Gas chromatography–mass spectrometry metabolic profiling and sensory evaluation of greenhouse mangoes (Mangifera indica L. ‘Irwin’) over multiple harvest seasons

Abstract

Compared to outdoor mango cultivation in the tropics, greenhouse cultivation in temperate regions is less reported due to its short history and small scale. Here, we evaluated for the first time the taste-focused quality of greenhouse-grown mangoes (Irwin) by GC–MS metabolic profiling and sensory evaluation for over three years (2021–2023). The relative standard deviation in sensory evaluation scores was approximately 15 % each year. Meanwhile, principal component analysis was performed on 45 identified metabolites and clustered with similar topology over three years. Orthogonal partial least squares regression analysis showed that a prediction model could be constructed with R₂, Q₂ = 0.9 or higher for all three years. Furthermore, the 2021 prediction model was used to evaluate the accuracy of sensory evaluation scores in a different year (2023). Compared to the model that used all 45 metabolites as explanatory variables, the accuracy of the model improved when using only 24 important metabolites which are common to both years (2021 and 2022), suggesting that these metabolites are highly reproducible across years. This study would contribute not only to fundamental greenhouse information, but also to the improvement of quality and cultivation methods in greenhouse in the future.