Whole molecule methods

In the past year, several groups have described methods that use genetic algorithms to flexibly dock ligands in a receptor active site. These methods, which tend to be much slower than those that use rigid structures, do an excellent job of locating low energy bound conformations. 

The connection/disconnection methods contain some features ot the sequential buildup procedures, as well as clever ways for altering bond connectivity of the ligand(s) during construction. Several of these techniques also sample the allowed conformational space of the fragments from which ligands are constructed. 

This category includes genetic algorithm methods,CONCEPTS, CONCERTS, the dynamic ligand design (OLD) approach, MCDNLG, 64 and Hahn s RECEPTOR program.  

Blaney, Weininger, and Dixon, The philosophy behind GAs and their use in various fields of computational chemistry are covered elsewhere.  

---

Whole molecule methods Fit known compounds into an active site in various orientations, assessing shape and/or electrostatic complementarity. 

Several Monte Carlo (MC) and simulated annealing (SA) approaches have been described that are well suited for determining low energy conformations of molecular fragments and in some cases entire ligands. - Interesting hybrids of MC and SA also have been explored these methods are discussed in the section on whole molecule methods. 

Whole molecule methods include DOCK and extensions to DOCK allowing ligand flexibility, " " the use of multiconformation databases, 2- molecular surface pattern matching, ellipsoid casting, " AUTODOCK, " " other Monte Carlo and/or simulated annealing methods, - and distance geometry.As already mentioned, excellent reviews of the field of docking are available. 

Alternate uses often make sense. We have repeatedly been struck with how useful some methods are when employed for functions outside the developers intentions. For example, whole molecule methods can be used to find the perfect replacement for one small portion of a large ligand. It makes sense to be creative and flexible in applying these techniques. 

In a contribution dealing with two related compound classes, space could be saved by treating them together in domains where they display close similarities. However, the only spheres where this applies to sulphones and sulphoxides are elemental sulphur determination and chromatography. The former is too unspecific to be considered for inclusion in this chapter. Chromatographic behaviour is determined by the whole molecule, but the widespread use of chromatographic methods does justify its treatment. At the risk of a very little duplication it has been deemed more suitable to provide separate accounts of the two compound classes. 

The volume, electron population, and charge of the whole molecule can also be obtained by the same method. Table 6.1 gives the atomic volumes, populations, and charges of the atoms in methanal. We see that the electron populations of the atoms and the corresponding charges are additive to four decimal places. 

Using the molecular orbital method, Coulson showed how certain electrons in benzene, namely, the p electrons, can move over the whole molecule instead of being restricted to the region between two particular atoms. 93 Coulson, collaborating later with Longuet-Higgins and the French theoreticians Pascaline and Raymond Daudel and Alberte and Bernard Pullman, was to become a major presence in quantum chemistry. But on the whole, Coulson said, he was inclined to characterize the period from 1933 until the end of the Second World War as the "Mulliken Era. "94... 

The use of a range of oxidative enzymes in synthesis is covered in Chapter 11, whilst the very powerful technique of regio- and stereo-specific biohydroxylation of even complex molecules by femienting whole-cell methods is covered in Chapter 12. 

The molecular orbital picture of benzene proposes that the six jt electrons are no longer associated with particular bonds, but are effectively delocalized over the whole molecule, spread out via orbitals that span all six carbons. This picture allows us to appreciate the enhanced stability of an aromatic ring, and also, in due course, to understand the reactivity of aromatic systems. There is an alternative approach based on Lewis structures that is also of particular value in helping us to understand chemical behaviour. Because this method is simple and easy to apply, it is an approach we shall use frequently. This approach is based on what we term resonance structures. 

According to this method, Fyles analyzed the transport rate of alkali metal cations for a series of 21 synthetic transporters (Figure 14). The whole molecules were designed to elucidate the structure-function relationship. They are composed of three parts core, wall, and head units. The core units were derived from tartaric acids so that the wall units may be fixed to provide structural control by incorporating both the polar and nonpolar functionality (Y and Z in Figure 14). The head groups (X) are attached to provide an overall amphiphilic nature. 

Another frequently used method to derive empirical relationships between structure and property is to divide the structure into chemically logic parts such as groups of atoms (functional groups) and to assign each group a contribution to the property of the whole molecule. This approach is termed the group contribution model (GCM). Since groups cannot be measured individually, it is necessary to derive... 

For the purpose of this calculation it was necessary to know the diffusion coefficient for ligand and whole molecules of hemocyanin (Table I) and the value of the diffusion coefficient for half molecules of hemocyanin. If the level of calcium ion is reduced in pH 9.6 glycine buffer to the point where no whole molecules remain, the half molecules always are contaminated with one-twelfth size (5S) subunits, so that it is impossible to measure directly the diffusion coefficient of half molecules. For a self-associating system, the Gouy method leads to an average diffusion coefficient which is the Z-average, defined by... 

---