Categorizing rice cultivars based on cluster analysis of amylose content, protein content and sensory attributes

Presently, rice cultivars are categorized according to grain dimensions, am ylose content, and alkali spreading value (gelatinization temperature type) . Categorization of rice cultivars based on total sensory impact is needed. This work endeavors to divide world rices into groups based on amylose, pr otein, flavor, and texture properties. Ninety-one rice samples representing 79 different cultivars and seven growing locations were separated into sev en groups with Ward's Cluster Analysis. Cluster 1 included a third of the r ice samples and had cultivars with a large diversity of grain shapes and am ylose contents. Mean attribute scores for this cluster were near the grand mean for the collective rice samples for nearly every sensory attribute. Cl uster 2 included conventional U.S. short- and medium-grain cultivars. Clust er 3 included conventional U.S. medium cultivars that were produced in Loui siana. Mean sensory scores for this cluster characterized these cultivars; as having relatively undesirable flavor and texture attributes. Cluster 4 i ncluded Japanese premium quality cultivars and U.S. medium-grain cultivars developed for the Japanese market. Ouster 5 included high-amylose, indica t ypes that had relatively firm textural properties. Cluster 6 included relat ively soft cooking, aromatic cultivars. Cluster 7 included waxy cultivars a nd other soft cooking grains. In several cases, the production environment (location, weather effects, etc.) influenced flavor and texture characteris tics and resulted in the cultivar falling into an unexpected cluster. This categorization serves as a catalyst for indexing rice cultivars for cooking and processing qualities.