Facilitating AI and Machine Learning in Chemistry R&D

This case study examines the integration of artificial intelligence (AI) and machine learning (ML) within high throughput experimentation (HTE) in chemistry research and development. It outlines how leading pharmaceutical organizations utilize Katalyst D2D software to enhance their experimental workflows, enabling efficient data generation for AI/ML applications. The study details the process by which one organization achieved a reduction in the number of reactions needed to optimize chemical processes by 40-50% through the use of ML models that predict reaction outcomes. Additionally, it describes the development of the Experiment Design Bayesian Optimizer (EDBO+) algorithm, which was integrated into Katalyst to streamline the design of experiments. The case study highlights the symbiotic relationship between HTE and AI/ML, demonstrating how these technologies collectively improve the efficiency of chemistry R&D and accelerate the drug discovery process. The findings suggest significant time and resource savings in experimental workflows.