Grouping of rice mutant lines based on morphological and agronomical traits under different moisture conditions using multivariate statistical methods

Abstract

To study the morphological and agronomical traits and determine heterotic patterns at the reproductive stage of 96 mutant rice genotypes (M₂), two separate experiments were carried out under drought stress and flooding conditions at Gonbad Kavous University, Iran, in 2016 using a randomized complete block design with three replications. A factor analysis was conducted to reduce the number of variables to fewer independent factors, and three and four factors were identified under the flooding and drought conditions, explaining 68.3% and 76.05% of the total variance, respectively. Cluster analysis using the selected factors by the Ward's minimum variance method and Euclidean distance led to grouping of the mutant rice genotypes. The best cutting points of the dendrograms were determined by the discriminant function analysis and four clusters were identified at both irrigation conditions, which were also significantly different. Genotypes of the first and second clusters had the highest yield and its components under the flooding condition. In contrast, under the drought condition, genotypes of the third and fourth clusters were higher yielding and more tolerant to the drought stress. Mutants No. 4, 18, 25, 28, 29, 43, 44, 48, 53, 72, 79, 88, 90, 91, 94 and 95 were present in the first and second clusters of the flooding condition, and in the third and fourth clusters of the drought-stress condition, indicating their superiority over other mutants in terms of grain yield and drought tolerance. The results obtained from this study can be used for selecting suitable parents for hybridization from different clusters to produce new rice cultivars.