Amur soybean varieties classification according to biochemical indicators by clustering method

The purpose of the study is to systematize soybean varieties of Amur breeding according to biochemical quality indicators based on cluster analysis for rational use in the production of food products and additives. The study was carried out in 2018-2020 in the laboratory for processing agricultural products of the Federal Scientific Center of the All-Russian Research Institute of Soybeans. The objects of study are 18 soybean varieties bred by the All-Russian Soybean Research Institute, grown in the conditions of the Amur Region and included in the State Register of Breeding Achievements of the Russian Federation. 13 indicators were selected as signs for cluster analysis: protein, essential amino acids (lysine, leucine, isoleucine, valine, threonine, phenylalanine, methionine + cysteine), oil, polyunsaturated fatty acids (oleic, linoleic, linolenic) and the activity of trypsin inhibitors. The use of cluster analysis methods made it possible to group soybean varieties according to the criterion of the greatest similarity of biochemical parameters and form five main clusters: 1 - Lebedushka, ZHuravushka, Pepelina, Zolushka, Alena; 2 - Sentyabrinka, Intriga; 3 - Kuhanna, Nevesta, Kruzhevnica, Kitrossa, Statnaya, Umka; 4 - Persona, Bonus; 5 - CHarodejka, Topaz, Nega 1. Based on the results of clustering, a comparative analysis of the selected clusters was carried out according to the average values of biochemical characteristics. According to the protein-oil composition, the identified clusters of soybean varieties are conditionally classified into three groups: universal (with protein content - at least 38 %, oils - at least 18 %), oilseeds (with protein content - less than 38 %, oils - more than 18 %) and high-protein (with a protein content of more than 40 %, oils - less than 18 %).