Proposed Quantitative Structure-activity Relationship Model Research Articles

Machine learning-driven QSAR models for predicting the cytotoxicity of five common microplastics

In the field of microplastics (MPs) toxicity prediction, machine learning (ML) computer simulation techniques are showing great potential. In this study, six ML algorithms were utilized to predict the toxicity of MPs on BEAS-2B cells based on quantitative structure-activity relationship (QSAR) models. Comparing the models of different algorithms, the extreme gradient boosting model showed the best fit and prediction performance (R2tra = 0.9876, R2test = 0.9286). Additionally, Williams plot analysis showed that the six models developed were able to predict stably within their applicability domain, with few outliers. Finally, the three feature importance methods—Embedded Feature Importance (EFI), Recursive Feature Elimination (RFE), and SHapley Additive exPlanations (SHAP)—consistently identified particle size as the most critical feature affecting toxicity prediction. The proposed QSAR model can be utilized for preliminary environmental exposure assessments of MPs and to better understand the associated health risks.

Application of QSAR models for acute toxicity of tetrazole compounds administrated orally and intraperitoneally in rat and mouse

Tetrazoles and their derivatives possess various biological activities, such as antibacterial, anti-fungal, and other activities. However, these compounds may induce specific cumulative and toxic effects in living organisms. Therefore, quantitative structure-activity relationship (QSAR) models were constructed to study the acute oral toxicity of tetrazoles in rats and mice. The toxicity data of 111 tetrazole compounds were collected using the ChemIDplus, ChEMBL and ECHA databases as response variables, while the PaDEL-descriptor generated the 2D descriptors as independent variables. The models were developed and validated following the OECD guidelines by the DTC-QSAR tool. Three QSAR models were successfully established for the oral routes of rat and mouse and the intraperitoneal route of mouse, respectively. The scatter plots showed high consistency between the training and test data sets. All the models successfully met the external and internal validation criteria. Most of the descriptors kept in the final models exhibited positive correlations with toxicity, whereas only 6 descriptors exhibited negative associations. Several chemicals were identified as response or structural outliers, based on the standardized residuals and leverage values. In conclusion, the findings of this investigation demonstrate that the proposed QSAR models hold promise in forecasting the acute toxicity of recently developed or synthesized tetrazole compounds, thereby mitigating potential risks to human health and the environment.

Novel and Predictive QSAR Model and Molecular Docking: New Natural Sulfonamides of Potential Concern against SARS-Cov-2

Background: Since the outbreak of the COVID-19 pandemic in 2019, the world has been racing to develop effective drugs for treating this deadly disease. Although there are now some vac-cines that have somewhat alleviated global panic, the lack of approved drugs remains a persistent challenge. Consequently, there is a pressing need to discover new therapeutic molecules. Methods: In this study, we explore the application of a quantitative structure−activity relationship (QSAR) model to predict the efficacy of 28 cyclic sulfonamide derivatives against SARS-CoV-2. The model was developed using multiple linear regression, and six molecular descriptors were identified as the most significant factors in determining the inhibitory activity. This proposed QSAR model holds the potential for aiding the virtual screening and drug design process in the development of new and more effective SARS-CoV-2 inhibitors. The model was also applied to seven natural products primary sulfonamides and sulfamates, demonstrating promising activity Results: The study results indicated that the atom count, as represented by the descriptor nCl, had the most significant impact on the inhibitory activity against SARS-CoV-2. The proposed model was val-idated using various statistical parameters, confirming its validity, robustness, and predictiveness, with a high correlation coefficient (R2) of 0.77 for the training group and 0.95 for the test group. Further-more, we predicted the activity of seven natural compounds, and among them, Dealanylascamycin exhibited the highest predicted activity. Subsequently, Dealanylascamycin was docked to SARS-CoV-2 and the results of the docking study further strengthened its potential as a promising candidate against COVID-19, suggesting that it should be considered for further optimization and validation. Conclusion: Our findings demonstrate promising predicted inhibitory activity against SARS-CoV-2 for seven natural products, primary sulfonamides, and primary sulfamates.

Prediction of molecular interactions and physicochemical properties relevant for vasopressin V2 receptor antagonism.

We have developed two ligand- and receptor-based computational approaches to study the physicochemical properties relevant to the biological activity of vasopressin V2 receptor (V2R) antagonist and eventually to predict the expected binding mode to V2R. The obtained quantitative structure activity relationship (QSAR) model showed a correlation of the antagonist activity with the hydration energy (EH2O), the polarizability (P), and the calculated partial charge on atom N7 (q6) of the common substructure. The first two descriptors showed a positive contribution to antagonist activity, while the third one had a negative contribution. V2R was modeled and further relaxed on a 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocoline (POPC) membrane by molecular dynamics simulations. The receptor antagonist complexes were guessed by molecular docking, and the stability of the most relevant structures was also evaluated by molecular dynamics simulations. As a result, amino acid residues Q96, W99, F105, K116, F178, A194, F307, and M311 were identified with the probably most relevant antagonist-receptor interactions on the studied complexes. The proposed QSAR model could explain the molecular properties relevant to the antagonist activity. The contributions to the antagonist-receptor interaction appeared also in agreement with the binding mode of the complexes obtained by molecular docking and molecular dynamics. These models will be used in further studies to look for new V2R potential antagonist molecules.

Performance of PLS, GA-PLS and OSC-GA-PLS Models in QSAR Study of Pyrazolone Derivatives Inhibitory Activity, and Image Processing for New Compounds Designing

Nowadays, Pyrazolone and its derivatives have gained a lot of attention due to their biological and medicinal applications. These compounds have antimicrobial, antifungal and anticancer properties. Therefore, using simple methods to prepare these compounds is important. Pyrazolone is one of the inhibitors of kinase domain containing receptor KDR or VEGFR-2. In this study, Quantitative Structure-Activity Relationship (QSAR) analysis was used to predict the inhibitory activity of new pyrazolone derivatives. Also, Bi-dimensional images were used to calculate pixels for QSAR modeling. Furthermore, the partial least squares (PLS) was used to establish a relationship between IC50-dependent variables and independent variables, i.e., pixels or hidden variables. In addition, Genetic Algorithm (GA) was used in PLS method (GA-PLS) to select the descriptors. In this method, the variables which selected to form the calibration model had negligible errors with acceptable characteristics. Pre-processing methods such as Orthogonal Signal Correction (OSC) were used to provide a suitable input for modeling as well as to improve the results of the GA (OSC-GA-PLS). Furthermore, Root Mean Squared Error of Prediction (RMSEP) was used to assess the performance of the models to predict the pIC50 of the studied compounds, the value of which was obtained equal to 0.30, 0.22, and 0.19 for PLS, GA-PLS and OSC-GA-PLS models, respectively. Finally, the proposed QSAR model was developed with the OSC-GA-PLS method to predict the inhibitory activity of the new compounds.

Ensemble hologram quantitative structure activity relationship model of the chromatographic retention index of aldehydes and ketones

色谱保留指数(retention index, RI)是色谱分析中的重要参数,不同化合物在不同极性固定相上具有不同的保留行为。醛酮化合物种类众多,实验测定其RI值的时间和经济成本高。该论文采用集成建模(ensemble modeling)结合全息定量构效关系(HQSAR)方法研究了醛酮化合物在2种固定相(DB-210和HP-Innowax)上色谱保留指数的定量构效关系(QSAR)模型。用外部测试集验证法和留一交叉验证法评估了所建立模型的预测能力。首先建立了34种被研究化合物的个体HQSAR模型。在固定相DB-210上,片段特性(FD)为“供体/受体原子(DA)”且片段尺寸(FS)为1~9时可得到最优个体模型,在固定相HP-Innowax上,FD为“DA”且FS为4~7时可得到最优个体模型,这两个模型的交叉验证相关系数( )分别为0.935和0.909,外部验证相关系数( )分别为0.925和0.927,一致性相关系数(CCC)分别为0.953和0.960,预测平方相关系数F2( )分别为0.922和0.918,预测平方相关系数F3( )分别为0.931和0.927。研究结果表明醛酮化合物的分子结构与RI值之间存在定量关系,用HQSAR方法可以建立二者之间的QSAR模型。其次,以4个预测准确度最高的个体HQSAR模型作为子模型通过算术平均建立了集成HQSAR模型。建立的集成HQSAR模型预测被研究化合物在DB-210和HP-Innowax固定相上RI值的 分别为0.927和0.919, 分别为0.929和0.963, CCC分别为0.956和0.979, 分别为0.927和0.958, 分别为0.935和0.963。与个体HQSAR模型相比,建立的集成HQSAR模型预测准确度更高。这说明集成建模是提高HQSAR模型预测能力的有效方法,HQSAR与集成建模方法相结合可以用于研究和预测醛酮化合物的RI值。

P.296 Interhemispheric imbalance of motor cortex excitability in affective disorders – a systematic review and meta-analysis

Tuberculosis (TB) is of one the most infectious disease caused by Mycobacterium tuberculosis which remains a serious public health problem. Emergence of multi-drug resistant strains of M. tuberculosis led to development of new and more potent anti-tuberculosis agents. The aim of this study was to correlate the chemical structures of the inhibitory compounds with their experimental activities. In this study, analogs of 2,4-disubstituted quinoline derivatives as potent anti-tubercular agents was subjected to quantitative structure–activity relationship (QSAR) analysis in order to build a QSAR model for predicting the activities of these compounds. In order to build the regression model, Genetic Function Approximation (GFA) and Multi-linear Regression approach were used to predict the activities of inhibitory compounds. Based on the prediction, the best validation model was found to have squared correlation coefficient (R2) of 0.9367, adjusted squared correlation coefficient (R2 adj) value of 0.9223 and cross validation coefficient (Qcv2) value of 0.8752. The chosen model was subjected to external validations and the model was found to have (R2 test) of 0.8215 and Y-randomization Coefficient (cRp2) of 0.6633. The proposed QSAR model provides a valuable approach for designing more potent anti-tubercular agents.

Molecular Docking and QSAR Studies of Coumarin Derivatives as NMT Inhibitors: Simple Structural Features as Potential Modulators of Antifungal Activity

Background:Treatments of fungal diseases, including Candidiasis, remain not up to scratch in spite of the mounting catalog of synthetic antifungal agents. These have served as the impetus for investigating new antifungal agents based on natural products. Consequently, genetic algorithm-multiple linear regression (GA-MLR) based QSAR (Quantitative Structure-Activity Relationship) studies of coumarin analogues along with molecular docking were carried out.Methods:Coumarin analogues with their MIC values were used to generate the training and test sets of compounds for QSAR models development; the analogues were also docked into the binding pocket of NMT (MyristoylCoA: protein N-myristoyltransferase).Results and Discussion:The statistical parameters for internal and external validation of QSAR analysis (R2= 0.830, Q2= 0.758, R2Pred= 0.610 and R2m overall= 0.683 ), Y Randomization, Ridge trace, VIF, tolerance and model criteria of Golbraikh and Tropsha data illustrate the robustness of the best proposed QSAR model. Most of the analogues bind to the electrostatic, hydrophobic clamp and display hydrogen bonding with amino acid residues of NMT. Interestingly, the most active coumarin analogue (MolDock score of -189.257) was docked deeply within the binding pocket of NMT, thereby displaying hydrogen bonding with Tyr107, Leu451, Leu450, Gln226, Cys393 and Leu394 amino acid residues.Conclusion:The combinations of descriptors from various descriptor subsets in QSAR analysis have highlighted the role of atomic properties such as polarizability and atomic van der Waals volume to explain the inhibitory activity. The models and related information may pave the way for important insight into the designing of putative NMT inhibitors for Candida albicans.

QSAR model for predicting neuraminidase inhibitors of influenza A viruses (H1N1) based on adaptive grasshopper optimization algorithm

ABSTRACT High-dimensionality is one of the major problems which affect the quality of the quantitative structure–activity relationship (QSAR) modelling. Obtaining a reliable QSAR model with few descriptors is an essential procedure in chemometrics. The binary grasshopper optimization algorithm (BGOA) is a new meta-heuristic optimization algorithm, which has been used successfully to perform feature selection. In this paper, four new transfer functions were adapted to improve the exploration and exploitation capability of the BGOA in QSAR modelling of influenza A viruses (H1N1). The QSAR model with these new quadratic transfer functions was internally and externally validated based on MSEtrain, Y-randomization test, MSEtest, and the applicability domain (AD). The validation results indicate that the model is robust and not due to chance correlation. In addition, the results indicate that the descriptor selection and prediction performance of the QSAR model for training dataset outperform the other S-shaped and V-shaped transfer functions. QSAR model using quadratic transfer function shows the lowest MSEtrain. For the test dataset, proposed QSAR model shows lower value of MSEtest compared with the other methods, indicating its higher predictive ability. In conclusion, the results reveal that the proposed QSAR model is an efficient approach for modelling high-dimensional QSAR models and it is useful for the estimation of IC50 values of neuraminidase inhibitors that have not been experimentally tested.

Theoretical investigation of a few selected compounds as potent anti-tubercular agents and molecular docking evaluation: A multi-linear regression approach

Emergence of multi-drug resistant strains of Mycobacterium tuberculosis to the available drugs has demanded for the development of more potent anti-tubercular agents with efficient pharmacological activities. Time consumed and expenses in discovering and synthesizing new drug targets with improved biological activity have been a major challenge toward the treatment of multi-drug resistance strain M. tuberculosis. To solve the above problem, Quantitative Structure Activity Relationship (QSAR) is a recent approach developed to discover a novel drug with a better biological against M. Tuberculosis. A validated QSAR model developed in this study to predict the biological activities of some anti-tubercular compounds and to design new hypothetical drugs is influenced with the molecular descriptors; AATS7s, VR1-Dzi, VR1-Dzs, SpMin7-Bhe and RDF110i. The internal validation test for the derived model was found to have correlation coefficient (R2) of 0.8875, adjusted correlation coefficient (R2adj) value of 0.8234 and leave one out cross validation coefficient (Qcv2) value of 0.8012 while the external validation test was found to have (R2test) of 0.7961 and Y-randomization Coefficient (cRp2) of 0.6832. Molecular docking shows that ligand 13 of 2,4-disubstituted quinoline derivatives have promising higher binding score of -18.8 kcal/mol compared to the recommended drugs; isoniazid -14.6 kcal/mol. The proposed QSAR model and molecular docking studies will provides valuable approach for the modification of the lead compound, designing and synthesis more potent anti-tubercular agents.

Combination of least absolute shrinkage and selection operator with Bayesian Regularization artificial neural network (LASSO-BR-ANN) for QSAR studies using functional group and molecular docking mixed descriptors

A combination of least absolute shrinkage and selection operator (LASSO) with Bayesian Regularization feed-forward artificial neural network (LASSO-BR-ANN) was used as a new approach in the quantitative structure-activity relationship (QSAR) studies. A mixture of the docking derived descriptors with the simple functional group (structural) features was also introduced as a new ensemble of descriptors for accurate QSAR modeling. The performance of introduced approaches was tested with QSAR modeling of the biological activities (pEC50) of 73 azine derivatives as new non-nucleoside reverse transcriptase inhibitors (NNRTIs) for treatment of HIV disease. Molecular docking descriptors (MDDs) were generated from ligand-receptor interactions and functional group features derived using Dragon 5.5 software. The dataset was divided into three sets of training, validation, and test data. LASSO, as a penalized regression method, was applied to the training data set for the selection of the most relevant descriptors among the mixture of the structural and MDDs. LASSO selected descriptors were used as inputs in the construction of the Bayesian Regularization artificial neural network (BR-ANN) model. The results showed that the addition of functional group properties to the MDDs improves the accuracy of the model. Under the optimum conditions, LASSO-BR-ANN was successfully applied for the prediction of PEC50 values for compounds in the external test set with mean square error (MSE) and coefficient of determination (R2) values of 0.07 and 0.88, respectively. Some of the prediction statistical parameters of the model were calculated and all of them were in their acceptable ranges, which confirm the validity of the proposed QSAR model.

IN-SILICO STUDY FOR PREDICTING THE INHIBITION CONCENTRATION OF SOME HETEROCYCLIC AND PHENYLIC COMPOUNDS AS POTENT HERBICIDES USING THE MLR - GFA APPROACH

The study of the quantitative structure-activity relationship (QSAR) was used in a set of data from 43 heterocyclic and phenylic inhibitor compounds in order to establish a correlation between the inhibitory concentrations of the compounds in question and their structures. The optimization method of the density functional theory (DFT) was used to minimize the energy of the 3D structures using the Becke functional hybrid Exchange (B3) parameter with the Lee, Yang, and Parr Functional Correlation (LYP), commonly called the B3LYP functional Hybrid and 6-31G* Basis Set (B3LYP/6-31G*) method, to discover their molecular Quantum descriptors. Five models of QSAR were generated with the technique of genetic function algorithm (GFA). Among the five models generated, model 1 was selected as the best model because of its statistical significance (Friedman's LOF = 0.3008, R 2 = 0.9784, R 2 adj = 0.9739, Q cv 2 = 0.9675 and R 2 pred = 0.7348). The meticulous model was evaluated by means of the Leave One out cross-validation (LOO-CV) approach, external validation of the compounds of the test set, Y -randomization test and applicability domain (Williams Plot). The proposed QSAR model was highly predictive and vigorous with good validation parameters. The molecular descriptors used in the model should be considered of great importance in improving the inhibitory concentrations of the herbicides and also in the conception of new herbicides with a higher concentration of inhibitor.

QSAR study of anti-Human African Trypanosomiasis activity for 2-phenylimidazopyridines derivatives using DFT and Lipinski's descriptors

The quantitative structure-activity relationship (QSAR) of sixty 2-phenylimidazopyridines derivatives with anti-Human African Trypanosomiasis (anti-HAT) activity has been studied by using the density functional theory (DFT) and statistical methods. Becke's three-parameter hybrid method and the Lee-Yang-Parr B3LYP functional employing 6–31G(d) basis set are used to calculate quantum chemical descriptors using Gaussian 03W software, and the five Lipinski's parameters were calculated using ChemOffice software.In order to obtain robust and reliable QSAR model, the original dataset was randomly divided into training and prediction sets comprising 48 and 12 compounds, respectively. An optimal model for the training set with significant statistical quality was established. The same model was further applied to predict pEC50 values of the 12 compounds in the test set, further showing that this QSAR model has high predictive ability. It is very interesting to find that the anti-HAT of these compounds appear to be mainly governed by four factors, i.e., the number of H-bond donors, the lowest unoccupied molecular orbital energy, the molecular weight and the octanol/water partition coefficient. Here the possible action mechanism of these compounds was analysed and discussed, in particular, important structural requirements for great anti-HAT activity will be by increasing molecular size and substitute the 2-phenylimidazopyridines derivatives with polar, ionic, stronger accepting electron ability group and heteroatoms attached to one or more hydrogen atoms. Based on this proposed QSAR model, some new compounds with higher anti-HAT activities have been theoretically designed. Such results can offer useful theoretical references for future experimental works.

Measurement and prediction of bioconcentration factors of organophosphate flame retardants in common carp (Cyprinus carpio)

The increase in the production and usage plus the toxicity nature of organophosphate flame retardants (OPFRs) has become a concern. However, limited information is available about the bioaccumulation potential of OPFRs in fish. In this study, we determined the 96 h LC50 s, and evaluated the bioaccumulation potential of six most frequently reported OPFRs in gill, kidney, liver, and muscle tissues of common carp (Cyprinus carpio) for 48 d, and a quantitative structure-activity relationship (QSAR) model was developed to predict bioconcentration factors (BCFs) for the remaining 16 OPFRs. The BCFs and half-lives (t1/2) in the tissues ranged from 6.54 (Tris (2-chloroisopropyl) phosphate, (TCPP)) to 528.15 (Tris (2-ethylhexyl) phosphate (TEHP)), and 2.25–5.78 days, respectively. The tissue-specific concentration and BCFs values followed the order of liver > kidney ≥ intestine >> muscle. The proposed QSAR model with a high cross-validated value (Q2(cum)) of 0.930 and a correlation coefficient of 0.94 was obtained and was able to predict log BCF from parameters related to molar volume and isotropic average static field polarizability. The results show that the model has a high level of accuracy, making the proposed approach a suitable method for predicting the log BCF.

Theoretical modeling for predicting the activities of some active compounds as potent inhibitors against Mycobacterium tuberculosis using GFA-MLR approach

Tuberculosis (TB) is of one the most infectious disease caused by Mycobacterium tuberculosis which remains a serious public health problem. Emergence of multi-drug resistant strains of M. tuberculosis led to development of new and more potent anti-tuberculosis agents. The aim of this study was to correlate the chemical structures of the inhibitory compounds with their experimental activities. In this study, analogs of 2,4-disubstituted quinoline derivatives as potent anti-tubercular agents was subjected to quantitative structure–activity relationship (QSAR) analysis in order to build a QSAR model for predicting the activities of these compounds. In order to build the regression model, Genetic Function Approximation (GFA) and Multi-linear Regression approach were used to predict the activities of inhibitory compounds. Based on the prediction, the best validation model was found to have squared correlation coefficient (R2) of 0.9367, adjusted squared correlation coefficient (R2 adj) value of 0.9223 and cross validation coefficient (Qcv2) value of 0.8752. The chosen model was subjected to external validations and the model was found to have (R2 test) of 0.8215 and Y-randomization Coefficient (cRp2) of 0.6633. The proposed QSAR model provides a valuable approach for designing more potent anti-tubercular agents.

Quantitative Structure-Activity Relationship Model for HCVNS5B inhibitors based on an Antlion Optimizer-Adaptive Neuro-Fuzzy Inference System

The global prevalence of hepatitis C Virus (HCV) is approximately 3% and one-fifth of all HCV carriers live in the Middle East, where Egypt has the highest global incidence of HCV infection. Quantitative structure-activity relationship (QSAR) models were used in many applications for predicting the potential effects of chemicals on human health and environment. The adaptive neuro-fuzzy inference system (ANFIS) is one of the most popular regression methods for building a nonlinear QSAR model. However, the quality of ANFIS is influenced by the size of the descriptors, so descriptor selection methods have been proposed, although these methods are affected by slow convergence and high time complexity. To avoid these limitations, the antlion optimizer was used to select relevant descriptors, before constructing a nonlinear QSAR model based on the PIC50 and these descriptors using ANFIS. In our experiments, 1029 compounds were used, which comprised 579 HCVNS5B inhibitors (PIC50 < ~14) and 450 non-HCVNS5B inhibitors (PIC50 > ~14). The experimental results showed that the proposed QSAR model obtained acceptable accuracy according to different measures, where {{boldsymbol{R}}}^{{boldsymbol{2}}} was 0.952 and 0.923 for the training and testing sets, respectively, using cross-validation, while {{boldsymbol{R}}}_{{boldsymbol{L}}{boldsymbol{O}}{boldsymbol{O}}}^{{boldsymbol{2}}} was 0.8822 using leave-one-out (LOO).

Probing the origin of estrogen receptor alpha inhibition via large-scale QSAR study.

Estrogen is an important component for the sustenance of normal physiological functions of the mammary glands, particularly for growth and differentiation. Approximately, two-thirds of breast cancers are positive for estrogen receptor (ERs), which is a predisposing factor for the growth of breast cancer cells. As such, ERα represents a lucrative therapeutic target for breast cancer that has attracted wide interest in the search for inhibitory agents. However, the conventional laboratory processes are cost- and time-consuming. Thus, it is highly desirable to develop alternative methods such as quantitative structure–activity relationship (QSAR) models for predicting ER-mediated endocrine agitation as to simplify their prioritization for future screening. In this study, we compiled and curated a large, non-redundant data set of 1231 compounds with ERα inhibitory activity (pIC50). Using comprehensive validation tests, it was clearly observed that the model utilizing the substructure count as descriptors, performed well considering two objectives: using less descriptors for model development and achieving high predictive performance (RTr2 = 0.94, QCV2 = 0.73, and QExt2 = 0.73). It is anticipated that our proposed QSAR model may become a useful high-throughput tool for identifying novel inhibitors against ERα.

Ligand recognition properties of the vasopressin V2 receptor studied under QSAR and molecular modeling strategies.

The design of new drugs that target vasopressin 2 receptor (V2R) is of vital importance to develop new therapeutic alternatives to treat diseases such as heart failure, polycystic kidney disease. To get structural insights related to V2R-ligand recognition, we have used a combined approach of docking, molecular dynamics simulations (MD) and quantitative structure-activity relationship (QSAR) to elucidate the detailed interaction of the V2R with 119 of its antagonists. The three-dimensional model of V2R was built by threading methods refining its structure through MD simulations upon which the 119 ligands were subjected to docking studies. The theoretical results show that binding recognition of these ligands on V2R is diverse, but the main pharmacophore (electronic and π-π interactions) is maintained; thus, this information was validated under QSAR results. QSAR studies were performed using MLR analysis followed by ANN analysis to increase the model quality. The final equation was developed by choosing the optimal combination of descriptors after removing the outliers. The applicability domains of the constructed QSAR models were defined using the leverage and standardization approaches. The results suggest that the proposed QSAR models can reliably predict the reproductive toxicity potential of diverse chemicals, and they can be useful tools for screening new chemicals for safety assessment.

2D-QSAR study of some 2,5-diaminobenzophenone farnesyltransferase inhibitors by different chemometric methods.

Quantitative structure activity relationship (QSAR) models can be used to predict the activity of new drug candidates in early stages of drug discovery. In the present study, the information of the ninety two 2,5-diaminobenzophenone-containing farnesyltranaferase inhibitors (FTIs) were taken from the literature. Subsequently, the structures of the molecules were optimized using Hyperchem software and molecular descriptors were obtained using Dragon software. The most suitable descriptors were selected using genetic algorithms-partial least squares and stepwise regression, where exhibited that the volume, shape and polarity of the FTIs are important for their activities. The two-dimensional QSAR models (2D-QSAR) were obtained using both linear methods (multiple linear regression) and non-linear methods (artificial neural networks and support vector machines). The proposed QSAR models were validated using internal validation method. The results showed that the proposed 2D-QSAR models were valid and they can be used for prediction of the activities of the 2,5-diaminobenzophenone-containing FTIs. In conclusion, the 2D-QSAR models (both linear and non-linear) showed good prediction capability and the non-linear models were exhibited more accuracy than the linear models.

Application of genetic algorithms for pixel selection in multivariate image analysis for a QSAR study of trypanocidal activity for quinone compounds and design new quinone compounds

Quantitative structure-activity relationship (QSAR) analysis has been directed to a series of 31 quinone compounds with trypanocidal activity that was performed by chemometrics methods. The trypanocidal activity of the quinones is related to their redox potential (Epcl). Bidimensional images were used to calculate some pixels. Multivariate image analysis was applied to QSAR modeling of the redox potential of quinones derivatives by means of multivariate calibration such as principal component regression (PCR) and partial least squares (PLS). In this paper we investigate the effect of pixel selection by application of genetic algorithms (GAs) for PLS model. GAs is very useful in the variable selection in modeling and calibration because of the strong effect of the relationship between presence/absence of variables in a calibration model and the prediction ability of the model itself. The subset of pixels, which resulted in the low prediction error, was selected by genetic algorithm. The resulted model showed high prediction ability with RMSEP of 0.0694, 0.0358 and 0.0059 for PCR, PLS and GA-PLS models, respectively. Furthermore, the proposed QSAR model with GA-PLS was used for modification of structure and their activity predicted.