Partial Least Squares Modeling and Genetic Algorithm Optimization in Quantitative Structure-Activity Relationships

Abstract

Quantitative structure-activity relationship (QSAR) studies based on chemometric techniques are reviewed. Partial least squares (PLS) is introduced as a novel robust method to replace classical methods such as multiple linear regression (MLR). Advantages of PLS compared to MLR are illustrated with typical applications. Genetic algorithm (GA) is a novel optimization technique which can be used as a search engine in variable selection. A novel hybrid approach comprising GA and PLS for variable selection developed in our group (GAPLS) is described. The more advanced method for comparative molecular field analysis (CoMFA) modeling called GA-based region selection (GARGS) is described as well. Applications of GAPLS and GARGS to QSAR and 3D-QSAR problems are shown with some representative examples. GA can be hybridized with nonlinear modeling methods such as artificial neural networks (ANN) for providing useful tools in chemometric and QSAR.     

Partial least squares modeling and genetic algorithm optimization in quantitative structure-activity relationships

Quantitative structure-activity relationship (QSAR) studies based on chemometric techniques are reviewed. Partial least squares (PLS) is introduced as a novel robust method to replace classical methods such as multiple linear regression (MLR). Advantages of PLS compared to MLR are illustrated with typical applications. Genetic algorithm (GA) is a novel optimization technique which can be used as a search engine in variable selection. A novel hybrid approach comprising GA and PLS for variable selection developed in our group (GAPLS) is described. The more advanced method for comparative molecular field analysis (CoMFA) modeling called GA-based region selection (GARGS) is described as well. Applications of GAPLS and GARGS to QSAR and 3D-QSAR problems are shown with some representative examples. GA can be hybridized with nonlinear modeling methods such as artificial neural networks (ANN) for providing useful tools in chemometric and QSAR.     

Qsar Models for Predicting the Acute Toxicity of Selected Organic Chemicals with Diverse Structures to Aquatic Non-Vertebrates and Humans

Abstract

The linear and non-linear relationships of acute toxicity (as determined on five aquatic non-vertebrates and humans) to molecular structure have been investigated on 38 structurally-diverse chemicals. The compounds selected are the organic chemicals from the 50 priority chemicals prescribed by the Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) programme. The models used for the evaluations are the best combination of physico-chemical properties that could be obtained so far for each organism, using the partial least squares projection to latent structures (PLS) regression method and backpropagated neural networks (BPN). Non-linear models, whether derived from PLS regression or backpropagated neural networks, appear to be better than linear models for describing the relationship between acute toxicity and molecular structure. BPN models, in turn, outperform non-linear models obtained from PLS regression. The predictive power of BPN models for the crustacean test species are better than the model for humans (based on human lethal concentration). The physico-chemical properties found to be important to predict both human acute toxicity and the toxicity to aquatic non-vertebrates are the n?octanol water partition coefficient (Pow) and heat of formation (HF). Aside from the two former properties, the contribution of parameters that reflect size and electronic properties of the molecule to the model is also high, but the type of physico-chemical properties differs from one model to another. In all of the best BPN models, some of the principal component analysis (PCA) scores of the ¹³C-NMR spectrum, with electron withdrawing/accepting capacity (LUMO, HOMO and IP) are molecular size/volume (VDW or MS1) parameters are relevant. The chemical deviating from the QSAR models include non-pesticides as well as some of the pesticides tested. The latter type of chemical fits in a number of the QSAR models. Outliers for one species may be different from those of other test organisms.     

3-D QSAR studies on histone deacetylase inhibitors. A GOLPE/GRID approach on different series of compounds

Docking simulation and three-dimensional quantitative structure-activity relationships (3D-QSARs) analyses were conducted on four series of HDAC inhibitors. The studies were performed using the GRID/GOLPE combination using structure-based alignment. Twelve 3-D QSAR models were derived and discussed. Compared to previous studies on similar inhibitors, the present 3-D QSAR investigation proved to be of higher statistical value, displaying for the best global model r2, q2, and cross-validated SDEP values of 0.94, 0.83, and 0.41, respectively. A comparison of the 3-D QSAR maps with the structural features of the binding site showed good correlation. The results of 3D-QSAR and docking studies validated each other and provided insight into the structural requirements for anti-HDAC activity. To our knowledge this is the first 3-D QSAR application on a broad molecular diversity training set of HDACIs.     

The Non-Grid Technique for Modeling 3D QSAR Using Self-Organizing Neural Network (SOM) and PLS Analysis: Application to Steroids and Colchicinoids

Abstract

A novel method for modeling 3D QSAR has been developed. The method involves a multiple training of a series of self-organizing networks (SOM). The obtained networks have been used for processing the data of one reference molecule. A scheme for the analysis of such data with the PLS analysis has been proposed and tested using the steroids data with corticosteroid binding globulin (CBG) affinity. The predictivity of the CBG models measured with the SDEP parameter is among the best one reported. Although 3-D QSAR models for colchicinoid series is far less predictive, it allows for a discussion on the relative influence of the structural motifs of these compounds.     

QSAR study and VolSurf characterization of anti-HIV quinolone library

Antiviral quinolones are promising compounds in the search for new therapeutically effective agents for the treatment of AIDS. To rationalize the SAR for this new interesting class of anti-HIV derivatives, we performed a 3D-QSAR study on a library of 101 6-fluoro and 6-desfluoroquinolones, taken either from the literature or synthesized by us. The chemometric procedure involved a fully semiempirical minimization of the molecular structures by the AMSOL program, which takes into account the solvatation effect, and their 3D characterization by the VolSurf/GRID program. The QSAR analysis, based on PCA and PLS methods, shows the key structural features responsible for the antiviral activity.     

Insight into crucial inhibitor–enzyme interaction of arylamides as novel direct inhibitors of the enoyl ACP reductase (InhA) from Mycobacterium tuberculosis: computer-aided molecular design

Abstract  

The enoyl ACP reductase enzyme (InhA) involved in the type II fatty acid biosynthesis pathway of Mycobacterium tuberculosis is an attractive target enzyme for antitubercular drug development. Arylamide derivatives are a novel class of InhA inhibitors used to overcome the drug-resistance problem of isoniazid, the frontline drug for tuberculosis treatment. Their remarkable property of inhibiting the InhA enzyme directly without requiring any coenzyme, makes them especially appropriate for the design of new antibacterials. In order to find a sound binding conformation for the different arylamide analogs, molecular docking experiments were performed with subsequent QSAR investigations. The X-ray conformation of one arylamide within its cocrystallized complex with InhA was used as a starting conformation for the docking experiments. The results thus obtained are perfectly consistent (rmsd = 0.73 ?) with the results from X-ray analysis. A thorough investigation of the arylamide binding modes with InhA provided ample information about structural requirements for appropriate inhibitor–enzyme interactions. Three different QSAR models were established using two three-dimensional (CoMFA and CoMSIA) and one two-dimensional (HQSAR) techniques. With statistically ensured models, the QSAR results obtained had high correlation coefficients between molecular structure properties of 28 arylamide derivatives and their biological activity. Molecular fragment contributions to the biological activity of arylamides could be obtained from the HQSAR model. Finally, a graphic interpretation designed in different contour maps provided coincident information about the ligand–receptor interaction thus offering guidelines for syntheses of novel analogs with enhanced biological activity.     

Modeling robust QSAR. 2. iterative variable elimination schemes for CoMSA: application for modeling benzoic acid pKa values

A variety of issues decide the efficiency of 3D QSAR methods, and their practical importance for drug design is still controversial. This refers both to the predictive ability and the possibility for the indication of these areas within 3D molecular representations that are responsible for biological or chemical effects. Technically, the latter comes down to the selection or elimination of the reliable variables during 3D QSAR modeling using the Partial Least-Squares (PLS) method. In this paper we used a series of benzoic acids to test the dependence between the predictive ability and variable selection performance of PLS with Iterative Variable Elimination (IVE-PLS) in the Comparative Molecular Surface Analysis (CoMSA) modeling of Hammett constant which correlates with the pKa values. Modeling this chemical effect allowed us to select the IVE-PLS variant that plots the contour maps indicating a carboxylic function, i.e., the region including the dissociation reaction center that determines the respective pKa values. In fact, it appeared that a novel robust IVE version is capable of the indication of the proper contour plots independent of the method used for the calculation of partial atomic charges (AM1 or Gasteiger-Marsili).     

Experimental design based 3-D QSAR analysis of steroid-protein interactions: Application to human CBG complexes

Summary An experimental design based 3-D QSAR analysis using a combination of principal component and PLS analysis is presented and applied to human corticosteroid-binding globulin complexes. The predictive capability of the created model is good. The technique can also be used as guidance when selecting new compounds to be investigated.