QSAR modeling for lipid peroxidation inhibition potential of flavonoids using topological and structural parameters

In the present study, Quantitative Structure-Activity Relationship (QSAR) modeling has been carried out for lipid peroxidation (LPO)-inhibition potential of a set of 27 flavonoids, using structural and topological parameters. For the development of models, three methods were used: (1) stepwise regression, (2) factor analysis followed by multiple linear regressions (FA-MLR) and (3) partial least squares (PLS) analysis. The best equation was obtained from stepwise regression analysis (Q² = 0.626) considering the leave-oneout prediction statistics.      

Experimental design,machine learning approaches for the optimization and modeling of caffeine adsorption

In the current research, the sorption of caffeine on fresh and calcined Cu–Al layered double hydroxide was comparatively studied based on adsorption parameters, adsorption kinetics, and adsorption isotherm. Response surface methodology (RSM), support vector machine (SVM) and artificial neural network (ANN), as data mining methods, were applied to develop models by considering various operating variables. Different characterization methods were exploited to conduct a comprehensive analysis of the characteristics of HDL in order to acquire a thorough understanding of its structural and functional features. The Langmuir model was employed to accurately describe the maximum monolayer adsorption capacity for calcined sample (q_max) of 152.99 mg/g mg/g with R² = 0.9977. The pseudo-second order model precisely described the adsorption phenomenon (R² = 0.999). The thermodynamic analysis also reveals a favorable and spontaneous process. The ANN model predicts adsorption efficiency result with R² = 0.989. The five-fold cross-validation was achieved to evaluate the validity of the SVM. The predication results revealed approximately 99.9% accuracy for test datasets and 99.63% accuracy for experiment data. Moreover, ANOVA analysis employing the central composite design-response surface methodology (CCD-RSM) indicated a good agreement between the quadratic equation predictions and the experimental data, which results in R² of 0.9868 and the highest removal percentages in optimized step were obtained for RSM (pH 5.05, mass of adsorbent 20 mg, time of 72 min, and caffeine concentrations of 22 mg/L). On the whole, the findings confirm that the proposed machine learning models provided reliable and robust computer methods for monitoring and simulating the adsorption of pollutants from aqueous solutions by Cu–Al–LDH.     

QSAR modeling of antiradical and antioxidant activities of flavonoids using electrotopological state (E-State) atom parameters

In the present paper QSAR modeling using electrotopological state atom (E-state) parameters has been attempted to determine the antiradical and the antioxidant activities of flavonoids in two model systems reported by Burda et al. (2001). The antiradical property of a methanolic solution of 1, 1-diphenyl-2-picrylhydrazyl (DPPH) and the antioxidant activity of flavonoids in a β-carotenelinoleic acid were the two model systems studied. Different statistical tools used in this communication are stepwise regression analysis, multiple linear regressions with factor analysis as the preprocessing step for variable selection (FA-MLR) and partial least squares analysis (PLS). In both the activities the best equation is obtained from stepwise regression analysis, considering, both equation statistics and predictive ability (antiradical activity: R ² = 0.927, Q² = 0.871 and antioxidant activity: R ² = 0.901, Q² = 0.841).      

Comparing various machine learning approaches in modeling the dynamic viscosity of CuO/water nanofluid

Nanofluids are broadly employed in heat transfer mediums to enhance their efficiency and heat transfer capacity. Thermophysical properties of nanofluids play a crucial role in their thermal behavior. Among various properties, the dynamic viscosity is one of the most crucial ones due to its impact on fluid motion and friction. Applying appropriate models can facilitate the design of nanofluidics thermal devices. In the present study, various machine learning methods including MPR, MARS, ANN-MLP, GMDH, and M5-tree are used for modeling the dynamic viscosity of CuO/water nanofluid based on the temperature, concentration, and size of nanostructures. The input data are extracted from various experimental studies to propose a comprehensive model, applicable in wide ranges of input variables. Moreover, the relative importance of each variable is evaluated to figure out the priority of the variables and their influences on the dynamic viscosity. Finally, the accuracy of the models is compared by employing the statistical criteria such as R-squared value. The models’ outputs disclosed that employing ANN-MLP approach leads to the most precise model. R-square value and average absolute percent relative error (AAPR) value of the model by using ANN-MLP model are 0.9997 and 1.312%, respectively. According to these values, ANN-MLP is a reliable approach for predicting the dynamic viscosity of the studied nanofluid. Additionally, based on the relative importance of the input variables, it is concluded that concentration has the highest relative importance; while the influence of size is the lowest one.

     

Non-linear QSAR modeling by using multilayer perceptron feedforward neural networks trained by back-propagation

The use of multilayer perceptrons (MLP) feedforward neural networks trained by back-propagation (BP) for non-linear QSAR model building is presented and explained in detail through a case study. This method was compared with others often used in this field, such as multiple linear regression (MLR), partial least squares (PLS) and quadratic PLS (QPLS). The case study deals with a series of 18 alpha adrenoreceptors agonists belonging to three different classes (alpha-1, alpha-2 and alpha-1,2) according to their different pharmacological effects. Each of them is described by 15 chemical features (the X block). Six pharmacological responses were also measured for each one to build the matrix of biological responses (the Y block). The results obtained indicated a slightly better performance of MLP against the other procedures, when using the correlation coefficient of the observed versus predicted response plots as an indicator of the goodness of the fit.     

Predictive modeling and computational machine learning simulation of adsorption separation using advanced nanocomposite materials

Adsorption process was simulated in this study for removal of Hg and Ni from water using nanocomposite materials. The used nanostructured material for the adsorption study was a combined MOF and layered double hydroxide, which is considered as MOF-LDH in this work. The data were obtained from resources and different machine learning models were trained. We selected three different regression models, including elastic net, decision tree, and Gradient boosting, to make regression on the small data set with two inputs and two outputs. Inputs are Ion type (Hg or Ni) and initial ion concentration in the feed solution (C₀), and outputs are equilibrium concentration (Ce) and equilibrium capacity of the adsorbent (Qe) in this dataset. After tuning their hyper-parameters, final models were implemented and assessed using different metrics. In terms of the R2-score metric, all models have more than 0.97 for Ce and more than 0.88 for Qe. The Gradient Boosting has an R2-score of 0.994 for Qe. Also, considering RMSE and MAE, Gradient Boosting shows acceptable errors and best models. Finally, the optimal values with the GB model are identical to dataset optimal: (Ion = Ni, C₀ = 250, Ce = 206.0). However, for Qe, it is different and is equal to (Ion = Hg, C0 = 121.12, Ce = 606.15). The results revealed that the developed methods of simulation are of high capacity in prediction of adsorption for removal of heavy metals using nanostructure materials.     

QSAR modeling of in vitro inhibition of cytochrome P450 3A4

We report the QSAR modeling of cytochrome P450 3A4 (CYP3A4) enzyme inhibition using four large data sets of in vitro data. These data sets consist of marketed drugs and drug-like compounds all tested in four assays measuring the inhibition of the metabolism of four different substrates by the CYP3A4 enzyme. The four probe substrates are benzyloxycoumarin, testosterone, benzyloxyresorufin, and midazolam. We first show that using state-of-the-art QSAR modeling approaches applied to only one of these four data sets does not lead to predictive models that would be useful for in silico filtering of chemical libraries. We then present the development and the testing of a multiple pharmacophore hypothesis (MPH) that is formulated as a conceptual extension of the traditional QSAR approach to modeling the promiscuous binding of a large variety of drugs to CYP3A4. In the simplest form, the MPH approach takes advantage of the multiple substrate data sets and identifies the binding of test compounds as either proximal or distal relative to that of a given substrate. Application of the approach to the in silico filtering of test compounds for potential inhibitors of CYP3A4 is also presented. In addition to an improvement in the QSAR modeling for the inhibition of CYP3A4, the results from this modeling approach provide structural insights into the drug-enzyme interactions. The existence of multiple inhibition data sets in the BioPrint database motivates the original development of the concept of a multiple pharmacophore hypothesis and provides a unique opportunity for formulating alternative strategies of QSAR modeling of the inhibition of the in vitro metabolism of CYP3A4.     

Predicting glycosylation stereoselectivity using machine learning

Predicting the stereochemical outcome of chemical reactions is challenging in mechanistically ambiguous transformations. The stereoselectivity of glycosylation reactions is influenced by at least eleven factors across four chemical participants and temperature. A random forest algorithm was trained using a highly reproducible, concise dataset to accurately predict the stereoselective outcome of glycosylations. The steric and electronic contributions of all chemical reagents and solvents were quantified by quantum mechanical calculations. The trained model accurately predicts stereoselectivities for unseen nucleophiles, electrophiles, acid catalyst, and solvents across a wide temperature range (overall root mean square error 6.8%). All predictions were validated experimentally on a standardized microreactor platform. The model helped to identify novel ways to control glycosylation stereoselectivity and accurately predicts previously unknown means of stereocontrol. By quantifying the degree of influence of each variable, we begin to gain a better general understanding of the transformation, for example that environmental factors influence the stereoselectivity of glycosylations more than the coupling partners in this area of chemical space.

A random forest algorithm, trained on a concise dataset and validated experimentally, accurately predicts the stereoselectivity of a complex organic coupling varying all reaction parameters as well as previously unknown mechanistic influences.     

人工神经网络预测非环状亚硝胺急性毒性的QSAR研究

在DFT-B3LYP/6-311+G(d,p)水平对60种非环状亚硝胺分子结构进行几何全优化,通过多元逐步线性回归(MSR)分析筛选出9个量子化学描述符作为自变量,log LD50(lethal dose 50％,LD50:大鼠口服急性毒性)作为因变量,采用人工神经网络(ANN)方法构建QSAR模型.经Levenber...     

Acetylcholinesterase inhibition by flavonoids from Agrimonia pilosa

In a bioassay-guided search for acetylcholinesterase (AChE) inhibitors from 180 medicinal plants, an ethyl acetate extract of whole plants of Agrimonia pilosa ledeb yielded tiliroside (1), 3-methoxy quercetin (2), quercitrin (3) and quercetin (4). We report herein for the first time that all four flavonol compounds showed significant inhibitory effects on AChE, particularly quercetin (4), which showed twice the activity of dehydroevodiamine (DHED).     

SAMPL6 logP challenge: machine learning and quantum mechanical approaches

Journal of Computer-Aided Molecular Design - Two different types of approaches: (a) approaches that combine quantitative structure activity relationships, quantum mechanical electronic structure...