Approach to the quantification of milk mixtures by partial least-squares, principal component and multiple linear regression techniques

Non-destructive prediction of yak meat freshness indicator by hyperspectral techniques in the oxidation process

This study examined the potential of hyperspectral techniques for the rapid detection of characteristic indicators of yak meat freshness during the oxidation of yak meat. TVB-N values were determined by significance analysis as the characteristic index of yak meat freshness. Reflectance spectral information of yak meat samples (400-1000 nm) was collected by hyperspectral technology. The raw spectral information was processed by 5 methods and then principal component regression (PCR), support vector machine regression (SVR) and partial least squares regression (PLSR) were used to build regression models. The results indicated that the full-wavelength based on PCR, SVR, and PLSR models were shown greater performance in the prediction of TVB-N content. In order to improve the computational efficiency of the model, 9 and 11 characteristic wavelengths were selected from 128 wavelengths by successive projection algorithm (SPA) and competitive adaptive reweighted sampling (CARS), respectively. The CARS-PLSR model exhibited excellent predictive power and model stability.

Authenticity assessment of dairy products by capillary electrophoresis

Milk and derivatives are a very important part in the diet of the world population. Products from goat, buffalo, and sheep species have a greater economic value than the cow ones, therefore, authenticity frauds by improperly adding cow's milk occur frequently: dairy products are among the seven more attractive foods for adulteration. Milk from each of the above-cited animal species has its own definite profile of whey proteins (variants of α-lactalbumin and β-lactoglobulin) and its definite profile of caseins (variants of α_S1 -, α_S2 -, β-, and κ-casein). Such proteins can be usefully exploited as markers of authenticity by using capillary electrophoresis which is the technique of choice for the analysis of proteins. Due to the multiple adjustable parameters that are unknown to other analytical techniques, capillary electrophoresis is able to detect frauds in milk mixtures and cheese with little use of solvents, fast analysis time, and ease of operation. This makes it attractive and competitive for routine checks that are very important to fight the adulteration market. Advantages and limitations are discussed.

A review of the application of near-infrared spectroscopy to rare traditional Chinese medicine

Near-infrared spectroscopy (NIRS) has become popular in the field of traditional Chinese medicine (TCM) due to its many practical advantages, such as providing rapid, accurate, and simultaneous analysis. This review is intended to provide up-to-date information on qualitative and quantitative NIRS analysis in TCM, especially rare TCM. By performing a substantial survey of the literature from China and abroad, we also combine our own studies on some rare TCMs using NIRS to review the application of NIRS in this field. Basic analytical methods and specific examples of NIRS demonstrates NIRS's ability in authenticity identification, species identification, geographic origin analysis, quantitative analysis, adulteration detection, rapid detection, and on-line monitoring of TCM, and illustrates the feasibility and effectiveness of NIRS applied in the quality control (QC) of TCM. Some disadvantages and prospects of NIRS are also discussed here in detail.

Chemometric characterization of fruit juices from Spanish cultivars according to their phenolic compound contents: I. Citrus fruits

The data set composed by phenolic compound profiles of 83 Citrus juices (determined by HPLC-DAD-MS/MS) was evaluated by chemometrics to differentiate them according to Citrus species (sweet orange, tangerine, lemon, and grapefruit). Cluster analysis (CA) and principal component analysis (PCA) showed natural sample grouping among Citrus species and even the Citrus subclass. Most of the information contained in the full data set can be captured if only 15 phenolic compounds (concentration ≥10 mg/L), which can be quantified with fast and accurate methods in real samples, are introduced in the models; a good classification which allows the confirmation of the authenticity of juices is achieved by linear discriminant analysis. Using this reduced data set, fast and routine methods have been developed for predicting the percentage of grapefruit in adulterated sweet orange juices using principal component regression (PCR) and partial least-squares regression (PLS). The PLS model has provided suitable estimation errors.

Detection of cow milk in donkey milk by chemometric procedures on triacylglycerol stereospecific analysis results

Stereospecific analysis is an important tool for the characterization of lipid fraction of food matrices, and also of milk samples. The results of a chemical-enzymatic-chromatographic analytical method were elaborated by chemometric procedures such as linear discriminant analysis (LDA) and artificial neural network (ANN). According to the total composition and intrapositional fatty acid distribution in the triacylglycerol (TAG) backbone, the obtained results were able to characterize pure milk samples and milk mixtures with 1, 3, 5% cow milk added to donkey milk. The resulting score was very satisfactory. Totally correct classified samples were obtained when the TAG stereospecific results of all the considered milk mixtures (donkey-cow) were elaborated by LDA and ANN chemometric procedures.

Identification of transgenic foods using NIR spectroscopy: a review

The utilization of chemometric methods in the quantitative and qualitative analysis of feeds, foods, medicine and so on has been accompanied with the great evolution in the progress and in the near infrared spectroscopy (NIRS). Hence, recently the application of NIR spectroscopy has extended on the context of genetics and transgenic products. The aim of this review was to investigate the application of NIR spectroscopy to identificate transgenic products and to compare it with the traditional methods. The results of copious researches showed that the application of NIRS technology was successful to distinguish transgenic foods and it has advantages such as fast, avoiding time-consuming, non-destructive and low cost in relation to the antecedent methods such as PCR and ELISA.

A comparative study between PCR and PLS in simultaneous spectrophotometric determination of diphenylamine, aniline, and phenol: Effect of wavelength selection

Partial least squares (PLS) and principal component regression (PCR) have received considerable attention in the chemometrics for multicomponent analysis where superiority of one over another is a challenging problem yet. Considering the effect of wavelength selection, a comparison was made between PCR and PLS methods by application those to simultaneous spectrophotometric determination of diphenylamine (DPA), a compound from the third European Union list of priority pollutants, and its environmentally related products aniline and phenol. The UV absorbance spectra of the methanolic solutions of the analytes were measured in the concentration ranges of 1.0-10.0 microg mL(-1) and then subjected to PCR and PLS. The models refinement procedure and validation was performed by cross-validation. A modified changeable size moving windows strategy, where optimized the intervals between the sensors in a selected windows, was also proposed to select the more informative spectral regions for each of the analytes. It was found that wavelength selection improved the quality of predictions for both regression methods whereas more reliable results were obtained by removing of the highly collinear neighboring wavelengths. The resultant data explained that PLS produced more or less better results when whole spectral data were used but in the case of selected wavelength regions both methods produced similar results and no comments could be given about the superiority of one against another. The major difference was obtaining the higher number of factors for PCR, which is not a significant problem.