Neural Network Modelling for Prediction of Zeta Potential

The study is focused on monitoring the influence of selected parameters on the zeta potential values of titanium dioxide nanoparticles. The influence of pH, temperature, ionic strength, and mass content of titanium dioxide in the suspension was assessed. More than a thousand samples were measured by combining these variables. On the basis of results, the model of artificial neural network was proposed and tested. The authors have rich experiences with neural networks applications and this case shows that the neural network model works with a very high prediction success rate of zeta potential. Clearly, pH has the greatest effect on zeta potential values. The influence of other variables is not so significant. However, it can be said that increasing temperature results in an increase in the value of the zeta potential of titanium dioxide nanoparticles. The ionic force affects the zeta potential depending on the pH; in the vicinity of the isoelectric point, its effect is negligible. The effect of the mass content of titanium dioxide in the suspension is absolutely minor.

Quantitative Structure Activity Relationship Models for the Antioxidant Activity of Polysaccharides

In this study, quantitative structure activity relationship (QSAR) models for the antioxidant activity of polysaccharides were developed with 50% effective concentration (EC50) as the dependent variable. To establish optimum QSAR models, multiple linear regressions (MLR), support vector machines (SVM) and artificial neural networks (ANN) were used, and 11 molecular descriptors were selected. The optimum QSAR model for predicting EC50 of DPPH-scavenging activity consisted of four major descriptors. MLR model gave EC50 = 0.033Ara-0.041GalA-0.03GlcA-0.025PC+0.484, and MLR fitted the training set with R = 0.807. ANN model gave the improvement of training set (R = 0.96, RMSE = 0.018) and test set (R = 0.933, RMSE = 0.055) which indicated that it was more accurately than SVM and MLR models for predicting the DPPH-scavenging activity of polysaccharides. 67 compounds were used for predicting EC50 of the hydroxyl radicals scavenging activity of polysaccharides. MLR model gave EC50 = 0.12PC+0.083Fuc+0.013Rha-0.02UA+0.372. A comparison of results from models indicated that ANN model (R = 0.944, RMSE = 0.119) was also the best one for predicting the hydroxyl radicals scavenging activity of polysaccharides. MLR and ANN models showed that Ara and GalA appeared critical in determining EC50 of DPPH-scavenging activity, and Fuc, Rha, uronic acid and protein content had a great effect on the hydroxyl radicals scavenging activity of polysaccharides. The antioxidant activity of polysaccharide usually was high in MW range of 4000-100000, and the antioxidant activity could be affected simultaneously by other polysaccharide properties, such as uronic acid and Ara.

Pattern recognition and classification of sediments according to their metal content using chemometric tools. A case study: the estuary of Nerbioi-Ibaizabal River (Bilbao, Basque Country)

Chemometrics are increasingly used in environmental monitoring studies, but are still far from being accepted as routine tools by field specialists. The multivariate character of usually highly correlated environmental data recommends the use of advanced chemometrics as part of the analytical methodology in order to get information on the basic structure of data. In this work, we have applied a battery of non-supervised (Principal Component Analysis (PCA)) and supervised (k-Nearest Neighbour (k-NN), Soft Independent Modelling of Class Analogies (SIMCA), Linear Discriminant Analysis (LDA) and Artificial Neural Networks (ANNs)) multivariate techniques on a specific environmental dataset. The dataset consists on the concentration of 14 elements (Al, As, Cd, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Sn, V and Zn) in 95 sediments collected at eight different locations of the estuary of the Nerbioi-Ibaizabal River (Bilbao, Basque Country) during 12 sampling campaigns conducted every 3 months between 2005 and 2008. The study aims to present a simple methodology of general applicability which may result in a flexible and practical tool to assess chemical pollution in sediments of a given specific site.

Prediction of carcinogenicity for diverse chemicals based on substructure grouping and SVM modeling

The Carcinogenicity Reliability Database (CRDB) was constructed by collecting experimental carcinogenicity data on about 1,500 chemicals from six sources, including IARC, and NTP databases, and then by ranking their reliabilities into six unified categories. A wide variety of 911 organic chemicals were selected from the database for QSAR modeling, and 1,504 kinds of different molecular descriptors were calculated, based on their 3D molecular structures as modeled by the Dragon software. Positive (carcinogenic) and negative (non-carcinogenic) chemicals containing various substructures were counted using atom and functional group count descriptors, and the statistical significance of ratios of positives to negatives was tested for those substructures. Very few were judged to be strongly related to carcinogenicity, among substructures known to be responsible for carcinogens as revealed from biomedical studies. In order to develop QSAR models for the prediction of the carcinogenicities of a wide variety of chemicals with a satisfactory performance level, the relationship between the carcinogenicity data with improved reliability and a subset of significant descriptors selected from 1,504 Dragon descriptors was analyzed with a support vector machine (SVM) method: the classification function (SVC) for weighted data in LIBSVM program was used to classify chemicals into two carcinogenic categories (positive or negative), where weights were set depending on the reliabilities of the carcinogenicity data. The quality and stability of the models presented were tested by performing a dual cross-validation procedure. A single SVM model as the first step was developed for all the 911 chemicals using 250 selected descriptors, achieving an overall accuracy level, i.e., positive and negative correct estimate, of about 70%. In order to improve the accuracy of the final model, the 911 chemicals were classified into 20 mutually overlapping subgroups according to contained substructures, a specific SVM model was optimized for each subgroup, and the predicted carcinogenicities of the 911 chemicals were determined by the majorities of the outputs of the corresponding SVM models. The model developed on the basis of grouping of chemicals into 20 substructures predicts the carcinogenicities of a wide variety of chemicals with a satisfactory overall accuracy of approximately 80%.

Artificial neural networks in biology and chemistry: the evolution of a new analytical tool

Once regarded as an eccentric and unpromising algorithm for the analysis of scientific data, the neural network has been developed in the last decade into a powerful computational tool. Its use now spans all areas of science, from the physical sciences and engineering to the life sciences and allied subjects. Applications range from the assessment of epidemiological data or the deconvolution of spectra to highly practical applications, such as the electronic nose. This introductory chapter considers briefly the growth in the use of neural networks and provides some general background in preparation for the more detailed chapters that follow.