Connection Table Generation

The CROSSBOW chemical search system (1-5) is a multilevel one. Search of the bit screens is first carried out and this quickly and cheaply reduces the file to ten per cent (or less) of its original size. There will almost certainly be many false drops but most of these can usually be removed by string search of the WLNs and/or molecular formulae and/or reference numbers. String searching is slower and more expensive than bit search. Connection table generation and atom-by-atom search of the connection tables (the third search level) are still slower and even more expensive, but the atom-by-atom search program is a very powerful tool which is used in about 80% of all searches. The CROSSBOW connection tables for the hits from any search are finally used as input to a structure display program. 

Bit and string searching is an interactive process. The paramaters are input on a VDU, the hit count is displayed on-line and, if necessary, the search parameters can be modified and the search repeated. At the end of a session, search hit files are merged as required and connection table generation, atom-by-atom search and structure display are run batchwise. 

An average bit and string search takes about five minutes (searching 190,000 compounds). The CROSSBOW connection table generation program handles 1500-2000 compounds a minute, and the atom-by-atom program searches 600 compounds a minute. Well over 90% of notations are amenable to connection table generation and over 90% of the compounds on the database can be structurally displayed directly from the connection table. 

Benzene has already been mentioned as a prime example of the inadequacy of a connection table description, as it cannot adequately be represented by a single valence bond structure. Consequently, whenever some property of an arbitrary molecule is accessed which is influenced by conjugation, the other possible resonance structures have to be at least generated and weighted. Attempts have already been made to derive adequate representations of r-electron systems [84, 85]. 

RAMSES is usually generated from molecular structures in a VB representation. The details of the connection table (localized charges, lone pairs, and bond orders) are kept within the model and are accessible for further processes. Bond orders are stored with the n-systems, while the number of free electrons is stored with the atoms. Upon modification oF a molecule (e.g., in systems dealing with reactions), the VB representation has to be generated in an adapted Form from the RAMSES notation. 

A descriptor for the 3D arrangement of atoms in a molceulc can be derived in a similar manner. The Cartesian coordinates of the atoms in a molecule can be calculated by semi-empirical quantum mechanical or molecular mechanics (force field) methods, For larger data sets, fast 3D structure generators are available that combine data- and rule-driven methods to calculate Cartesian coordinates from the connection table of a molecule (e.g., CORINA [10]). 

An important approach to the graphic representation of molecules is the use of a connection table. A connection table is a data base that stores the available bond types and hybridizations for individual atoms. Using the chemical formula and the connection table, molecular stmctures may be generated through interactive graphics in a menu-driven environment (31—33) or by using a linear input of code words (34,35). The connection table approach may be carried to the next step, computer-aided molecular design (CAMD) (36). 

Leatherface is a 2-D molecular editor that modifies properties of atoms and bonds in molecular connection tables according to rules specified by the user. Unlike Permute, Leatherface encodes no chemical knowledge and neither processes nor generates 3-D structures. Its real strength is that it allows the user to impose a very detailed and precisely specified chemical view on large numbers of connection tables. 

The DARC code [10] resembles a connection table, since it expresses or implies the nature of each atom and bond, but it is generated in a concise, linear form. The description begins with one atom which is chosen as the "focus" of the structure and then proceeds outward, describing the "environment" of the "focus."... 

Algorithm I - Registration - Canonicalization of Connection Tables. A connection table for a chemical substance with n atoms can be numbered in as many as n different ways. The problem of generating a canonical form involves selecting a... 

Compare the newly generated compact connection table to the retained compact connection table. 

Figure 11. Generation of a coordinate representation from a connection table ( continued on facing page)...

Once the desired structure is generated the user should be able to use its representation (the connection table) in many different ways to store it, to combine it with other structures, supplement it with textual information, to decompose it to fragments, add it to a collection, use it as a target or query compound in different searches or procedures, use it in different applications such as simulation of spectra, determination of properties, etc. calculate molecular formula, draw it on a plotter, etc. 

Yoneda [189] has devised a program which generates elementary reaction networks for reactions involving free radicals, ions or active sites on heterogeneous catalysts. Reactants and products are represented by connectivity tables and reactions by matrices equal to the difference of the matrices of products and reactants, respectively. As soon as a set of reaction matrices has been defined, by applying them to the initial reactants, a new set of intermediates and products is obtained, which are themselves submitted to reactions and so on. Restrictions are necessary to avoid the appearance of unrealistic steps or compounds. 

---