Characterization of sweet and bitter cassava (Manihot esculenta Crantz) genotypes through multivariate analysis

Abstract

Cassava has importance as a source of human and animal food. With the objectives to select promising sweet and bitter cassava varieties for breeding programs, 27 genotypes were characterized in terms of their quantitative and qualitative properties. Roots were harvested from three plants per genotype, washed, peeled, sanitized. Regarding the yield, the storage root number (SRN), and the fresh storage root weight (FSRW), were determined, as well as the root fresh matter content (RFMC), and root dry matter content (RDMC), both expressed as a percentage. Among the cassava genotypes, the protein content ranged from 0.1-0.7%; lipids 0.3-2.1%; moisture 58.0-65.2%; 0.1-1.0% ash; fibers 0.9-1.9%; acidity 1,1-2,7%; pH 6.3-6.8; TSS between 0.8-1.2 ºBrix; glucose 0.1-0.8% and sucrose 0.5-1.0%, except for the fructose and starch contents, which did not vary significantly. The principal component analysis showed that the factors explain 84.2% of the total variability and through cluster analysis, evidencing cluster III for the highest starch yield and cluster I for the highest average of lipids and proteins