Prediction of total phenolics, anthocyanins and antioxidant capacity of blackberry (Rubus sp.), blueberry (Vaccinium sp.) and jaboticaba (Plinia cauliflora (Mart.) Kausel) skin using colorimetric parameters

Abstract The food industry should have methods that allow identifying bioactive compounds quickly, efficiently and at low cost. The objective of this research was to obtain mathematical models for predicting the amount of total phenolics, anthocyanins and antioxidant capacity of blackberry, blueberry and jaboticaba skin, based on colorimetric parameters (L*, a*, b*, C* and h*). All analyses were done in the phenolic extracts obtained from these fruits. The mathematical models were obtained through multiple linear regression using the stepwise method. The significant correlation between the colorimetric parameters and chemical analyses resulted in models with a high predictive ability (high r2), such as in the determination of the contents of anthocyanins (r2 = 0.99), total phenolics (r2 = 0.86) and for the antioxidant capacity (r2 = 0.99). Thus, the prediction of levels of these bioactives compound, using colorimetric parameters, is very promising for cost and time savings.