Akademik Veri Yönetim Sistemi
LWT-FOOD SCIENCE AND TECHNOLOGY, cilt.237, 2025 (SCI-Expanded, Scopus)
Olive (Olea europaea L.) oil quality is intricately influenced by the biochemical composition, fatty acid profile, and pomological traits of the cultivar. Given the increasing consumer demand for functional foods and healthpromoting dietary fats, the comprehensive evaluation of these traits using multivariate and chemometric approaches is crucial. Traditional analyses often fall short of revealing the complex interactions between phenolic compounds, fatty acid composition, and oxidative stability. Therefore, integrative methods such as multivariate analyses and artificial neural network (ANN)-based modeling offer powerful tools for optimizing varietal selection aligned with nutritional value and oil quality. In this study, 22 olive cultivars were assessed under uniform agroecological conditions for fruit weight (2.15-6.55 g), oil content (16.23-30.92 %), total phenolic content (125.00-284.83 mg CAE/kg), and antioxidant activity. The dominant fatty acids included oleic acid (67.20-77.50 %), palmitic acid (9.20-15.55 %), and linoleic acid (7.20-13.50 %). Substantial genotypic variation was observed in UFAs/SFAs ratios (4.36-8.00) and oxidative stability indicators such as peroxide value and free fatty acid content. PCA revealed that the first two principal components explained 60.68 % of the total variance, primarily driven by oleic acid, total phenolics, and oil content. Hierarchical cluster analysis (HCA) grouped cultivars into four main clusters with distinct compositional traits, while correlation matrix analysis (CMA) indicated a strong positive correlation between oleic acid and total phenolic content (r = 0.88**). Multiple regression analysis (MRA) further identified oleic acid and free fatty acids as significant predictors (p <= 0.01) of overall oil quality. In parallel, an ANN model based on 10 input traits achieved a prediction accuracy of 92.3 % during validation, with an R2 of 0.91 and a mean squared error (MSE) of 0.024. Notably, cultivars such as 'Savrani', 'Memecik', 'Memeli', 'Halhal & imath;', and 'Ayval & imath;k' demonstrated superior profiles with high oleic acid, phenolic content, and favorable unsaturation indices. The integration of chemometric techniques and ANN-based modeling provided a comprehensive framework for elucidating trait interrelationships and predicting olive oil quality. The study highlights key biochemical markers and compositional traits as reliable predictors of nutritional and oxidative stability. The identification of superior cultivars supports evidence-based selection strategies for high-value olive oil production with enhanced health benefits. These findings not only contribute to precision breeding and functional food development but also address growing public health concerns regarding dietary fat quality.