AI Tool Designs Desired Molecules From Text, Boosting Drug Discovery

Researchers developed a multimodal AI tool that combines language models and graph-based models to efficiently design new molecules with desired properties from natural language queries, outperforming existing methods and potentially automating the entire molecule design and synthesis process.

• Researchers developed a multimodal tool that combines a large language model (LLM) with graph-based AI models to efficiently design new molecules with desired properties based on natural language queries.
• The method outperformed existing LLM-based approaches, generating molecules that better matched user specifications and were more likely to have a valid synthesis plan, improving the success ratio from 5% to 35%.
• The approach could potentially automate the entire process of designing and making molecules, providing a significant time-saver for pharmaceutical companies and material science applications.