Of 3D QSAR Models

3D QSAR study on the structural requirements for inhibiting AChE 

FIGURE 29.3 Example of minaprine-derived AQiE inhibitors used in the study by Sippl et 

FIGURE 29.4 Comparison of the predicted position of the aminopyri-dazine 3y (dark-gray) and the X-ray structure of the AChE-decamethonium 

FIGURE 29.6 PLS coefficient maps obtained using the water probe (left side) and the methyl probe (right side). Green and cyan fields are contoured at —0.003, yellow and oange fields are contoured at -tO.003 (compound 4j is shown for comparison). 

FIGURE 29.5 Receptor-based alignment of all investigated inhibitors obtained by docking analyses. The solvent accessible surface of the binding pocket is displayed. 

---

The major hurdle to overcome in the development of 3D-QSAR models using steric, electrostatic, or lipophilic fields is related to both conformation selection and subsequent suitable overlay (alignment) of compounds. Therefore, it is of some interest to provide a conformation-ally sensitive lipophilicity descriptor that is alignment-independent. In this chapter we describe the derivation and parametrization of a new descriptor called 3D-LogP and demonstrate both its conformational sensitivity and its effectiveness in QSAR analysis. The 3D-LogP descriptor provides such a representation in the form of a rapidly computable description of the local lipophilicity at points on a user-defined molecular surface. 

Hopfinger AJ, Wang S, Tokarski JS et al. (1997) Construction of 3D-QSAR models using the 4D-QSAR analysis formalism. J Am Chem Soc 119 10509-10524. 

From the studies above, it was shown that the use of 3D-QSAR models led to the identification of binding site regions, where steric, electronic, or hydrophobic effects played a dominant role. Although cationic interactions in both SI and S4 subsites were favorable for in vitro affinity, they might be detrimental for oral bio availability. Thus, the CoMFA steric field contribution could be seen as a balance between pure steric plus hydrophobic effects. The contributions for steric, electrostatic and hydrophobic fields from the CoM-SIA studies were 25, 44, and 31%, respectively. 

Jayatilleke P, Nair A, Zauhar R, Welsh WJ. Computational studies on HIV-1 protease inhibitors Influence of calculated inhibitor-enzyme binding affinities on the statistical quality of 3D-QSAR Models. J Med Chem 2000 43 4446. 

Figure 4. Pictorial representation of 3D-QSAR models. The color code is as follows sterically favourable and unfavourable interactions, green and red regions, respectively favourable and unfavourable influence of high electron density, cyan and yellow zones respectively. To aid interpretation the template 26, idazoxan compounds 35 and 40 have been added to the electrostatic map, whereas clonidine, compounds 5, 8 and 34 are shown in the steric map. n, number of data points q and r, cross-validated and non-cross-validated correlation coefficient, respectively s, standard deviation one, optimal number of components.

Inductive Inference Module Performs generation of rules on the basis of structure and activity data based on algorithms of the logical structural approach [20] and provides tools for automated selection of biophores (pharmacophores) and interactive building of 3D QSAR models. The module performs statistical evaluation of the predictive and discriminating power of selected biophores and models. 

Stored in a table where columns are descriptors, and rows are compounds (or conformers), QSAR data sets contain separate columns for the measured target property (Y), attributed to the training set, as well as computed descriptors for (external) reference compounds on which the QSAR model is tested—the test set. Statistical procedures, e.g., multiple linear regression (MLR), projection to latent structures (PLS), or neural networks (NN) [38], are then used to establish a mathematical soft model relating the observed measurement(s) in the Y column(s) with some combination of the properties represented in the subsequent columns. PLS, NN, and AI (artificial intelligence) techniques have been explored by Green and Marshall in the context of 3D-QSAR models [39], and were shown to extract similar information. A problem that may lead to spurious (chance) correlations when using MLR techniques, the colinearity between various descriptors, or cross-correlation, is usually dealt with in PLS [40],... 

Cross-validation estimates model robustness and predictivity to avoid overfitting in QSAR [27]. In 3D-QSAR models, PLS and NN model complexity are established by testing the significance of adding a new dimension to the current QSAR, i.e., a PLS component or a hidden neuron, respectively. The optimal number of PLS components or hidden neurons is usually chosen from the analysis with the highest q2 (cross-validated r2) value, Eq. (3). The most popular cross-validation technique is leave-one-out (LOO), where each compound is left out of the model once and only once, yielding reproducible results. An extremely fast LOO method, SAMPLS [42], which evaluates the covariance matrix only, allows the end user to rapidly estimate the robustness of 3D-QSAR models. Randomly repeated cross-validation rounds using leave 20% out (L5G), or leave 50% out (L2G), are routinely used to check internal... 

Correlation of experimental and calculated activities assesses the quality of 3D-QSAR models. The squared correlation coefficient (r ) yielded by this statistics is a measure of the goodness of fit. The robustness of the model is tested via cross-validation techniques (leave-x%-out), indicating the goodness of prediction q ). Models with > 0.4—0.5 are considered to yield reasonable predictions for hypo-... 

This chapter is based on the authors 12 years of combined experience regarding quantitative structure-activity relationship (QSAR) modeling. The intent is to present a discussion of principles and caveats aimed at the occasional end user, while offering some in-depth comments for those experienced in the area of three-dimensional (3D) QSAR. More than 200 CoMFA papers have been published since the initial inclusion of comparative molecular field analysis in SYBYL in 1988. It would have been beyond the scope of this chapter to critique all these reports. Instead, we focus on providing a working knowledge on the generation, critical evaluation, and meta-analysis of 3D-QSAR models. 

HOMO fields have been shown to be beneficial for the refinement of 3D-QSAR models for data sets such as the angiotensin converting enzyme (ACE) inhibitors. HOMO fields appear to describe the interaction between the ionized ligand and the metal ion in the molecular binding domain. More recently, molecular orbital fields have been used in the construction of 3D-QSAR models for molecular reactivity end points (e.g., metabolic rate constants). ... 

The Walters group reports that it typically takes many generations, but not much computer time, to produce a population of 3D-QSAR models with a good correlation between the observed and predicted aaivity. The results can be displayed graphically to show favorable and unfavorable points of interaaion. 

---