Study on grain yield stability, molecular diversity and multi-trait relationships among elite soybean lines.

Abstract

The demand for soybean production has increased in recent years, due to its multipurpose use for human food, livestock feed and industrial purposes. The soybean crop is one of the important source of oil and protein of the world, and is used as a source of high quality edible oil and protein. For a quantitative trait, yield is known to be influenced by changes in the environment in which the crop is grown, suggesting the need to evaluate soybean lines in different growing regions to assess their adaptability and stability. In plant breeding, selection is one of the most important stages in the breeding cycle. Multi-location testing of soybean genotypes precedes selection while genetic characterisation of germplasm enhances selection due to the variation realised and it is the basis for genetic improvement. The objectives of the study were: 1) to determine yield stability and adaptability of elite soybean lines across six locations, 2) to study genotype by trait associations and multiple trait relationships among soybean elite lines across six locations and 3) to assess the level of genetic diversity among the soybean elite lines using single nucleotide polymorphisms (SNP) markers. The stability and adaptation study was carried out to investigate genotype by environment interaction (GEI) for grain yield of 26 elite soybean lines along with four checks grown in 6 environments spreading across three countries (Zambia, Malawi and Mozambique) in a 6 x 5 alpha lattice design. The additive main effect and multiplicative interaction model (AMMI) indicated that environments, genotypes and GEI significantly affected grain yield (P<0.001) and contributed 3.8%, 17% and 78%, respectively, to the total variation. Three AMMI interaction principal components (IPCA1, IPCA2 and IPCA3) were significant (P<0.01). Genotype plus GEI (GGE) biplots were created based on the first two principal components, PC1 and PC2, which accounted for 39.23% and 26.86% of genotype plus GEI variation, respectively. The GGE biplot analysis ranked the genotypes for yield and stability, and environments for representativeness and discriminativeness. The relationships between genotypes and environments were also demonstrated. Genotype TGX 2001-3FM was identified as the ideal genotype with high yield mean performance and high stability. Therefore, it could be recommended for cultivar release if the study can be repeated to verify these findings. Chitedze in Malawi was the most informative test environment hence it is ideal for selecting generally adapted genotypes. Genotypes TGX 2002-4FM and TGX 2001-15DM were low yielding but with high stability hence can be recommended for further improvements. For the second objective, a study was conducted using 30 genotypes to determine the correlation and path coefficient of secondary traits on yield. The genotype by trait biplot is a tool that graphically compares genotypes on the basis of multiple traits and graphically visualises trait relationships, and genotype-trait associations. Trait profiling of genotypes through genotype-trait association analysis helps in making decisions on which genotypes to use as parents for a breeding programme. Significant differences among genotypes were observed for all studied traits. Correlation coefficient analysis presented that grain yield had a significant and negative correlation with days to 50% flowering. However, grain yield had a significant and positive correlation with plant height. Path coefficient analysis indicated that plant height and early vigour had a positive direct effect on yield while days to 50% flowering and days to 50% podding had negative indirect effects on yield via days to maturity. The genotype by trait biplot graphically showed consistent trait relationships and identified TGX 2001-3FM, TGX 2001-26DM and TGX 2002-3DM as genotypes that can be used as parents in breeding programmes for yield improvement. Estimation of genetic diversity among 48 soybean lines from the International Institute for Tropical Agriculture (IITA) was conducted using 348 SNP markers. The average gene diversity and genetic distance ranged from 0.42 to 0.55 with an average of 0.47 and 0.61 to 0.87, respectively. The polymorphic information content ranged from 0.44 to 0.50 with a mean of 0.48. Genotypes TGX 2002-3DM and TGX 2002-3FM had the highest genetic distance between them indicating that they were highly diverse. The AMOVA indicated highly significant differences at F=0.001 with among individuals, among populations and within individuals contributing 45%, 28% and 26%, respectively. The 48 soybean lines were clustered in three main groups. The study indicated that genetic diversity exists among the IITA tested lines. The information obtained from the study, can be fully utilised in future soybean breeding programmes through crossing of diverse parents in order to incorporate new alleles to develop improved cultivars. In general, the study showed the existence of genotype by environment of soybean grain yield across the selected locations in southern Africa. Based on the SNP markers, the study confirmed the existence of wide genetic diversity among the soybean lines. The lines with superior performances can be used for future breeding programmes or recommended for cultivar release.