Predicting the antioxidant capacity and total phenolic content of bearberry leaves by data fusion of UV–Vis spectroscopy and UHPLC/Q-TOF-MS

The total phenolic content (TPC) and antioxidant capacity have been considered as important quality parameters for plant extracts. In this study, bearberry leaves were regarded as studied subject and a reliable method was established to predict the TPC and antioxidant capacity of bearberry leaves. Ultraviolet–visible spectrometry (UV–Vis) and ultra high pressure liquid chromatography coupled to time-of-flight mass spectrometry (UHPLC/Q-TOF-MS) were used to provide spectral fingerprinting and metabolomic profiling. The data obtained (separately and merged) were used to build partial least squares (PLS) regression model. The PLS model built by using ultraviolet–visible spectra provided a satisfactory prediction result. Mid-level data fusion using the scores significantly improved the performance of PLS regression model, the residual predictive deviations (RPDs) for TPC and α, α-diphenyl-β-picrylhydrazyl (DPPH) were 6.258 and 6.699, respectively, showing an excellent predictive ability. This study proved the potential of combination of UV–Vis spectrometry and UHPLC/Q-TOF-MS in the prediction of TPC and antioxidant capacity of plant extracts. Model built from mid-level data fusion. [Display omitted] •UV–Vis can provide good predictions for TPC and antioxidant capacity of bearberry.•The combination of UV–Vis and UPLC-Q/TOF-MS was used for the prediction of TPC and antioxidant capacity of bearberry.•Mid-level data fusion demonstrated to be more efficient on the prediction of the antioxidant capacity than a single technique.