Structural formulae descriptors

Most of the models and descriptors discussed so far are based on the two-dimensional representation of the compounds, i.e. on their structural formula. 

Although the term vitamin A has been used to denote specific chemical compounds, such as retinol or its esters, this term now is used more as a generic descriptor for compounds that exhibit the biological properties of retinol. Retinoid refers to the chemical entity retinol or other closely related naturally occurring derivatives. Retinoids also include structurally related synthetic analogues, which need not have retinol-like (vitamin A) activity. The structural formulas for the vitamin A family of retinoids are shown in Figure 66.3. Retinol (vitamin Aj), a primary alcohol, is present in esterihed form in the tissues of animals and saltwater fish, mainly in the liver. A closely related compound, 3-dehydroretinol (vitamin A2), is obtained from the tissues of freshwater fish and usually occurs mixed with retinol. 

Full details of the QSAR training set should be provided, including details of chemical structure (names and structural formulae, and chemical abstract service [CAS] numbers if available) and, in the case of a QSAR, data for all descriptor and response variables. If the data used to develop the model were based upon the processing of raw data (e.g., the averaging of replicate values), it is preferable if all of the raw data are supplied. 

The structural formula of this new cytostatic (picoplatin) is shown below. In determining the stereo descriptor, the situation arises here that the highest ranking ligand occurs twice. In such cases the priority number of the ligand coordinated trans to it with the lowest priority (i.e. with the highest priority number) is used as the configuration index (principle of trans maximum difference). 

These are deduced from a topological picture (2D picture) of the molecules. The picture carries information on how the atoms are connected and what is the nature of bonds (structural formula of a molecule). Mathematically, the topology picture is described with the connectivity matrix. Pioneering work in this field was published in 1947 by Wiener on paraffin hydrocarbons [31]. It is defined as a half sum of the off-diagonal elements in the topological distance matrix. In the last few decades dozens of descriptors have been deduced... 

The FDA is represented on the US. Adopted Names CouncU, in cooperation with the American Medical Association, the American Pharmaceutical Association, and the US. Pharmacopeia (USP). This group has adopted a variety of consistent rules for established names. Proposals for the adoption of nonproprietary names submitted undergo routine review for stereochemical accuracy in both systematic names and structural formulas. As part of the ongoing review of the compendial literature, the USP has a project in process that will result in the correction of systematic names which are incomplete with regard to stereochemical descriptors. 

Fig. 5 Structural formulas are the best representation of molecules for chemists - numerical descriptors are preferred by computers.

Substructure descriptors describe the two-dimensional structural formulas and are therefore very valuable formal representations of chemical stmctures. About 0.1 to 1 s is needed to compute an SSD for one compound. They can thus be appHed to huge virtual libraries. 

Fig. 10 Use of pharmacophore descriptors in order to find the correct superposition of dihydrofolate and methotrexate [73] (see legend of Fig. 9 for the pharmacophore types), a) According to the structural formula the molecules are very similar and can be superimposed easily, b) However, visualization of hydrogen bonding acceptor and donor sites turns up major differences in the hydrogen bonding pattern, c) One-by-one mapping of the pharmacophore vectors leads to a different superposition the heterocyclic ring system of methotrexate is rotated to reproduce the pharmacophore of dihydrofolate.

Topological descriptors can be computed very quickly for a huge number of compounds. These descriptors are able to reproduce the intuitive classification of structural formulas. For this reason, topological descriptors are a good choice to describe the diversity of huge virtual libraries in an automated way. 

Classification based on three-dimensional molecular structures requires more time-consuming computations, but results in more reliable information. These descriptors are able to identify different pharmacological profiles of compounds and thus provide the chemist with novel information that is often not recognizable from the structural formulas directly. 

The National Cancer Institute (NCI) database at http // 129.43.27.140/ncidb2/ contains over 250000 entries, which are searchable by structure, formula, CAS registry number or name. The results are analysed in terms of physical molecular descriptors (such as log P) and predicted biological activities against around 100 biological activities. 

Before embarking on a description of the computational methods involved, and how well they perform, we should address the goals of crystal structure prediction. At its most ambitious level, the aim is to start from nothing more than the structural formula of a molecule and to predict, with perfect reliability, the structure of the resulting solid, with no input from experimental observations. (Here, by structure, we mean the space group, unit cell parameters and a fiiU specification of all atomic positions.) This goal is, of course, unrealistic polymorphism in molecular crystals tells us that there is often not just one crystal structure for a molecule and we know that the crystal that is produced in an experiment depends on a variety of factors, from thermodynamic descriptors of the system (temperature and pressure) to the method of crystallization, solvent used and the presence of impurities. Without a detailed description of the crystallization conditions, prediction of the resulting structure cannot be the aim. Furthermore, many of these factors are not sufficiently well understood to be represented in a computational procedure for crystal structure prediction. 

Certainly, complete generation of all structures is not the only way to solve the problem. A review of different methods can be found in [13], where in particular the evaluation of molecular descriptors for large libraries specified by generic structural formulas is described. In any case canonization of data and normal forms play a central role. Hence it is time to consider this problem and to describe a canonizer. 

In the Fischer convention, the ermfigurations of other molecules are described by the descriptors d and L, which are assigned comparison with the reference molecule glyceraldehyde. In ertqrloying the Fischer convention, it is convenient to use projection formulas. These are planar representations defined in such a w as to convey three-dimensional structural information. The molecule is oriented with the major carbon chain aligned vertically in such a marmer that the most oxidized terminal carbon is at the top. The vertical bonds at each carbon are directed back, away fiom the viewer, and the horizontal bonds are directed toward the viewer. The D and L forms of glyceraldehyde are shown below with the equivalent Fischer projection formulas. 

Attempts to produce descriptors similar to cis and trims for stereochemicidly more complicated coordination entities have tailed to achieve generality, and labels such as foe and mer are no longer recommended. Nevertheless, a diastereoisomeric structure may be indicated for any polyhedron using a configuration index as an affix to the name or formula. Finally, the chiralities of enantiomeric structures can be indicated using chirality symbols. 

Nowadays, more than 4000 types of descriptors are known.17 There exist different ways to classify them. With respect to the type of molecular representation used for their calculations—chemical formula, molecular graph, or spatial positions of atoms—one speaks about ID, 2D, and 3D descriptors, respectively. Descriptors can be global (describing the molecule as a whole) and local (only selected parts are considered). One could distinguish information-based descriptors, which tend to code the information stored in molecular structures, and knowledge-based (or semiempir-ical) descriptors issued from the consideration of the mechanism of action. Most of those descriptors can be obtained with the DRAGON, CODESSA PRO, and ISIDA programs. 

---