Repository logo

Characterisation of sweet sorghum germplasm based on agro-morphological traits, molecular markers and juice related traits.

Thumbnail Image



Journal Title

Journal ISSN

Volume Title



There is rising interest for alternative energy sources because of the decline in fossil fuel production and concern over environmental pollution. Currently most biofuel is based on maize and sugar cane as raw materials. However, the use of feedstocks has triggered concerns related to food security, while sugar cane has a high-water consumption and high production requirements amongst other drawbacks. A crop which meets several requirements for biofuel (such as high biomass yield and growth rate, perennial growth, low input requirements, adaptation to the marginal areas, and tolerance to multiple stresses) is sweet sorghum. This study, therefore, aimed at characterising sweet sorghum germplasm using agro-morphological traits and molecular markers (single nucleotide polymorphisms (SNP) during the 2016-2017 summer season at two sites (Ukulinga farm and Potchefstroom). Fourteen quantitative traits were evaluated in an alpha lattice (10 x 5) design with three replications. Analysis of variance for the quantitative traits revealed high levels of genetic variability. This implies that morphological traits differed greatly with a significant G x E interaction across the two sites. Most of the accessions yielded high at Ukulinga than Potchefstroom on juice yield and %brix with a mean yield of 9 605 l/ha and 16.3%, respectively. Most of the accessions studied were early to medium maturing, as evidenced by the mean number of days to 50% flowering (74 days). Analysis of principle components showed that the first four principle components (PC) accounted for 79.12% of the total variation and that some quantitative traits were significantly positively correlated. The estimates for phenotypic coefficient of variation (PCV) were higher than those of genetic coefficient of variation (GCV) for all the traits, indicating the influence of the environment on these traits. However, GCV values for days to 50% flowering, plant height, stalk diameter and stalk yield were very close to PCV. This indicated minimal influence of the environment on the phenotypic expression of these traits. The highest broad sense heritability (H2) of 99.2% was recorded for plant height. Juice volume had the highest expected genetic advance, expressed as a percentage of mean (GAM) of 131.2%. Days to 50% flowering were significantly and positively correlated to plant height, stalk diameter, number of leaves, stalk yield, brix, juice volume and bagasse weight, but negatively significantly correlated to panicle length, panicle width, panicle weight and 1000 grain weight. Plant height was significantly positively correlated to stalk diameter, number of leaves per plant, stalk yield, juice volume and fresh bagasse weight. Bagasse weight, brix, stalk diameter, plant height and number of leaves had a highly positive and direct contribution on juice yield. Several traits had a highly positively and indirect contribution on juice yield via these traits which had a direct contribution. This revealed primary and secondary traits with practical relevance to sweet sorghum improvement programme, because they showed direct and indirect effects on juice yield (volume), which ultimately translates to sugar yield for ethanol production. Kompetitive Allele Specific Polymorphism (KASP) genotyping using 137 SNP markers revealed a considerable level of genetic diversity among the sweet sorghum accessions. Three populations were generated from the analysis. The expected heterozygosity (He) values ranged from 0.236 to 0.291 with a mean of 0.266. The mean of effective alleles across populations was of 1.438. The percentage of polymorphic loci ranged from 80.29% to 91.24% with a mean of 86.86%. Dissimilarity indices ranged from 0.000 to 0.583 with a mean of 0.296. The highest dissimilarity index was observed between SA 2193 and SA 2014, which implied a considerable amount of genetic diversity. Accessions were clustered into three main groups based on dissimilarity indices. The study identified SA 4490, SA 2400, SA 4495, SA 2193 and SA 4479 as superior accessions in juice yield. These accessions should be used as parents in sweet sorghum improvement programme.


Master of Science in Plant Breeding. University of KwaZulu-Natal, Pietermaritzburg, 2018.