A Bayesian sparse reconstruction method for fault detection and isolation

This article develops a Bayesian method for fault detection and isolation using a sparse reconstruction framework. The normal/training data is assumed to follow a signal‐plus‐noise model, and an indicator matrix is used to show whether the test data is from a faulty process. The distribution of the indicator matrix is modeled by a Laplacian distribution, which forces the indicator matrix to be a sparse one, and a Gibbs sampler is derived to obtain the estimation/reconstruction of the indicator matrix, the unobserved signals, and other parameters like signal mean, covariance, and noise variance. The faulty variables can then be detected and isolated by inspecting whether corresponding rows of the indicator matrix are zero. The proposed Bayesian approach is data driven; it allows for simultaneous fault detection and isolation. A simulation study and an industrial case study are used to test the performance of the proposed method. Copyright © 2015 John Wiley & Sons, Ltd.     

A novel multi‐mode data processing method and its application in industrial process monitoring

Multi‐mode process monitoring is a key issue often raised in industrial process control. Most multivariate statistical process monitoring strategies, such as principal component analysis (PCA) and partial least squares, make an essential assumption that the collected data follow a unimodal or Gaussian distribution. However, owing to the complexity and the multi‐mode feature of industrial processes, the collected data usually follow different distributions. This paper proposes a novel multi‐mode data processing method called weighted k neighbourhood standardisation (WKNS) to address the multi‐mode data problem. This method can transform multi‐mode data into an approximately unimodal or Gaussian distribution. The results of theoretical analysis and discussion suggest that the WKNS strategy is more suitable for multi‐mode data normalisation than the z‐score method is. Furthermore, a new fault detection approach called WKNS‐PCA is developed and applied to detect process outliers. This method does not require process knowledge and multi‐mode modelling; only a single model is required for multi‐mode process monitoring. The proposed method is tested on a numerical example and the Tennessee Eastman process. Finally, the results demonstrate that the proposed data preprocessing and process monitoring methods are particularly suitable and effective in multi‐mode data normalisation and industrial process fault detection. Copyright © 2014 John Wiley & Sons, Ltd.     

The statistics of ToF‐SIMS data revisited and introduction of the empirical Poisson correction

Generation of time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS) data involves two overarching processes: secondary ion production and secondary ion detection. The interpretation of ToF‐SIMS data is facilitated if the intensities of the as‐measured mass spectra are proportional to the abundances of the species under investigation. While secondary ion yield is normally taken to be a linear process, ion detection is not owing to detector dead‐time effects. Consequently, methods have been devised that attempt to linearize, or correct, data that are affected by the dead time. In this article, we review the statistics of ToF‐SIMS data generation and confirm a report in the literature that abundance estimates from so‐called Poisson corrections are biased. We show that these corrections are only unbiased asymptotically and that a rigorous probabilistic analysis can quantitatively account for the observed bias. Two sources of bias are identified, one having a statistical basis and one due to the form of the correction equation at high ion detection rates. Based on insights gained from this analysis, we propose a new correction equation, the empirical Poisson correction, which largely eliminates the statistical bias. The performance of the proposed correction is illustrated by reanalyzing 14 experimentally measured datasets that suffer from varying levels of dead‐time effects. Copyright © 2016 John Wiley & Sons, Ltd.     

Differentially expressed genes selection via Laplacian regularized low-rank representation method

With the rapid development of DNA microarray technology and next-generation technology, a large number of genomic data were generated. So how to extract more differentially expressed genes from genomic data has become a matter of urgency. Because Low-Rank Representation (LRR) has the high performance in studying low-dimensional subspace structures, it has attracted a chunk of attention in recent years. However, it does not take into consideration the intrinsic geometric structures in data.In this paper, a new method named Laplacian regularized Low-Rank Representation (LLRR) has been proposed and applied on genomic data, which introduces graph regularization into LRR. By taking full advantages of the graph regularization, LLRR method can capture the intrinsic non-linear geometric information among the data. The LLRR method can decomposes the observation matrix of genomic data into a low rank matrix and a sparse matrix through solving an optimization problem. Because the significant genes can be considered as sparse signals, the differentially expressed genes are viewed as the sparse perturbation signals. Therefore, the differentially expressed genes can be selected according to the sparse matrix. Finally, we use the GO tool to analyze the selected genes and compare the P-values with other methods.The results on the simulation data and two real genomic data illustrate that this method outperforms some other methods: in differentially expressed gene selection.     

Multimode process monitoring based on Bayesian method

Multimode process monitoring has recently attracted much attention both in academy and industry. Conventional methods assume that either the process data are Gaussian in each operation mode, or some process knowledge should be incorporated, thus making the methods supervised. In this paper, a new unsupervised method is developed for multimode process monitoring, which is based on Bayesian inference and two‐step independent component analysis–principal component analysis (ICA–PCA) feature extraction strategy. ICA–PCA is first introduced for feature extraction and dimension reduction. By transferring the traditional monitoring statistic to fault probability in each operation mode, monitoring results in different operation modes can be easily combined by the Bayesian inference. Another contribution of the present paper is the development of a new fault identification method. Through analyses of the posterior probability and the joint probability for the monitored data sample, the correct operation mode or fault scenario can be identified. Three case studies are demonstrated to evaluate the feasibility and efficiency of the proposed method. Copyright © 2009 John Wiley & Sons, Ltd.     

Probabilistic model‐based discriminant analysis and clustering methods in chemometrics

In chemometrics, the supervised and unsupervised classification of high‐dimensional data has become a recurrent problem. Model‐based techniques for discriminant analysis and clustering are popular tools, which are renowned for their probabilistic foundations and their flexibility. However, classical model‐based techniques show a disappointing behaviour in high‐dimensional spaces, which up to now have been limited in their use within chemometrics. The recent developments in model‐based classification overcame these drawbacks and enabled the efficient classification of high‐dimensional data, even in the ‘small n / large p’ condition. This work presents a comprehensive review of these recent approaches, including regularization‐based techniques, parsimonious modelling, subspace classification methods and classification methods based on variable selection. The use of these model‐based methods is also illustrated on real‐world classification problems in chemometrics using R packages. Copyright © 2013 John Wiley & Sons, Ltd.     

Using Unfold‐PCA for batch‐to‐batch start‐up process understanding and steady‐state identification in a sequencing batch reactor

In chemical and biochemical processes, steady‐state models are widely used for process assessment, control and optimisation. In these models, parameter adjustment requires data collected under nearly steady‐state conditions. Several approaches have been developed for steady‐state identification (SSID) in continuous processes, but no attempt has been made to adapt them to the singularities of batch processes. The main aim of this paper is to propose an automated method based on batch‐wise unfolding of the three‐way batch process data followed by a principal component analysis (Unfold‐PCA) in combination with the methodology of Brown and Rhinehart 2 for SSID. A second goal of this paper is to illustrate how by using Unfold‐PCA, process understanding can be gained from the batch‐to‐batch start‐ups and transitions data analysis. The potential of the proposed methodology is illustrated using historical data from a laboratory‐scale sequencing batch reactor (SBR) operated for enhanced biological phosphorus removal (EBPR). The results demonstrate that the proposed approach can be efficiently used to detect when the batches reach the steady‐state condition, to interpret the overall batch‐to‐batch process evolution and also to isolate the causes of changes between batches using contribution plots. Copyright © 2007 John Wiley & Sons, Ltd.     

A new strategy of outlier detection for QSAR/QSPR

The crucial step of building a high performance QSAR/QSPR model is the detection of outliers in the model. Detecting outliers in a multivariate point cloud is not trivial, especially when several outliers coexist in the model. The classical identification methods do not always identify them, because they are based on the sample mean and covariance matrix influenced by the outliers. Moreover, existing methods only lay stress on some type of outliers but not all the outliers. To avoid these problems and detect all kinds of outliers simultaneously, we provide a new strategy based on Monte‐Carlo cross‐validation, which was termed as the MC method. The MC method inherently provides a feasible way to detect different kinds of outliers by establishment of many cross‐predictive models. With the help of the distribution of predictive residuals such obtained, it seems to be able to reduce the risk caused by the masking effect. In addition, a new display is proposed, in which the absolute values of mean value of predictive residuals are plotted versus standard deviations of predictive residuals. The plot divides the data into normal samples, y direction outliers and X direction outliers. Several examples are used to demonstrate the detection ability of MC method through the comparison of different diagnostic methods. © 2009 Wiley Periodicals, Inc. J Comput Chem, 2010     

Aligning retention time shifts in HPLC three‐dimensional spectra by icoshift approach combined with data arrangement methods and the release of a graphical user interface

High‐performance liquid chromatography coupled with photodiode array detection has been extensively applied in many fields and the peaks among the analyzed samples can be shifted due to the variations of instrumental and experimental conditions. In multivariate analysis, retention time alignment is an important pretreatment step. Hence, the shifted peaks in high‐performance liquid chromatography coupled with photodiode array detection three‐dimensional spectra should be aligned for further analysis. Being motivated by this purpose, the interval correlated shifting method combined with the proposed data arrangement methods are recommended and employed on high‐performance liquid chromatography coupled with photodiode array detection data as a demonstration. We validate the alignment performance of the proposed method through comparison the consistency of the retention time before and after alignment. The obtained results demonstrated that the proposed method is capable of successful aligning the employed data. Additionally, the interval correlated shifting method combined with the data arrangement modes is implemented in an easy‐to‐use graphical user interface environment and so can be operated easily by users not familiar with programming languages.     

A novel fast method for aqueous derivatization of THC,OH‐THC and THC‐COOH in human whole blood and urine samples for routine forensic analyses

A novel aqueous in situ derivatization procedure with propyl chloroformate (PCF) for the simultaneous, quantitative analysis of Δ⁹‐tetrahydrocannabinol (THC), 11‐hydroxy‐Δ⁹‐tetrahydrocannabinol (OH‐THC) and 11‐nor‐Δ⁹‐tetrahydrocannabinol‐carboxylic acid (THC‐COOH) in human blood and urine is proposed. Unlike current methods based on the silylating agent [N,O‐bis(trimethylsilyl)trifluoroacetamide] added in an anhydrous environment, this new proposed method allows the addition of the derivatizing agent (propyl chloroformate, PCF) directly to the deproteinized blood and recovery of the derivatives by liquid–liquid extraction. This novel method can be also used for hydrolyzed urine samples. It is faster than the traditional method involving a derivatization with trimethyloxonium tetrafluoroborate. The analytes are separated, detected and quantified by gas chromatography–mass spectrometry in selected ion monitoring mode (SIM). The method was validated in terms of selectivity, capacity of identification, limits of detection (LOD) and quantification (LOQ), carryover, linearity, intra‐assay precision, inter‐assay precision and accuracy. The LOD and LOQ in hydrolyzed urine were 0.5 and 1.3 ng/mL for THC and 1.2 and 2.6 ng/mL for THC‐COOH, respectively. In blood, the LOD and LOQ were 0.2 and 0.5 ng/mL for THC, 0.2 and 0.6 ng/mL for OH‐THC, and 0.9 and 2.4 ng/mL for THC‐COOH, respectively. This method was applied to 35 urine samples and 50 blood samples resulting to be equivalent to the previously used ones with the advantage of a simpler method and faster sample processing time. We believe that this method will be a more convenient option for the routine analysis of cannabinoids in toxicological and forensic laboratories.     

Lab‐on‐a‐Chip Reactor Imaging with Unprecedented Chemical Resolution by Hadamard‐Encoded Remote Detection NMR

The development of microfluidic processes requires information‐rich detection methods. Here we introduce the concept of remote detection exchange NMR spectroscopy (RD‐EXSY), and show that, along with indirect spatial information extracted from time‐of‐flight data, it provides unique information about the active regions, reaction pathways, and intermediate products in a lab‐on‐a‐chip reactor. Furthermore, we demonstrate that direct spatial resolution can be added to RD‐EXSY efficiently by applying the principles of Hadamard spectroscopy.     

Screening free radical scavengers in Xiexin Tang by HPLC‐ABTS‐DAD‐Q‐TOF/MS

Xiexin Tang (XXT) is a traditional Chinese medicine (TCM) that has been used in herbal clinics for more than 1800 years. Many studies have shown that XXT has therapeutic effects on patients with arteriosclerosis owing to its antioxidant activity. However, there is little information about the relationship between the chemical composition of XXT and its antioxidant activity. In this study, the HPLC‐ABTS‐DAD‐Q‐TOF/MS method, which can simultaneously identify individual components and rapidly screen for antioxidant compounds, was used to screen and identify antioxidant components in XXT. The 15 compounds identified were gluco‐syringic acid, adenine, gallic acid, biflorin, cularine, 6‐C ‐arabinose‐8‐C ‐glucose‐chrysin, 6‐C ‐glucose‐8‐C ‐arabinose–chrysin, baicalin, rhein‐8‐O‐β ‐d ‐glucopyranoside, coptisine, epiberberine, jatrorrhizine, norwogonin, 5,7,2′‐trihydroxy‐6‐ methoxyflavone and baicalein. In addition, the data showed that the antioxidant activity of peaks 4, 6, and 11 was lower in XXT than in its constituent herbs, while the activity of peaks 1, 2, 3, 5, 7, 8, 10, 12, 13, 14 and 15 was higher in XXT. Compound 5 had the strongest antioxidant activity in XXT, while compound 1 showed the strongest antioxidant activity among its constituent herb. The differences between antioxidant activities of major components of XXT and those of its constituent herbs might be due to the interaction of crude drugs that changes the solubility of active components during the decoction process. The results show that the HPLC‐ABTS‐DAD‐Q‐TOF/MS method can successfully combine on‐line mass spectrometry with activity detection system. It is a useful tool for the rapid detection and identification of antioxidants, and for quantitative analysis of individual antioxidants in complex mixtures such as plant extracts. Furthermore, this method does not require extensive extract purification and fraction collection.     

Complete determination of plant tissues based only on auto‐fluorescence and the advanced image analysis – study of needles and stamens

Proper determination of tissues is one of the challenging problems in modern medicine and histology. Currently, interpretation of the results mainly depends on the experience of a histologist, leading to high percentage of results misinterpretation. Bearing in mind potential application, we proposed the set of procedures that allow us to obtain precise, mathematically determined parameters for tissue discrimination. First, the method was tested on simulated set of images and compared with several other algorithms. As the set of experimentally obtained input data, auto‐fluorescence images of needle cross sections (Picea omorika) and stamens of common centaury (Centaurium erythraea) were used. Determination of cell types is based on inherent features of plant cells – auto‐fluorescence. As each cell type consists of various fluorescent components in different quantities for each type of tissue, its integral emission spectrum can be used as the fingerprint for identification. Cross sections were imaged using four sets of filters for detection of fluorescence (both excitation and emission). Such filter set is standard equipment for most fluorescence microscopes. One additional image was transmission image using the same optics. By applying ℓ₀‐norm‐constrained nonnegative matrix factorization in a space induced by explicit feature maps, it is possible to identify up to 11 tissues in needles and five in stamens (actual number of tissues). In comparison to other image analysis methods, the greatest advantage is the fact that the number of extracted components significantly exceeds the number of initial images while most other techniques can extract only as much components as the number of initial images. Copyright © 2015 John Wiley & Sons, Ltd.     

Fault diagnosis using kNN reconstruction on MRI variables

In industrial processes, investigating the root causes of abnormal events is a crucial task when process faults are detected; isolating the faulty variables provides additional information for investigating the root causes of the faults. The traditional contribution plot is a popular and perspicuous tool to isolate faulty variables. However, this method can only determine one faulty variable (the biggest contributor) when there are several variables out of control at the same time. In the presented work, a novel fault diagnosis method is derived using k‐nearest neighbor (kNN) reconstruction on maximize reduce index (MRI) sensors; it is aimed at identifying all fault variables precisely. This method can identify the faulty variables effectively through reconstructing MRI variables one by one. A numerical example focuses on validating the performance of kNN missing data analysis method firstly, then multi‐sensors fault identification results are also given. Tennessee Eastman process is provided to demonstrate that the proposed approach can identify the responsible variables for the multiple sensors fault. Copyright © 2015 John Wiley & Sons, Ltd.     

Determination of chlorogenic acid,polyphenols and antioxidants in green coffee by thin‐layer chromatography,effect‐directed analysis and dot blot – comparison to HPLC and spectrophotometry methods

The great prevalence of thin‐layer chromatography over high‐performance liquid chromatography is connected with the possibility of analyzing many samples in parallel. Therefore, the method is often used in screening and/or effect directed analysis to compare composition and chemical/biological properties of many samples in one run. It was already proved, that high performance thin‐layer chromatography, in many cases, can replace high‐performance liquid chromatography for quantitative analysis. The main aim of the paper is to show that simple thin‐layer chromatography can also be used as a quantitative or at least as a semi‐quantitative method, even when it concerns effect directed analysis e.g. direct bioautography. Chlorogenic acid content was measured in four methanol extracts of various green coffees and in one extract of black coffee using thin‐layer chromatography with ultraviolet detection and thin‐layer chromatography with effect directed detection. High‐performance liquid chromatography was used as a reference method. Additionally, total contents of polyphenols and antioxidants were estimated using thin‐layer chromatography or dot‐blot on chromatography plates. These results were compared to spectrophotometric methods. It was proved that thin‐layer chromatography can be used as a quantitative (using densitometry) or semi‐quantitative method (using other detection methods including effect directed detection) as well as for estimating total antioxidants or polyphenols content.     

Isotope pattern vector based tandem mass spectral data calibration for improved peptide and protein identification

Tandem mass spectra contain noisy peaks which make peak picking for peptide identification difficult. Moreover, all spectral peaks can be shifted due to systematic measurement errors. In this paper, a novel use of an isotope pattern vector (IPV) is proposed for denoising and systematic measurement error prediction. By matching the experimental IPVs with the theoretical IPVs of candidate fragment ions, true ionic peaks can be identified. Furthermore, these identified experimental IPVs and their corresponding theoretical IPVs are used in an optimization process to predict the systematic measurement error associated with the target spectrum. In return, the subsequent spectral data calibration based on the predicted systematic measurement error enhances the data quality. We show that such an integrated denoising and calibration process leads to significantly improved peptide and protein identification. Different from the commonly employed chemical calibration methods, our IPV‐based method is a purely computational method for individual spectra analysis and globally optimizes the use of spectral data. Copyright © 2009 John Wiley & Sons, Ltd.     

Interrelationships between generalized Tikhonov regularization,generalized net analyte signal,and generalized least squares for desensitizing a multivariate calibration to interferences

Orthogonal pre‐processing (orthogonal projection) of spectral data is a common approach to generate analyte‐specific information for use in multivariate calibration. The goal of this pre‐processing is to remove from each spectrum the respective sample interferent contributions (spectral interferences from overlap, scatter, noise, etc.). Two approaches to accomplish orthogonal pre‐processing are net analyte signal (NAS) and generalized least squares (GLS). Developed in this paper is the mathematical relationship between NAS and GLS. It is also realized that orthogonal NAS pre‐processing can remove too much analyte signal and that the degree of interferent correction can be regulated. Similar to GLS, the degree of correction is accomplished by using a regularization (tuning) parameter to form generalized NAS (GNAS). Also developed in this paper is an alternative to GNAS and GLS based on generalized Tikhonov regularization (GTR). The mathematical relationships between GTR, GNAS, and GLS are derived. A result is the ability to express the model vector as the sum of two contributions: the orthogonal NAS contribution and a non‐NAS contribution from the interferent components. Thus, rather than the usual situation of sequentially pre‐processing data by either GNAS or GLS followed by model building with the pre‐processed data, the methods of GTR, GNAS, and GLS are expressed as direct computations of model vectors allowing concurrent pre‐processing and model building to occur. Simultaneous pre‐processing and model forming are shown to be natural to the GTR process. Two near‐infrared spectroscopic data sets are studied to compare the theoretical relationships between GTR, GNAS, and GLS. One data set covers basic calibration, and the other data set is for calibration maintenance. Filter factor representation is key to developing the interprocess relationships. Copyright © 2013 John Wiley & Sons, Ltd.     

Precursor ion scans for the targeted detection of stable‐isotope‐labeled peptides

Stable isotope labels are routinely introduced into proteomes for quantification purposes. Full labeling of cells in varying biological states, followed by sample mixing, fractionation and intensive data acquisition, is used to obtain accurate large‐scale quantification of total protein levels. However, biological processes often affect only a small group of proteins for a short time, resulting in changes that are difficult to detect against the total proteome background. An alternative approach could be the targeted analysis of the proteins synthesized in response to a given biological stimulus. Such proteins can be pulse‐labeled with a stable isotope by metabolic incorporation of ‘heavy’ amino acids. In this study we investigated the specific detection and identification of labeled proteins using acquisition methods based on Precursor Ion Scans (PIS) on a triple‐quadrupole ion trap mass spectrometer. PIS‐based methods were set to detect unique immonium ions originating from labeled peptides. Different labels and methods were tested in standard mixtures to optimize performance. We showed that, in comparison with an untargeted analysis on the same instrument, the approach allowed a several‐fold increase in the specificity of detection of labeled proteins over unlabeled ones. The technique was applied to the identification of proteins secreted by human cells into growth media containing bovine serum proteins, allowing the preferential detection of labeled cellular proteins over unlabeled bovine ones. However, compared with untargeted acquisitions on two different instruments, the PIS‐based strategy showed some limitations in sensitivity. We discuss possible perspectives of the technique. Copyright © 2009 John Wiley & Sons, Ltd.     

Integrative graph regularized matrix factorization for drug-pathway associations analysis

Pathway-based drug discovery can give full consideration to the efficacy of compounds in the systemic physiological environment. The recently emerged drug-pathway association identification approaches gain popularity due to its potential to decipher the mechanism of action and the targets of compounds. In this study, we propose a novel drug-pathway association identification method: Integrative Graph regularized Matrix Factorization (IGMF). It employs graph regularization to encode data geometrical information and prevent possible overfitting in prediction. Furthermore, it achieves parts-based and sparse data representation by imposing L₁-norm regularization on the objective function.Empirical studies demonstrate that IGMF has strong advantages in identifying more new drug-pathway associations compared to its peer methods. It further shows a good capability to unveil the intrinsic structures of data. As an effective drug-pathway discovery method, it will inspire new analytics methods in this subfield.     

Rapid and sensitive identification of ricin in environmental samples based on lactamyl agarose beads using LC‐MS/MS (MRM)

Ricin, a plant‐derived toxin extracted from the seeds of Ricinus communis (castor bean plant), is one of the most toxic proteins known. Ricin's high toxicity, widespread availability, and ease of its extraction make it a potential agent for bioterrorist attacks. Most ricin detection methods are based on immunoassays. These methods may suffer from low efficiency in matrices containing interfering substances, or from false positive results due to antibody cross reactivity, with highly homologous proteins. In this study, we have developed a simple, rapid, sensitive, and selective mass spectrometry assay, for the identification of ricin in complex environmental samples. This assay involves three main stages: (a) Ricin affinity capture by commercial lactamyl‐agarose (LA) beads. (b) Tryptic digestion. (c) LC‐MS/MS (MRM) analysis of tryptic fragments. The assay was validated using 60 diverse environmental samples such as soil, asphalt, and vegetation, taken from various geographic regions. The assay's selectivity was established in the presence of high concentrations of competing lectin interferences. Based on our findings, we have defined strict criteria for unambiguous identification of ricin. Our novel method, which combines affinity capture beads followed by MRM‐based analysis, enabled the identification of 1 ppb ricin spiked into complex environmental matrices. This methodology has the potential to be extended for the identification of ricin in body fluids from individuals exposed (deliberately or accidentally) to the toxin, contaminated food or for the detection of the entire family of RIP‐II toxins, by applying multiplex format.