Open AccessGenetic characterization and quantitative trait relationship using multivariate techniques reveal diversity among tomato germplasms
Author(s) -
Ene Chikezie Onuora,
Abtew Wosene Gebreselassie,
Oselebe Happiness Ogba,
Ozi Friday Ugadu,
Ikeogu Ugochukwu Nathaniel
Publication year - 2022
Publication title -
food science and nutrition
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.614
H-Index - 27
ISSN - 2048-7177
DOI - 10.1002/fsn3.2850
Subject(s) - biology , randomized block design , biplot , trait , principal component analysis , horticulture , quantitative trait locus , botany , genotype , genetics , mathematics , statistics , computer science , gene , programming language
Tomato accessions collected from different sources were evaluated to study their diversity, genotype–traits association, as well as pinpoint most selective trait(s) in a controlled environment in Jimma, Ethiopia. The two terms pot experiments were carried out in randomized complete block design with three replications. The genotype–trait (GT) biplot revealed high percentage variability above 70% in related growth traits for the first and second principal components (PC) summed up, in the two trials, whereas related floral and fruit traits association indicated medium to high (55%–65%) total explained variations in both seasons. It further showed that ‘wild parent’, ‘CLN2498D’, ‘CLN2498F’, ‘UC Dan India’, ‘Ruma’, ‘PT4722A’, ‘CLN2679F’, ‘CLN2585C’ and ‘CLN2585D’ were the best performers in most of the related growth, floral, and fruit traits in those seasons. Principal component analysis showed that traits, such as plant height, number of branches, leaves, nodes, internodes, stem girth, style length, stigma length and diameter, flower length and width, number of flowers per truss, number of fruits per truss, and fruit weight per plant, in the first dimension were positively related to yield and consistent with high loading factors in both seasons and could be underpinned highly important in breeding for increased fruit yield. Clustering and its comparison of means showed that ‘CLN2498D’, ‘PT4722A’, ‘Ruma’, ‘Tropimech’, and ‘UC Dan India’ of cluster I in both trials expressed the best traits including related growth, floral, and fruit traits. Therefore, selection for any trait would favor accessions in this cluster.