Marko Razinger

Graph automorphism perception algorithms in computer-enhanced structure elucidation.

The concept of graph symmetry is explained in terms of the vertex automorphism group, which is a subgroup of the complete vertex permutation group. The automorphism group can be deduced from the automorphism partition of graph vertices. An algorithm is described which constructs the automorphism group of a graph from the automorphism vertex partitioning. The algorithm is useful especially for graphs which contain more than one vertex-partition set. Several well-known topological symmetry perception algorithms that yield automorphism partitions are compared. The comparison is favorable to the Shelley-Munk algorithm, developed in the framework of the SESAMI system for computer-enhanced structure elucidation.

Modelling of gas chromatographic retention indices using counterpropagation neural networks

Abstract Unspecific fragmentation of organic substances in the ion source of MS detector hinders identification of organic substances in gas chromatographic separation. In such instances theoretical prediction of the retention indices could be a useful tool. A new method for theoretical prediction of gas chromatographic retention indices is described. Artificial neural networks were trained in counterpropagation mode to predict retention data. Extensive data sets of simple organic compounds with known retention indices taken from the literature were serving for training and test sets. The structure of molecules was described with a 12-dimensional vector the components of which were topological and chemical parameters. Various geometries of artificial neural networks were tested and different divisions into training and testing sets tried. The ANN with the configuration of 15 × 15 neurons has been chosen for routine work. The average RMS value was 36.6 retention time units.

On calculation of the detour index

An algorithm for the detection of the longest path between any two vertices of a graph was proposed. The method was used to calculate the detour index of fused bicyclic structures. Analytical formulas for the detour index of fused bicyclic structures were derived, and it was shown the these can only be third-order polynomials.

Molecular Shapes and Chirality

In this report we consider quantitative characterization of molecular chirality based on the binary code that describes the shape of the molecular periphery. We have limited our attention to chirality in two-dimensional space and consider benzenoid forms that are chiral and have constant perimeter P, i.e., we consider planar structures constructed from fused regular hexagons. The binary codes used assign either zero or one to CC bonds on the molecular periphery depending on whether we turn right or left, respectively, at each branching site of graphite lattice as we move around the molecular periphery. We have examined all benzenoid structures having molecular perimeter of 22 and fewer CC bonds (over 30 pairs of chiral structures). Since enantiomers have distinct codes, we based our quantitative measure of chirality on the degree of dissimilarity between the codes for a structure and its mirror image. As discussed in the text, the measure of dissimilarity depends on the length of the segments of the codes...

Stereoisomer generation in computer-enhanced structure elucidation

The existence of stereoisomerism can be related to the presence of structural features referred to as stereocenters. On the basis of this relationship, computer software capable of generating all stereoisomers of a given constitutional isomer has been developed. Input to the program is a canonical connection table. In an initial step the stereocenters present in the structure are identified and characterized either as true-stereocenters or para-stereocenters. In the presence of n different true-stereocenters, structure generation is trivial and leads to 2n stereoisomers, each of which can be represented by a parity vector. In the presence of topological symmetry, the number of stereoisomers may be less than 2n; i.e., some of the 2n parity vectors correspond to equivalent configurations. An algorithm for determining equivalences among the set of 2n parity vectors is described. Program output includes the parity vector of each valid stereoisomer and the relationship between them, i.e., enantiomeric or diastereomeric. A simple procedure for representing the parity vector of the stereoisomer as a three-dimensional representation is given.

Kisik — a combined chemical information system for a minicomputer

Abstract A minicomputer-oriented chemical information system (CIS) based on three different Spectrometries, i.e. infrared, mass, and 13C-n.m.r., is described and discussed.The system has roughly the same characteristics as CISs implemented on large mainframe computers: substructure search, library searches on various files, file manipulation, statistical handling of retrieved data, etc. The source package is very suitable and simple for moving the entire CIS from one computer to another. In addition, the system has the Option UPDATE that enables the user to create his own feles and modify them easily; this is rather difficult and expensive to implement on larger systems because of the very high disk-space price to frequency-of-access ratio. However, the quantity of data is strongly dependent on the disk space. At the moment the system handles 1016 compounds, each of which is described by a chemical name, molecular formula and weight, two-dimensional structure image, infrared, mass, and 13C-n.m.r. spectra. All these data for one compound are linked on-line via the identity number of the compound so there is no delay in accessing any of the items mentioned. The entire data bank together with the program package has a 1.8 Mbyte requirement which fits well within the 2.5 Mbyte space available on the small disk used by a PDP 11 34 minicomputer.

Stereochemistry and sequence rules a proposal for modification of Cahn-Ingold-Prelog system

Abstract A modification of the CIP system is proposed to combine the advantages of the CIP system and the methods for computerised treatment of stereogenicity. The proposal is based strictly on the topology. The basis for ranking of ligands in the computer application is a method for the canonical numbering of the molecular graphs, therefore no transformations to the hierarchical digraphs are necessary. For the manual application of the proposed system, a transformation of rings and multiple bonds to chains which is equivalent to the canonical numbering, can be used. New terms constitutional stereogenic centre and configurational stereogenic centre are introduced to specify whether a stereogenic centre has the ligands with identical constitution or not. The term configurational stereogenic centre is a topological alternative to the geometrical term pseudoasymmetric stereogenic centre . Set of interdependent stereogenic centres is proposed as a basis for the treatment of the stereochemical differences between the ligands. The topological definition of the interdependence of the stereogenic centres is also presented. The modified system of the sequence rules consists of four rules.

Interactive Simulation of Infrared, Mass, and 13C NMR Spectra

(1 ) “Gebietsbeschreibung Anorganika”, 2nd ed., IDC Internationale Dokumentationsgesellschaft fur Chemie mbH Fachinformationszentrum Chemie, 6000 Frankfurt 90, Germany, March 1979, p 3. (2) M. Isenberg and H . Kaltenhauser, “Systematische Speicherung anorganischer Stoffe auf dem GREMAS-Erganzungsband” , Unveroffentlichter Bericht der IDC Internationale Dokumentations(3) gesellschaft fur Chemie mbH, 6000 Frankfurt 90, West Germany, 1969. W. Simmler, “Wissenschaftlich-technische Dokumentation in der chemischen Industrie-ein Beitrag zum Umweltschutz”, In Deutscher Dokumentartag, 1976 (publiziert 1977). Verlag Dokumentation Miinchen, pp 164-77. “Information uber die anorganisch-chemische Literatur”, Nachr. Chem. Tech., 23, 524-5 (1975). H. Griinewald. “Information und Dokumentation auf dem Gebiet der anorganischen’Chemie”, Nachr. Dok., 28, 19-25 (1977). “Magnetbandsystem Anorganische Chemie”, Nachr. Chem. Tech., 19, 182-3 (1971); Nachr. Dok., 22, 172-3 (1971). F. Ehrhardt, “Utilization of Chemical Abstracts Service (CAS) Data Bases. Application for the IDC-lnorganica-Documentationsystem”, Bundesministerium fur Forschung und Technologie, Report B M R F B ID 79-03, 1979, p 53. C . Weiske, “Das Klassifikationssystem des Chemischen Informationsdienstes”. DK-Mitt. 21. 11-5 (1977). C. Weiske, “Chemical’Information Services;’, Angew. Chem., Inr . Ed. Engl., 9, 5 5 0 5 (1970). E. Meyer, R. Jansen, and E. Sens, “Das IDC-Thesaurus-System”, Nachr. Dok., 23, 203-11 (1972).

Computer generation of chemical structures from known fragments

The interactive generation of chemical structures from given fragments is described and discussed. It is implemented as a part of our expert system CARBON, based on C-13 NMR spectra. As it is designed, this program can also be a useful tool in the structure elucidation process when information on parts of the structure is obtained by other means (IR, mass and other spectrometries, chemical analysis, other relevant information). The topological characteristics of candidate fragments are first chosen interactively and then the elements are connected in all topologically possible ways. In the following step, the topological building blocks are substituted by chemical structural fragments resulting in a set of all chemical structures consistent with the input information.

5 Reduction of the Information Space for Data Collections

Publisher Summary The chapter discusses reduction of the information space for data collections. In spite of the fact that on large hard disks quite large quantities of complex data can be stored, the actual bottleneck is still computation time for handling large amounts of complex objects. The computation time depends on the number of variables (components) with which each object is represented. The general form of a multivariate object that is stored in a collection of equally represented items is the vector form. The chapter discusses the use of two methods, Fast Fourier Transformation (FFT) and Fast Hadamard-Transformation (FHT), for the reduction of object representations and presents some examples in one- and two-dimensional patterns (spectra, images) the way explained procedures can be used to estimate the most appropriate reduction scheme for a given problem. It has to be emphasized that any reduction of information space means a loss of information; hence, a great care has to be paid in each application to avoid losing the relevant ones.