This study compares the application of two variable selection methods in partial least squares regression (PLSR), the variable importance in projection (VIP) method and the selectivity ratio (SR) method. For this purpose, three different data sets were analysed: (a) physiochemical water quality parameters related to sensorial data, (b) gas chromatography–mass spectrometry (GC-MS) chemical (organic compound) profiles from fossil sea sediment samples related to sea surface temperature (SST) changes, and (c) exposed genes of Daphnia magna female samples related to their total offspring production. Correlation coefficients (r), levels of significance (p-value) and interpretation of the underlying experimental phenomena allowed the discussion about the best approach for variable selection in each case. The comparison of the two variable selection methods in the first water quality data set showed that the SR method is more accurate for sensorial prediction. For the climate data set, when raw total ion current (TIC) GC-MS chromatograms were considered, variables selected using the VIP method were easier to interpret compared with those selected by the SR method. However, when only some chromatographic peak areas (concentrations) were considered, the SR method was more efficient for prediction, and the VIP method selected the most relevant variables for the interpretation of SST changes. Finally, for the transcriptomic data set, the SR method was found again to be more reliable for prediction purposes. Copyright © 2015 John Wiley & Sons, Ltd.

Comparison of the variable importance in projection (VIP) and of the selectivity ratio (SR) methods for variable selection and interpretation

Flavonoids and their glycosides, including anthocyanins

Background Food adulteration , commonplace throughout human history, remains a concern today, with several notable instances involving the agro-food industry. Scope and approach In this review, we introduce four vibrational spectroscopic techniques (near-infrared, mid-infrared, Raman spectroscopy, and hyperspectral imaging) for the determination of food authenticity and adulteration. The characteristics and applications of these techniques, along with the major barriers and limitations, are discussed, with an emphasis on the treatment of spectral data using chemometrics . Key findings and conclusions Vibrational spectroscopic techniques have potential to fulfill the industrial need for food quality and authenticity analysis, however, still requires measurement accessories and dynamic chemomatric analytical methods for modern food inspection. We believe this review will be an effective guide for food industry researchers and engineers to aid in the selection of spectroscopic methods to analyze food quality and authenticity.

A review of vibrational spectroscopic techniques for the detection of food authenticity and adulteration

Metformin Enhances Autophagy and Normalizes Mitochondrial Function to Alleviate Aging-Associated Inflammation

Potential source contributions and risk assessment of PAHs in sediments from Taihu Lake, China: comparison of three receptor models.

Determination of adulteration and authenticity of extra virgin olive oil (EVOO) was investigated by means of infrared spectroscopy and chemometric methods. The study was focused on the detection and quantification of extra virgin olive oil adulteration by soybean (SB) and sunflower (SF) oils using FT-IR spectroscopy based on the use of PLS modeling and variable importance of projection (VIP) scores. A PLS model, using orthogonal signal correction and mean centering data pretreatments, and VIP scores variable preselection, was able to predict the concentration of sunflower and soybean oil adulterants in the 1–24 % weight ratio range with relative prediction errors lower than 3 % (w/w), for external validation samples. Moreover, the PLS-DA (discriminant analysis) model using the same preselected wavelengths was able to explain 99.9 % of variance and to predict with 100 % accuracy both classes of adulteration (EVOO–SB and EVOO–SF) in the external validation.

Detection of Olive Oil Adulteration Using FT-IR Spectroscopy and PLS with Variable Importance of Projection (VIP) Scores

Fourier transform infrared (FTIR) coupled to chemometrics was shown to be a useful method to classify and predict the quality of four commercial grade virgin olive oils (VOO). FTIR and physicochemical data were collected using a set of 70 samples representing extra virgin (EV), virgin (V), ordinary virgin (OV), and lampante (L) commercial grade olive oils collected in Beni Mellal region (central Morocco). Two partial least squares discriminant analysis (PLS-DA) models using physicochemical data and FTIR data were established and compared. The PLS-DA model using only physicochemical data was not accurate enough to distinguish satisfactorily among OV, V, and EV olive oil grades. On the contrary, the PLS-DA model on FTIR data was better in the calibration, able to describe 98 % of the spectral information and predicting 93 % of the VOO grades. In the external validation, this PLS-DA model accurately classified VOO commercial grades with prediction accuracy of 100 %. The proposed procedure is fast, nondestructive, simple, and easy to operate, and it is recommended for the quick monitoring of olive oil’s quality.

Stefan Platikanov

Papers

Comparison of the variable importance in projection (VIP) and of the selectivity ratio (SR) methods for variable selection and interpretation

Detection of Olive Oil Adulteration Using FT-IR Spectroscopy and PLS with Variable Importance of Projection (VIP) Scores

FTIR Spectroscopy and PLS-DA Classification and Prediction of Four Commercial Grade Virgin Olive Oils from Morocco

Study of motor oil adulteration by infrared spectroscopy and chemometrics methods

FTIR-ATR determination of solid non fat (SNF) in raw milk using PLS and SVM chemometric methods