Variable Selection Applied to the Development of a Robust Method for the Quantification of Coffee Blends Using Mid Infrared Spectroscopy

New genetic markers for 100% arabica coffee demonstrate high discriminatory potential for InDel-HRM-based coffee authentication

Food authenticity is crucial in today's society, given the heightened consumer awareness and attention to the products they consume. Reliable and efficient techniques are needed to quickly detect potential food adulterations that can negatively impact product quality and economic value. Coffee, a globally traded agricultural product, holds immense economic importance, with an estimated value of USD 83 billion. It is widely consumed and recognized as a functional food that provides minerals (K, Mg, Mn, Cr), niacin, and antioxidants. However, the preferred coffee species, Coffea arabica, known for its superior drink quality, is often adulterated with Coffea canephora (Robusta and Conilon) beans, even in 100% Arabica coffee. To distinguish between these two coffee species, a comprehensive study was conducted using a robust approach to identify differences in Single-Ortholog Copy (SOC) based on InDel regions in these gene pairs. These differences were validated using a meticulous methodology that considered variations in amplicon size: electrophoretic profile, and high-resolution melting (HRM). The innovative combination of InDels and HRM resulted in highly distinctive HRM profiles, outperforming SNP-based methods previously used. The targeted InDel approach utilized in this study facilitated precise quantification of Coffea species beans with a detection sensitivity of 0.5%. The study's findings establish the reliability and accuracy in distinguishing between the two coffee species, showcasing the valuable application of InDels for quality control and ensuring the authenticity of coffee beans. This pioneering research contributes to the advancement of authenticity verification methods for both imported and exported coffee beans, as well as in future studies that require significant genetic differences between these species, such as C. arabica and C. canephora.

SHS-GC-MS applied in Coffea arabica and Coffea canephora blend assessment

Considering the great economic significance of Coffea arabica (arabica) associated with the lower production cost of C. canephora (conilon), blends of these coffees are commercially available to reduce costs and combine sensory attributes. Thus, analytical tools are required to ensure consistency between real and labeled compositions. In this sense, chromatographic methods based on volatile analysis using static headspace-gas chromatography-mass spectrometry (SHS-GC-MS) and Fourier transform infrared (FTIR) spectroscopy associated with chemometric tools were proposed for the identification and quantification of arabica and conilon blends. The peak integration from the total ion chromatogram (TIC) and extracted ion chromatogram (EIC) was compared in multivariate and univariate scenarios. The optimized partial least squares (PLS) models with uninformative variable elimination (UVE) and chromatographic data (TIC and EIC) have similar accuracy according to a randomized test, with prediction errors between 3.3% and 4.7% and R_p² > 0.98. There was no difference between the univariate models for the TIC and EIC, but the FTIR model presented a lower performance than GC-MS. The multivariate and univariate models based on chromatographic data had similar accuracy. For the classification models, the FTIR, TIC, and EIC data presented accuracies from 96% to 100% and error rates from 0% to 5%. Multivariate and univariate analyses combined with chromatographic and spectroscopic data allow the investigation of coffee blends.

NIRS and Aquaphotomics Trace Robusta-to-Arabica Ratio in Liquid Coffee Blends

Coffee is both a vastly consumed beverage and a chemically complex matrix. For a long time, an arduous chemical analysis was necessary to resolve coffee authentication issues. Despite their demonstrated efficacy, such techniques tend to rely on reference methods or resort to elaborate extraction steps. Near infrared spectroscopy (NIRS) and the aquaphotomics approach, on the other hand, reportedly offer a rapid, reliable, and holistic compositional overview of varying analytes but with little focus on low concentration mixtures of Robusta-to-Arabica coffee. Our study aimed for a comparative assessment of ground coffee adulteration using NIRS and liquid coffee adulteration using the aquaphotomics approach. The aim was to demonstrate the potential of monitoring ground and liquid coffee quality as they are commercially the most available coffee forms. Chemometrics spectra analysis proved capable of distinguishing between the studied samples and efficiently estimating the added Robusta concentrations. An accuracy of 100% was obtained for the varietal discrimination of pure Arabica and Robusta, both in ground and liquid form. Robusta-to-Arabica ratio was predicted with R2CV values of 0.99 and 0.9 in ground and liquid form respectively. Aquagrams results accentuated the peculiarities of the two coffee varieties and their respective blends by designating different water conformations depending on the coffee variety and assigning a particular water absorption spectral pattern (WASP) depending on the blending ratio. Marked spectral features attributed to high hydrogen bonded water characterized Arabica-rich coffee, while those with the higher Robusta content showed an abundance of free water structures. Collectively, the obtained results ascertain the adequacy of NIRS and aquaphotomics as promising alternative tools for the authentication of liquid coffee that can correlate the water-related fingerprint to the Robusta-to-Arabica ratio.

Identification of Leaf-Scale Wheat Powdery Mildew (Blumeria graminis f. sp. Tritici) Combining Hyperspectral Imaging and an SVM Classifier

Powdery mildew (PM, Blumeria graminis f. sp. tritici) is a devastating disease for wheat growth and production. It is highly meaningful that the disease severities can be objectively and accurately identified by image visualization technology. In this study, an integral method was proposed based on a hyperspectral imaging dataset and machine learning algorithms. The disease severities of wheat leaves infected with PM were quantitatively identified based on hyperspectral images and image segmentation techniques. A technical procedure was proposed to perform the identification and evaluation of leaf-scale wheat PM, specifically including three primary steps of the acquisition and preprocessing of hyperspectral images, the selection of characteristic bands, and model construction. Firstly, three-dimensional reduction algorithms, namely principal component analysis (PCA), random forest (RF), and the successive projections algorithm (SPA), were comparatively used to select the bands that were most sensitive to PM. Then, three diagnosis models were constructed by a support vector machine (SVM), RF, and a probabilistic neural network (PNN). Finally, the best model was selected by comparing the overall accuracies. The results show that the SVM model constructed by PCA dimensionality reduction had the best result, and the classification accuracy reached 93.33% by a cross-validation method. There was an obvious improvement of the identification accuracy with the model, which achieved an 88.00% accuracy derived from the original hyperspectral images. This study can provide a reference for accurately estimating the disease severity of leaf-scale wheat PM and other plant diseases by non-contact measurement technology.

A data fusion model merging information from near infrared spectroscopy and X-ray fluorescence. Searching for atomic-molecular correlations to predict and characterize the composition of coffee blends

This article aims to develop and validate a multivariate model for quantifying Robusta-Arabica coffee blends by combining near infrared spectroscopy (NIRS) and total reflection X-ray fluorescence (TXRF). For this aim, 80 coffee blends (0.0-33.0%) were formulated. NIR spectra were obtained in the wavenumber range 11100-4950 cm-1 and 14 elements were determined by TXRF. Partial least squares models were built using data fusion at low and medium levels. In addition, selection of predictive variables based on their importance indices (SVPII) improved results. The best model reduced the number of variables from 1114 to 75 and root mean square error of prediction from 4.1% to 1.7%. SVPII selected NIR regions correlated with coffee components, and the following elements were chosen: Ti, Mn, Fe, Cu, Zn, Br, Rb, Sr. The model interpretation took advantage of the data fusion between atomic and molecular spectra in order to characterize the differences between these coffee varieties.