Similarity Principal Component Research Articles

Multivariate statistical methods for Port Salut Argentino cheese analysis based on ripening time, storage conditions, and sampling sites

The objective of the present work was to compare multivariate statistical methods for the classification of Port Salut Argentino cheese samples based on ripening time (1, 6, 13, 27, and 56 days), storage conditions (traditionally ripened and ripened after frozen storage) and sampling sites (internal and external zones) using the contents of caseins, peptides and amino acids measured by chromatographic analysis as well as textural and physical parameters. In particular, two linear methods, principal component analysis (PCA) and principal component similarity (PCS), and a non-linear method, the Kohonen self-organizing artificial neural network (Kohonen ANN), were compared. The two linear methods showed the same grouping of cheese samples according to ripening time, sampling site and storage condition. These methods are closely related in their mathematical basis and the similar grouping showed by both methods can be explained by the fact that the first three principal components explained 89.3% of the data set variation. The non-linear Kohonen ANN uses a mathematical procedure completely different from PCA; however, only slight differences were observed in the grouping of cheese samples. Those differences may be related to the weight that each model gives to every variable. One interesting feature of Kohonen ANN is that weight maps (contour plots) sometimes are superior to principal component loadings (vectors) for the understanding of relationships between the groups and the original variables.

A Computer‐Aided Strategy for Structure‐Function Relation Study of Food Proteins Using Unsupervised Data Mining

Data mining (DM) based on Bayesian neural networks (BNN) is popular for exploring consistent patterns and/or systematic relationships of variables in chemometrics. A combination of the unsupervised principal component similarity (PCS) analysis with the random‐centroid optimization for site‐directed mutagenesis of amino acid sequences (RCG) is proposed to correlate the sequence data with functions of proteins. Principal component similarity based on similarity was superior to classifications based on dissimilarity represented by multidimensional distances. Important factors (independent variables) to be used in the optimization could be selected through PCS processing of data to improve reliability of function prediction. Dimensionality reduction using PCS by eliminating minor factors and/or trend‐line drawing on response surface maps in RCG to determine the direction of search shift toward the global optimum are useful for approximating the underlying response surfaces. Application of the sequence PCS was useful in elucidating usually unknown mechanisms of underlying functions of a protein based on its amino acid sequence. The functions in question may be predicted using a modern version of neural networks.

DEFINITION OF OUTLIERS USING UNSUPERVISED PRINCIPAL COMPONENT SIMILARITY ANALYSIS FOR SENSORY EVALUATION OF FOODS

ABSTRACT As unsupervised classifications, principal component similarity (PCS) and cluster analysis (CA) were compared for outlier detectability in panel evaluation. By rotating the reference, PCS can define outlying panelists based on the similarity of their evaluation patterns with that of the reference panelist. As a result, the outliers detected on PCS scattergrams are dependent on the reference selected, whereas, outliers detected by CA are based on dissimilarity, thus being rather unilateral. The definition of outliers in PCS is new as it is different from the currently most popular definitions based on dissimilarity. For verifying the outliers thus obtained, random-centroid optimization (RCO) was applied for selecting the best samples by each cluster of panelists. This combination of PCS/RCO may be useful in finding the likeness distribution among consumers and then in creating food products to correctly respond to the demands of different consumer groups.

Solid phase microextraction/gas chromatography of Salmonella-infected beef.

Eight strains of Salmonellae were incubated in TSB culture medium at 37 degrees C for 24 h. Volatile compounds derived from the bacteria were collected using solid-phase microextraction fibers and then applied to gas chromatography (GC). Similarity analysis of the GC patterns thus obtained could separate these strains on principal component similarity (PCS) scattergrams. Five major food-related pathogenic bacteria and 10 other bacteria (including one Salmonella strain) were also classified by growing in the same medium. It is then proposed to utilize this approach to improve the GC/PCS method of Nakai et al. [Nakai, S.; Wang, Z. H.; Dou, J.; Nakamura, S.; Ogawa, M.; Nakai, E.; Vangerstoep, J. J. Agric. Food Chem. 1999, 47, 576-583] that has been developed for screening safe foods by detecting bacteria contaminated foods. Inoculating food samples pre-enriched through preliminary incubation into a culture medium and then subjecting to the GC/PCS method after secondary incubation enhances the detectability of pathogenic bacteria.

New multivariate strategy for panel evaluation using principal component similarity

Advantages of principal component similarity (PCS) as an unsupervised classification techniques compared to supervised methods are the detectability of outliers or anomalies, a capacity to classify continuum, easy identification of causes for grouping, and a potential of discovering new groups. The ability of detecting outliers or anomalies was utilized to eliminate panel members who were not appropriate for the classification purposes by using PCS scattergrams. After eliminating the unqualified panelists, the evaluation score tables were realigned from the one based on panelists to that based on samples. Accumulated principal component score were computed for the samples in a form of SPCi × Si, where PCi was the principal component score of sample i and Si was its proportion within the total variation. The conventional averaging algorithms of evaluated scores are useful, as this techniques absorbs the effect of unjustifiable negative scores reported by outliers. However, more reasonable summary scores for the samples could be obtained by eliminating unreasonable evaluation scores made by the outliers as well as by using the above new summary values. Differences in deviations of evaluation among attributes, such as those in likenesses of color and taste of meat products, were normalized. Specific training or selecting qualified panelists prior to panel evaluation is unnecessary when the information obtained from panel evaluation is required to reflect the variable broad patterns of consumer preference. Information obtained from consumer preference test based on cluster analysis, which was applied to a beverage, might be more readily recovered by using this new strategy.

Solid phase extraction for GC analysis of beany flavours in soymilk

An octadecyl sorbent column was selected for solid phase extraction of volatile components of soymilk. The volatiles thus adsorbed were eluted with a mixture of hexane and diethyl ether (3:1 v v ) and then applied directly to GC column. The detection limit of the method was 0.1 ppm for both hexanal and hexanol. Total content of these two compounds in soymilk was linearly correlated to the lipoxygenase activity in soybeans. Data points on principal component similarity (PCS) scattergrams for soymilk samples were grouped toward two directions depending on the grit size of ground beans heated by microwave. Sensory panels detected beany flavour in these samples but without distinction between these two types of GC patterns as classified on the PCS scattergram. Principal components 1 (PC1) and 2 (PC2) had high loadings of pentanol and hexanal and hexanol and hexanal, respectively. Pentanol was produced upon heating finely ground soybeans.

Classification of Microbial Defects in Milk Using a Dynamic Headspace Gas Chromatograph and Computer-Aided Data Processing. 2. Artificial Neural Networks, Partial Least-Squares Regression Analysis, and Principal Component Regression Analysis

Objective, yet cost-effective evaluation of flavor is difficult in quality control of milk. Inexpensive gas chromatographs in conjunction with computer models make it feasible to construct an objective flavor evaluation system for routine quality control purposes. The purpose of this study was to classify milk with microbial off-flavors using a low-cost headspace gas chromatograph and computer-aided data processing. Principal component similarity (PCS) analysis was discussed in part 1. In part 2, artificial neural networks (ANN), partial least-squares regression (PLS) analysis, and principal component regression (PCR) analysis are examined. UHT milk was inoculated with various bacteria (Pseudomonas fragi, Pseudomonas fluorescens, Lactococcus lactis, Enterobactor aerogenes, and Bacillus subtilis) and a mixed culture (P. fragi:E. aerogenes:L. lactis = 1:1:1) to approximately 4.0 log10 CFU mL-1. ANN were able to make better predictions than PLS and PCR. The prediction ability of PLS was better than PCR. The ...

Reverse‐Phase HPLC Analysis of Reference Cheddar Cheese Samples for Assessing Accelerated Cheese Ripening

ABSTRACTApplying water extracts from Cheddar cheese to an octadecyl vinyl alcohol copolymer column using a reliable auto‐sampling system provided highly repeatable HPLC patterns. Two batches of standard samples of mild, medium, sharp, and extra‐sharp and one batch of abused samples (rapidly aged under abnormal conditions) were analyzed. Principal component similarity (PCS) analysis indicated similar shifts of plots due to aging of the standard batches, while the plot for the abused batch deviated from the pathway of normal aging. PCS may be useful for analysis of accelerated ripening effects as well as finding unusual samples during quality control.

Reverse‐Phase HPLC Analysis of Cheese Samples Aged by a Fast‐Ripening Process

ABSTRACTCheese ripening accelerated by adding enzyme extracts derived from lactic bacteria promoted proteolysis and produced an acceptable novel cheese based on sensory tests. However, HPLC profiles for watersoluble compounds were considerably different from those for control Cheddar cheese without the added enzyme extracts, probably indicating a different mechanism of proteolysis. The samples did not follow the usual aging pathway on a similarlity scattergram, when HPLC data were processed by principal component similarity (PCS) analysis. Such analysis was useful for evaluating effects of accelerated cheese ripening.

Principal component similarity analysis for classification and its application to GC data of mango

As a method for classification, principal component similarity (PCS) analysis reported previously was modified. It was found by using model computation that PCS was easy to use to interpret the result of discrimination, especially when purposes of classification were not clearly defined in advance (unsupervised). A portable, low-cost headspace gas chromatograph (GC) was used for routine analysis of volatile compounds recovered from mango samples. PCS could classify the mango samples based on cultivars which were ripened in storage and on the tree. PCS was, as a result, found to be useful in processing the obtained GC data for crude classification, preliminary to detailed analysis, e.g. obtained GC data for crude classification, preliminary to detained analysis, e.g. stepwise discriminant analysis.