This presentation provides an update on the use of chemical structure-based machine learning for predicting which chiral HPLC column will be useful for separating the enantiomers of a particular compound. Historically a prolonged trial and error process enabled by laboratory automation, chiral column selection is now becoming predictable, thanks to the help of machine learning and artificial intelligence.
Learning Objectives:
Understand the basic priciples of the 3DMol 'point cloud' approach to chemical structure-based machine learning predictions
Understand the use of metrics for comparing the accuracy of different chemical structure-based machine learning models
Understand the structure and function of the NSF Center for Bioanalytic Metrology and how it helps to focus academic innovation at Purdue, Indiana University and Notre Dame on measurement science challenges of importance to our industry members, Abbvie, Agilent, BASF, BMS, Corteva, Evonik, Exxon Mobil, Genentech, Lilly, Merck, Moderna, Pfizer, Proctor & Gamble and Takeda
