Rapid classification of four Ficus species by UV-spectral fingerprinting coupled with chemometrics

Accurate discrimination of plant species within the same genus remains a challenge due to morphological similarities and chemical complexity. This study investigated Ultra-Violet metabolic fingerprinting combined with chemometric analysis for rapid classification of fruits of four Ficus species (Ficus variegata, Ficus fulva, Ficus padana, and Ficus racemosa) and its correlation with antioxidant activity. Methanolic extracts of Ficus fruits were analyzed at a wavelength of 200-400 nm. Multivariate analysis including Principal Component Analysis (PCA) achieved a clear separation of species, explaining 94.3% of the total variance (PC1:57.6%; PC2: 36.7%). Partial Least Squares-Regression (PLSR) identified critical variable contributing to antioxidant potency, with a high correlation (R² = 0.9945) and RMSECV 64.2316 μg/mL. The antioxidant activity test of the four Ficus species showed that F. racemosa, F. padana and F. fulva exhibited strong antioxidant activity with IC₅₀ values of 70.46 ± 5, 75.98 ± 3 and 85.71 ± 4 μg/mL, respectively. F. variegata showed the weakest antioxidant activity with an IC₅₀ value of 204.4 ± 5 μg/mL. The most influential spectral region for antioxidant prediction was 239 to 323 nm, suggesting that aromatic and conjugated phenolic-related chromophores contributed substantially to the observed spectral differences. These findings indicate that UV-spectral fingerprinting coupled with chemometrics can provide a rapid, low-solvent, and practical screening tool for preliminary species classification and antioxidant activity prediction, although structural confirmation of individual metabolites requires complementary chromatographic or mass spectrometric analysis.