent architecture with a graphical user interface.
Main Results:
- Successful integration of the ChatInvent agentic system into a major pharmaceutical company's discovery pipeline.
- Demonstration of the system's utility in molecular design and synthesis planning.
- Establishment of an extensible, robust, and scalable multi-agent architecture.
Conclusions:
- Agentic systems represent a significant advancement in AI-driven drug discovery.
- The ChatInvent project provides a practical model for implementing LLM-based tools in pharmaceutical research.
- Continued development and addressing persistent challenges are crucial for realizing the full potential of agentic systems in chemistry.
Related Concept Videos
Drug Discovery: Overview
08:21Monitoring GPCR-β-arrestin1/2 Interactions in Real Time Living Systems to Accelerate Drug Discovery
Antidepressant Drugs: MAOIs and Other Agents
Sedatives and Hypnotics Drugs: Miscellaneous Agents
Melatonin congeners like ramelteon (Rozerem) and tasimelteon (Hetlioz) selectively bind to melatonin receptors (MT1 and MT2) and thus mimic the actions of melatonin, a hormone that regulates sleep-wake cycles. Tasimelteon is primarily used for non-24-hour sleep-wake disorder, common in blind patients. They are also used to treat conditions like insomnia...
Heart Failure Drugs: Inotropic Agents
Antipsychotic Drugs: Typical and Atypical Agents
