Factor analysis-based evaluation of stalk yield and quality traits in sweet sorghum genotypes

Abstract

Sweet sorghum is one of the most important global crops for energy, food, and feed, valued for its climate resilience and adaptability to diverse growing conditions. A total number of 25 sweet sorghum genotypes were evaluated to identify the best genotype (s) with high stalk yield and quality during the 2024 and 2025 growing seasons in alpha lattice design. The results showed significant variation among the tested genotypes across the eight evaluated traits. The superior genotypes exhibiting high average values for stalk and quality yield traits were MN4080, Rex, Honey, Cukarcirak, AGSC2, and Sugar drip, whereas Dall, Grcoba, Atlas, Williams, Roma, and MN1500 and Rio recorded the lowest performance across the evaluated traits. The stalk yield was significantly correlated with stalk diameter, juice extraction percentage and juice yield. To address multicollinearity and clarify complex trait relationships, factor analysis was applied, reducing the variables to three main factors that accounted for over 80 % of the total variation. The first factor, which included stalk diameter, stalk yield, juice extraction percentage, and juice yield, accounted for 40.17 % of the total variation. Both factor and cluster heat map analyses consistently grouped MN4080, Rex, Honey, Cukarcirak, AGSC2 and Sugar drip, as promising genotypes for enhancing sweet sorghum stalk yield and improving quality yield. Using factor scores in ANOVA proved more effective than relying on the original data, making these scores more reliable for selection purposes and applicable in sweet sorghum breeding programs.