Abstract
Catalyst optimization for enantioselective transformations has traditionally relied on empirical evaluation of catalyst properties. Although this approach has been successful in the past it is intrinsically limited and inefficient. To address this problem, our laboratory has developed a fully informatics guided workflow to leverage the power of artificial intelligence (AI) and machine learning (ML) to accelerate the discovery and optimization of any class of catalyst for any transformation. This approach is mechanistically agnostic, but also serves as a discovery platform to identify high performing catalysts that can be subsequently investigated with physical organic methods to identify the origins of selectivity.
Highlights
Efficient, catalytic, enantioselective reactions have a transformative impact on chemical synthesis, and these are important components of a synthetic chemist’s toolbox
Over the past decade our laboratory has focused on the development of tools which merge the power of modern computing, data science, and machine learning with chemoinformatics in an effort to create models which make reliable predictions of catalyst enantioselectivity.[4]
By ensuring that maximum catalyst structure diversity is represented in a dataset, we have found that the predictions made from models trained on that dataset can be reliable at predicting new catalyst selectivities – a much more difficult task than predicting the selectivity of a catalyst that the model has has already seen
Summary
Catalytic, enantioselective reactions have a transformative impact on chemical synthesis, and these are important components of a synthetic chemist’s toolbox. State-of-the-art enantioselective catalyst development has relied on empiricism and the chemical intuition of a proficient chemist. This approach presents undesirable limitations, and many strategies have been developed to accelerate this process, including increasing throughput with advanced screening protocols,[1] making high-throughput computation of transition state energies feasible,[2] and using mechanism-guided correlations between Linear Free Energy Relationships (LFERs) and enantioselectivity.[3]. Catalysts are often screened from commercially available libraries, relying on the assumption that adequate catalyst diversity is found in commercially available compounds
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
