Multivariate analysis of variance of designed chromatographic data. A case study involving fermentation of rooibos tea.

An ultimate goal of investigations of rooibos plant material subjected to different stages of fermentation is to identify the chemical changes taking place in the phenolic composition, using an untargeted approach and chromatographic fingerprints. Realization of this goal requires, among others, identification of the main components of the plant material involved in chemical reactions during the fermentation process. Quantitative chromatographic data for the compounds for extracts of green, semi-fermented and fermented rooibos form the basis of preliminary study following a targeted approach. The aim is to estimate whether treatment has a significant effect based on all quantified compounds and to identify the compounds, which contribute significantly to it. Analysis of variance is performed using modern multivariate methods such as ANOVA-Simultaneous Component Analysis, ANOVA - Target Projection and regularized MANOVA. This study is the first one in which all three approaches are compared and evaluated. For the data studied, all tree methods reveal the same significance of the fermentation effect on the extract compositions, but they lead to its different interpretation.

[1]  Olav M. Kvalheim,et al.  Interpretation of latent-variable regression models , 1989 .

[2]  M. Glomb,et al.  Chemistry of color formation during rooibos fermentation. , 2012, Journal of agricultural and food chemistry.

[3]  L. Buydens,et al.  Regularized MANOVA (rMANOVA) in untargeted metabolomics. , 2015, Analytica chimica acta.

[4]  Beata Walczak,et al.  Analysis of variance of designed chromatographic data sets: The analysis of variance-target projection approach. , 2015, Journal of chromatography. A.

[5]  C. Muller,et al.  Potential of rooibos, its major C-glucosyl flavonoids, and Z-2-(β-D-glucopyranosyloxy)-3-phenylpropenoic acid in prevention of metabolic syndrome , 2018, Critical reviews in food science and nutrition.

[6]  A. Porzel,et al.  Oxidation of the dihydrochalcone aspalathin leads to dimerization. , 2009, Journal of agricultural and food chemistry.

[7]  M. Glomb,et al.  Characterization of phenolic compounds in rooibos tea. , 2008, Journal of agricultural and food chemistry.

[8]  Rabia Johnson,et al.  Amelioration of palmitate-induced insulin resistance in C₂C₁₂ muscle cells by rooibos (Aspalathus linearis). , 2013, Phytomedicine : international journal of phytotherapy and phytopharmacology.

[9]  Olivier Ledoit,et al.  A well-conditioned estimator for large-dimensional covariance matrices , 2004 .

[10]  A. Stalmach,et al.  Bioavailability of C-linked dihydrochalcone and flavanone glucosides in humans following ingestion of unfermented and fermented rooibos teas. , 2009, Journal of agricultural and food chemistry.

[11]  Nico A. Walters,et al.  Improved HPLC method for rooibos phenolics targeting changes due to fermentation , 2017 .

[12]  Tarja Rajalahti,et al.  Discriminating variable test and selectivity ratio plot: quantitative tools for interpretation and variable (biomarker) selection in complex spectral or chromatographic profiles. , 2009, Analytical chemistry.

[13]  E. Joubert.  HPLC quantification of the dihydrochalcones, aspalathin and nothofagin in rooibos tea (Aspalathus linearis) as affected by processing , 1996 .

[14]  D. Ferreira,et al.  Occurrence and sensory perception of Z-2-(β-d-glucopyranosyloxy)-3-phenylpropenoic acid in rooibos (Aspalathus linearis). , 2013, Food chemistry.

[15]  Age K. Smilde,et al.  ANOVA-simultaneous component analysis (ASCA): a new tool for analyzing designed metabolomics data , 2005, Bioinform..