Elucidating genetic diversity and population structure in rice germplasm and identification of high yielding stable genotypes using MGIDI and GGE biplot analysis

The north-eastern region of India, being considered part of the Indo-Malayan biodiversity hotspot, suffered from the extinction of a large number of landraces including rice (Oryza sativa L.). Recent changes, specifically rapid urbanization, extreme climate events, and the introduction of profitable agriculture using high-yielding varieties, have also supplemented the process. Systematic evaluation of diverse genotypes is urgently required for their proper utilization. By evaluating 148 rice genotypes from the north-eastern hilly region of India over three successive years, key traits—namely, the number of filled grains per panicle, test weight, and yield per plant were identified. These traits were found to be governed by additive gene action, with lesser environmental effects on their expression. Based on the multi-trait stability index and GGE (genotype main effect plus genotype by environment interaction) biplot analysis, eight genotypes, namely, Beoidhan 2, Jalbudi, Motadhan, Salidhan, Tapolea, Lypyagopal, Badalsali, and Dagum, were identified. The first five principal components cumulatively explained 79.51% of the total variance. Genotyping of rice germplasm using 50 Generation Challenge Programme markers resulted in a total of 94 alleles. Polymorphic information content ranged from 0.14 to 0.69 with an average of 0.36. Likewise, Shannon's information index ranged from 0.20 to 1.33, with an average of 0.57. Nei's genetic distance-based clustering has grouped the genotypes into four major clusters, whereas the Bayesian model-based approach has resulted in two groups with 142 pure lines and 06 admixtures. Based on an analysis of molecular variance, 57.63% of the variance was due to genetic differentiation among the individuals within populations, while 41.92% of the variance was accounted for within individuals. Wright's statistics indicated that the genotypes of Sikkim were highly differentiated from those of Tripura. The second-highest level of differentiation was observed among the genotypes of Mizoram and Tripura. In principal coordinate analysis, the first three axes explained 25.79% of the total variation. The landraces with a high level of differentiation and enhanced yield potential in multi-environments, namely, Motadhan, Salidhan, Tapolea, and Lypyagopal, would help increase the yield potential vis-à-vis farmers' income.