scispace - formally typeset
Search or ask a question

Showing papers in "Sar and Qsar in Environmental Research in 2017"


Journal ArticleDOI
TL;DR: It is demonstrated that the performance of machine learning methods surpassed those of chemical similarity assessments, particularly in the case of novel repurposed indications.
Abstract: Drug repurposing provides a non-laborious and less expensive way for finding new human medicines. Computational assessment of bioactivity profiles shed light on the hidden pharmacological potential of the launched drugs. Currently, several freely available computational tools are available via the Internet, which predict multitarget profiles of drug-like compounds. They are based on chemical similarity assessment (ChemProt, SuperPred, SEA, SwissTargetPrediction and TargetHunter) or machine learning methods (ChemProt and PASS). To compare their performance, this study has created two evaluation sets, consisting of (1) 50 well-known repositioned drugs and (2) 12 drugs recently patented for new indications. In the first set, sensitivity values varied from 0.64 (TarPred) to 1.00 (PASS Online) for the initial indications and from 0.64 (TarPred) to 0.98 (PASS Online) for the repurposed indications. In the second set, sensitivity values varied from 0.08 (SuperPred) to 1.00 (PASS Online) for the initial indications and from 0.00 (SuperPred) to 1.00 (PASS Online) for the repurposed indications. Thus, this analysis demonstrated that the performance of machine learning methods surpassed those of chemical similarity assessments, particularly in the case of novel repurposed indications.

45 citations


Journal ArticleDOI
TL;DR: The experimental results of classifying the anti-hepatitis C virus activity of thiourea derivatives demonstrate that the proposed descriptor selection method in the QSAR classification model performs effectively and competitively compared with other existing penalized methods in terms of classification performance on both the training and the testing datasets.
Abstract: A high-dimensional quantitative structure–activity relationship (QSAR) classification model typically contains a large number of irrelevant and redundant descriptors. In this paper, a new design of descriptor selection for the QSAR classification model estimation method is proposed by adding a new weight inside L1-norm. The experimental results of classifying the anti-hepatitis C virus activity of thiourea derivatives demonstrate that the proposed descriptor selection method in the QSAR classification model performs effectively and competitively compared with other existing penalized methods in terms of classification performance on both the training and the testing datasets. Moreover, it is noteworthy that the results obtained in terms of stability test and applicability domain provide a robust QSAR classification model. It is evident from the results that the developed QSAR classification model could conceivably be employed for further high-dimensional QSAR classification studies.

35 citations


Journal ArticleDOI
TL;DR: To build QSAR models for natural lipase inhibitors by using the Monte Carlo method, the molecular structures were represented by the simplified molecular input line entry system (SMILES) notation and molecular graphs and the design of new potential lipase inhibitor is presented.
Abstract: Obesity is one of the most provoking health burdens in the developed countries. One of the strategies to prevent obesity is the inhibition of pancreatic lipase enzyme. The aim of this study was to build QSAR models for natural lipase inhibitors by using the Monte Carlo method. The molecular structures were represented by the simplified molecular input line entry system (SMILES) notation and molecular graphs. Three sets - training, calibration and test set of three splits - were examined and validated. Statistical quality of all the described models was very good. The best QSAR model showed the following statistical parameters: r2 = 0.864 and Q2 = 0.836 for the test set and r2 = 0.824 and Q2 = 0.819 for the validation set. Structural attributes for increasing and decreasing the activity (expressed as pIC50) were also defined. Using defined structural attributes, the design of new potential lipase inhibitors is also presented. Additionally, a molecular docking study was performed for the determination of binding modes of designed molecules.

33 citations


Journal ArticleDOI
TL;DR: External validated quantitative structure–toxicity relationship models were developed for toxicity of cosmetic ingredients on three different ecotoxicologically relevant organisms, namely Pseudokirchneriella subcapitata, Daphnia magna and Pimephales promelas following the OECD guidelines by partial least squares (PLS) regression technique.
Abstract: In this study, externally validated quantitative structure–toxicity relationship (QSTR) models were developed for toxicity of cosmetic ingredients on three different ecotoxicologically relevant organisms, namely Pseudokirchneriella subcapitata, Daphnia magna and Pimephales promelas following the OECD guidelines. The final models were developed by partial least squares (PLS) regression technique, which is more robust than multiple linear regression. The obtained model for P. subcapitata shows that molecular size and complexity have significant impacts on the toxicity of cosmetics. In case of P. promelas and D. magna, we found that the largest contribution to the toxicity was shown by hydrophobicity and van der Waals surface area, respectively. All models were validated using both internal and test compounds employing multiple strategies. For each QSTR model, applicability domain studies were also performed using the “Distance to Model in X-space” method. A comparison was made with the ECOSAR predic...

32 citations


Journal ArticleDOI
TL;DR: The experimental results from classifying the neuraminidase inhibitors of influenza A viruses (H1N1) demonstrate that the proposed method in the QSAR classification model performs effectively and competitively compared with other existing penalized methods in terms of classification performance and the number of selected descriptors.
Abstract: Descriptor selection is a procedure widely used in chemometrics The aim is to select the best subset of descriptors relevant to the quantitative structure-activity relationship (QSAR) study being considered In this paper, a new descriptor selection method for the QSAR classification model is proposed by adding a new weight inside L1-norm The experimental results from classifying the neuraminidase inhibitors of influenza A viruses (H1N1) demonstrate that the proposed method in the QSAR classification model performs effectively and competitively compared with other existing penalized methods in terms of classification performance and the number of selected descriptors

31 citations


Journal ArticleDOI
TL;DR: On the basis of obtained structural attributes, 11 new compounds were designed, out of which five compounds were found to have better activity than the best active compound in the series.
Abstract: Application of HIV-1 protease inhibitors (as an anti-HIV regimen) may serve as an attractive strategy for anti-HIV drug development. Several investigations suggest that there is a crucial need to develop a novel protease inhibitor with higher potency and reduced toxicity. Monte Carlo optimized QSAR study was performed on 200 hydroxyethylamine derivatives with antiprotease activity. Twenty-one QSAR models with good statistical qualities were developed from three different splits with various combinations of SMILES and GRAPH based descriptors. The best models from different splits were selected on the basis of statistically validated characteristics of the test set and have the following statistical parameters: r2 = 0.806, Q2 = 0.788 (split 1); r2 = 0.842, Q2 = 0.826 (split 2); r2 = 0.774, Q2 = 0.755 (split 3). The structural attributes obtained from the best models were analysed to understand the structural requirements of the selected series for HIV-1 protease inhibitory activity. On the basis of obtained structural attributes, 11 new compounds were designed, out of which five compounds were found to have better activity than the best active compound in the series.

27 citations


Journal ArticleDOI
TL;DR: This study illustrates the power of ML-based QSAR approaches for the screening and discovery of potent, isoform-selective HDACIs.
Abstract: Histone deacetylases (HDAC) are emerging as promising targets in cancer, neuronal diseases and immune disorders. Computational modelling approaches have been widely applied for the virtual screening and rational design of novel HDAC inhibitors. In this study, different machine learning (ML) techniques were applied for the development of models that accurately discriminate HDAC2 inhibitors form non-inhibitors. The obtained models showed encouraging results, with the global accuracy in the external set ranging from 0.83 to 0.90. Various aspects related to the comparison of modelling techniques, applicability domain and descriptor interpretations were discussed. Finally, consensus predictions of these models were used for screening HDAC2 inhibitors from four chemical libraries whose bioactivities against HDAC1, HDAC3, HDAC6 and HDAC8 have been known. According to the results of virtual screening assays, structures of some hits with pair-isoform-selective activity (between HDAC2 and other HDACs) were revealed. This study illustrates the power of ML-based QSAR approaches for the screening and discovery of potent, isoform-selective HDACIs.

25 citations


Journal ArticleDOI
TL;DR: Twelve new compounds were designed for better PDE10A inhibitory activity on the basis of the observations of the 3D QSAR studies, and may help in further designing new PDE 10A inhibitors with promising activity.
Abstract: Schizophrenia is a complex disorder of thinking and behaviour (0.3−0.7% of the population is affected). The over-expression of phosphodiesterase 10A (PDE10A) enzyme may be a potential target for schizophrenia and Huntington’s disease. Because 3D QSAR analysis is one of the most frequently used modelling techniques, in the present study, five different 3D QSAR tools, namely CoMFA, CoMSIA, kNN-MFA, Open3DQSAR and topomer CoMFA methods, were used on a dataset of pyrimidine-based PDE10A inhibitors. All developed models were validated internally and externally. The non-commercial Open3DQSAR produced the best statistical results amongst 3D QSAR tools. The structural interpretations obtained from different methods were thoroughly analysed and were justified on the basis of information obtained from the crystal structure. Information from one method was mostly validated by the results of other methods and vice versa. In the current work, the use of multiple tools in the same analysis revealed more complet...

25 citations


Journal ArticleDOI
TL;DR: In this paper, quantitative structure-activity (toxicity) relationships (QSA(T)R) analysis has been performed for a dataset of 54 6-arylpyrazine-2-carboxamides as trypanosoma brucei inhibitors to identify the important structural features required for future optimization of lead candidates.
Abstract: Human African trypanosomiasis (HAT) is prevalent in African countries, covering 37 countries, mostly sub-Saharan. A limited number of drugs are available to cure this neglected disease. In the present work, quantitative structure–activity (toxicity) relationships (QSA(T)R) analysis has been performed for a dataset of 54 6-arylpyrazine-2-carboxamides as Trypanosoma brucei inhibitors to identify the important structural features required for future optimization of lead candidates. The QSA(T)R models satisfy OECD guidelines and have high statistical robustness. The QSA(T)R models are based on easily interpretable molecular descriptors. The QSA(T)R models indicate that Trypanosoma brucei inhibitory activity of 6-arylpyrazine-2-carboxamides has correlation with the presence of N-sec-butylformamide and substituted benzene. The results could be beneficial for further optimization of 6-arylpyrazine-2-carboxamides as Trypanosoma brucei inhibitors. Some potential candidate molecules have been proposed.

22 citations


Journal ArticleDOI
TL;DR: Public domain and commercial in silico tools were compared for their performance in predicting the skin sensitization potential of chemicals, with positive and negative predictive values up to 80% and 84%, respectively.
Abstract: Public domain and commercial in silico tools were compared for their performance in predicting the skin sensitization potential of chemicals. The packages were either statistical based (Vega, CASE ...

22 citations


Journal ArticleDOI
TL;DR: The easy availability of the involved structural descriptors and the simplicity of the MLR model make the corresponding model attractive at an exploratory level for proposing, from this limited dataset, guidelines in the design of new potentially active molecules.
Abstract: QSAR models are proposed for predicting the toxicity of 33 piperidine derivatives against Aedes aegypti. From 2D topological descriptors, calculated with the PaDEL software, ordinary least squares multilinear regression (OLS-MLR) treatment from the QSARINS software and machine learning and related approaches including linear and radial support vector machine (SVM), projection pursuit regression (PPR), radial basis function neural network (RBFNN), general regression neural network (GRNN) and k-nearest neighbours (k-NN), led to four-variable models. Their robustness and predictive ability were evaluated through both internal and external validation. Determination coefficients (r2) greater than 0.85 on the training sets and 0.8 on the test sets were obtained with OLS-MLR and linear SVM. They slightly outperform PPR, radial SVM and RBFNN, whereas GRNN and k-NN showed lower performance. The easy availability of the involved structural descriptors and the simplicity of the MLR model make the corresponding model attractive at an exploratory level for proposing, from this limited dataset, guidelines in the design of new potentially active molecules.

Journal ArticleDOI
TL;DR: This study shows that whilst dividing the training set into subsets (i.e. clusters) improves prediction accuracy, it may not matter which method (expert based or purely machine learning) is used to divide the dataset into subset.
Abstract: Previous modelling of the median lethal dose (oral rat LD50) has indicated that local class-based models yield better correlations than global models. We evaluated the hypothesis that dividing the dataset by pesticidal mechanisms would improve prediction accuracy. A linear discriminant analysis (LDA) based-approach was utilized to assign indicators such as the pesticide target species, mode of action, or target species - mode of action combination. LDA models were able to predict these indicators with about 87% accuracy. Toxicity is predicted utilizing the QSAR model fit to chemicals with that indicator. Toxicity was also predicted using a global hierarchical clustering (HC) approach which divides data set into clusters based on molecular similarity. At a comparable prediction coverage (~94%), the global HC method yielded slightly higher prediction accuracy (r2 = 0.50) than the LDA method (r2 ~ 0.47). A single model fit to the entire training set yielded the poorest results (r2 = 0.38), indicating that there is an advantage to clustering the dataset to predict acute toxicity. Finally, this study shows that whilst dividing the training set into subsets (i.e. clusters) improves prediction accuracy, it may not matter which method (expert based or purely machine learning) is used to divide the dataset into subsets.

Journal ArticleDOI
TL;DR: A novel approach provides a theoretical foundation applicable to the evaluation and prediction of biogenic NER formation during pesticide degradation experiments, and may also be employed for the interpretation of NER data from regulatory studies.
Abstract: In biodegradation studies with isotope-labelled pesticides, fractions of non-extractable residues (NER) remain, but their nature and composition is rarely known, leading to uncertainty about their risk. Microbial growth leads to incorporation of carbon into the microbial mass, resulting in biogenic NER. Formation of microbial mass can be estimated from the microbial growth yield, but experimental data is rare. Instead, we suggest using prediction methods for the theoretical yield based on thermodynamics. Recently, we presented the Microbial Turnover to Biomass (MTB) method that needs a minimum of input data. We have estimated the growth yield of 40 organic chemicals (31 pesticides) using the MTB and two existing methods. The results were compared to experimental values, and the sensitivity of the methods was assessed. The MTB method performed best for pesticides. Having the theoretical yield and using the released CO2 as a measure for microbial activity, we predicted a range for the formation of b...

Journal ArticleDOI
TL;DR: The results indicate that random selection could lead to erroneous results and a rational selection allows for obtaining more reliable conclusions.
Abstract: This study performed an analysis of the influence of the training and test set rational selection on the quality and predictively of the quantitative structure-activity relationship (QSAR) model. The study was carried out on three different datasets of Influenza Neuraminidase (H1N1) inhibitors. The three datasets were divided into training and test sets using three rational selection methods: based on k-means, Kennard-Stone algorithm and Activity and the results were compared with Random selection. Then, a total of 31,490 mathematical models were developed and those models that presented a determination coefficient higher than: r2train > 0.8, r2loo > 0.7, r2test > 0.5 and minimum standard deviation (SD) and minimum root-mean square error (RMS) were selected. The selected models were validated using the internal leave-one-out method and the predictive capacity was evaluated by the external test set. The results indicate that random selection could lead to erroneous results. In return, a rational selection allows for obtaining more reliable conclusions. The QSAR models with major predictive power were found using the k-means algorithm and selection by activity.

Journal ArticleDOI
TL;DR: QSAR models to predict the inhibitory activity of a diverse and heterogeneous set of 88 organic molecules against the bromodomains BRD2, BRD3 and BRD4 are presented and suggest that the models proposed will be a good tool for studying the inhibitatory activities of drug candidates againstThe bromidomains considered during epigenetic drug discovery.
Abstract: Epigenetic drug discovery is a promising research field with growing interest in the scientific community, as evidenced by the number of publications and the large amount of structure-epigenetic activity information currently available in the public domain. Computational methods are valuable tools to analyse and understand the activity of large compound collections from their structural information. In this manuscript, QSAR models to predict the inhibitory activity of a diverse and heterogeneous set of 88 organic molecules against the bromodomains BRD2, BRD3 and BRD4 are presented. A conformation-dependent representation of the chemical structures was established using the RDKit software and a training and test set division was performed. Several two-linear and three-linear QuBiLS-MIDAS molecular descriptors ( www.tomocomd.com ) were computed to extract the geometric structural features of the compounds studied. QuBiLS-MIDAS-based features sets, to be used in the modelling, were selected using dimensionality reduction strategies. The multiple linear regression procedure coupled with a genetic algorithm were employed to build the predictive models. Regression models containing between 6 to 9 variables were developed and assessed according to several internal and external validation methods. Analyses of outlier compounds and the applicability domain for each model were performed. As a result, the models against BRD2 and BRD3 with 8 variables and the model with 9 variables against BRD4 were those with the best overall performance according to the criteria accounted for. The results obtained suggest that the models proposed will be a good tool for studying the inhibitory activities of drug candidates against the bromodomains considered during epigenetic drug discovery.

Journal ArticleDOI
TL;DR: This study proposes about nine potential hits which are expected to be promising molecules, having not only drug-like properties, but also being effective against multiple Mtb targets, which is expected to been effective against resistant strains by simultaneously inhibiting several potential Mtb drug targets.
Abstract: Developing effective inhibitors against Mycobacterium tuberculosis (Mtb) is a challenging task, primarily due to the emergence of resistant strains. In this study, we have proposed and implemented an in silico guided polypharmacological approach, which is expected to be effective against resistant strains by simultaneously inhibiting several potential Mtb drug targets. A combination of pharmacophore and QSAR based virtual screening strategy taking three key targets such as InhA (enoyl-acyl-carrier-protein reductase), GlmU (N-acetyl-glucosamine-1-phosphate uridyltransferase) and DapB (dihydrodipicolinate reductase) have resulted in initial 784 hits from Asinex database of 435,000 compounds. These hits were further subjected to docking with 33 Mtb druggable targets. About 110 potential polypharmacological hits were taken by integrating the aforementioned screening protocols. Further screening was conducted by taking various parameters and properties such as cell permeability, drug-likeness, drug-induced phospholipidosisand structural alerts. A consensus analysis has yielded 59 potential hits that pass through all the filters and can be prioritized for effective drug-resistant tuberculosis. This study proposes about nine potential hits which are expected to be promising molecules, having not only drug-like properties, but also being effective against multiple Mtb targets.

Journal ArticleDOI
TL;DR: Anticancer activity predicted by PASS with a probability Pa > 30% was confirmed by the experiment and relatively small Pa values estimated by PASS indicated the novelty of the considered derivatives comparing to the compounds from the PASS training set.
Abstract: Anticancer activity as an associated action for a series of dithiocarbamates of 9,10-anthracenedione was predicted using the PASS computer program and analysed with PharmaExpert software. The predicted cytotoxic activity of the dithiocarbamate derivatives of 9,10-anthracenedione was evaluated in vitro on cancer cells of the human lung (A549), prostate (PC3), colon (HT29) and human breast (MCF7) using the sulforhodamine B (SRB) cell viability assay. Among these compounds, 9,10-dioxo-9,10-dihydroanthracen-1-yl pyrrolidin-1-carbodithioate and 9,10-dioxo-9,10-dihydroanthracen-2-yl pyrrolidin-1-carbodithioate were identified as the most potent anticancer agents with cytotoxic activity against the MCF-7 human breast cell line with GI50 values of 1.40 μM and 1.52 μM, whereas the GI50 value for the reference anticancer drug mitoxantrone was 3.93 μM. Thus, anticancer activity predicted by PASS with a probability Pa > 30% was confirmed by the experiment. Relatively small Pa values estimated by PASS indicat...

Journal ArticleDOI
TL;DR: The residues such as Lys85, Phe147, Tyr148, Leu198, Val211, and Asp212 were found to be the most important interacting residues for possessing PfCDPK-1 inhibitory activity.
Abstract: Current research on antimalarial protein kinases has provided an opportunity to design kinase-based antimalarial drugs We have developed a common feature-based pharmacophore model from a set of multiple chemical scaffolds including derivatives of 3,6-imidazopyridazines, pyrazolo[2,3-d]pyrimidines and imidazo[1,5-a]pyrazines, in order to incorporate the maximum structural diversity information in the model for the Plasmodium falciparum calcium-dependent protein kinase-1 (PfCDPK-1) target The best pharmacophore model (Hypo-1) with the essential features of two hydrogen bond donors (HBD), one hydrophobic aromatic (HYAr) and one ring aromatic (RA) showed the classification accuracies of 8627%, 7843% and 10000% in labelling the training and test set (test set-1 and test set-2) compounds into more active and less active classes In order to identify the crucial interaction between multiple scaffold ligands and the target protein, we first developed the homology model using a template structure of P bergheii (PbCDPK1; PDB ID: 3Q5I), and thereafter performed the docking studies The residues such as Lys85, Phe147, Tyr148, Leu198, Val211, and Asp212 were found to be the most important interacting residues for possessing PfCDPK-1 inhibitory activity

Journal ArticleDOI
J. H. Lee1, Shaherin Basith1, Minghua Cui1, B. Kim1, S. Choi1 
TL;DR: The results demonstrated that the developed MCM using Laplacian-modified naïve Bayesian method was successful in classifying inhibitors and non-inhibitors for each CYP isoform.
Abstract: The cytochrome P450 (CYP) enzyme superfamily is involved in phase I metabolism which chemically modifies a variety of substrates via oxidative reactions to make them more water-soluble and easier to eliminate. Inhibition of these enzymes leads to undesirable effects, including toxic drug accumulations and adverse drug-drug interactions. Hence, it is necessary to develop in silico models that can predict the inhibition potential of compounds for different CYP isoforms. This study focused on five major CYP isoforms, including CYP1A2, 2C9, 2C19, 2D6 and 3A4, that are responsible for more than 90% of the metabolism of clinical drugs. The main aim of this study is to develop a multiple-category classification model (MCM) for the major CYP isoforms using a Laplacian-modified naive Bayesian method. The dataset composed of more than 4500 compounds was collected from the PubChem Bioassay database. VolSurf+ descriptors and FCFP_8 fingerprint were used as input features to build classification models. The results demonstrated that the developed MCM using Laplacian-modified naive Bayesian method was successful in classifying inhibitors and non-inhibitors for each CYP isoform. Moreover, the accuracy, sensitivity and specificity values for both training and test sets were above 80% and also yielded satisfactory area under the receiver operating characteristic curve and Matthews correlation coefficient values.

Journal ArticleDOI
TL;DR: This study highlights the potential of CORAL software for rapid screening of prospective compounds from a large chemical space and provides information that could aid in the design and development of potential Pgp inhibitors.
Abstract: P-glycoprotein (Pgp) inhibition has been considered as an effective strategy towards combating multidrug-resistant cancers. Owing to the substrate promiscuity of Pgp, the classification of its interacting ligands is not an easy task and is an ongoing issue of debate. Chemical structures can be represented by the simplified molecular input line entry system (SMILES) in the form of linear string of symbols. In this study, the SMILES notations of 2254 Pgp inhibitors including 1341 active, and 913 inactive compounds were used for the construction of a SMILE-based classification model using CORrelation And Logic (CORAL) software. The model provided an acceptable predictive performance as observed from statistical parameters consisting of accuracy, sensitivity and specificity that afforded values greater than 70% and MCC value greater than 0.6 for training, calibration and validation sets. In addition, the CORAL method highlighted chemical features that may contribute to increased and decreased Pgp inhi...

Journal ArticleDOI
TL;DR: The proposed QSPR model improves previous reported models of the bioconcentration factor in the present dataset, and the best multivariable linear regression models were found with the Replacement Method variable sub-set selection technique.
Abstract: The ANTARES dataset is a large collection of known and verified experimental bioconcentration factor data, involving 851 highly heterogeneous compounds from which 159 are pesticides. The BCF ANTARES data were used to derive a conformation-independent QSPR model. A large set of 27,017 molecular descriptors was explored, with the main intention of capturing the most relevant structural characteristics affecting the studied property. The structural descriptors were derived with different freeware tools, such as PaDEL, Epi Suite, CORAL, Mold2, RECON, and QuBiLs-MAS, and so it was interesting to find out the way that the different descriptor tools complemented each other in order to improve the statistical quality of the established QSPR. The best multivariable linear regression models were found with the Replacement Method variable sub-set selection technique. The proposed QSPR model improves previous reported models of the bioconcentration factor in the present dataset.

Journal ArticleDOI
TL;DR: The empirical lipophilicity (RM) was compared with the corresponding log P characteristics calculated using alternative methods for deducing the lipophilic features, and the mean values of the selected molecular descriptors that were averaged over the chosen calculation methods were subsequently correlated with the RM parameter.
Abstract: Finding a balance between a desired drug's potency and its physicochemical properties that are important for its molecule pharmacokinetic or pharmacodynamics profile is still a challenging issue in rational drug discovery. Quantitative assessment of the lipophilic characteristics of potential drug molecules is indispensable for efficient development of Absorption, Distribution, Metabolism, Excretion, Toxicity-tailored structure-activity models; therefore reliable procedures for deriving log P from molecular structure are desirable. In the current work a range of various software log P predictors for estimation of the numerical lipophilic values for a set of cholic acid derivatives were employed and subsequently cross-compared with the experimental parameters. Thus, the empirical lipophilicity (RM) was compared with the corresponding log P characteristics calculated using alternative methods for deducing the lipophilic features. The mean values of the selected molecular descriptors that were averaged over the chosen calculation methods (consensus clog P) were subsequently correlated with the RM parameter. As an additional experiment, the iterative variable elimination partial least squares (IVE-PLS) methodology for an ensemble of descriptors retrieved from Dragon 6.0 software was applied for a set of drug transporters. To investigate the variations within the ensemble of cholic acid derivatives principal component analysis (PCA) and self-organizing neural network (SOM) procedures were used to visualize the major differences in the performance of drug promoters with respect to their lipophilic profile.

Journal ArticleDOI
TL;DR: Several QSAR models for the prediction of MOA of 221 phenols to the ciliated protozoan Tetrahymena pyriformis, using Chemistry Development Kit descriptors are reported, and the improvement was significant.
Abstract: The phenols are structurally heterogeneous pollutants and they present a variety of modes of toxic action (MOA), including polar narcotics, weak acid respiratory uncouplers, pro-electrophiles, and soft electrophiles. Because it is often difficult to determine correctly the mechanism of action of a compound, quantitative structure-activity relationship (QSAR) methods, which have proved their interest in toxicity prediction, can be used. In this work, several QSAR models for the prediction of MOA of 221 phenols to the ciliated protozoan Tetrahymena pyriformis, using Chemistry Development Kit descriptors, are reported. Four machine learning techniques (ML), k-nearest neighbours, support vector machine, classification trees, and artificial neural networks, have been used to develop several models with higher accuracies and predictive capabilities for distinguishing between four MOAs. They showed global accuracy values between 95.9% and 97.7% and area under Receiver Operator Curve values between 0.978 ...

Journal ArticleDOI
TL;DR: After molecular dynamics simulations, one hit has been predicted to possess good binding affinity for both ALK and EGFR (T790M), which can be further investigated for its experimental in-vitro/in-vivo activities.
Abstract: Extensively validated 3D pharmacophore models for ALK (anaplastic lymphoma kinase) and EGFR (T790M) (epithelial growth factor receptor with acquired secondary mutation) were developed. The pharmacophore model for ALK (r2 = 0.96, q2 = 0.692) suggested that two hydrogen bond acceptors and three hydrophobic groups arranged in 3-D space are essential for the binding affinity of ALK inhibitors. Similarly, the pharmacophore model for EGFR (T790M) (r2 = 0.92, q2 = 0.72) suggested that the presence of a hydrogen bond acceptor, two hydrogen bond donors and a hydrophobic group plays vital role in binding of an inhibitor of EGFR (T790M). These pharmacophore models allowed searches for novel ALK and EGFR (T790M) dual inhibitors from multiconformer 3D databases (Asinex, Chembridge and Maybridge). Finally, the eight best hits were selected for molecular dynamics simulation, to study the stability of their complexes with both proteins and final binding orientations of these molecules. After molecular dynamics simulations, one hit has been predicted to possess good binding affinity for both ALK and EGFR (T790M), which can be further investigated for its experimental in-vitro/in-vivo activities.

Journal ArticleDOI
TL;DR: The models developed in this study provide an excellent prediction of the inhibitory concentration of a new family of AChE inhibitors, and all criteria used to validate these models revealed the superiority of the GFA model.
Abstract: A new family of AChE inhibitors, N-benzylpiperidines, showed exceptional efficacy in vitro and in vivo, minimal side effects and high selectivity for acetylcholinesterase (AChE). Three regression methods were chosen in this work to develop robust predictive models, namely multiple linear regression (MLR), genetic function approximation (GFA) and multilayer perceptron network (MLP). Ten descriptors were selected for a dataset of 99 molecules, using a genetic algorithm. The best results were obtained for MLP with a 10-6-1 artificial neural network model trained with the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm. Statistical prediction for MLR and GFA were r2 = 0.882 and r2 = 0.875, respectively. Because internal and external validation strategies play an important role, we adopted all available validation strategies to check the robustness of the models. All criteria used to validate these models revealed the superiority of the GFA model. Therefore, the models developed in this study provide an excellent prediction of the inhibitory concentration of a new family of AChE inhibitors.

Journal ArticleDOI
TL;DR: The prediction of SOE is based on a combination of LMNA (Labelled Multilevel Neighbourhood of Atom) descriptors and Bayesian-like algorithm implemented in PASS software and MetaTox web-service.
Abstract: Biotransformation is a process of the chemical modifications which may lead to the reactive metabolites, in particular the epoxides. Epoxide reactive metabolites may cause the toxic effects. The prediction of such metabolites is important for drug development and ecotoxicology studies. Epoxides are formed by some oxidation reactions, usually catalysed by cytochromes P450, and represent a large class of three-membered cyclic ethers. Identification of molecules, which may be epoxidized, and indication of the specific location of epoxide functional group (which is called SOE - site of epoxidation) are important for prediction of epoxide metabolites. Datasets from 355 molecules and 615 reactions were created for training and validation. The prediction of SOE is based on a combination of LMNA (Labelled Multilevel Neighbourhood of Atom) descriptors and Bayesian-like algorithm implemented in PASS software and MetaTox web-service. The average invariant accuracy of prediction (AUC) calculated in leave-one-out and 20-fold cross-validation procedures is 0.9. Prediction of epoxide formation based on the created SAR model is included as the component of MetaTox web-service ( http://www.way2drug.com/mg ).

Journal ArticleDOI
TL;DR: This work investigates the estimated reduction and oxidation potentials of 34 nitroaromatic compounds using quantum chemical approaches and identifies relationships between electron affinity (ELUMO) and hydrophobicity (log P).
Abstract: Nitroaromatic compounds and the products of their degradation are toxic to bacteria, cells and animals. Various studies have been carried out to better understand the mechanism of toxicity of aromatic nitrocompounds and their relationship to humans and the environment. Recent data relate cytotoxicity of nitroaromatic compounds to their single- or two-electron enzymatic reduction. However, mechanisms of animal toxicity could be more complex. This work investigates the estimated reduction and oxidation potentials of 34 nitroaromatic compounds using quantum chemical approaches. All geometries were optimized with density functional theory (DFT) using the solvation model based on density (SMD) and polarizable continuum model (PCM) solvent model protocols. Quantitative structure–activity/property (QSAR/QSPR) models were developed using descriptors obtained from quantum chemical optimizations as well as the DRAGON software program. The QSAR/QSPR equations developed consist of two to four descriptors. Cor...

Journal ArticleDOI
TL;DR: Computational studies based on pharmacophore modelling, atom-based QSAR, molecular docking, free binding energy calculation and dynamics simulation were performed on a series of pyridine-3-carboxamide-6-yl-urea derivatives, revealing structural and chemical features necessary for these molecules to inhibit GyrB.
Abstract: DNA gyrase subunit B (GyrB) is an attractive drug target for the development of antibacterial agents with therapeutic potential. In the present study, computational studies based on pharmacophore modelling, atom-based QSAR, molecular docking, free binding energy calculation and dynamics simulation were performed on a series of pyridine-3-carboxamide-6-yl-urea derivatives. A pharmacophore model using 49 molecules revealed structural and chemical features necessary for these molecules to inhibit GyrB. The best fitted model AADDR.13 was generated with a coefficient of determination (r²) of 0.918. This model was validated using test set molecules and had a good r² of 0.78. 3D contour maps generated by the 3D atom-based QSAR revealed the key structural features responsible for the GyrB inhibitory activity. Extra precision molecular docking showed hydrogen bond interactions with key amino acid residues of ATP-binding pocket, important for inhibitor binding. Further, binding free energy was calculated by the MM-GBSA rescoring approach to validate the binding affinity. A 10 ns MD simulation of inhibitor #47 showed the stability of the predicted binding conformations. We identified 10 virtual hits by in silico high-throughput screening. A few new molecules were also designed as potent GyrB inhibitors. The information obtained from these methodologies may be helpful to design novel inhibitors of GyrB.

Journal ArticleDOI
TL;DR: Investigation of the acid/base properties and physicochemical characteristics of three classes of compounds: fungicides, herbicides and insecticides found that the pesticides were generally smaller, possessed a neutral charge state and were more lipophilic.
Abstract: Drug-likeness has long been studied in the pursuit of finding new medicines. Similarly, in the agrochemical field there is value in exploring the properties of the chemicals involved. Patterns that emerge can potentially influence future discovery campaigns to improve the probability of commercial success. In this study we investigate the acid/base properties and physicochemical characteristics of three classes of compounds: fungicides, herbicides and insecticides. In comparison with FDA-approved drugs, it was noted that the pesticides were generally smaller, possessed a neutral charge state and were more lipophilic. The results are discussed in the light of their intended targets.

Journal ArticleDOI
TL;DR: Both ligand and structure-based drug design approaches are deployed to hunt novel drug-like candidates as mPGES-1 inhibitors and the validated pharmacophore model is used as a three-dimensional query to screen the FDA-approved Lopac database.
Abstract: COX-2 inhibitors exhibit anticancer effects in various cancer models but due to the adverse side effects associated with these inhibitors, targeting molecules downstream of COX-2 (such as mPGES-1) has been suggested. Even after calls for mPGES-1 inhibitor design, to date there are only a few published inhibitors targeting the enzyme and displaying anticancer activity. In the present study, we have deployed both ligand and structure-based drug design approaches to hunt novel drug-like candidates as mPGES-1 inhibitors. Fifty-four compounds with tested mPGES-1 inhibitory value were used to develop a model with four pharmacophoric features. 3D-QSAR studies were undertaken to check the robustness of the model. Statistical parameters such as r2 = 0.9924, q2 = 0.5761 and F test = 1139.7 indicated significant predictive ability of the proposed model. Our QSAR model exhibits sites where a hydrogen bond donor, hydrophobic group and the aromatic ring can be substituted so as to enhance the efficacy of the inhibitor. Furthermore, we used our validated pharmacophore model as a three-dimensional query to screen the FDA-approved Lopac database. Finally, five compounds were selected as potent mPGES-1 inhibitors on the basis of their docking energy and pharmacokinetic properties such as ADME and Lipinski rule of five.