Trait analysis of short duration boro rice (Oryza sativa L.) varieties in northern region of Bangladesh: Insights from heatmap, correlation and PCA

Abstract

The role of genetic variation is utmost importance in the selection of a desirable genotype for breeding programs. In the northern region of Bangladesh, the performance of five short duration boro rice varieties (BRRI dhan28, BRRI dhan74, BRRI dhan81, BRRI dhan88, and Bangabandhu dhan100) on yield and yield-contributing traits were evaluated in four locations. Yield attributes tiller hill-1 (TN), panicle hill-1 (PN), filled grain panicle-1 (FG), panicle length (PL), 1000-grain weight (TGW), and harvest index (HI), showed a positive correlation with grain yield (GY). Conversely, spikelet sterility (SSP) and unfilled grains panicle-1 (UFG) showed negative associations with grain yield. Principal component analysis (PCA) revealed that the first three components accounted for 89.60% of variability. The highest positive eigenvalue was observed in DF, DM, BY, HI, PN, SY, TN, PL and GY in PC1, indicating their significant influence on the overall genotype variations. PC2 was primarily driven by plant height (PH) and FG, while PC3 driven by TGW and UFG. The results suggest that grain yield is highly influenced by FG, PH, DM, DF, PN, TN, PL, TGW, SY, BY, and HI. These components highlighted the importance in distinguishing rice genotypes with higher grain yield potential. By considering these traits, BRRI dhan74 have outperformed other varieties, followed by BRRI dhan88 and Bangabandhu dhan100. BRRI dhan74 is most suitable for all four location followed by BRRI dhan88. The traits such as PN, PL, FG, TGW, PH, & HI significantly influence rice yield and should be prioritized in breeding programs aimed at developing high-yielding rice varieties.

Keywords

• biplot
• cluster
• grain yield
• HYV
• multi-location
• scree plot