Principal component and cluster analysis in sorghum(sorghum bicolor (l.) moench)

In the current set of an experiment, thirty two sorghum genotypes were assessed for different yield and yield component traits. For evaluation of these traits, basic statistics, correlation, Principal component (PC) and diversity analyses were employed to obtain suitable parents that can be further exploited in future breeding programmes. The estimation of descriptive statistics of nine quantitative traits indicated the existence of variability among the genotypes. Correlation analysis revealed that grain yield was positively correlated with panicle length, leaf length and leaf width whereas fodder yield was positively correlated with number of leaves/plant, leaf width, leaf length, plant height and days to maturity. The positive correlation among these yield contributing traits suggested that these characters are important for direct selection of high yielding genotypes. Principal component (PC) analysis showed first 3 PCs having Eigen value >1 explaining 70.89% of the total variation with different yield and yield component traits. In the biplot analysis between PCs 1 and 2, the genotypes remained scattered in all four quadrants, showing large genetic variability in quantitative traits. The thirty two genotypes were grouped into five clusters on the basis of average linkage and dendogram. The cluster-I having 13 genotypes followed by cluster-III (9), cluster-II (7), cluster-V (2) and cluster-IV (1). Distribution pattern of all the genotypes into five clusters showed the presence of considerable genetics diversity among the genotypes for most of the traits under consideration. Various useful correlations and aforementioned information extracted from cluster and PC analysis will be helpful in designing breeding programmes to obtain high yielding genotypes in sorghum for seed as well as fodder yield.