AI-Enabled Retrosynthesis for Drug Development

AI methods and systems for predicting reactants and their synthesis paths to support drug design and chemical synthesis.

A time-consuming and costly step in drug development is the identification of drug-like small molecules that display desired properties against a specific biomolecular target and then the synthesis of such molecules if they do not exist. Retrosynthesis is a procedure where such a desired molecule is transformed into potential reactants, and thus, the synthesis routes are identified. The success and efficiency of retrosynthesis of drug-like small molecules immensely impact the entire drug development process and affect its success rate, costs, and speed.

The Need

Drug design involves manual synthesis planning by medicinal chemists that is slow and dependent on the knowledge and experience of the medicinal chemists. The process is not scalable to include all possible synthesis paths or all possible molecule products to be synthesized. Process automation is needed to leverage a broader network of expertise and insights.

The Technology

Researchers at The Ohio State University have developed a set of AI tools and systems to automate the identification of synthesis paths for any given molecule. Machine learning is used to conduct retrosynthesis: given a molecule, the AI tools and systems will predict the most possible, probably multiple, reactants and their synthesis paths that have high probability of truly happening in de novo drug design and chemical synthesis.

Commercial Applications

The algorithm behind AI-based retrosynthesis for drug design has great potential outside of pharma. In fact, biotech, healthcare, and chemical industries all rely on understanding reactant behaviors to enable synthesis path optimization.


The proposed AI-based software and technology can significantly accelerate the conventional drug development process by automating synthesis path search over a huge chemical space and chemical reaction space; speeding up the process and reducing the cost; and potentially discovering new knowledge from a huge amount of chemical and reaction data that medicinal chemists may not be able to catch.


Provisional Patent Application Filed.

Loading icon