Development of a model for quality evaluation of litchi fruit

Abstract

Sensory quality of litchi fruit is closely related to the appearance, flavor and nutritional quality. Principle component analysis (PCA) and multi-linear regression analysis (MLR) were performed to select the crucial quality factors in determining litchi sensory quality, and then a mathematical model was established to evaluate the litchi quality. Sensory evaluation value was conducted by a trained panel while sixteen instrumental quality parameters were used, including color (L*, a* and b*), single fruit weight (SFW), total soluble solids (TSS), pH, crude protein (CP), vitamin C (Vc), titratable acid (TA), glucose, fructose, sucrose, malic acid (MA), tartaric, edible rate (ER), and juice yield (JY). The PCA result demonstrated that six principle components exhibited high relationship with sweet and sour degrees, color, ER, TSS, Vc, CA and JY, which accounted for 82.359% of the original variables, and MLR results indicated that the crucial quality parameters were SFW, TA, a*, and ER. The predicted model was established in which validity QLOO2=0.603>0.5 in leave-one-out cross validation. Cluster analysis was used to select the cultivars suitable for eating quality which can provided a guide for litchi production and breeding.