Conformational analysis general approach

The accurate spatial location of these atoms generally needs a sophisticated approach, for example, the study of a complete deuterated set of isotopic derivatives in microwave spectroscopy or the use of neutron diffraction techniques. We shall see below that a set of CNDO/2 calculations combined with suitable experiments (microwave spectroscopy and/or electron diffraction) may help to solve the geometrical and conformational analysis of compounds containing many hydrogen atoms. 

When a validated hit is selected as a promising lead compound, its physicochemical profile must be studied in detail. Sophisticated in silica approaches such as 3D lipophilicity predictions coupled with extensive conformational analysis [49, 50,135,146] and molecular field interactions (MIFs) [147-150] could be helpful to better interpret the detailed experimental investigations of their ionization constants by capillary electrophoresis or potentiometric titrations [151, 152] and their lipophilicity profiles by potentiometry [153]. However, these complex approaches cannot be performed easily on large number of compounds and are generally applied only on the most promising compounds. 

An obvious approach to the problem of conformational analysis in these complexes is the use of NMR spectroscopy. However, several general problems immediately arise, namely, that most metal ions have nuclear spin, many are labile to substitution of the ligands, and they are frequently paramagnetic. In addition the ligands often contain N14. All these properties potentially hinder the observation of fine structure and thus the analysis of the NMR spectra. However, in some complexes the problems are either eliminated or reduced. For example Co(III) compounds are usually diamagnetic and inert to substitution, and in some complexes the ligand-proton fine structure is observed to be relatively uncomplicated by Co or N quad-rupole relaxation effects, for example, (CoEDTA)", Fig. 18. 

Current methods take root in the early 1960s, when the conformational analysis of macromolecules became of general interest [29-30]. Anderson et al. [31] used model building and X-ray diffraction studies to determine the double helical structures of polysaccharides using crystalline structure data as an initial set of coordinates followed by computational sampling of new structures by rotation around selected covalent bonds. The details of these so-called hard-sphere calculations are described in Rees and Skerrett [32] and Rees and Smith [33]. This approach was also applied to carbohydrate conformations in the analysis of bacteria and polysaccharidic structures and linkages [34-35]. 

In general, five different approaches can be distinguished and are applied to explore the conformational space of a molecule systematic searches, rule-based and data-based approaches (model building), random methods, genetic algorithms, distance geometry, and simulation methods. Some of the basic principles and ideas behind these concepts have already been described in the Secs. 2 and 3 of this article. In the following, the application of these concepts to conformational analysis and searches will be discussed. 

Bach year usually sees the publication of a handful of new methods for exploring conformational space. Many can be considered as variants on one of the approaches discussed thus far but which may provide some advantage in terms of the efficiency and effectiveness with which they explore conformational space. Some of these alternative methods are designed for quite specific types of molecule (such as ring systems) and as may not be particularly general approaches to conformational analysis. Here we describe in more detail two of these newer methods, one which extends the systematic search and one which uses an alternative approach to generating the initial structure prior to energy minimisation. 

Taking into account all the subtleties associated with the steady-state NOE presented above, it should be clear that it is unwise to place too much significance on the absolute magnitudes of steady-state NOE enhancements. In reality, differences of a few percent mean little when taken on their own, and it is generally necessary to consider a collection of enhancements when undertaking structural or conformational analysis to be certain of an unambiguous conclusion. A qualitative interpretation of many measiu-ements is the most appropriate approach to interpreting steady-state NOE data. 

---