Structures Browsing

Fig. 1. Empiric observation that more than approx 20 compounds from any scaffold need to be tested in order to be sure that an active will be found. (A) This shows the results from a whole cell assay in which the compounds have been classified using the Level 1 Ring System of the Structural Browsing Index (SBI) also known as molecular equivalence numbers or meqnums, which is the most-detailed level. The most-populated SBI containing only inactives is labeled 1, it contains 16 compounds. The most populated SBI containing actives is labeled 2. (B) Looking at only the compounds present in the most-populated SBI (2 above) and arranging the compounds in this SBI randomly, the smallest set of compounds in which an active (indicated by +) can be found is approx 30 compounds.

Fig. 4. Comparision of molecular equivalence numbers or structural browsing indices (SBI) containing active molecules from four different sublibraries. This figure shows how, even though the libraries described in Fig. 3 occupy similar chemistry spaces, they still undersample certain areas so that some active classes are only found when screening one particular sublibrary and not the others. This figure takes the active compounds represented in Fig. 3B using structural Browsing Indices and shows which library the actives came from. This serves to emphasize how active compounds with particular structural features may be identified in only one of the three sublibraries, e.g., compounds containing the SBI 7704 are only found in the CAC sublibrary.

Based Application for Clustering, Structure Browsing, and Structure-Activity Relationship... 

This structure browsing procedure has proved very popular with users at Pfizer, both as a simple alternative to substructure search, and as a means of identif5dng analogous structures (lateral thinking ). Examples of the sort of similar structures produced from browse searches are shown in Figures 1 and 2 (not ranked in order) other examples are given in reference 4. Some of these would clearly be retrieved by conventional substructure searches, while others would not. 

In a similar fashion to structure browsing, this is used in two ways within SOCRATES. A complete file of structures may be divided into clusters, systematically reflecting the structural variation within the file, and one compoimd chosen from within each cluster, so as to create a subset of structural representatives , e.g., for use in large-scale screening programmes. Alternatively, a substantial output from a substructure search may be clustered, and one exemplifying structure from each cluster examined. In this way, the structural variation within the output may be readily appreciated. 

Sandy Lawson s contribution on chemical structure browsing (Chapter 4) may seem peripheral to the main topic of the symposium, but the algorithm he employs could, he hopes, be used for data bases other than Beilstein in future. 

Some of the aspects of structure browsing with the Lawson Number (LN) are described including limitations. Use of several LN in combination, single LN, and range-searching is demonstrated for the retrieval of various analogues, including positional isomers. 

The purpose of this paper is to describe the implementation of a structure browsing tool in computerized data bases, with particular reference to the Beilstein data base as an example. The retrieval term which will form the main subject of this paper is the so-called Lawson-Number (LN). For obvious reasons, it is somewhat embarrassing for the present author to give an account of a descriptor which bears his own name. However, the driving force behind the choice of that particular name was not the author himself, but rather a combination of circumstances, the prime being that all sensible names for numbers of all descriptions had already been used up in the early days of the Beilstein Online venture. There were (and still are) the following terms in daily use in the production processes at the Beilstein Institute ... 

---