Swiss Chemical Society (SCS) Lectureship - Prof. Scott Denmark (University of Illinois)
Event details
| Date | 20.09.2023 |
| Hour | 16:15 › 17:15 |
| Speaker | Prof. Scott Denmark, University of Illinois at Urbana-Champaign (UIUC), USA |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Link to poster
Title: Application of Chemoinformatics and Machine Learning to Enantioselective Catalysis
Abstract:
The development of synthetic methods in organic chemistry has historically been driven by Edisonian empiricism. Catalyst design is no exception wherein experimentalists attempt to qualitatively recognize patterns in catalyst structures to improve catalyst selectivity and efficiency. However, this approach is hindered by the inherent limitations of the human brain to find patterns in large collections of data, and the lack of quantitative guidelines to aid catalyst selection. Chemoinformatics provides an attractive alternative for several reasons: no mechanistic information is needed; catalyst structures can be characterized by 3D-descriptors which quantify the steric and electronic properties of thousands of candidate molecules; and the suitability of a given catalyst candidate can be quantified by comparing its properties to a computationally derived model on the basis of experimental data. The ability to accurately predict a selective catalyst using a set of non-optimal data remains a Grand Challenge of machine learning with respect to asymmetric catalysis.
This lecture will describe a newly developed, chemoinformatic workflow that consists of the following components: (A) construction of an in silico library of a large collection of conceivable, synthetically accessible catalysts of a particular scaffold; (B) calculation of robust chemical descriptors for each scaffold (C) selection of a representative subset of the catalysts in this space. This subset is termed the Universal Training Set (UTS), so named because it is agnostic to reaction or mechanism. (D) Collection of the training data, and (E) application of modern machine learning methods to generate models that predict the enantioselectivity of each member of the in silico library. These models are evaluated with an external test set of catalysts. The validated models can then be used to select the optimal catalyst for a given reaction. Recent illustrations include enantioselective addition of thiols to acyl imines, atropselective iodination of pyridines and enantioselective vinylogous Mukaiyama aldol reactions. Solutions to challenges presented by small data sets and failed regression analysis will be presented.
Speaker's biography:
Scott E. Denmark was born in Lynbrook, NY in 1953. He obtained a S. B. from the Massachusetts Institute of Technology in 1975 and a D.Sc. Tech. from the ETH-Zürich under the direction of Albert Eschenmoser in 1980. That same year he was appointed as assistant professor at the University of Illinois and since 1991 has been the Reynold C. Fuson Professor of Chemistry.
Professor Denmark has won a number of honors including the Pedler and Robert Robinson Medals (RSC), the Aldrich Award for Creative Work in Synthetic Organic Chemistry, Brown Award for Creative Research in Synthetic Methods, the F. S. Kipping Award in Silicon Chemistry (ACS), the Prelog Medal, Noyori Prize and the Paracelsus Prize. He is currently on the Board of Directors of Organic Reactions and Organic Syntheses and serves on the Editorial Advisory Boards of the Journal of the American Chemical Society, the Journal of Organic Chemistry and Organic Letters.
Lab website: https://denmarkgroup.illinois.edu/
Title: Application of Chemoinformatics and Machine Learning to Enantioselective Catalysis
Abstract:
The development of synthetic methods in organic chemistry has historically been driven by Edisonian empiricism. Catalyst design is no exception wherein experimentalists attempt to qualitatively recognize patterns in catalyst structures to improve catalyst selectivity and efficiency. However, this approach is hindered by the inherent limitations of the human brain to find patterns in large collections of data, and the lack of quantitative guidelines to aid catalyst selection. Chemoinformatics provides an attractive alternative for several reasons: no mechanistic information is needed; catalyst structures can be characterized by 3D-descriptors which quantify the steric and electronic properties of thousands of candidate molecules; and the suitability of a given catalyst candidate can be quantified by comparing its properties to a computationally derived model on the basis of experimental data. The ability to accurately predict a selective catalyst using a set of non-optimal data remains a Grand Challenge of machine learning with respect to asymmetric catalysis.
This lecture will describe a newly developed, chemoinformatic workflow that consists of the following components: (A) construction of an in silico library of a large collection of conceivable, synthetically accessible catalysts of a particular scaffold; (B) calculation of robust chemical descriptors for each scaffold (C) selection of a representative subset of the catalysts in this space. This subset is termed the Universal Training Set (UTS), so named because it is agnostic to reaction or mechanism. (D) Collection of the training data, and (E) application of modern machine learning methods to generate models that predict the enantioselectivity of each member of the in silico library. These models are evaluated with an external test set of catalysts. The validated models can then be used to select the optimal catalyst for a given reaction. Recent illustrations include enantioselective addition of thiols to acyl imines, atropselective iodination of pyridines and enantioselective vinylogous Mukaiyama aldol reactions. Solutions to challenges presented by small data sets and failed regression analysis will be presented.
Speaker's biography:
Scott E. Denmark was born in Lynbrook, NY in 1953. He obtained a S. B. from the Massachusetts Institute of Technology in 1975 and a D.Sc. Tech. from the ETH-Zürich under the direction of Albert Eschenmoser in 1980. That same year he was appointed as assistant professor at the University of Illinois and since 1991 has been the Reynold C. Fuson Professor of Chemistry.
Professor Denmark has won a number of honors including the Pedler and Robert Robinson Medals (RSC), the Aldrich Award for Creative Work in Synthetic Organic Chemistry, Brown Award for Creative Research in Synthetic Methods, the F. S. Kipping Award in Silicon Chemistry (ACS), the Prelog Medal, Noyori Prize and the Paracelsus Prize. He is currently on the Board of Directors of Organic Reactions and Organic Syntheses and serves on the Editorial Advisory Boards of the Journal of the American Chemical Society, the Journal of Organic Chemistry and Organic Letters.
Lab website: https://denmarkgroup.illinois.edu/
Practical information
- Informed public
- Free