Computable molecular descriptors physicochemical properties

In both cases, a structural representation of a small molecule is the input parameter to a conceptual set of operations that give rise to numerical outputs such as molecular descriptors, physicochemical properties, or biological outcomes (Fig. 13.1-l(a)). However, to be useful in predictive ways, such as when used to support prospective decisions about the investment of synthetic chemistry resources, at least some of these numerical outputs must be computable given only a structure representation. Only this situation allows relationships between experimentally determined values and computed values to be used to predict experimental outcomes for new molecules, based on their structural similarity to molecules that have already been experimentally tested (Fig. 13.1-l(b)). Most broadly, chemical space is a colloquialism that refers to the ranges and distributions of computed or measured outputs based on chemical structure inputs, and serves as a mathematical framework for quantitative comparisons of similarities and differences between small molecules (Fig. 13.1-l(c)). 

An extensive series of studies for the prediction of aqueous solubility has been reported in the literature, as summarized by Lipinski et al. [15] and jorgensen and Duffy [16]. These methods can be categorized into three types 1 correlation of solubility with experimentally determined physicochemical properties such as melting point and molecular volume 2) estimation of solubility by group contribution methods and 3) correlation of solubility with descriptors derived from the molecular structure by computational methods. The third approach has been proven to be particularly successful for the prediction of solubility because it does not need experimental descriptors and can therefore be applied to collections of virtual compounds also. 

Molecular descriptors vary gready in both their origins and their applications. They come from both experimental measurements and theoretical computations. Typical molecular descriptors from experimental measurements include logP, aqueous solubility, molar refractivity, dipole moment, polarizability, Hammett substituent constants, and other empirical physicochemical properties. Notice that the majority of experimental descriptors are for entire molecules and come directly from experimental measurements. A few of them, such as various substituent constants, are for molecular fragments attached to certain molecular templates and they are derived from experimental results. 

Molecular Connectivity. A class of molecular descriptors derived from the connection table of a structure. For increasing path lengths (1-, 2-, 3-bonds, etc.), the molecular connectivity values are computed as the sum of functions of the connectivity values (number of attachments) of the atoms in the path. Molecular connectivity descriptors can be used to distinguish structures. As such, they can be correlated with physicochemical properties that are functions of structure size, linearity, and degree of branching. 

Theoretical (computational) calculations can also offer quantitative descriptors of physicochemical properties of the molecular structures, molecular interactions, and thermodynamics of interactions. Principally, extensive studies on the catalytic site of GP have been exploited in theoretical QSAR studies [4]. The techniques engaged correlate biochemical behaviors with the known crystallographic structures, and map regions around the inhibitor molecule and added water molecules to improve the in silico prediction [106-110]. 

Only a few compounds screened in early lead identification phases are synthesized in-house. More flexible and cost effective is to purchase chemicals from external suppliers. Most vendors provide lists of some ten to himdred thousand chemicals on compact discs and guarantee delivery within days to weeks. To explore this huge amount of data with the aid of computers, chemical information is transformed to computer-readable strings, e.g., smiles code, and different descriptors are determined. 1-dimensional (1-D) descriptors encode chemical composition and physicochemical properties, e.g., molecular weight, stoichiometry (C O Hj,), hydrophobicity, etc. 2-D descriptors reflect chemical topology, e.g., connectivity indices, degree of branching, number of aromatic bonds, etc. 3-D descriptors consider 3-D shape, volume or surface area. 

One can describe a molecule in many ways and the same applies to bioisosteres. Molecular descriptor methods are covered in the third part by the application of different representations. A number of computational approaches to bioisosteric replacement are covered in chapters on physicochemical properties, molecular topology, molecular shape, and the use of protein structure information. Each chapter covers many common methods and overviews of when best to apply these methods, and where they have been successfully applied. 

The compound s stmcture is entered using, typically, computer graphics. Each stmcture is optimized by molecular mechanics, which is followed by molecular orbital calculations. Topological indices, electronic parameters, physicochemical properties, indicator variables, and so forth are used as molecular stmcture descriptors. 

We believe that much more efficient descriptors can be discovered and this is a focus of our work. Until recently QSAR analyses have used relatively simple molecular descriptors based on substituent constants (e.g., Hammett constants, n, or molar refractivities), physicochemical properties (e.g., partition coefficients), or topological indices (e.g., Randic, Weiner, or Kier and Hall indices). Recently we and others have developed several new information rich, computationally efficient representations. The most interesting of these are the molecular eigenvalue indices. 

---