Volume 17, Issue 4 (11-2025)
J Crop Breed 2025, 17(4): 1-9 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Vahed Semeskandeh M, Kiani G, Ramezani R, Aghajani Qara M. (2025). Evaluation and Selection of Desirable Genotypes using Multivariate Statistical Methods in the F2 Generation of Rice. J Crop Breed. 17(4), 1-9. doi:10.61882/jcb.2025.1573
URL: http://jcb.sanru.ac.ir/article-1-1573-en.html
URL: http://jcb.sanru.ac.ir/article-1-1573-en.html
1- Sari University of Agricultural Sciences and Natural Resources, Sari, Iran
2- Department of Plant Breeding, Sari University of Agricultural Sciences and Natural Resources, Sari, Iran
3- Gorgan University of Agricultural Sciences and Natural Resources, Tabarestan Agricultural Biotechnology and Genetics Research Institute, Gorgan, Iran
2- Department of Plant Breeding, Sari University of Agricultural Sciences and Natural Resources, Sari, Iran
3- Gorgan University of Agricultural Sciences and Natural Resources, Tabarestan Agricultural Biotechnology and Genetics Research Institute, Gorgan, Iran
Abstract: (804 Views)
Extended Abstract
Background: In many countries of the world, food supply, especially the food security of the countries, is a priority due to the biological and non-biological challenges and the growth of the world population. Rice is among the plants that have turned into a strategic plant due to its daily consumption and food needs of many countries. A suitable breeding program is needed to increase the potential of rice production. Genetic diversity in selected populations plays an important role in optimizing the breeding process, but genetic diversity has decreased due to the increase in genetic erosion and the intensity of breeding activities. One of the most important ways to increase rice production potential is to use a crossing system. Crossbreeding in rice is a very widely used method to increase yield due to the heterosis phenomenon and leads to the production of superior progeny from the parents or the commercial variety of the region. Multivariate statistics is a suitable method for analyzing different relationships between traits, identifying existing diversity, and evaluation accuracy in choosing the superior genotype to improve grain yield in rice. Cluster analysis and regression are powerful tools in multivariate statistics, being used as a selection index in determining traits affecting performance. Correlation by expressing the relationship between variables is a very effective method in determining the process of choosing the best plant in statistical analyses. This research aims to use classical breeding and multivariate statistics in an F2 rice population to select lines with high yield potential for the next generation.
Methods: This research was conducted in the rice research farm of the Tabarestan Agricultural Genetics and Biotechnology Research Institute, Sari University of Agricultural Sciences and Natural Resources. In total, 2000 F2 seeds obtained from the crossing of Nemat as the maternal base with the local variety Seng Tarem as the paternal base were planted in 50 rows, 10 meters long, and a planting distance of 25x25 in an area of about 125 square meters. The best plants (n = 18) in terms of morphological traits were selected for the analysis of morphological traits, such as plant height, the number of fertile tillers, spike length, the number of hollow seeds, the number of full seeds, total number of seeds, hundred-seed weight, seed length, seed width, and plant yield using multivariate statistics. Cluster analysis of traits with the Euclidean distance scale was used to group the plants, and a detection function was used for grouping accuracy. Correlation and regression analysis were used to investigate the relationship between the traits.
Results: The result of the cluster analysis divided the genotypes into three groups. The first group had the highest average seed width (2.69), total number of seeds (168.5), and the number of full seeds (117.22), and the lowest average plant height (121.83). The third group had the highest average values of 100-seed weight (2.94), the number of fertile tillers (35), and spike length (29.78). The lowest average number of hollow seeds (17.22) and the highest average seed length (11) belonged to Group 2. The third group was considered the yield group with the maximum 100-seed weight, the number of fertile tillers, and plant height. The correctness of the grouping of the plants belonging to the groups was confirmed by the detection function. The function introduced by the diagnostic analysis captured 94% of the variance of the test data and could distinguish the groups from each other with an eigenvalue of 32.67. The average traits of the number of fertile tillers, the number of hollow seeds, total number of seeds, 100-seed weight, and the yield of a single plant were significantly different between the groups and were the main factors that differentiated the plants. In the simple linear correlation of the traits spike length with plant height (r = 0.566), total number of seeds with the number of full seeds
(r = 0.533), 100-seed weight with the number of full seeds (r = -0.497), and seed width with spike length (r = -0.489) were significantly correlated at the 5% level. Total number of seeds with number of hollow seeds (r = 0.782), plant height, and number of fertile tillers with yield
(r = 0.789 and 0.789, respectively) were significantly related at the 1% level (99% confidence level). The regression analysis output of the model with an adjusted explanation coefficient (93.3%) shows the high correlation of these traits for the justification of yield changes. In the interpretation of yield changes, using regression analysis with the combined method of plant height and the number of fertile tillers showed positive and very significant effects on yield variability. Regarding the influential traits in the grouping and separation of genotypes, the highest yields were obtained for plants 5 (87.83 g), 10 (80.45), 13 (78.93), 4 (77.69), 15 (72.24), and 12 (70.75).
Conclusion: The results of the analysis showed that the plants 5 (87.83 g/plant), 10 (80.45), 13 (78.93), 4 (77.69), 15 (72.24), and 12 (70.75) having desirable morphological traits were selected as the best genotypes for producing promising lines. Multivariate statistics, as a suitable technique in the selection of the superior genotype, could distinguish between the F2 population of rice plants 4, 5, 10, 12, 13, and 15 with the desired morphological traits and high-yield potential as superior plants for producing promising lines.
Background: In many countries of the world, food supply, especially the food security of the countries, is a priority due to the biological and non-biological challenges and the growth of the world population. Rice is among the plants that have turned into a strategic plant due to its daily consumption and food needs of many countries. A suitable breeding program is needed to increase the potential of rice production. Genetic diversity in selected populations plays an important role in optimizing the breeding process, but genetic diversity has decreased due to the increase in genetic erosion and the intensity of breeding activities. One of the most important ways to increase rice production potential is to use a crossing system. Crossbreeding in rice is a very widely used method to increase yield due to the heterosis phenomenon and leads to the production of superior progeny from the parents or the commercial variety of the region. Multivariate statistics is a suitable method for analyzing different relationships between traits, identifying existing diversity, and evaluation accuracy in choosing the superior genotype to improve grain yield in rice. Cluster analysis and regression are powerful tools in multivariate statistics, being used as a selection index in determining traits affecting performance. Correlation by expressing the relationship between variables is a very effective method in determining the process of choosing the best plant in statistical analyses. This research aims to use classical breeding and multivariate statistics in an F2 rice population to select lines with high yield potential for the next generation.
Methods: This research was conducted in the rice research farm of the Tabarestan Agricultural Genetics and Biotechnology Research Institute, Sari University of Agricultural Sciences and Natural Resources. In total, 2000 F2 seeds obtained from the crossing of Nemat as the maternal base with the local variety Seng Tarem as the paternal base were planted in 50 rows, 10 meters long, and a planting distance of 25x25 in an area of about 125 square meters. The best plants (n = 18) in terms of morphological traits were selected for the analysis of morphological traits, such as plant height, the number of fertile tillers, spike length, the number of hollow seeds, the number of full seeds, total number of seeds, hundred-seed weight, seed length, seed width, and plant yield using multivariate statistics. Cluster analysis of traits with the Euclidean distance scale was used to group the plants, and a detection function was used for grouping accuracy. Correlation and regression analysis were used to investigate the relationship between the traits.
Results: The result of the cluster analysis divided the genotypes into three groups. The first group had the highest average seed width (2.69), total number of seeds (168.5), and the number of full seeds (117.22), and the lowest average plant height (121.83). The third group had the highest average values of 100-seed weight (2.94), the number of fertile tillers (35), and spike length (29.78). The lowest average number of hollow seeds (17.22) and the highest average seed length (11) belonged to Group 2. The third group was considered the yield group with the maximum 100-seed weight, the number of fertile tillers, and plant height. The correctness of the grouping of the plants belonging to the groups was confirmed by the detection function. The function introduced by the diagnostic analysis captured 94% of the variance of the test data and could distinguish the groups from each other with an eigenvalue of 32.67. The average traits of the number of fertile tillers, the number of hollow seeds, total number of seeds, 100-seed weight, and the yield of a single plant were significantly different between the groups and were the main factors that differentiated the plants. In the simple linear correlation of the traits spike length with plant height (r = 0.566), total number of seeds with the number of full seeds
(r = 0.533), 100-seed weight with the number of full seeds (r = -0.497), and seed width with spike length (r = -0.489) were significantly correlated at the 5% level. Total number of seeds with number of hollow seeds (r = 0.782), plant height, and number of fertile tillers with yield
(r = 0.789 and 0.789, respectively) were significantly related at the 1% level (99% confidence level). The regression analysis output of the model with an adjusted explanation coefficient (93.3%) shows the high correlation of these traits for the justification of yield changes. In the interpretation of yield changes, using regression analysis with the combined method of plant height and the number of fertile tillers showed positive and very significant effects on yield variability. Regarding the influential traits in the grouping and separation of genotypes, the highest yields were obtained for plants 5 (87.83 g), 10 (80.45), 13 (78.93), 4 (77.69), 15 (72.24), and 12 (70.75).
Conclusion: The results of the analysis showed that the plants 5 (87.83 g/plant), 10 (80.45), 13 (78.93), 4 (77.69), 15 (72.24), and 12 (70.75) having desirable morphological traits were selected as the best genotypes for producing promising lines. Multivariate statistics, as a suitable technique in the selection of the superior genotype, could distinguish between the F2 population of rice plants 4, 5, 10, 12, 13, and 15 with the desired morphological traits and high-yield potential as superior plants for producing promising lines.
Keywords: Multivariate statistics, Rice, Segregating generation of the F2 population, Selection, Yield
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
