Synthesis Design Systems

Clearly, for symmetry reasons, the reverse process should also be considered. In fact, early versions of our reaction prediction and synthesis design system EROS [21] contained the reaction schemes of Figures 3-13, 3-15, and 3-16 and the reverse of the scheme shown in Figure 3-16. These four reaction schemes and their combined application include the majority of reactions observed in organic chemistry. Figure 3-17 shows a consecutive application of the reaction schemes of Figures 3-16 and 3-13 to model the oxidation of thioethers to sulfoxides. 

To recognize the different levels of representation of biochemical reactions To understand metabolic reaction networks To know the principles of retrosynthetic analysis To understand the disconnection approach To become familiar with synthesis design systems... 

Because of the complexity of the problem and the large amount of program development work that has to go into a synthesis design system, only a few groups worldwide have been active in this area. Here, we mention only a selection of the major ideas and achievements in this area. It is not the intention to give a comprehensive overview. For this, interested readers can consult Chapter X, Section 3.2 in the Handbook. This chapter presents one such system, the WODCA program, in greater detail. 

Several concepts for the implementation of synthesis design system have appeared since the early 1970s. 

Just as a researcher will perform a literature synthesis for a compound, there are computer programs for determining a synthesis route. These programs have a number of names, among them synthesis design systems (SDS) or computer-aided organic synthesis (CAOS) or several other names. 

The interrelation of reaction databases and synthesis design systems is brought into close focus this week by a variety of presentations, not least from the originator of many of the ideas on automatic reaction analysis and synthesis design, George Vladutz. 

In 1967, work was presented from a Sheffield group on indexing chemical reactions for database budding. In 1969, a Harvard group presented its first steps in the development of a system for computer-assisted synthesis design. Soon afterwards, groups at Brandeis University and TU Munich, Germany, presented their work in this area. 

To be able to define reaction planning, reaction prediction, and synthesis design To know how to acquire knowledge from reaction databases To understand reaction simulation systems... 

Computer-aided process synthesis systems do not mean completely automated design systems (57). Process synthesis should be carried out by interactive systems, in which the engineer s role is to carry out synthesis and the machine s role is to analy2e the performance of synthesized systems. Computet apphcations in the future will probably deal with the knowledge-based system in appHed artificial intelligence. Consequendy, research on computer-aided process synthesis should be directed toward the realization of such systems with the collaboration of experienced process engineers. 

Masso, A. H. and Rudd, D. F. (1969) AIChEJI 15, 10. The synthesis of system design heuristic structures. 

A New Treatment of Chemical Reactivity Development of EROS, an Expert System for Reaction Prediction and Synthesis Design... 

Synthesis design and reaction prediction can draw benefits from all these features of a computer. Our own work in this area began in 1974, and in 1978 the computer program system EROS (Elaboration of Reactions for Organic Synthesis) was first presented 6. Since then, several reports on certain aspects of the system development have appeared, but sometimes in less easily available journals or books7. Moreover, there has been no description of the overall system as it now stands. This article is intended to rectify this situation. 

A system for synthesis design working with a database of reactions will initially contain only a few reactions and thus produce only a few alternatives. Thus, evaluation and selection are not very important. However, as the size of the database is expanded, more and more alternatives will be obtained for a given target structure. Here again, the development of a general evaluation and selection package will be required to handle the various transforms of the database. 

We have already emphasized our view that the evaluation of chemical reactions and synthetic pathways is of preeminent importance in any system for computer-assisted synthesis design or reaction prediction. The quality of the evaluation process will determine to a large extent the overall quality of such a system. 

---