Fig. 6: Application of our machine learning (ML) optimisation framework to pharmaceutical process development with two active pharmaceutical ingredient (API) synthesis case studies.
From: Highly parallel optimisation of chemical reactions through automation and machine intelligence
The scatter plots show the area percent (AP) yield and AP selectivity of experiments selected by our ML Bayesian optimisation (BO) workflow at each iteration for each campaign. Select high-performing reaction conditions for each case study are presented in the tables. a Nickel-catalysed Suzuki reaction in an API synthesis. b Palladium-catalysed Buchwald–Hartwig coupling reaction in an API synthesis.
