| dc.description.abstract | Background: Mung bean is a pulse crop principally grown in the tropic and subtropic parts of the world for its
nutrient-rich seeds. Seven mung beans accessions from Eastern Kenya were evaluated using thirteen phenotypic
traits. In addition, 10 SSR markers were used to determine their genetic diversity and population structure. This
aimed at enhancing germplasm utilization for subsequent mung bean breeding programs.
Results: Analysis of variance for most of the phenology traits showed significant variation, with the yield traits
recording the highest. The first three principal components (PC) explained 83.4% of the overall phenotypic
variation, with the highest (PC1) being due to variation of majority of the traits studied such as pod length, plant
height, and seeds per pod. The dendogram revealed that the improved genotypes had common ancestry with the
local landraces. The seven mung beans were also genotyped using 10 microsatellite markers, eight of which
showed clear and consistent amplification profiles with scorable polymorphisms in all the studied genotypes.
Genetic diversity, allele number, and polymorphic information content (PIC) were determined using powermarker
(version 3.25) and phylogenetic tree constructed using DARWIN version 6.0.12. Analysis of molecular variance
(AMOVA) was calculated using GenALEx version 6.5. A total of 23 alleles were detected from the seven genotypes
on all the chromosomes studied with an average of 2.875 across the loci. The PIC values ranged from 0.1224
(CEDG056) to 0.5918 (CEDG092) with a mean of 0.3724. Among the markers, CEDG092 was highly informative while
the rest were reasonably informative except CEDG056, which was less informative. Gene diversity ranged from
0.1836 (CEDG050) to 0.5102 (CDED088) with an average of 0.3534. The Jaccards dissimilarity matrix indicated that
genotypes VC614850 and N26 had the highest level of dissimilarity while VC637245 and N26 had lowest
dissimilarity index. The phylogenetic tree grouped the genotypes into three clusters as revealed by population
structure analysis (K = 3), with cluster III having one unique genotype (VC6137B) only. AMOVA indicated that the
highest variation (99%) was between individual genotype. In addition, marker traits association analysis revealed 18
significant associations (P < 0.05).
Conclusion: These findings indicate sufficient variation among the studied genotypes that can be considered for
germplasm breeding programs. | en_US |
