Precedent Finder: Locating Pareto-Optimal Reactions.

Area of Science:

• Chemistry
• Computer Science
• Data Science

Background:

• Chemical reaction databases are vast and complex.
• Efficiently searching for specific reaction information, like optimal conditions or catalytic systems, is challenging for synthetic chemists.
• Existing search tools may not effectively handle the multiobjective nature of reaction optimization.

Purpose of the Study:

• To introduce Precedent Finder, a novel cheminformatics search tool.
• To demonstrate how Precedent Finder utilizes multiobjective optimization for retrieving relevant chemical reaction data.
• To showcase the tool's utility in supporting synthetic chemists by identifying successful catalytic systems and optimizing synthetic routes.

Main Methods:

• Developed Precedent Finder, a cheminformatics search tool.
• Implemented multiobjective optimization to retrieve Pareto-optimal data points.
• Utilized reaction similarity, yield, and date as primary search parameters.

Main Results:

• Precedent Finder successfully retrieves Pareto-optimal data points from chemical reaction databases.
• The tool can highlight successful catalytic systems for specific query reactions.
• Demonstrated the application of Precedent Finder in supporting synthetic chemists' workflow.

Conclusions:

• Precedent Finder is an effective tool for locating relevant reaction information.
• The tool aids in optimizing synthetic routes for complex pharmaceutical substances by identifying efficient catalytic systems.
• Precedent Finder enhances the efficiency and success rate of chemical synthesis.