Pharmacophoric atoms

Fig. 10. Pharmacophores for angiotension-converting enzyme. Distances in nm. (a) The stmcture of a semirigid inhibitor and distances between essential atoms from which one pharmacophore was derived (79). (b) In another pharmacophore, atom 1 is a potential zinc ligand (sulfhydryl or carboxylate oxygen), atom 2 is a neutral hydrogen bond acceptor, atom 3 is an anion (deprotonated sulfur or charged oxygen), atom 4 indicates the direction of a hydrogen bond to atom two, and atom 5 is the central atom of a carboxylate, sulfate, or phosphate of which atom 3 is an oxygen, or atom 5 is an unsaturated carbon when atom 3 is a deprotonated sulfur. The angle 1- -2- -3- -4 is —135 to —180° or 135 to 180°, and 1- -2- -3- -5 is —90 to 90°.

Step 3. Overlay all agonists in their most common confonnations using dihydrexidine as a template compound by superimposition of equivalent pharmacophoric atoms of all the agonists and those of DFDC. 

A wide variety of features have been used in bit vectors, including molecular fragments, 3-D potential pharmacophores, atom pairs, 2-D pharmacophores, topological torsions, and variety of topological indices. 

Sheridan et al.54 pharmacophore. Atom labeled D is a dummy atom along the bond angle bisector and 1.2 A from the atom to which it is attached. Three essential groups are needed the cationic center (A), an electronegative atom (B), and an atom (C) that forms a dipole with B. 

In 1996, Sheridan et al. [16] were the first to use pharmacophoric atom types for an autocorrelation approach. This technique is suited to characterize ligand-receptor interactions in a general way, allowing for more different but equally interacting molecules to be identified as similar. Sheridan et al. also extended the topological Carhart approach to the 3D case, and this was soon followed up by a binary representation of such a descriptor [17]. In 2003, Stiefl and Baumann [18] reported an autocorrelation approach using surface points representing pharmacophoric features. 

The work of Schneider et al. [6] first focused on the scaffold-hopping ability of autocorrelation descriptors, in this case topological pharmacophores. The general description of the atoms with pharmacophore atom types in combination with the decomposition of molecules into atom pairs was shown to be especially successful in finding new molecules with significant different molecular scaffolds, maintaining the desired biological effect. 

Another way to view similarity between 3-D structures is to focus on the pharmacophore atoms and the direction, or points, of their interaction with a target protein. The program FAMILY (142) assigns 3-D structures to families of compounds in which the variation in all distances between the points of interest are within a specified tolerance, usually 0.3-0.5 A. FAMILY uses the Bron-Kerbosh clique detection algorithm (143,144) to find these common 3-D substructures, and is rapid in execution since a typical test found that 384 compounds could be matched over seven points in under a minute on a VAX 9000. The points that are considered in the analysis are selected in an initial run of ALADDIN, and are typically the pharmacophore atoms and all heavy atoms that are attached to them. In a classification of dopaminergics, the atoms attached to these attached atoms were also used to increase the number of families found. In this example of compounds that met the pharmacophore requirements, it was shown that the set of computer-designed compounds (97) sorted itself into 36 families whereas compounds in a definitive review (145) sorted themselves into 15 families. 

If molecules are thought of as wire models, then one would select a geometric definition of similarity that depended on the location of the atomic nuclei. Similarity might be based on the distances between atoms of a specified type or between all atoms in the molecule. In medicinal chemistry one would typically measure the distances between the pharmacophoric atoms or their projected binding points. Such a description is rather straightforward and obvious for small molecules however, it is also a very powerful descriptor of protein structure. For example, one can easily recognize elements of secondary struaure such as helices or sheets from a shaded diagram of the distance between every Co of a protein. ... 

These points, lines, and planes may be defined by the coordinates of atoms in the molecule or calculated from them. This is under the control of the user. Thus, for example, one can specify a point to be the center of mass of a ring, or the calculated location of the proposed atom in the receptor that binds to a particular atom in the molecule, or along a line at a certain distance from specified atoms or a plane to be the least-squares plane of the whole molecule, or the plane defined by the pharmacophoric atoms. 

The power of the context-sensitive execution of GCL becomes apparent when one considers the example in Figure 7, the ACL that was used to remove the unwanted substituents from database structures in the D2 design problem. This ACL automatically AXEs and NIBBLES until no FRGs (freely rotatable groups) or TERMINALFRG s (terminal freely rotatable groups) are present. Only then are the hit atoms transformed into the target pharmacophore atoms. 

---