Multiblock Principal Component Analysis Research Articles

A Novel Decentralized Process Monitoring Scheme Using a Modified Multiblock PCA Algorithm

Decentralized process monitoring based on purely data-based methods has recently gained considerable attention in multivariate statistical process monitoring circle. Although the process variables can be divided into several blocks automatically according to their statistical preferences, most of the existing multiblock modeling strategies tends to build local monitoring models individually, where the relevance among different blocks is ignored, and this leaves a room for enhancing process monitoring performance. Inspired by the recognition of this lack, a modified multiblock principal component analysis (MBPCA) algorithm is proposed for extracting block scores with respect to both specificity in each block and relevance among different blocks. Based on this sort of modeling strategy, a novel decentralized process monitoring is formulated by incorporating a PCA-based process decomposition strategy for block division, Bayesian inference to achieve decision fusion of fault detection, and reconstruction-based contribution plots for fault diagnosis. The superiority and validity of the proposed method is finally demonstrated through comparison studies on two simulated examples.

Metabolic analysis of the response of Pseudomonas putida DOT-T1E strains to toluene using Fourier transform infrared spectroscopy and gas chromatography mass spectrometry.

IntroductionAn exceptionally interesting stress response of Pseudomonas putida strains to toxic substances is the induction of efflux pumps that remove toxic chemical substances from the bacterial cell out to the external environment. To exploit these microorganisms to their full potential a deeper understanding of the interactions between the bacteria and organic solvents is required. Thus, this study focuses on investigation of metabolic changes in P. putida upon exposure to toluene.ObjectiveInvestigate observable metabolic alterations during interactions of three strains of P. putida (DOT-T1E, and its mutants DOT-T1E-PS28 and DOT-T1E-18) with the aromatic hydrocarbon toluene.MethodsThe growth profiles were measured by taking optical density (OD) measurement at 660 nm (OD660) at various time points during incubation. For fingerprinting analysis, Fourier-transform infrared (FT-IR) spectroscopy was used to investigate any phenotypic changes resulting from exposure to toluene. Metabolic profiling analysis was performed using gas chromatography-mass spectrometry (GC–MS). Principal component—discriminant function analysis (PC-DFA) was applied to the FT-IR data while multiblock principal component analysis (MB-PCA) and N-way analysis of variance (N-way ANOVA) were applied to the GC–MS data.ResultsThe growth profiles demonstrated the effect of toluene on bacterial cultures and the results suggest that the mutant P. putida DOT-T1E−18 was more sensitive (significantly affected) to toluene compared to the other two strains. PC-DFA on FT-IR data demonstrated the differentiation between different conditions of toluene on bacterial cells, which indicated phenotypic changes associated with the presence of the solvent within the cell. Fifteen metabolites associated with this phenotypic change, in P. putida due to exposure to solvent, were from central metabolic pathways. Investigation of MB-PCA loading plots and N-way ANOVA for condition | strain × time blocking (dosage of toluene) suggested ornithine as the most significant compound that increased upon solvent exposure.ConclusionThe combination of metabolic fingerprinting and profiling with suitable multivariate analysis revealed some interesting leads for understanding the mechanism of Pseudomonas strains response to organic solvent exposure.Electronic supplementary materialThe online version of this article (doi:10.1007/s11306-016-1054-1) contains supplementary material, which is available to authorized users.

Evaluation of metabolomics profiles of grain from maize hybrids derived from near-isogenic GM positive and negative segregant inbreds demonstrates that observed differences cannot be attributed unequivocally to the GM trait.

IntroductionPast studies on plant metabolomes have highlighted the influence of growing environments and varietal differences in variation of levels of metabolites yet there remains continued interest in evaluating the effect of genetic modification (GM).ObjectivesHere we test the hypothesis that metabolomics differences in grain from maize hybrids derived from a series of GM (NK603, herbicide tolerance) inbreds and corresponding negative segregants can arise from residual genetic variation associated with backcrossing and that the effect of insertion of the GM trait is negligible.MethodsFour NK603-positive and negative segregant inbred males were crossed with two different females (testers). The resultant hybrids, as well as conventional comparator hybrids, were then grown at three replicated field sites in Illinois, Minnesota, and Nebraska during the 2013 season. Metabolomics data acquisition using gas chromatography–time of flight-mass spectrometry (GC–TOF-MS) allowed the measurement of 367 unique metabolite features in harvested grain, of which 153 were identified with small molecule standards. Multivariate analyses of these data included multi-block principal component analysis and ANOVA-simultaneous component analysis. Univariate analyses of all 153 identified metabolites was conducted based on significance testing (α = 0.05), effect size evaluation (assessing magnitudes of differences), and variance component analysis.ResultsResults demonstrated that the largest effects on metabolomic variation were associated with different growing locations and the female tester. They further demonstrated that differences observed between GM and non-GM comparators, even in stringent tests utilizing near-isogenic positive and negative segregants, can simply reflect minor genomic differences associated with conventional back-crossing practices.ConclusionThe effect of GM on metabolomics variation was determined to be negligible and supports that there is no scientific rationale for prioritizing GM as a source of variation.Electronic supplementary materialThe online version of this article (doi:10.1007/s11306-016-1017-6) contains supplementary material, which is available to authorized users.

Combining DI-ESI-MS and NMR datasets for metabolic profiling.

Metabolomics datasets are commonly acquired by either mass spectrometry (MS) or nuclear magnetic resonance spectroscopy (NMR), despite their fundamental complementarity. In fact, combining MS and NMR datasets greatly improves the coverage of the metabolome and enhances the accuracy of metabolite identification, providing a detailed and high-throughput analysis of metabolic changes due to disease, drug treatment, or a variety of other environmental stimuli. Ideally, a single metabolomics sample would be simultaneously used for both MS and NMR analyses, minimizing the potential for variability between the two datasets. This necessitates the optimization of sample preparation, data collection and data handling protocols to effectively integrate direct-infusion MS data with one-dimensional (1D) 1H NMR spectra. To achieve this goal, we report for the first time the optimization of (i) metabolomics sample preparation for dual analysis by NMR and MS, (ii) high throughput, positive-ion direct infusion electrospray ionization mass spectrometry (DI-ESI-MS) for the analysis of complex metabolite mixtures, and (iii) data handling protocols to simultaneously analyze DI-ESI-MS and 1D 1H NMR spectral data using multiblock bilinear factorizations, namely multiblock principal component analysis (MB-PCA) and multiblock partial least squares (MB-PLS). Finally, we demonstrate the combined use of backscaled loadings, accurate mass measurements and tandem MS experiments to identify metabolites significantly contributing to class separation in MB-PLS-DA scores. We show that integration of NMR and DI-ESI-MS datasets yields a substantial improvement in the analysis of neurotoxin involvement in dopaminergic cell death.

Plant-wide process monitoring based on mutual information–multiblock principal component analysis

Multiblock principal component analysis (MBPCA) methods are gaining increasing attentions in monitoring plant-wide processes. Generally, MBPCA assumes that some process knowledge is incorporated for block division; however, process knowledge is not always available. A new totally data-driven MBPCA method, which employs mutual information (MI) to divide the blocks automatically, has been proposed. By constructing sub-blocks using MI, the division not only considers linear correlations between variables, but also takes into account non-linear relations thereby involving more statistical information. The PCA models in sub-blocks reflect more local behaviors of process, and the results in all blocks are combined together by support vector data description. The proposed method is implemented on a numerical process and the Tennessee Eastman process. Monitoring results demonstrate the feasibility and efficiency.

Monitoring multi-mode plant-wide processes by using mutual information-based multi-block PCA, joint probability, and Bayesian inference

A multi-mode plant-wide process monitoring scheme that integrates mutual information (MI)-based multi-block principal component analysis (PCA), joint probability, and Bayesian inference is developed. Given that the prior process is not always available, an MI-based block division method is proposed to divide blocks automatically by considering both cross-relations and high-order statistical information among variables. The PCA monitoring model is established in each sub-block and each mode, and a joint probability based on T2 statistics is defined to identify running-on mode. Then, the statistics in different sub-blocks are combined by using Bayesian inference to provide an intuitive indication. Finally, an improved contribution plot method is proposed to identify the root cause of faults. The feasibility and efficiency of the proposed method are evaluated by case studies on a numerical process and the Tennessee Eastman benchmark process. Monitoring results and comparisons with conventional PCA methods indicate the superiority of the proposed method.

Fault detection and identification using a Kullback-Leibler divergence based multi-block principal component analysis and bayesian inference

Considering the huge number of variables in plant-wide process monitoring and complex relationships (linear, nonlinear, partial correlation, or independence) among these variables, multivariate statistical process monitoring (MSPM) performance may be deteriorated especially by the independent variables. Meanwhile, whether related variables keep high concordance during the variation process is still a question. Under this circumstance, a multi-block technology based on mathematical statistics method, Kullback-Leibler Divergence, is proposed to put the variables having similar statistical characteristics into the same block, and then build principal component analysis (PCA) models in each low-dimensional subspace. Bayesian inference is also employed to combine the monitoring results from each sub-block into the final monitoring statistics. Additionally, a novel fault diagnosis approach is developed for fault identification. The superiority of the proposed method is demonstrated by applications on a simple simulated multivariate process and the Tennessee Eastman benchmark process.

Progressive multi-block modelling for enhanced fault isolation in batch processes

A multi-block progressive modelling approach is proposed for enhanced fault isolation in batch processes. The unfolding of batch data typically leads to matrices with a large number of columns and this complicates contribution analysis. In order to rapidly focus fault isolation in batch processes, it would be desirable to employ multi-block modelling techniques. Multi-block model such as consensus principal component analysis (CPCA) can produce multiple monitoring charts for sub-blocks and block loadings and block scores can be obtained which can represent unique behaviour of each sub-block. CPCA model uses super score which is the same as score from normal principal component analysis (PCA) model and it does not produce enhanced monitoring performance. Multi-block PCA (MBPCA) model using block score for model calculation can represent sub-blocks’ character but block scores are obtained from super loading so it may not be the best way to describe sub-blocks. A new MBPCA model is proposed for better expression of each sub-block. Through progressive modelling and contribution analysis, variables related to or affected by the fault, as well as the associated time information, are gradually identified. This enables a fault propagation path being established. The proposed method is applied to a benchmark simulated penicillin production process, PenSim.

Integrating multiple analytical platforms and chemometrics for comprehensive metabolic profiling: application to meat spoilage detection

Untargeted metabolic profiling has become a common approach to attempt to understand biological systems. However, due to the large chemical diversity in the metabolites it is generally necessary to employ multiple analytical platforms so as to encompass a wide range of metabolites. Thus it is beneficial to find chemometrics approaches which can effectively integrate data generated from multiple platforms and ideally combine the strength of each platform and overcome their inherent weaknesses; most pertinent is with respect to limited chemistries. We have reported a few studies using untargeted metabolic profiling techniques to monitor the natural spoilage process in pork and also to detect specific metabolites associated with contaminations with the pathogen Salmonella typhimurium. One method used was to analyse the volatile organic compounds (VoCs) generated throughout the spoilage process while the other was to analyse the soluble small molecule metabolites (SMM) extracted from the microbial community, as well as from the surface of the spoiled/contaminated meat. In this study, we exploit multi-block principal component analysis (MB-PCA) and multi-block partial least squares (MB-PLS) to combine the VoCs and SMM data together and compare the results obtained by analysing each data set individually. We show that by combining the two data sets and applying appropriate chemometrics, a model with much better prediction and importantly with improved interpretability was obtained. The MB-PCA model was able to combine the strength of both platforms together and generated a model with high consistency with the biological expectations, despite its unsupervised nature. MB-PLS models also achieved the best over-all performance in modelling the spoilage progression and discriminating the naturally spoiled samples and the pathogen contaminated samples. Correlation analysis and Bayesian network analysis were also performed to elucidate which metabolites were correlated strongly in the two data sets and such information could add additional information in understanding the meat spoilage process.

Deflation strategies for multi-block principal component analysis revisited

Within the framework of multi-block data sets, multi-block principal component analysis has been successfully used as a tool to investigate the structure of spectroscopy, -omics and sensory data. The determination of the successive principal components involves a deflation procedure which can be performed according to several strategies. We discuss the respective interest of these strategies and show orthogonality properties related to the vectors of loadings or to the scores. Reconstruction formulas for the data blocks are established for each deflation strategy. Interpretational aspects of the different deflation strategies are discussed and illustrated on the basis of a real and a simulated data set.

Quantitative Metabolomic Profiling of Serum, Plasma, and Urine by 1H NMR Spectroscopy Discriminates between Patients with Inflammatory Bowel Disease and Healthy Individuals

Serologic biomarkers for inflammatory bowel disease (IBD) have yielded variable differentiating ability. Quantitative analysis of a large number of metabolites is a promising method to detect IBD biomarkers. Human subjects with active Crohn’s disease (CD) and active ulcerative colitis (UC) were identified, and serum, plasma, and urine specimens were obtained. We characterized 44 serum, 37 plasma, and 71 urine metabolites by use of 1H NMR spectroscopy and “targeted analysis” to differentiate between diseased and non-diseased individuals, as well as between the CD and UC cohorts. We used multiblock principal component analysis and hierarchical OPLS-DA for comparing several blocks derived from the same “objects” (e.g., subject) to examine differences in metabolites. In serum and plasma of IBD patients, methanol, mannose, formate, 3-methyl-2-oxovalerate, and amino acids such as isoleucine were the metabolites most prominently increased, whereas in urine, maximal increases were observed for mannitol, allantoin, xylose, and carnitine. Both serum and plasma of UC and CD patients showed significant decreases in urea and citrate, whereas in urine, decreases were observed, among others, for betaine and hippurate. Quantitative metabolomic profiling of serum, plasma, and urine discriminates between healthy and IBD subjects. However, our results show that the metabolic differences between the CD and UC cohorts are less pronounced.

Multiblock principal component analysis: an efficient tool for analyzing metabolomics data which contain two influential factors

Principal component analysis (PCA) is probably one of the most used methods for exploratory data analysis. However, it may not be always effective when there are multiple influential factors. In this paper, the use of multiblock PCA for analysing such types of data is demonstrated through a real metabolomics study combined with a series of data simulating two underlying influential factors with different types of interactions based on 2 × 2 experiment designs. The performance of multiblock PCA is compared with those of PCA and also ANOVA-PCA which is another PCA extension developed to solve similar problems. The results demonstrate that multiblock PCA is highly efficient at analysing such types of data which contain multiple influential factors. These models give the most comprehensive view of data compared to the other two methods. The combination of super scores and block scores shows not only the general trends of changing caused by each of the influential factors but also the subtle changes within each combination of the factors and their levels. It is also highly resistant to the addition of ‘irrelevant’ competing information and the first PC remains the most discriminant one which neither of the other two methods was able to do. The reason of such property was demonstrated by employing a 2 × 3 experiment designs. Finally, the validity of the results shown by the multiblock PCA was tested using permutation tests and the results suggested that the inherit risk of over-fitting of this type of approach is low.

VOC-based metabolic profiling for food spoilage detection with the application to detecting Salmonella typhimurium-contaminated pork

In this study, we investigated the feasibility of using a novel volatile organic compound (VOC)-based metabolic profiling approach with a newly devised chemometrics methodology which combined rapid multivariate analysis on total ion currents with in-depth peak deconvolution on selected regions to characterise the spoilage progress of pork. We also tested if such approach possessed enough discriminatory information to differentiate natural spoiled pork from pork contaminated with Salmonella typhimurium, a food poisoning pathogen commonly recovered from pork products. Spoilage was monitored in this study over a 72-h period at 0-, 24-, 48- and 72-h time points after the artificial contamination with the salmonellae. At each time point, the VOCs from six individual pork chops were collected for spoiled vs. contaminated meat. Analysis of the VOCs was performed by gas chromatography/mass spectrometry (GC/MS). The data generated by GC/MS analysis were initially subjected to multivariate analysis using principal component analysis (PCA) and multi-block PCA. The loading plots were then used to identify regions in the chromatograms which appeared important to the separation shown in the PCA/multi-block PCA scores plot. Peak deconvolution was then performed only on those regions using a modified hierarchical multivariate curve resolution procedure for curve resolution to generate a concentration profiles matrix C and the corresponding pure spectra matrix S. Following this, the pure mass spectra (S) of the peaks in those region were exported to NIST 02 mass library for chemical identification. A clear separation between the two types of samples was observed from the PCA models, and after deconvolution and univariate analysis using N-way ANOVA, a total of 16 significant metabolites were identified which showed difference between natural spoiled pork and those contaminated with S. typhimurium.

Statistical Monitoring of Complex Chemical Processes Using Agent-Based Systems

It is highly desirable to have a statistical process monitoring (SPM) system that detects the abnormalities in process operations quickly with as few missed and false alarms as possible while the process operates under various operating conditions. An agent-based combined monitoring and fault detection framework is proposed in this study. In this framework, different SPM techniques compete with and complement each other to enhance detection speed and accuracy. SPM techniques from literature such as principal component analysis (PCA), multiblock PCA (MBPCA), and dynamic PCA (DPCA) techniques are implemented in this agent-based process supervision system. An agent performance assessment and agent management layer provides dynamic adaptation of the supervision system and improves the performance of SPM. The statistical information coming from each of the statistical techniques is summarized through a consensus mechanism. The performance of the agent-based consensus mechanism using different consensus criteri...

Assessment of adaptive focused acoustics versus manual vortex/freeze-thaw for intracellular metabolite extraction from Streptomyces lividans producing recombinant proteins using GC-MS and multi-block principal component analysis

We compared the gas chromatography-mass spectrometry (GC-MS) metabolite profiles of mouse tumour necrosis factor alpha (mTNF-alpha) secreting Streptomyces lividans TK24 to the non-secreting wild type and the wild type harbouring the empty pIJ486 plasmid by multi-block principal component analysis (PCA). The multi-block PCA model successfully identified peaks that were statistically different between the protein secreting and non-secreting strains, and at the same time also uncovered the efficiency of intracellular metabolite extraction by an ultrasonic adaptive focused acoustics (AFA) technique compared to a manual vortex/freeze-thaw method. Fifty-one metabolites were significantly different between the three biological strains and 17 of these were abundant in the mTNF-alpha secreting strain compared to the non-secreting strains. No significant differences in the number of detected metabolite peaks were observed between the two extraction techniques. However, from the loadings of the multi-block PCA model, as well as univariate statistical analysis, we observed that the relative peak response ratios to the internal standard of 10 metabolites were higher for the AFA extraction, suggesting a more efficient recovery of these metabolites than achieved with the manual vortex/freeze thaw method.

Revealing Covariance Structures in Fourier Transform Infrared and Raman Microspectroscopy Spectra: A Study on Pork Muscle Fiber Tissue Subjected to Different Processing Parameters

The aim of this study was to investigate the correlation patterns between Fourier transform infrared (FT-IR) and Raman microspectroscopic data obtained from pork muscle tissue, which helped to improve the interpretation and band assignment of the observed spectral features. The pork muscle tissue was subjected to different processing factors, including aging, salting, and heat treatment, in order to induce the necessary degree of variation of the spectra. For comparing the information gained from the two spectroscopic techniques with respect to the experimental design, multiblock principal component analysis (MPCA) was utilized for data analysis. The results showed that both FT-IR and Raman spectra were mostly affected by heat treatment, followed by the variation in salt content. Furthermore, it could be observed that IR amide I, II, and III band components appear to be effected to a different degree by brine-salting and heating. FT-IR bands assigned to specific protein secondary structures could be related to different Raman C-C stretching bands. The Raman C-C skeletal stretching bands at 1,031, 1,061, and 1,081 cm(-1) are related to the IR bands indicative of aggregated beta-structures, while the Raman bands at 901 cm(-1) and 934 cm(-1) showed a strong correlation with IR bands assigned to a alpha-helical structures. At the same time, the IR band at 1,610 cm(-1), which formerly was assigned to tyrosine in spectra originating from pork muscle, did not show a correlation to the strong tyrosine doublet at 827 and 852 cm(-1) found in Raman spectra, leading to the conclusion that the IR band at 1,610 cm(-1) found in pork muscle tissue is not originating from tyrosine.

Multiblock Principal Component Analysis Based on a Combined Index for Semiconductor Fault Detection and Diagnosis

The purposes of multivariate statistical process control (MSPC) are to improve process operations by quickly detecting when process abnormalities have occurred and diagnosing the sources of the process abnormalities. In the area of semiconductor manufacturing, increased yield and improved product quality result from reducing the amount of wafers produced under suboptimal operating conditions. This paper presents a complete MSPC application method that combines recent contributions to the field, including multiway principal component analysis (PCA), recursive PCA, fault detection using a combined index, and fault contributions from Hotelling's T/sup 2/ statistic. In addition, a method for determining multiblock fault contributions to the combined index is introduced. The effectiveness of the system is demonstrated using postlithography metrology data and plasma stripper processing tool data.

ICs going on a 3-V diet : Betty Prince and Roelof H. W. Salters. IEEE Spectrum, 22 (May 1992)

A multi-block progressive modelling approach is proposed for enhanced fault isolation in batch processes. The unfolding of batch data typically leads to matrices with a large number of columns and this complicates contribution analysis. In order to rapidly focus fault isolation in batch processes, it would be desirable to employ multi-block modelling techniques. Multi-block model such as consensus principal component analysis (CPCA) can produce multiple monitoring charts for sub-blocks and block loadings and block scores can be obtained which can represent unique behaviour of each sub-block. CPCA model uses super score which is the same as score from normal principal component analysis (PCA) model and it does not produce enhanced monitoring performance. Multi-block PCA (MBPCA) model using block score for model calculation can represent sub-blocks’ character but block scores are obtained from super loading so it may not be the best way to describe sub-blocks. A new MBPCA model is proposed for better expression of each sub-block. Through progressive modelling and contribution analysis, variables related to or affected by the fault, as well as the associated time information, are gradually identified. This enables a fault propagation path being established. The proposed method is applied to a benchmark simulated penicillin production process, PenSim.