Principal Component-artificial Neural Network Modeling Research Articles

Competitive adsorption of methylene blue and brilliant green onto graphite oxide nano particle following: Derivative spectrophotometric and principal component-artificial neural network model methods for their simultaneous determination

In this work, the competitive adsorption of methylene blue (MB) and brilliant green (BG) onto graphite oxide (GO) nanoparticles followed by their accurate and reproducible determination by second order derivative spectrophotometry (SODS) and principal component–artificial neural network model (PCA–ANN) model has been studied. The evaluation of kinetic and isotherm studies was investigated at optimum experimental conditions set as pH=7.0, 8mg of GO and 14min contact time in binary systems. The equilibrium amounts of MB and BG dyes in binary mixture adsorbed onto GO-NP has opposite correlation with their initial concentration. Principal component analysis (PCA) used to minimize the dimensionality of large data sets via reducing the number of spectral data by a three-layered feed-forward artificial neural network (ANN) trained by Levenberg–Marquardt back-propagation algorithm. The ANN model was able to predict the concentrations of both dyes in mixtures with a tangent sigmoid transfer function (tansig) at hidden layer with 20 neurons and a linear transfer function (purelin) at output layer. Several isotherm models were applied to experimental data and the isotherm constants were calculated for BG and MB dyes. Among the applied models, the extended Freundlich isotherm model adequately predicts the multi-component adsorption equilibrium data at moderate ranges of concentration.

Application of silver nanoparticles and principal component-artificial neural network models for simultaneous determination of levodopa and benserazide hydrochloride by a kinetic spectrophotometric method

A multicomponent analysis method based on principal component analysis-artificial neural network model (PC-ANN) is proposed for the simultaneous determination of levodopa (LD) and benserazide hydrochloride (BH). The method is based on the reaction of levodopa and benserazide hydrochloride with silver nitrate as an oxidizing agent in the presence of PVP and formation of silver nanoparticles. The reaction monitored at analytical wavelength 440 nm related to surface plasmon resonance band of silver nanoparticles. Differences in the kinetic behavior of the levodopa and benserazide hydrochloride were exploited by using principal component analysis, an artificial neural network (PC-ANN) to resolve concentration of analytes in their mixture. After reducing the number of kinetic data using principal component analysis, an artificial neural network consisting of three layers of nodes was trained by applying a back-propagation learning rule. The optimized ANN allows the simultaneous determination of analytes in mixtures with relative standard errors of prediction in the region of 4.5 and 6.3 for levodopa and benserazide hydrochloride respectively. The results show that this method is an efficient method for prediction of these analytes.

Fast Nondestructive Measurement and Discrimination of Tablet Based on Vis/NIR Spectroscopy and Chemometrics

This paper presents methods based on chemometrics analysis to select the optimal model for variety discrimination of ginkgo (Ginkgo biloba L.) tablets by using a visible/short-wave near-infrared spectroscopy (Vis/NIRS) system. The tablet varieties used in the research include Da na kang, Xin bang, Tian bao ning, Yi kang, Hua na xing, Dou le, Lv yuan, Hai wang, and Ji yao. All samples (n=270) were scanned in the Vis/NIR region between 325-1075nm using a spectrograph. Principal component artificial neural network (PC-ANN) was used to identify the tablet varieties. In PC-ANN models, the scores of the principal components were chosen as the input nodes for the input layer of ANN. Independent component analysis (ICA) was executed to select several optimal wavelengths based on loading weights. The absorbance values log (1/R), corresponding to the wavelengths of 481nm, 1000nm, 460nm, 572nm, 658nm, 401nm, 998nm, 996nm, 468nm and 661nm were then chosen as the input data of artificial neural network (IC-ANN), and the discrimination rate was reached at 95.6%, which was better than PC-ANN. The results indicated that ginkgo tablets discrimination was good based on the both methods.

Application of multivariate image analysis in QSPR study of 13C chemical shifts of naphthalene derivatives: A comparative study

A new implemented QSPR method, whose descriptors achieved from bidimensional images, was applied for predicting 13C NMR chemical shifts of 25 mono substituted naphthalenes. The resulted descriptors were subjected to principal component analysis (PCA) and the most significant principal components (PCs) were extracted. MIA-QSPR (multivariate image analysis applied to quantitative structure–property relationship) modeling was done by means of principal component regression (PCR) and principal component-artificial neural network (PC-ANN) methods. Eigen value ranking (EV) and correlation ranking (CR) were used here to select the most relevant set of PCs as inputs for PCR and PC-ANN modeling methods. The results supported that the correlation ranking-principal component-artificial neural network (CR-PC-ANN) model could predict the 13C NMR chemical shifts of all 10 carbon atoms in mono substituted naphthalenes with R 2 ≥ 0.922 for training set, R 2 ≥ 0.963 for validation set and R 2 ≥ 0.936 for the test set. Comparison of the results with other existing factor selection method revealed that less accurate results were obtained by the eigen value ranking procedure.

Correlation ranking and stepwise regression procedures in principal components artificial neural networks modeling with application to predict toxic activity and human serum albumin binding affinity

The performance of the two novel QSAR algorithms, stepwise regression (SR) and correlation ranking (CR) principal component-artificial neural network modeling methods named (SR-PC-ANN and CR-PC-ANN, respectively), combined with two factor selection approaches is compared. These algorithms are applied to predict the toxic activity of 278 substituted benzenes belonging to diverse types of compounds as well as human serum albumin (HSA) binding affinity of a set of 95 compounds. A total number of 1233 and 698 theoretical descriptors are calculated for each molecule for the toxicity and binding affinity data sets, respectively. In PCA analysis, the principal components were extracted by applying two different approaches named the individual PCA approach, PCA(I), and the combined PCA approach, PCA(C). In the PCA(I) approach, descriptors were gathered into groups and then the PCA analysis was performed on each group. In the combined PCA approach, PCA(C), PCA analysis was applied on the pool of all calculated descriptors combined in one group. One is able to see that the models obtained by the CR-PC-ANN(I) approach are superior to those obtained from the other approaches. Both the external and cross-validation methods were used to validate the performances of the resulting models. Randomization test was employed to check the suitability of the models and to investigate the possibility of obtaining chance models.

Exploring QSARs of Some Analgesic Compounds by PC-ANN

Quantitative structure-activity relationship study was performed to understand analgesic activity for a set of 95 heterogeneous analgesic compounds. This study was performed by using the principal component-artificial neural network modeling method, with application of eigenvalue ranking factor selection procedure. The results obtained by principal component-artificial neural network give advanced regression models with good prediction ability using a relatively low number of principal components. A 0.834 correlation coefficient was obtained using principal component-artificial neural network with six extracted principal components.

Simultaneous determination of thorium and uranyl ions by optode spectra and chemometric techniques

A sensitive optode was fabricated from a recently synthesized ligand (triazene-1,3-di(2-methoxyphenyl)). Simultaneous determination of thorium and uranium in aqueous samples were performed by this optode in conjunction with partial least squares (PLS) and artificial neural network (ANN) techniques. The measured absorbances at 27 selected wavelengths were used for training. A PLS1 model with 4 and 3 latent variables for thorium and uranium respectively, a PLS2 model with 4 latent variables and a principal component ANN model (4-2-2) with linear transfer function after hidden and output layers were created. Relative error of prediction for PLS1, PLS2 and ANN models in synthetic mixtures were 7.40%, 7.12% and 5.04% respectively.

Application of principal component-artificial neural network models for simultaneous determination of phenolic compounds by a kinetic spectrophotometric method

A multicomponent analysis method based on principal component analysis-artificial neural network models (PC-ANN) is proposed for the determination of phenolic compounds. The method relies on the oxidative coupling of phenols (phenol, 2 chlorophenol, 3-chlorophenol and 4-chlorophenol) to N, N-diethyl- p-phenylenediamine in the presence of hexacyanoferrate(III). The reaction monitored at analytical wavelength 680 nm of the dye formed. Phenols can be determined individually over the concentration range 0.1–7.0 μg ml −1. Differences in the kinetic behavior of the four species were exploited by using PC-ANN, to resolve mixtures of phenol. After reducing the number of kinetic data using principal component analysis, an artificial neural network consisting of three layers of nodes was trained by applying a back-propagation learning rule. The optimized ANN allows the simultaneous quantitation of four analytes in mixtures with relative standard errors of prediction in the region of 5% for four species. The results show that PC-ANN is an efficient method for prediction of the four analytes.

Optical system for tablet variety discrimination using visible/near-infrared spectroscopy

An optical system based on visible/near-infrared spectroscopy (Vis/NIRS) for variety discrimination of ginkgo (Ginkgo biloba L.) tablets was developed. This system consisted of a light source, beam splitter system, sample chamber, optical detector (diffuse reflection detector), and data collection. The tablet varieties used in the research include Da na kang, Xin bang, Tian bao ning, Yi kang, Hua na xing, Dou le, Lv yuan, Hai wang, and Ji yao. All samples (n=270) were scanned in the Vis/NIR region between 325 and 1075 nm using a spectrograph. The chemometrics method of principal component artificial neural network (PC-ANN) was used to establish discrimination models of them. In PC-ANN models, the scores of the principal components were chosen as the input nodes for the input layer of ANN, and the best discrimination rate of 91.1% was reached. Principal component analysis was also executed to select several optimal wavelengths based on loading values. Wavelengths at 481, 458, 466, 570, 1000, 662, and 400 nm were then used as the input data of stepwise multiple linear regression, the regression equation of ginkgo tablets was obtained, and the discrimination rate was researched 84.4%. The results indicated that this optical system could be applied to discriminating ginkgo (Ginkgo biloba L.) tablets, and it supplied a new method for fast ginkgo tablet variety discrimination.

Simultaneous Spectrophotometric Determination of Phosphate and Silicate by Using Principal Component Artificial Neural Network

A very sensitive, simple and selective spectrophotometric method for simultaneous determination of phosphate and silicate based on formation of phospho- and silicomolybdenum blue complexes in the presence of ascorbic acid is described. Although the complexes of phosphate and silicate with reagent in the presence of ascorbic acid show a spectral overlap, they have been simultaneously determined by principal component artificial neural network (PC-ANN). The PC-ANN architectures were different for phosphate and silicate. The output of phosphate PC-ANN architecture was used as an input for silicate PC-ANN architecture. This modification improves the capability of silicate PC-ANN model for prediction of silicate concentrations. The linear range was 0.01-3.00 microg mL(-1) for phosphate and 0.01-5.00 microg mL(-1) for silicate. Interference effects of common anions and cations were studied and the proposed method was also applied satisfactorily to the determination of phosphate and silicate in detergents.

ANN‐QSAR Model of Drug‐binding to Human Serum Albumin

Quantitative structure-activity relationship study was performed to understand drug binding to human serum albumin. This study was performed on 94 different human serum albumin drug and drug-like compounds by using the principal component-artificial neural network modeling method, with application of eigenvalue ranking factor selection procedure. The results obtained by principal component-artificial neural network gives better regression models with good prediction ability using a relatively low number of principal components in comparison to other quantitative structure-activity relationship studies on the same data set of compounds. A 0.8497 coefficient of determination was obtained using principal component-artificial neural network with six extracted principal components.

Spectrophotometric resolution of ternary mixtures of tryptophan, tyrosine, and histidine with the aid of principal component–artificial neural network models

A simple and sensitive spectrophotometric method to resolve ternary mixtures of tryptophan (Trp), tyrosine (Tyr), and histidine (His) in synthetic and water samples is described. It relies on the different kinetic rates of the analytes in their oxidative reaction with potassium ferricyanide (K 3Fe(CN) 6) in alkaline medium. The absorbance data were monitored on the analytical wavelength (420 nm) of K 3Fe(CN) 6 spectrum. Synthetic mixtures of the three amino acids were analyzed, and the data obtained were processed by principal component–artificial neural network (PC–ANN) models. After reducing the number of spectral data using principal component analysis, an artificial neural network consisting of three layers of nodes was trained by applying a back-propagation learning rule. Tangent and sigmoidal transfer function were used in the hidden and output layers, respectively. The analytical performance of this method was characterized by relative standard error. The method allowed the determination of Trp, Tyr, and His at concentrations between 10 and 55, 10 and 60, and 10 and 40 μg ml −1, respectively. The results show that the PC–ANN is an efficient method for prediction of the three analytes.

Measurement of yogurt internal quality through using Vis/NIR spectroscopy

Abstract A nondestructive optical method for determining the sugar content and acidity of yogurt was investigated. Three types of preprocessing, S. Golay smoothing with multiplicative scatter correction (S. Golay smoothing with MSC), S. Golay 1st-Der and wavelet package transform (WPT), were used before the data were analyzed with chemometrics methods of partial least square (PLS). Spectral data sets as the logarithms of the reflectance reciprocal were analyzed to build a best model for predicting the sugar content and acidity of yogurt. A model using preprocessing of WPT with a correlation coefficient of 0.91 and 0.90, a root mean square error of prediction (RMSEP) of 0.36 and 0.04 showed an excellent prediction performance to sugar content and acidity. S. Golay smoothing with MSC was also finer, combined with the calibration and validation results. S. Golay 1st-Der was the worse preprocessing method in this experiment. In the paper, a multivariate calibration method of principal component artificial neural network (PC-ANN) was also established. In PC-ANN models, the scores of the principal components were chosen as the input nodes for the input layer of ANN. After adjusting the number of input nodes (principal components), hidden nodes, as well as learning rate and momentum of the network, a model with a correlation coefficient of 0.92 and 0.91, a root mean square error of prediction (RMSEP) of 0.33 and 0.04 showed an excellent prediction performance on sugar content and acidity. At the same time, the sensitive wavelengths corresponding to the sugar content and acidity of yogurt were proposed on the basis of regression coefficients by PLS.

Toward an Optimal Procedure for PC‐ANN Model Building: Prediction of the Carcinogenic Activity of a Large Set of Drugs.

The performances of the three novel QSAR algorithms, principal component-artificial neural network modeling method combining with three factor selection procedures named eigenvalue ranking, correlation ranking, and genetic algorithm (ER-PC-ANN, CR-PC-ANN, PC-GA-ANN, respectively), are compared by application of these model to the prediction of the carcinogenic activity of a large set of drugs (735 drugs) belonging to a diverse type of compounds. A total number of 1350 theoretical descriptors are calculated for each molecule. The matrix of calculated descriptors (with 735 x 1350 dimension) is subjected to PCA. 95% of the variances in the matrix are explained by the first 137 principal components (PC's). From the pool of 137 PC's, the factor selection methods (ER, CR, and GA) are employed to select the best set of PC's for PC-ANN modeling. In the ER-PC-ANN, the PC's are successively entered into the ANN based on their decreasing eigenvalue. In the CR-PC-ANN, the ANN is first employed to model the nonlinear relationship between each one of the PC's and the carcinogen activity separately. Then, the PC's are ranked based on their decreasing correlating ability and entered to the input layer of the network one after another. Finally, a search algorithm (i.e. genetic algorithm) is used to find the best set of PC's. Both the external and cross-validation methods are used to validate the performances of the resulting models. One is able to see that the results obtained by the PC-GA-ANN and CR-PC-ANN procedures are superior to those resulted from the EV-PC-ANN. Comparison of the results reveals that the results produced by the PC-GA-ANN algorithm are better than those produced by CR-PC-ANN. However, the difference is not significant.

Toward an Optimal Procedure for PC-ANN Model Building: Prediction of the Carcinogenic Activity of a Large Set of Drugs

The performances of the three novel QSAR algorithms, principal component-artificial neural network modeling method combining with three factor selection procedures named eigenvalue ranking, correlation ranking, and genetic algorithm (ER-PC-ANN, CR-PC-ANN, PC-GA-ANN, respectively), are compared by application of these model to the prediction of the carcinogenic activity of a large set of drugs (735 drugs) belonging to a diverse type of compounds. A total number of 1350 theoretical descriptors are calculated for each molecule. The matrix of calculated descriptors (with 735 x 1350 dimension) is subjected to PCA. 95% of the variances in the matrix are explained by the first 137 principal components (PC's). From the pool of 137 PC's, the factor selection methods (ER, CR, and GA) are employed to select the best set of PC's for PC-ANN modeling. In the ER-PC-ANN, the PC's are successively entered into the ANN based on their decreasing eigenvalue. In the CR-PC-ANN, the ANN is first employed to model the nonlinear relationship between each one of the PC's and the carcinogen activity separately. Then, the PC's are ranked based on their decreasing correlating ability and entered to the input layer of the network one after another. Finally, a search algorithm (i.e. genetic algorithm) is used to find the best set of PC's. Both the external and cross-validation methods are used to validate the performances of the resulting models. One is able to see that the results obtained by the PC-GA-ANN and CR-PC-ANN procedures are superior to those resulted from the EV-PC-ANN. Comparison of the results reveals that the results produced by the PC-GA-ANN algorithm are better than those produced by CR-PC-ANN. However, the difference is not significant.

Correlation ranking procedure for factor selection in PC-ANN modeling and application to ADMETox evaluation

A correlation ranking procedure is proposed for selection of factors in principal component-artificial neural network (PC-ANN). The model was applied in the ADMETox evaluation to predict the carcinogenesis activity of 60 organic solvents and the blood–brain barrier (BBB) partitioning of 115 diverse organic molecules. A total of 150 molecular descriptors, including quantum chemical, constitutional, topological and chemical descriptors were calculated. The resulted descriptors were subjected to principal component analysis (PCA), and a three-layered feed forward artificial neural network (ANN) model was employed to model the nonlinear relationship between the extracted principal components (PCs) and the activities. A correlation ranking procedure is proposed here to select the most relevant set of PCs. First, the nonlinear relationship between each one of the PCs was modeled by separate neural networks and the correlation ability of each PC with the activity data was determined. Then, the PCs were entered to the ANN model based on their decreasing correlation ability. The results supported that the proposed model could predict the carcinogenesis activity and logBBB of the organic compounds with percent relative error lower than 4%. Comparison of the results with two other existing factor selection methods named eigenvalue ranking (EV-PC-ANN) and genetic algorithm (PC-GA-ANN) revealed that the proposed model gave results near to the PC-GA-ANN method, while less accurate results were obtained by the EV-PC-ANN procedure.

SIMULTANEOUS DETERMINATION OF Os(VIII) AND Ru(IV) AS CATALYSTS THROUGH A SINGLE CATALYTIC KINETIC RUN USING PRINCIPAL COMPONENT ARTIFICIAL NEURAL NETWORK

ABSTRACT In this study, principal component artificial neural network (PC-ANN) is proposed for the simultaneous determination of Os(VIII) and Ru(IV) on the basis of their catalytic effects on the indicator reaction between Ce(IV) and As(III). The differential equations involving negative synergistic effects were solved numerically by the PC-ANN. The PC-ANN architectures were different for Os(VIII) and Ru(IV). The output of Os PC-ANN architecture was used as an input node for Ru PC-ANN architecture. This modification improves the capability of Ru PC-ANN model for prediction of Ru(IV) concentrations. Detection limits were 0.05 and 0.08 ng/mL for Os(VIII) and Ru(IV), respectively. Dynamic ranges for Os(VIII) and Ru(IV) were 0.22 to 4.50 and 0.45 to 9.00 ng/mL, respectively. The method was used for the determination of Os(VIII) and Ru(IV) in synthetic samples with the compositions similar to the real alloys.