The physicochemical traits of highland barley prominently affect the quality of Tsampa. To find out the relevance between the physicochemical properties of raw material and the texture parameters of processed products, twenty-five physicochemical traits and ten quality parameters for seventy-six varieties of highland barley were measured and analyzed. The results showed that there was a significant difference between the physicochemical indexes for highland barleys of various colors. The dark highland barley generally has more fat, protein, total dietary fiber, phenolic, Mg, K, Ca, and Zn and less amylose, Fe, Cu, and Mo than light colored barley. Then, these highland barleys were made into Tsampa. A comprehensive quality evaluation model based on the color and texture parameters of Tsampa was established through principal component analysis. Then, cluster analysis was used to classify the tested samples into three edible quality grades predicated on the above evaluation model. At last, the regression analysis was applied to establish a Tsampa quality predictive model according to the physicochemical traits of the raw material. The results showed that amylose, protein, β-Glucan, and a* and b* could be used to predict the comprehensive quality of Tsampa. The predicted results indicated that 11 of 14 validated samples were consistent with the actual quality, and the accuracy was above 78.57%. Our study built the approach of the appropriate processing varieties evaluation. It may provide reference for processing specific highland barley.
Keywords: Tsampa; highland barley; physicochemical traits; predictive model; quality evaluation.