Capillary electrophoretic profiling and pattern recognition analysis of urinary nucleosides from thyroid cancer patients

Metabolic profiling analysis by capillary electrophoresis (CE) was combined with pattern recognition methods to see some correlation between urinary nucleoside levels and thyroid cancer. A total of 15 nucleosides were identified in urines from 12 female thyroid cancer patients and 12 healthy females studied. From the scatter plot evaluation, inosine showed the highest estimated diagnostic power with ca. 97.725% confidence level, followed by N²-methylguanosine. Star symbol graphs showed differences in levels of both minor and major nucleosides between cancer and normal groups more efficiently, compared with histogram. The stepwise discriminant analysis (SDA) selected N²-methylguanosine, N²,N²-dimethylguanosine and 1-methylguanosine as the most discriminating variables between thyroid cancer and normal groups. The canonical discriminant analysis (CDA) correctly classified all urine specimens studied into two separate clusters of cancer and normal groups in a canonical plot. The principal component analysis (PCA) distinguished cancer patients from normal controls in a principal component plot. The cluster analysis (CA) yielded a dendrogram displaying group separation without any single wrong linkage.     

Artificial neural networks based on genetic input selection for quantification in overlapped capillary electrophoresis peaks

The application of multilayer perceptron artificial neural networks (MLP ANN) based on genetic input selection for quantification of the unresolved peaks in micellar electrokinetic capillary chromatography (MECC) is reported. An optimization strategy for genetic input selection was also proposed. When the corresponding CE peaks cannot be resolved completely only by separation techniques, MLP ANN based on genetic input selection can be a suitable tool to resolve the problem. Both the spectra and the electrophoretograms of the unseparated analytes were used as the multivariate input data. The two kinds of the data were suitable for quantification of overlapped CE peaks by MLP ANN based on genetic input selection. The study also shows that the applying of genetic input selection in MLP ANN can improve the precision of quantification in both completely and partially overlapped CE peaks to some extent.     

Recognizing chromatographic peaks with pattern recognition methods III. Application of the algorithm for peak recognition in trace analysis

Summary The application of a newly developed peak recognition algorithm is shown. This algorith is based on the KNN method, one of the pattern recognition methods. It is shown that peaks with a S/N-ratio down to one can be safely recognized. This is also possible if the baseline has not only detector noise, but has other disturbances, e.g., noise signals which are generated by a reaction detector. The recognition ability of the algorithm is demonstrated by a standard chromatogram with three different concentrations and with two different sampling rates. The improvement against the classical algorithm is demonstrated. Some properties of the algorithm are discussed.     

Improved neural networks based on genetic algorithm for pulse recognition

Pulse diagnosis is an important part of Chinese medicine and has played an important role in the development of Chinese medical science. However, the pulse is traditionally determined by cutting it off, which leads to a lack of objective standard pulse identification methods and affects their accuracy and feasibility. This research has studied and discussed the processing and identification of four kinds of pulse: normal pulse, wiry pulse, smooth pulse, and thready pulse. Four frequency-domain characteristics of the pulse wave and six kinds of wavelet scale energy characteristic information were extracted, and a three-layer BP (backprocessing) neural network was established. The LM (Levenberg–Marquard) algorithm and a genetic algorithm were used to improve the BP neural network, to train on and predict experimental samples, and to obtain classification accuracies of 90% and 95% respectively. Moreover, improved BP neural network based on a genetic algorithm has shown highly superior performance in terms of convergence speed and low error rate.     

Classification and pattern recognition of acyclic octenes based on mass spectra

Two SIMCA models were developed for the classification of acyclic octene isomers, which only form a fraction of a very complex product mixture obtained, for example, from the transformation of 1-butene. The effects of spectral transformation, namely autocorrelation and logarithmic intensity ratios transforms, and (square-root) scaling of the octane isomers mass-spectral data were investigated. Both the spectral-features preprocessing methods and scaling were found to be vital for an adequate development and improvement of the classification models. The best SIMCA models were successfully applied on gas-chromatography-mass spectroscopy (GC-MS) analysis collected from the dimerization of 1-butene over heterogeneous catalysts in the liquid phase.     

Reduced bio basis function neural network for identification of protein phosphorylation sites: comparison with pattern recognition algorithms

Protein phosphorylation is a post-translational modification performed by a group of enzymes known as the protein kinases or phosphotransferases (Enzyme Commission classification 2.7). It is essential to the correct functioning of both proteins and cells, being involved with enzyme control, cell signalling and apoptosis. The major problem when attempting prediction of these sites is the broad substrate specificity of the enzymes. This study employs back-propagation neural networks (BPNNs), the decision tree algorithm C4.5 and the reduced bio-basis function neural network (rBBFNN) to predict phosphorylation sites. The aim is to compare prediction efficiency of the three algorithms for this problem, and examine knowledge extraction capability. All three algorithms are effective for phosphorylation site prediction. Results indicate that rBBFNN is the fastest and most sensitive of the algorithms. BPNN has the highest area under the ROC curve and is therefore the most robust, and C4.5 has the highest prediction accuracy. C4.5 also reveals the amino acid 2 residues upstream from the phosporylation site is important for serine/threonine phosphorylation, whilst the amino acid 3 residues upstream is important for tyrosine phosphorylation.     

A pattern recognition approach to 14-epi-hydrophenanthrene core of the morphine alkaloids based on shikimic acid

An expeditious and stereoselective construction of C-14-epimer of the tetracyclic hydrophenanthrene framework of the morphine alkaloids is described. The core structure is obtained in nine steps in 11% overall yield from shikimic acid via three key transformations: (i) coupling of shikimic acid with 2-iodoisovanillin at C-3 by double S_N2 inversion to form the aryl ether 5; (ii) an intramolecular Heck reaction to construct the dihydrobenzofuran ring and (iii) a McMurry coupling to furnish the hydrophenanthrene core.     

ToF‐SIMS for the characterization of hyperbranched aliphatic polyesters: probing their molecular weight on surfaces based on principal component analysis (PCA)

A series of 2,2‐bis(hydroxymethyl)propionic acid (Bis‐MPA) hyperbranched aliphatic polyesters with different molecular weights (generations) is analysed for the first time by time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS). The main negative and positive low‐mass fragments are identified in the fingerprint part of the spectra (m/z < 400) and are principally assigned to fragmentation of the Bis‐MPA repeating units. In addition, it is shown that the fragmentation pattern is highly affected by the functional end‐groups. This is illustrated for a phthalic acid end‐capped hyperbranched polymer and for an acetonide‐terminated dendrimer analog. Also, typical fragments assigned to the ethoxylated pentaerythritol core molecule are detected. These ions show decreasing intensities with increasing molecular weight. This intensity dependency on the generation is used to calibrate the molecular weight of hyperbranched polyesters on the surface. To obtain quantitative information, a principal component analysis (PCA) multivariate statistical method is applied to the ToF‐SIMS data. The influence of different normalization procedures prior to PCA calculation is tested, e.g. normalization to the total intensity, to the intensities of ions assigned to the Bis‐MPA repeating unit or to intensities of fragments due to the core molecule. It is shown that only one principal component (PC1) is needed to describe most of the variance between the samples. In addition, PC1 takes into account the generation effect. However, different relationships between the PC1 scores and the hyperbranched mass average molecular weights are observed depending on the normalization procedure used. Normalization of data set ion intensities by ion intensities from the core molecule allows linearization of the SIMS intensities versus the molecular weight and allows the hyperbranched polymers to be discriminated up to the highest generations. In addition, PCA applied to ToF‐SIMS data provides an extended interpretation of the spectra leading to further identification of the correlated mass peaks, such as those of the Bis‐MPA repeating unit (terminal, dendritic and linear) and those of the core molecule. Finally, the work presented demonstrates the extreme potential of the static ToF‐SIMS and PCA techniques in the analysis of dendritic molecules on solid surfaces. Copyright © 2003 John Wiley & Sons, Ltd.     

Monitoring and classification of wastewater quality using supervised pattern recognition techniques and deterministic resolution of molecular absorption spectra based on multiwavelength UV spectra deconvolution

A field flow approach for the in situ monitoring of wastewater quality is developed and assessed in this work, based on a combination of methods employing deconvolution of molecular absorption spectra and in situ/on-line analysis of wastewater effluent of various origin. The approach involves in situ immersion probes to monitor basic physicochemical parameters followed by UV spectrum deconvolution in order to provide a rapid estimate of organic matter, suspended solids and nitrate and on-line analysis of phosphates in a fully automated setup. The collected data are then treated with a series of supervised pattern recognition techniques in order to classify wastewater effluent according to their origin in three major categories namely municipal, industrial and hospital. The results suggest that the method affords a good approximation of realistic concentrations, as determined by reference methods, while it affords a good classification among various wastewater effluents of different origin. In that manner, the method enables a rapid inference of treated wastewater quality and a robust assessment of treatment process state, especially with regards to violations of effluent quality parameters.     

Discrimination of Alismatis Rhizoma based on chromatographic fingerprint,multiple components quantification,and pattern recognition analysis

A simple, sensible, and reliable HPLC–DAD method was first developed for fingerprint analysis of Alismatis Rhizoma, and then applied to analyze 85 samples from three main cultivated areas. In all, 40 common fingerprint peaks were designated, and six of which were definitely identified. Then, the combinatory analysis using similarity evaluation, principal component analysis, and orthogonal partial least square discriminant analysis revealed clear chemical consistency between samples from Fujian and Jiangxi provinces and substantial differences between those from Fujian/Jiangxi and Sichuan provinces. Furthermore, six components were dug out as potential chemical markers for distinguishing Alismatis Rhizoma from different areas, among which five were qualified for quantitative analysis. In conclusion, the combination of chemical fingerprint, multiple components quantification, and pattern recognition analysis was rather powerful and useful in discriminating Alismatis Rhizoma from different regions, which was a benefit for quality control.     

Generalized Dice's coefficient‐based multi‐block principal component analysis with Bayesian inference for plant‐wide process monitoring

Plant‐wide process monitoring is challenging because of the complex relationships among numerous variables in modern industrial processes. The multi‐block process monitoring method is an efficient approach applied to plant‐wide processes. However, dividing the original space into subspaces remains an open issue. The loading matrix generated by principal component analysis (PCA) describes the correlation between original variables and extracted components and reveals the internal relations within the plant‐wide process. Thus, a multi‐block PCA method that constructs principal component (PC) sub‐blocks according to the generalized Dice coefficient of the loading matrix is proposed. The PCs corresponding to similar loading vectors are divided within the same sub‐block. Thus, the PCs in the same sub‐block share similar variational behavior for certain faults. This behavior improves the sensitivity of process monitoring in the sub‐block. A monitoring statistic T² corresponding to each sub‐block is produced and is integrated into the final probability index based on Bayesian inference. A corresponding contribution plot is also developed to identify the root cause. The superiority of the proposed method is demonstrated by two case studies: a numerical example and the Tennessee Eastman benchmark. Comparisons with other PCA‐based methods are also provided. Copyright © 2014 John Wiley & Sons, Ltd.     

基于主成分分析的骨碎补药材乙醇和环己烷提取物的高效液相色谱指纹图谱及指标成分的定量分析

建立了骨碎补药材乙醇和环己烷提取物的高效液相色谱(HPLC)指纹图谱,并利用主成分分析法(PCA)对指纹图谱进行统计分析,以各主要色谱峰的保留时间和峰面积为变量得到score图和loading图。在score图和loading图中,骨碎补的正品和非正品可明显区分,且揭示出对此区分贡献最大的4个潜在指标成分,其中已知成分为柚皮苷、新北美圣草苷和E-4-O-β-D-葡萄糖酰咖啡酸。同时测定了这3种成分在19批正品和非正品骨碎补药材中的含量,其中10批骨碎补药材正品中3种成分的含量为: 柚皮苷6.36～10.1 mg/g,新北美圣草苷5.14～9.21 mg/g,E-4-O-β-D-葡萄糖酰咖啡酸1.87～3.19 mg/g。该方法更全面地反映了药材的化学成分信息,并能从定性和定量两方面控制骨碎补药材的内在质量。     

Direct and two-stage data analysis procedures based on PCA, PLS-DA and ANN for ISE-based electronic tongue-Effect of supervised feature extraction

A novel strategy of data analysis for artificial taste and odour systems is presented in this work. It is demonstrated that using a supervised method also in feature extraction phase enhances fruit juice classification capability of sensor array developed at Warsaw University of Technology. Comparison of direct processing (raw data processed by Artificial Neural Network (ANN), raw data processed by Partial Least Squares-Discriminant Analysis (PLS-DA)) and two-stage processing (Principal Components Analysis (PCA) outputs processed by ANN, PLS-DA outputs processed by ANN) is presented. It is shown that considerable increase of classification capability occurred in the case of the new method proposed by the authors.     

High‐performance liquid chromatography–based fingerprint analysis with chemical pattern recognition for evaluation of Mahonia bealei (Fort.) Carr

A simple and efficient high‐performance liquid chromatography method combined with chemical pattern recognition was established for quality evaluation of Mahonia bealei (Fort.) Carr. A common pattern of 30 characteristic peaks was applied for similarity analysis, hierarchical cluster analysis, principal component analysis, and partial least squares discriminant analysis in the 37 batches of M. bealei (Fort.) Carr. to discriminate wild M. bealei (Fort.) Carr., cultivated M. bealei (Fort.) Carr., and its substitutes. The results showed that partial least squares discriminant analysis was the most effective method for discrimination. Eight characteristics peaks with higher variable importance in projection values were selected for pattern recognition model. A permutation test and 26 batches of testing set samples were performed to validate the model that was successfully established. All of the training and testing set samples were correctly classified into three clusters (wild M. bealei (Fort.) Carr., cultivated M. bealei (Fort.) Carr., and its substitutes) based on the selected chemical markers. Moreover, 26 batches of unknown samples were used to predict the accuracy of the established model with a discrimination accuracy of 100%. The obtained results indicated that the method showed great potential application for accurate evaluation and prediction of the quality of M. bealei (Fort.) Carr.     

The determination for the three genotypes of D16S539 locus based on near-infrared spectroscopy and chemical pattern recognition

The paper has established an approach of typing short tandem repeats (STRs) based on the near-infrared spectroscopy (NIRS)-chemical pattern recognition. Taking the three genotypes 9-9, 9-11 and 11-11 of D16S539 locus as example, which have a middle degree of difference, DNA fragments containing the polymorphism sites were amplified by a pair of primers to obtain three genotypes samples; these samples were tested by the NIRS directly; using their spectra as recognition variables, the chemical pattern recognition models of the three genotypes were respectively established by using the principal discriminant variate (PDV) and support vector machine (SVM). The two models have a good fitting ability and strong prediction (i.e. the predicting accuracy was 100%). They are robust for these strong collinear spectra and the small number of the calibration samples. Without any preprocessing for the analyzed samples after PCR, the three genotypes of D16S539 locus could be indirectly determined by using the NIRS-s of the samples with the help of the models. This method is simple, rapid and low cost.     

A method for the detection of alcohol vapours based on optical sensing of magnesium 5,10,15,20-tetraphenyl porphyrin thin film by an optical spectrometer and principal component analysis

In this work we have proposed a method for the detection of alcohol vapours, i.e. methanol, ethanol and isopropanol, based on the optical sensing response of magnesium 5,10,15,20-tetraphenyl porphyrin (MgTPP) thin films, as measured by optical spectrometry with the assistance of chemometric analysis. We have implemented a scheme which allows a laboratory UV–vis spectrometer to act as a so-called “electronic nose” with very little modification. MgTPP thin films were prepared by a spin coating technique, using chloroform as the solvent, and then subjected to thermal annealing at 280 °C in an argon atmosphere. These MgTPP optical gas sensors presented significant responses with methanol compared to ethanol and isopropanol, based on the dynamic flow of alcohol vapours at the same mol% of alcohol concentration. Density functional theory (DFT) calculations were performed to model the underlying mechanism of this selectivity. The performance of the optical gas sensors was optimised by varying the fabrication parameters. It is hoped that the MgTPP thin film together with an off-the-shelf optical spectrometer and a simple chemometrics algorithm can be a valuable tool for the analysis of alcoholic content in the beverage industry.