Unsupervised pattern recognition chemometrics for distinguishing different Egyptian Olive varieties using a new integrated densitometric reversed-phase high-performance thin-layer chromatography-image analysis technique

Abstract

The merits of chemometrics in categorizing diff erent Egyptian olive chemovarieties based on their compositional integrity were implemented in this study. Fingerprints of 9 diff erent olive leaves varieties cultivated in Egypt were established using reversed-phase high-performance thin-layer chromatography (RP-HPTLC) prior to and after post-chromatographic derivatization with natural product-polyethylene glycol (NP/PEG) reagent and image analysis using ImageJ® software in order to build 2 separate data matrices. The chromatographic fi ngerprints were separately subjected to unsupervised pattern recognition multivariate analysis to build 2 separate models using principal component analysis (PCA) and hierarchical clustering analysis (HCA) algorithms to explore the distribution pattern of diff erent chemovarieties. The second model which involved olive samples' fi ngerprints after post-chromatographic derivatization exhibited greater ability to reveal a broader spectrum of phytoconstituents with enhanced sensitivity. Densitometric RP-HPTLC quantifi cation of oleuropein marker was compared to image analysis approach using Sorbfi l TLC Videodensitometer ® by newly developed and validated methods. Densitometry exhibited better performance characteristics than image analysis method and therefore was executed for determination of oleuropein concentration in the 9 Egyptian olive varieties. Oleuropein marker solely was found to be inadequate for standardization of olive leaves varieties. This study demonstrated a comprehensive approach for the rapid classifi cation of diff erent Egyptian olive varieties, which is crucial to warranting their chemical-consistency and, thereafter, eff ective consistency.