Hansch validation

The applicability of Eq. (45) to a broad range of biological (i.e., toxic, geno-toxic) structure-activity relationships has been demonstrated convincingly by Hansch and associates and many others in the years since 1964 [60-62, 80, 120-122, 160, 161, 195, 204-208, 281-285, 289, 296-298]. The success of this model led to its generalization to include additional parameters in attempts to minimize residual variance in such correlations, a wide variety of physicochemical parameters and properties, structural and topological features, molecular orbital indices, and for constant but for theoretically unaccountable features, indicator or dummy variables (1 or 0) have been employed. A widespread use of Eq. (45) has provided an important stimulus for the review and extension of established scales of substituent effects, and even for the development of new ones. It should be cautioned here, however, that the general validity or indeed the need for these latter scales has not been established. 

Calculated descriptors have generally fallen into two broad categories those that seek to model an experimentally determined or physical descriptor (such as ClogP or CpKJ and those that are purely mathematical [such as the Kier and Hall connectivity indices (4)]. Not surprisingly, the latter category has been heavily populated over the years, so much so that QSAR/QSPR practitioners have had to rely on model validation procedures (such as leave-k-out cross-validation) to avoid models built upon chance correlation. Of course, such procedures are far less critical when very few descriptors are used (such as with the Hansch, Leo, and Abraham descriptors) it can even be argued that they are unnecessary. 

Be able to critically evaluate the validity of a Hansch equation... 

Hansch equations may be used to predict the activity of an as yet unsynthesized analogue. This enables the medicinal chemist to make an informed choice as to which analogues are worth synthesizing. However, these predictions should only be regarded as valid if they are made within the range of parameter values used to establish the Hansch equation. Furthermore, when the predicted activity is widely different from the observed value, it indicates that the activity is affected by factors, such as the ease of metabolism, that were not included in the derivation of the Hansch equation. 

Although there is a strong negative correlation between partition coefficient and aqueous solubility (Hansch et al., 1968 Chiou et al., 1977), and a strong positive correlation between % and molecular volume (Dearden et al., 1988), the use of the partial least squares (PLS) method in this study allows the simultaneous use of intercorrelated descriptors. Nevertheless, the use of four descriptors to model the bioconcentration factor of only 11 compounds contravenes the Topliss and Costello (1972) rule, and renders the QSAR of dubious validity. 

Figure 1 illustrates a common pitfall when log P is not measured. Apart from the fact that tt and log P are not additive, one may be tempted to take tr values from Hansch s work on phenoxyacetic acids, for example, and apply these without checking by log P measurement their validity for the series being studied. When tt is taken from one series and applied... 

Kim, K.H. (1995b). Comparison of Classical QSAR and Comparative Molecular Field Analysis Toward Lateral Validations. In Classical and Three-Dimensional QSAR in Agrochemistry (Hansch, C. and Fujita, T, eds.), American Chemical Society, Washington (DC), pp. 302-317. 

Hansch QSAR and related approaches belong to the world of numbers conceptually, knowledge-based expert system approaches do not. Three areas of debate about expert systems have arisen from the distinction. Can you devise ways to generate qualified output from computer-based expert systems without hiding quantitative methods inside them Assuming you can, how do you validate an expert system How can you usefully combine output from different systems—some quantitative and some qualitative—to make predictions more reliable The first of the questions is more a historical than a current one some of the systems described earlier in this chapter demonstrate... 

Chiu, T.-L. and So, S.-S. (2004) Development of neiual network QSPR models for Hansch substituent constants. 1. Method and validation. J. Chem. Inf. Comput. Sci., 44, 147-153. 

Kim KH. Comparison of classical QSAR and comparative molecular field analysis Toward lateral validations. In Hansch C, Fujita T, eds. Classical and Three-Dimensional QSAR In Agro-Chemistry, ACS Symposium series. Vol. 606. Washington, DC American Chemical Society, 1995 302-317. 

Correspondingly, eq. 64 was criticized by Unger and Hansch [307] in a noteworthy paper, which constitutes a milestone in the development of Hansch analysis. They formulated rules for the derivation of extrathermodynamic equations which are summarized here because of their general validity (supplementary comments are given in parentheses) ... 

With the evidence on hand it is impossible to differentiate between both equations on a rational basis and to prefer either eq. 65 or 68. The equations are derived from a less well designed group of compounds with the exception of hydrogen and the methyl group, all other substituents are halogens. A validation of all these equations can only be achieved after synthesis and testing of additional compounds with larger we/n-substituents, hydrophilic, and/or electron donor substituents (the importance of a proper selection of substituents in Hansch analysis is discussed in chapter 6). 

Choplin, F (1990) Computers and the medicinal chemist. In Hansch C, Sammes PG, Taylor JB (eds) Comprehensive Medicinal Chemistry Pergamon Press, UK 4 33-58 Tropsha A, Gramatica P, Gombar V (2003) The importance of being earnest validation is the absolute essential for successful application and interpretation of QSPR models. Mol Inform... 

The most appropriate and widely used method for extracting information from large data sets is QSAR and its relatives, quantitative structure-property relationships (QSPR) for property modeling, and quantitative structure-toxicity relationships (QSTR) for toxicity modeling. QSAR is a simple, well validated, computationally efficient method of modeling first developed by Hansch and Fujita several decades ago (30). QSAR has proven to be very effective for discovery and optimization of drug leads as well as prediction of physical properties, toxicity, and several other important parameters. QSAR is capable of accounting for some transport and metabolic (ADMET) processes and is suitable for analysis of in vivo data. 

The concept of lateral validation was first formulated by Hansch for classical QSARs. In this approach, the choice of parameters, their sign, and the size of their coefficients are compared with those from other QSARs. A comparison is illustrated in Table 5 for the Hammett equation ... 

Hansch recommended a lateral validation of QSAR results, i.e., the comparison of models of closely related series of compounds in one biological test system (cf. equations 16 and 17) or trie comparison of the QSAR models derived for one series of compounds in several related biological test models (e.g., serine and cysteine proteases). If all models are of comparable quality, and if they show similar regression coefficients of the physicochemical terms, the results can be accepted. However, in most cases the required effort will be too large to do this routinely. In addition, even closely related enzymes or receptors may have significantly different binding sites. 

Ob-6 AvatyCO 1) In the QSAR-context, it is necessary and sufficient that p verifies conditions C2), i.e., to be a dissimilarity index. In this sense MSD was used as a steric parameter (instead of the Taft Eg constant, for example) in Hansch type structure-biological activity correlations. 2) These results obtained for MSD are valid... 

---