Perić Vesna, Kravić Natalija, Tabaković Marijenka, Mladenović Drinić Snežana, Nikolić Valentina, Simić Marijana, Nikolić Ana
Maize Research Institute Zemun Polje, Slobodana Bajića 1, 11185 Belgrade, Serbia.
Plants (Basel). 2025 Jan 12;14(2):201. doi: 10.3390/plants14020201.
Driven by the growing demands for plant-based protein in Europe and attempts of soybean breeding programs to improve the productivity of created varieties, this study aimed to enhance genetic resource utilization efficiency by providing information relevant to well-focused breeding targets. A set of 90 accessions was subjected to a comprehensive assessment of genetic diversity in a soybean working collection using three marker types: morphological descriptors, agronomic traits, and SSRs. Genotype grouping patterns varied among the markers, displaying the best congruence with pedigree data and maturity for SSRs and agronomic traits, respectively. The clear origin-related grouping pattern was not observed for any of the marker types. For the diversity assessed by morphological descriptors, Homogeneity Analysis by Means of Alternating Least Squares (HOMALS) yielded the most efficient classification by identifying the traits with the highest discriminative power and separating the genotypes into homogeneous groups. According to genetic distances (GDs), the highest diversity was found for morphological descriptors (GD = 517), followed by SSRs (GD = 0.317) and agronomic traits (GD = 0.244). The analysis of molecular variance (AMOVA) revealed a weak differentiation between geographic groups (Φ = 0.061), emphasizing the highest differentiation for Canadian genotypes (Φ = 0.148 **). A low correlation was found between molecular and morphological, i.e., agronomic trait-based matrices (0.061 *, i.e., -0.027, respectively). The overall assessed diversity highlighted the importance of introducing new sources of variation to promote long-term improvement in soybean breeding.
受欧洲对植物性蛋白质需求不断增长以及大豆育种计划提高育成品种生产力的尝试推动，本研究旨在通过提供与重点明确的育种目标相关的信息来提高遗传资源利用效率。使用三种标记类型（形态描述符、农艺性状和简单序列重复标记（SSRs））对一组90份种质进行了大豆工作收集品遗传多样性的综合评估。标记间的基因型分组模式各不相同，其中SSR标记和农艺性状分别与系谱数据和成熟度显示出最佳一致性。对于任何一种标记类型，均未观察到明显的与起源相关的分组模式。对于通过形态描述符评估的多样性，交替最小二乘法均值同质性分析（HOMALS）通过识别具有最高判别力的性状并将基因型分为同质组，产生了最有效的分类。根据遗传距离（GDs），形态描述符的多样性最高（GD = 517），其次是SSR标记（GD = 0.317）和农艺性状（GD = 0.244）。分子方差分析（AMOVA）显示地理组间的分化较弱（Φ = 0.061），强调加拿大基因型的分化最高（Φ = 0.148 **）。分子与形态学（即基于农艺性状的矩阵）之间的相关性较低（分别为0.061 *和 -0.027）。总体评估的多样性突出了引入新变异来源以促进大豆育种长期改良的重要性。
