Andrea Bootsma
Andrea Bootsma
Department of Chemistry
University of Georgia
Chemistry Building, Room 400
Organic Seminar

 

Abstract: Retrosynthetic analysis and the development of a synthetic plan is a key step in the synthesis of natural and most pharmaceutically relevant targets. The algorithmic nature of this process makes it appear amenable to automation, and in fact the development of a synthesis planning computer program was first proposed by E. J. Corey in 1967, in the same paper in which he first formalized the principles of retrosynthetic analysis. In the half-century since that proposal there have been multiple attempts to create such a program, that could take a synthetic target as input and provide for the user a full synthetic pathway, beginning from available starting materials. It is only recently that two programs have achieved this goal. The first such program, Chematica, uses a library of reaction rules programmed by expert organic chemists to identify potential pathways. The synthetic routes generated by this program have been used in the successful synthesis of industrially relevant targets. The second program, Mentor, which was developed more recently, uses neural networks to determine the probability of each potential retrosynthetic step, and through recursive application develops likely synthetic routes. These two programs have made E. J. Corey’s original idea a reality for today’s organic chemist