Testing the potential of Surface-Enhanced Raman Spectroscopy for varietal and growing system discrimination of frozen berry fruits

A new, cost-effective method combining chemometrics with Surface-Enhanced Raman Scattering (SERS) was developed to differentiate various small berry fruits from Romanian markets and classify them according to the growing system (i.e. organic or conventional). Utilizing SERS data with Partial Least Squares-Discriminant Analysis (PLS-DA) we distinguished among four botanical groups (strawberry, raspberry, blackberry, blueberry) and identified 132 effective spectral markers, achieving 100 % accuracy in cross-validation. The PLS-DA analysis of SERS data yielded an 87 % accuracy score for classifying organic versus conventional farming systems, with sensitivity, specificity, and precision scores greater than 84 %. This classification model correctly predicted the farming system for 29 out of 33 samples, underscoring the relevance of the identified markers and the methodology’s efficacy for the rapid assessment of unknown samples. [Display omitted] •Innovative SERS application for berries discrimination•A perfect simultaneous discrimination of all four berry varieties•Fast and reliable tool for the discrimination of organic vs conventional berries•New approach for a fast berry safety and quality screening