Biometric methods in studying the gene pool Indian pea

Russian Research Institute for Sorghum and Maize «Rossorgo», 410050, Saratov, st. 1st Institutskiy proezd, 4, Russia
Federal State Budgetary Scientific Organization «Federal Center of Agriculture Research of the South- East Region», 410010, Saratov, Tulaikov Street, 7, Russia

Abstract

The article presents the results of using biometric methods for assessing the gene pool (using the example of the indian pea). The variation of traits of the model population of indian pea, as well as the correlation coefficients and factor loadings on the variables, is considered. The model population includes 25 samples of different ecological and geographical origin. The nursery was founded according to the scheme of a one-factor complex. Statistical indicators of the sample, coefficients of variation of signs were revealed: ˃20% - fat content in seeds, fat gather, fiber gather, 10-20% - height of attachment of the lower bean, number of twigs of the 1st order, number of beans per plant, number of seeds in a bean , number of seeds per plant, seed yield, pod width, gather of protein, ash, BEV per unit area, ˂10% - stem length, weight of 1000 seeds, plant productivity, pod length, content of protein, fiber, ash, BEV in seeds analysis of the experimental data, a matrix of correlation coefficients was calculated, including indicators of vegetative traits and the content of nutrients in the grain. In order to optimize the interpretation of the matrix of correlation coefficients, a factor analysis was performed using the method of principal components. Signs that determine the first, second and third hypothetical factors (stem length, lower bean attachment height, content of protein, fat, fiber, BEV, ash, number of seeds in a bean, weight of 1000 seeds, plant productivity, seed yield, gather per hectare of protein, fiber, ash, BEV, the number of seeds from one plant) determine 63,72% of the accumulated dispersion.