Novel Boronlectins Based on Bispyridium Salt with a Flexible Linker: Discriminative Sensing of Lactose and Other Monosaccharides and Disaccharides in Aqueous Solution

N,N‐Di‐2‐picolylamine (DPA)‐derived diboronic acid receptors (NHBAs) with a flexible linker were designed and synthesized in this study, and two‐component sensing ensembles based on cationic NHBAs and an anionic fluorescent indicator 8‐hydroxypyrene‐1,3,6‐trisulfonic acid trisodium salt (HPTS) were successfully developed for both monosaccharides and disaccharides sensing. The dibranched ortho‐substituted receptor NHoBA exhibited unexpected selectivity towards lactose among five disaccharides used. The discrimination of five disaccharides and six monosaccharides was finally achieved by the integrated sensor array through linear discriminant analysis (LDA).     

Discrimination and molecular design of new theoretical hypolipaemic agents using the molecular connectivity functions

The molecular topology model and discriminant analysis have been applied to the prediction and QSAR interpretation of some pharmacological properties of hypolipaemic drugs using multivariable regression equations with their statistical parameters. Regression analysis showed that the molecular topology model predicts these properties. The corresponding stability (cross-validation) studies done on the selected prediction models confirmed the goodness of the fits. The method used for hypolipaemic activity selection was a linear discriminant analysis (LDA). We make use of the pharmacological distribution diagrams (PDDs) as a visualizing technique for the identification and design of new hypolipaemic agents.     

Molecular activity prediction by means of supervised subspace projection based ensembles of classifiers

This paper proposes a method for molecular activity prediction in QSAR studies using ensembles of classifiers constructed by means of two supervised subspace projection methods, namely nonparametric discriminant analysis (NDA) and hybrid discriminant analysis (HDA). We studied the performance of the proposed ensembles compared to classical ensemble methods using four molecular datasets and eight different models for the representation of the molecular structure. Using several measures and statistical tests for classifier comparison, we observe that our proposal improves the classification results with respect to classical ensemble methods. Therefore, we show that ensembles constructed using supervised subspace projections offer an effective way of creating classifiers in cheminformatics.     

Boosting partial least‐squares discriminant analysis with application to near infrared spectroscopic tea variety discrimination

In the present study, boosting has been combined with partial least‐squares discriminant analysis (PLS‐DA) to develop a new pattern recognition method called boosting partial least‐squares discriminant analysis (BPLS‐DA). BPLS‐DA is implemented by firstly constructing a series of PLS‐DA models on the various weighted versions of the original calibration set and then combining the predictions from the constructed PLS‐DA models to obtain the integrative results by weighted majority vote. Coupled with near infrared (NIR) spectroscopy, BPLS‐DA has been applied to discriminate different kinds of tea varieties. As comparisons to BPLS‐DA, the conventional principal component analysis, linear discriminant analysis (LDA), and PLS‐DA have also been investigated. Experimental results have shown that the inter‐variety difference can be accurately and rapidly distinguished via NIR spectroscopy coupled with BPLS‐DA. Moreover, the introduction of boosting drastically enhances the performance of an individual PLS‐DA, and BPLS‐DA is a well‐performed pattern recognition technique superior to LDA. Copyright © 2012 John Wiley & Sons, Ltd.     

Recognition of benzene, toluene and xylene using TGS array integrated with linear and non-linear classifier

Three volatile organic compounds (VOCs): benzene, toluene and xylene were measured with an array of six Taguchi gas sensors in the air with variable humidity content. The recognition of single compounds was performed, based on measurement results. The principal component analysis (PCA) pointed at humidity as the main classification factor in the measurement data set. The linear discriminant analysis (LDA) was applied to overcome this drawback and enforce classification with respect to benzene, toluene or xylene. It was shown that discriminant function analysis (DFA), which is an LDA method allowed for 100% success rate in test samples recognition of benzene. It did not allow for accurate recognition of test samples of toluene or xylene. Following, the non-linear classifier, radial basis function neural network (RBFNN) was applied. A specific configuration of input ‘s was found, which provided for successful recognition of each single compound: benzene, toluene or xylene in air with variable humidity content.     

Prediction of pesticide acute toxicity using two-dimensional chemical descriptors and target species classification

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.     

QSAR analysis of hypoglycemic agents using the topological indices

The molecular topology model and discriminant analysis have been applied to the prediction of some pharmacological properties of hypoglycemic drugs using multiple regression equations with their statistical parameters. Regression analysis showed that the molecular topology model predicts these properties. The corresponding stability (cross-validation) studies performed on the selected prediction models confirmed the goodness of the fits. The method used for hypoglycemic activity selection was a linear discriminant analysis (LDA). We make use of the pharmacological distribution diagrams (PDDs) as a visualizing technique for the identification and selection of new hypoglycemic agents, and we tested on rats the predictive ability of the model.     

Combining visible and near-infrared spectroscopy with chemometrics to trace muscles from an autochthonous breed of pig produced in Uruguay: a feasibility study

Visible (Vis) and near-infrared reflectance (NIR) spectroscopy combined with chemometrics was explored as a tool to trace muscles from autochthonous and crossbreed pigs from Uruguay. Muscles were sourced from two breeds, namely, the Pampa-Rocha (PR) and the Pampa-Rocha x Duroc (PRxD) crossbreed. Minced muscles were scanned in the Vis and NIR regions (400–2,500 nm) in a monochromator instrument in reflectance. Principal component analysis (PCA), discriminant partial least square regression (DPLS), linear discriminant analysis (LDA) based on PCA scores and soft independent modelling of class analogy (SIMCA) were used to identify the origin of the muscles based on Vis and NIR data. Full cross validation was used as validation method when classification models were developed. DPLS correctly classified 87% of PR and 78% of PRxD muscle samples. LDA calibration models correctly classified 87 and 67% of muscles as PR and PRxD, respectively. SIMCA correctly classified 100% of PR muscles. The results demonstrated the usefulness of Vis and NIR spectra combined with chemometrics as rapid method for authentication and identification of muscles according to the breed of pig.     

Qualitative analysis of green coffee by infrared spectrometry

Principal component analysis and linear discriminant analysis (LDA) were applied to the mid-infrared spectra for the qualitative analysis of the variety of green coffee (Arabica and Robusta). It is shown that the KBr pellet technique in combination with the LDA method can successfully be used for the identification of sample origin.     

Qualitative analysis of green coffee by infrared spectrometry

Principal component analysis and linear discriminant analysis (LDA) were applied to the mid-infrared spectra for the qualitative analysis of the variety of green coffee (Arabica and Robusta). It is shown that the KBr pellet technique in combination with the LDA method can successfully be used for the identification of sample origin.     

基于近红外光谱技术与化学计量学的绿茶无损鉴别方法研究

该文利用近红外光谱技术结合化学计量学方法开发了不同品种绿茶的无损鉴别方法。通过近红外光谱技术得到了8个品种绿茶样品的近红外光谱,比较了单一以及优化组合光谱预处理方法对光谱的影响,利用无监督的主成分分析(PCA)与有监督的线性判别分析方法(LDA)分别构建了茶叶品种鉴别模型。结果表明：对比单一预处理方法,优化组合预处理具有更优的鉴别准确性。标准正态变量变换预处理消除了茶叶样品大小不均造成的光谱散射影响,一阶导数预处理实现了变动背景的消除,减少了基线漂移的影响,突出了图谱中的有效信息,采用二者相结合的预处理方式并结合无监督的主成分分析法可实现较为准确的绿茶样品种类鉴别分析,准确率达75.0%。此外,采用有监督的线性判别分析方法处理原始光谱数据,可达到100%的鉴别准确率,但该方法需提供类别的先验知识。因此,采用近红外光谱技术和化学计量学相结合的手段可实现不同品种绿茶的快速无损鉴别。     

Comparative assessment of multiresponse regression methods for predicting the mechanisms of toxic action of phenols

The use of regression methods for classifying and predicting the mechanisms of toxic action of phenols was investigated in this study. Multiresponse regression was conducted using a total of six linear and nonlinear regression methods: simple linear regression (LinReg), logistic regression (LogReg), generalized additive model (GAM), locally weighted regression scatter plot smoothing (LOWESS), multivariate adaptive regression splines (MARS), and projection pursuit regression (PPR). A database containing phenols acting by four mechanisms (polar narcosis, weak acid respiratory uncoupling, proelectrophilicity, and soft electrophilicity) was used to assess the performances of the six regression methods in the multiresponse regression approach. For comparison purposes, traditional linear discriminant analysis (LDA) was also conducted as a baseline method to study the potential improvement of prediction accuracy by the multiresponse regression approach. Results showed that compared to LDA, the overall mechanism prediction error rate could be reduced to below 10% by multiresponse regression based on PPR. In addition to prediction accuracy, interpretability of the resultant models was discussed.     

核分段逆回归集成线性判别分析用于质谱数据分类

针对高维小样本质谱数据在构造模型时易产生的过拟合现象、变量间的严重共线性、及结构与性质间的非线性关系,采用了核分段逆回归(KSIR)特征提取集成线性判别分析(LDA)新技术。首先以KSIR算法完成质谱数据的非线性特征提取,然后在由新特征矢量张成的低维空间构造样本类别的线性判别函数,负责各样本个体类别的判定。将KSIR-LDA方法应用于软饮料的质谱数据分类,结果表明:该方法不仅适应质谱数据与性质间的非线性关系,而且可以更少、解释能力更强的特征变量取得更高的分类精度,并能实现在低维特征空间对数据的解释及可视化。     

A chromatochemometric approach for evaluating and selecting the perfume maceration time

A chemometric treatment of the data obtained by gas chromatography (GC) with flame ionization detector (FID) has been proposed to study the maceration time involved in perfumes manufacture with the final purpose of reducing this time but preserving the organoleptic characteristics of the perfume that is being elaborated. In this sense, GC–FID chromatograms were used as a fingerprint of perfume samples subjected to different maceration times, and data were treated by linear discriminant analysis (LDA), by comparing to a set of samples known to be macerated or not, which were used as calibration objects. The GC–FID methodology combined with the treatment of data by LDA has been applied successfully to seven different perfumes. The constructed LDA models exhibited excellent Wilks’ lambdas (0.013–0.118, depending on the perfume), and up to a reduction of 57% has been achieved with respect to the maceration time initially established.     

Computational chemistry approach for the early detection of drug-induced idiosyncratic liver toxicity

Idiosyncratic drug toxicity (IDT), considered as a toxic host-dependent event, with an apparent lack of dose response relationship, is usually not predictable from early phases of clinical trials, representing a particularly confounding complication in drug development. Albeit a rare event (usually <1/5000), IDT is often life threatening and is one of the major reasons new drugs never reach the market or are withdrawn post marketing. Computational methodologies, like the computer-based approach proposed in the present study, can play an important role in addressing IDT in early drug discovery. We report for the first time a systematic evaluation of classification models to predict idiosyncratic hepatotoxicity based on linear discriminant analysis (LDA), artificial neural networks (ANN), and machine learning algorithms (OneR) in conjunction with a 3D molecular structure representation and feature selection methods. These modeling techniques (LDA, feature selection to prevent over-fitting and multicollinearity, ANN to capture nonlinear relationships in the data, as well as the simple OneR classifier) were found to produce QSTR models with satisfactory internal cross-validation statistics and predictivity on an external subset of chemicals. More specifically, the models reached values of accuracy/sensitivity/specificity over 84%/78%/90%, respectively in the training series along with predictivity values ranging from ca. 78 to 86% of correctly classified drugs. An LDA-based desirability analysis was carried out in order to select the levels of the predictor variables needed to trigger the more desirable drug, i.e. the drug with lower potential for idiosyncratic hepatotoxicity. Finally, two external test sets were used to evaluate the ability of the models in discriminating toxic from nontoxic structurally and pharmacologically related drugs and the ability of the best model (LDA) in detecting potential idiosyncratic hepatotoxic drugs, respectively. The computational approach proposed here can be considered as a useful tool in early IDT prognosis.     

New discrimination method combining hit quality index based spectral matching and voting

A new discrimination method, called hit quality index (HQI)-voting, that uses the HQI for discriminant analysis has been developed. HQI indicates the degree of spectral matching between two spectra as known. In this method, a library sample yielding the highest HQI value for an unknown sample was initially searched and a group containing this sample was chosen as the group for the unknown sample. When overall spectral features of two groups are quite close to each other, many library samples with similar HQI values could be available for an unknown sample. In this situation, the simultaneous consideration of multiple votes (several library samples with close HQI values) for final decision would be more robust. In order to evaluate the discrimination performance of HQI-voting, three different near-infrared (NIR) spectroscopic datasets composed of two sample groups were used: (1) domestic and imported sesame samples, (2) domestic and imported Angelica gigas samples, and (3) diesel and light gas oil (LGO) samples. For the purpose of comparison, principal component analysis–linear discriminant analysis (PCA–LDA), partial least squares–discriminant analysis (PLS–DA) as well as k-nearest neighbor (k-NN) were also performed using the same datasets and the resulting accuracies were compared. The discrimination performances improved with the use of HQI-voting in comparison with those resulted from PCA–LDA and PLS–DA. The overall results support that HQI-voting is a comparable discrimination method to that of existing factor-based multivariate methods.     

Recognition for avian influenza virus proteins based on support vector machine and linear discriminant analysis

Total 200 properties related to structural characteristics were employed to represent structures of 400 HA coded proteins of influenza virus as training samples. Some recognition models for HA proteins of avian influenza virus (AIV) were developed using support vector machine (SVM) and linear discriminant analysis (LDA). The results obtained from LDA are as follows: the identification accuracy (Ria) for training samples is 99.8% and Ria by leave one out cross validation is 99.5%. Both Ria of 99.8% for training samples and Ria of 99.3% by leave one out cross validation are obtained using SVM model, respectively. External 200 HA proteins of influenza virus were used to validate the external predictive power of the resulting model. The external Ria for them is 95.5% by LDA and 96.5% by SVM, respectively, which shows that HA proteins of AIVs are preferably recognized by SVM and LDA, and the performances by SVM are superior to those by LDA.