Representation of Chemical Structures

Retrieving bioactive hits from compound databases requires the analysis of the relationship between the stracture of compounds and their biological activity, that is, their binding 

Structural keys are used in order to transform structural information of different molecules into normalized 

BOX 10.1 SMILES, WLN, and InChl Notation of (2E -3-cyclohexyl 2 -[(R) hydroxy(phenyl)methyl]acrylonltrlle 

Molecular ID fingerprints were introdnced to overcome the inherent problems and shortcomings of structural keys. 

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. 

Connectivity matrix and connection table. The most frequently used forms for representing chemical structures in the computer are the connectivity matrix (CM) and the connection table (CT). In the CM, the diagonal element cii is a chemical symbol of the i-th atoms, while the off-diagonal elements cjj represent bond orders 

It can be seen that a number of information in CM is redundant (each bond is listed twice) and that a large portion of matrix is empty (elements are equal to zero). This indicates the structure can be represented more economically with a table of constant width w. Such representation requires only wN instead of N variables. In the i-th row of the new representation w data associated with i-th atom (chemical symbol of the element, sequential numbers and bond types to its neighbors) are stored. Such representation is called the connection table of a chemical structure or CT (Fig. 4.2). 

Another approach is to label the selected chemical fragments of easily recognizable features (for example -OH or -C( = 0)- group, aromatic ring, structural skeletons, etc.) with consecutive numbers, letters, or special characters. The full structure representation is than either a list of all present or a list of all different structural features. If different fragments are labeled as fi, then structure S can be written as a set of m fragments  

For a given structure the number and type of different (or of all) fragments depends entirely on the definition of fragments. For any kind of structure handling procedure the definition of fragments should be unique for all structures in the collection. 

---

To learn more about connection tables and matrix representations of chemical structures... 

Figure 2-23. Transformation of representations of chemical structures between chemists and computers,...

Over and beyond the representations of chemical structures presented so far, there are others for specific applications. Some of the representations discussed in this section, e.g., fragment coding or hash coding, can also be seen as structure descriptors, but this is a more philosophical question. Structure descriptors are introduced in Chapter 8. 

This ambiguous representation of chemical structures as a string allows a veiy efficient similarity search,... 

In the mid 1970s, Ugi and co-workers developed a scheme based on treating reactions by means of matrices - reaction (R-) matrices [16, 17]. The representation of chemical structures by bond and electron (BE-) matrices was presented in Section 2.4. BE-matrices can be constructed not only for single molecules but also for ensembles of them, such as the starting materials of a reaction, e.g., formaldehyde (methanal) and hydrocyanic add as shown with the B E-matrix, B, in Figure 3-12. Figure 3-12 also shows the BE-matrix, E, of the reaction product, the cyanohydrin of formaldehyde. 

Finding the adequate descriptor for the representation of chemical structures is one of the basic problems in chemical data analysis. Several methods have been developed in the most recent decades for the description of molecules including their chemical or physicochemical properties [1]. 

With better hardware and software, more exact methods can be used for the representation of chemical structures and reactions. More and more quantum mechanical calculations can be utilized for chemoinformatics tasks. The representation of chemical structures will have to correspond more and more to our insight into theoretical chemistry, chemical bonding, and energetics. On the other hand, chemoinformatics methods should be used in theoretical chemistry. Why do we not yet have databases storing the results of quantum mechanical calculations. We are certain that the analysis of the results of quantum mechanical calculations by chemoinformatics methods could vastly increase our chemical insight and knowledge. 

Two-Dimensional Representation of Chemical Structures. The lUPAC standardization of organic nomenclature allows automatic translation of a chemical s name into its chemical stmcture, or, conversely, the naming of a compound based on its stmcture. The chemical formula for a compound can be translated into its stmcture once a set of semantic rules for representation are estabUshed (26). The semantic rules and their appHcation have been described (27,28). The inverse problem, generating correct names from chemical stmctures, has been addressed (28) and explored for the specific case of naming condensed benzenoid hydrocarbons (29,30). 

Chemoinformatics (or cheminformatics) deals with the storage, retrieval, and analysis of chemical and biological data. Specifically, it involves the development and application of software systems for the management of combinatorial chemical projects, rational design of chemical libraries, and analysis of the obtained chemical and biological data. The major research topics of chemoinformatics involve QSAR and diversity analysis. The researchers should address several important issues. First, chemical structures should be characterized by calculable molecular descriptors that provide quantitative representation of chemical structures. Second, special measures should be developed on the basis of these descriptors in order to quantify structural similarities between pairs of molecules. Finally, adequate computational methods should be established for the efficient sampling of the huge combinatorial structural space of chemical libraries. 

Four main approaches have been suggested for the representation of chemical structures in machine-readable form fragment codes, systematic nomenclature, linear notations, and connection tables. 

The central point here is that the two representations carry the same information. You just need to remember the rules to understand the simpler ones. To provide more examples of the meaning of these concise representations of chemical structures, I have collected a number of additional examples in the Appendix. 

Table 1 shows an example of markup, generated using the OSCAR 3 system. The abstract of a polymer research paper has been parsed by OSCAR and the resulting markup for the first sentence of the abstract is shown in-line with the text (Table IB). The first chemical entity encountered in the sentence is oleic acid , which has been marked up as type = CM (Chemical Moiety) and a number of other annotations, such as in-line representations of chemical structure (InChl, SMILES) have been attached. 

Figure 1.2. Scheele isolated a family of naturally occurring sour substances. Subsequently, the elemental composition of each substance was determined using Lavoisier s combustion method and later the different structures were proposed. Each member contains common structural element, a carboxylic acid group which gives each its sour taste. The two-dimensional representation of chemical structures as shown is convenient but can be misleading. Also shown is the structure of urea, the first naturally occurring substance to be made in the laboratory by Wohler (shown), who provided the first experimental challenge to the concept of vitalism.

Sachse, S., Rappert, A. and Galizia, C. G. (1999). The spatial representation of chemical structures in the antennal lobe of honeybees steps towards the olfactory code. European Journal of Neuroscience 11 3970-3982. 

A satisfactory theoretical model for ethylene oxide should take into account as many as possible of the physical properties discussed above, but should be able to predict or explain its chemical properties as well. Three such ogodels have been proposed which are based on molecular-orbital theory s 1-3W.1 ° and two more which conform rather to tho valence-bond representation of chemical structure.1M, W 7 The relative merits of all these models have been discussed in recent reviews.8 7 1301... 

A second initiative is being developed by Dr. Ann Richard and coworkers at the EPA. The Distributed Structure-Searchable Toxicity (DSSTox) public database network is a flexible community-supported, web-based approach for the collation of data. It is based on the SDF format for the representation of chemical structure. It is intended to enable decentralized, free public access to toxicity data files. This should allow users from different disciplines to be linked. Public, commercial, industry, and academic groups have also been asked to contribute to, and expand, the DSSTox public database network. Data from potentially any toxicological endpoint can be collated in the DSSTox public database network, including both human health, and environmental endpoints (Richard et al 2002 Richard and Williams, 2002). 

These identifiers were developed as an lUPAC project in 2000-2004. They are the most recent technology aimed at an unambiguous text-string representation of chemical structures. (Earlier technologies included Wiswesser line notation, which is not described here, and SMILES, described below.)... 

Common planar representations of chemical structures account at best qualitatively for the basic properties of such encoded compounds. For example, one can expect a sour taste for a compound formula, which contains a carboxylic group. However, to deduce which one of two given compounds, possessing the same carboxylic group, will be more acidic and how many times more, is a matter of a more or less scientific guess. Three-dimensional representations readily provided by modern molecular modelling software are not those helpful in this respect. 

FIGURE 2 Mathematical representation of chemical structure as (a) a graph (b) a molecular graph (c) an adjacency matrix. 

Molecular Gngerprints are string representations of chemical structures designed to enhance the efficiency of chemical database searching and analysis. They can encode... 

Takahashi, Y, Sukekawa, M. and Sasaki, S.I. (1992). Automatic Identification Molecular Similarity Using Reduced-Graph Representation of Chemical Structure. J.Chem.Inf.Comput.ScL, 32, 639-643. 

Zupan, J. and Novic, M. (1997). General Type of a Uniform and Reversible Representation of Chemical Structures. Anal.Chim.Acta, 348,409-418. 

---