Abstract

Sweet potatoes serve as an alternative carbohydrate source to mitigate rice shortages resulting from climate change. This study aimed to evaluate the genetic diversity of sweet potato agro-morphological traits. Seventy genotypes, along with two controlled trial treatments, were evaluated using augmented design at the Mechara Agricultural Research Center. Among the total genotypes assessed, 40.28% exhibited a commercial yield significantly higher than the overall average, ranging from 15.83 to 44.17 tons/ha, with 29 genotypes showing a total yield/ha significantly surpassing the average for all genotypes.  The yield of sweet potato (ton/ha) shows a positive and high correlation with the number of roots per plant, root diameter, number of sold roots, and sales yield (ton/ha). Principal component analysis revealed that five principal components with eigenvalues greater than one contributed to 79.1% of the total variation. Traits that contributed the most to diversity included total yield, number of marketable tubers per hectare, marketable yield, number of roots per plant, and number of unsold roots per hectare. The meta-analysis based on agro-morphological characteristics grouped genotypes into five clusters. The Euclidean genetic distance between clusters ranged from 4.3 to 9.5 among the evaluated sweet potato genotypes, with the highest intergroup distances observed between groups V and I, followed by between groups IV and V. These high intergroup distances suggest a greater likelihood of achieving violation and maximal dissociation heterozygosis by crossing genotypes belonging to different groups due to the greater distance between them. Overall, the results of this study suggest the presence of significant genetic diversity among sweet potato genotypes for agro-morphological traits, which could serve as an important gene source for the genetic improvement of sweet potato in the future.

Key words: Cluster analysis, genetic diversity, principal component analysis.