Molecular descriptors, used relationships

The octanol/water coefficient (log P) is the standard molecular descriptor used to provide the chemical property of the hydrophobicity of a molecule. Compounds with high partition coefficients usually have very low aqueous solubility. This will decrease the chance of attack by hydroxyl radicals and lead to a lower rate constant. Nevertheless, the linear relationships with log P do not reproduce similar trends of several chemical classes such as alkane and phenol. They could be either positive or negative linear relationships. Alkene (R2 < 0.67), benzene (R2 < 0.78), carboxylic acid (R2 < 0.74), and halide (R2 < 0.55) classes do not provide significant correlations. 

The first set of aliphatic compounds included methane, acetic acid, 2-butanone, and acetamide, which were analyzed using EHOMO as the molecular descriptor. The relationship between the kinetic rates of these compounds and EHomo reflected a very good correlation (r2 = 0.96). Figure 10.14 shows the correlation of the set of aliphatic compounds. From this figure, it can be seen that the kinetic rate decreases as the EHOMO increases. Application of the F test showed that the level of significance was 2.5%. That is, there is a 2.5% chance of erroneously concluding that they are not related. Based on Table 10.12, the calculated F(12) was 40.5, which is larger than the F(u)a0 025... 

Dimov, N., Osman, A., Mekenyan, O. and Papazova, D. (1994). Selection of Molecular Descriptors Used in Quantitative Structure Gas Chromatographic Retention Relationships. 1. Application to Alkylbenzenes and Naphthalenes. Anal.Chim.Acta, 298,303-317. 

DT Does not make any assumption of the type of relationship between target property and molecular descriptors Models are easy to interpret Fast classification speed Multi-class classification May have over fitting when training set is small and number of molecular descriptors is large Ranks molecular descriptors using information gain which may not be the best for some problems... 

Quantitative structure-activity relationships QSAR. The QSAR approach pioneered by Hansch and co-workers relates biological data of congeneric structures to physical properties such as hydrophobicity, electronic, and steric effects using linear regression techniques to estimate the relative importance of each of those effects contributing to the biological effect. The molecular descriptors used can be 1-D or 3-D (3D-QSAR). A statistically sound QSAR regression equation can be used for lead optimization. 

Figure 1. Mechanism-based QSAR model and its relationship to chemical and biological data of the molecules used in its derivation. A quantitative prediction of the toxicity of an untested chemical requires the availability of a QSAR model for related compounds acting by a putative common molecular mechanism and the ability to measure or predict the required mechanism-based molecular descriptors used in the QSAR model.

Chemical structures can be described by binary molecular descriptors (used as the Y-matrix in multivariate data analysis). In the case of yes/no-classifications a single binary y-variable can be used to indicate whether a particular structural property is present or not. The type of molecular descriptors (small or large fragments, atom-centered fragments, functional groups or classes of compounds) is essential to obtain a close relationship between structures and spectra. 

Bagchi, M. C., Mills, D., Basak, S. C. Quantitative structure-activity relationship (QSAR) studies of quinolone antibacterials against M. fortuitum and M. smegmatis using theoretical molecular descriptors. 

Molecules with a large molecular weight or size are confined to the transcellular route and its requirements related to the hydrophobicity of the molecule. The transcellular pathway has been evaluated for many years and is thought to be the main route of absorption of many drugs, both with respect to carrier-mediated transport and passive diffusion. The most well-known requirement for the passive part of this route is hydrophobicity, and a relationship between permeability coefficients across cell monolayers such as the Caco-2 versus log P and log D 7.4 or 6.5 have been established [102, 117]. However, this relationship appears to be nonlinear and reaches a plateau at around log P of 2, while higher lipophilicities result in reduced permeability [102, 117, 118]. Because of this, much more attention has recently been paid towards molecular descriptors other than lipophilicity [86, 119-125] (see section 5.5.6.). The relative contribution between the para-cellular and transcellular components has also been evaluated using Caco-2 cells, and for a variety of compounds with different charges [110, 112] and sizes [112] (see Section 5.4.5). 

The molecular descriptors refer to the molecular size and shape, to the size and shape of hydrophilic and hydrophobic regions, and to the balance between them. Hydrogen bonding, amphiphilic moments, critical packing parameters are other useful descriptors. The VolSurf descriptors have been presented and explained in detail elsewhere [8]. The VolSurf descriptors encode physico-chemical properties and, therefore, allow both for a design in the physico-chemical property space in order to rationally modulate pharmacokinetic properties, and for establishing quantitative structure-property relationships (QSPR). 

Because of the large number of chemicals of actual and potential concern, the difficulties and cost of experimental determinations, and scientific interest in elucidating the fundamental molecular determinants of physical-chemical properties, considerable effort has been devoted to generating quantitative structure-property relationships (QSPRs). This concept of structure-property relationships or structure-activity relationships (QSARs) is based on observations of linear free-energy relationships, and usually takes the form of a plot or regression of the property of interest as a function of an appropriate molecular descriptor which can be calculated using only a knowledge of molecular structure or a readily accessible molecular property. 

The usual approach is to compile data for the property in question for a series of structurally similar molecules and plot the logarithm of this property versus molecular descriptors, on a trial-and-error basis seeking the descriptor which best characterizes the variation in the property. It may be appropriate to use a training set to obtain a relationship and test this relationship on another set. Generally a set of at least ten data points is necessary before a reliable QSPR can be developed. 

Pollutants with high VP tend to concentrate more in the vapor phase as compared to soil or water. Therefore, VP is a key physicochemical property essential for the assessment of chemical distribution in the environment. This property is also used in the design of various chemical engineering processes [49]. Additionally, VP can be used for the estimation of other important physicochemical properties. For example, one can calculate Henry s law constant, soil sorption coefficient, and partition coefficient from VP and aqueous solubility. We were therefore interested to model this important physicochemical property using quantitative structure-property relationships (QSPRs) based on calculated molecular descriptors [27]. 

Basak, S. C., Natarajan, R., Mills, D. Structure-activity relationships for mosquito repellent aminoamides using the hierarchical QSAR method based on calculated molecular descriptors, 2005, pp. 958-963. 

This example belongs to the area quantitative structure-property relationships (QSPR) in which chemical-physical properties of chemical compounds are modeled by chemical structure data—mostly built by multivariate calibration methods as described in this chapter und using molecular descriptors (Todeschini and Consonni... 

Arrays of biological data can form the basis for uniquely informative molecular descriptors. By defining the relationships between compounds using biological descriptors (in vitro profiles) in addition to chemical structures, medicinal chemists are given new perspectives to support lead optimization. 

---