Abstract:

A quantitative structure-retention relationship (QSRR) model to predict the chromatographic retention times of
fatty acids in edible vegetable oil was established by multiple linear regression (MLR) following methyl esterification and
characterization using the steric and electronic descriptors (SEDs). The prediction stability and capability of the model were
analyzed by internal and external validation. The correlation coefficient (R) for the established model and the correlation
coefficient (RLOO) of leave-one-out (LOO) cross-validation were 0.9990 and 0.9970, respectively, and the Q2 ext was 0.9860.
Results indicated that the molecular structures of fatty acid methyl esters were well characterized by SEDs parameters, and
this model had good stability and predictability. The proposed model could provide an important support to analyze and
identify the fatty acids in cooking oil indirectly.

Key words: edible vegetable oil, fatty acids, relative retention time, quantitative structure-retention relationship (QSRR), steric and electronic descriptors (SEDs)