Multivariate characterization of mutant ginger genotypes using cluster and principal component analyses for yield and trait improvement

Abstract

Understanding phenotypic diversity is essential for effective ginger (Zingiber officinale Roscoe) improvement. This study evaluated 15 mutant ginger lines at the Cross River Basin Development Authority, Calabar, Nigeria during the 2023 cropping season using a Randomized Complete Block Design (RCBD) with three replications. Phenotypic variation was analysed using Principal Component Analysis (PCA) and hierarchical clustering. The first four principal components (PCs) explained 93.64% of total variability, with PC1 contributing 72.87%, PC2 13.37%, PC3 6.59%, and PC4 1.14%. Rhizome length showed the highest positive loading on PC1 (0.9912), while establishment count (−0.8189) and plant height (−0.4558) dominated PC2. The genotypes were grouped into four clusters, indicating substantial genetic diversity. Cluster 2 (six genotypes) exhibited higher establishment counts (up to 91%), more rhizome fingers (8–12 per plant), longer rhizomes (15.3–18.6 cm), and higher yields (up to 32.7 t ha⁻¹), making them ideal candidates for yield-focused breeding programmes. Cluster 3 (two genotypes) showed high emergence rates (>90%), taller plants, 
(up to 102 cm) and broad leaves (5.2–6.1 cm), suggesting potential for improving vegetative vigour. The integration of PCA and clustering provides a robust framework for elite genotype selection and targeted trait improvement, offering a strategic pathway to develop high-yielding and regionally adaptable ginger cultivars.