Bioscience Evidence 2025, Vol.15, No.5, 237-248 http://bioscipublisher.com/index.php/be 241 the distribution of proteins and oils (Mo et al., 2024; Huang et al., 2025). Metabolomics studies further revealed that key metabolites such as glucose, citric acid, and α -ketoglutaric acid showed significant differences in content among different phenotypes, reflecting the competition of carbon sources between amino acid and fatty acid synthesis. Systems biology analysis has also identified a number of core genes that may simultaneously negatively regulate proteins and oils, providing new ideas for breaking through this trade-off (Kumar et al., 2021; Mo et al., 2024; Huang et al., 2025). Figure 1 Genetic regulatory network of seed size (weight), oil accumulation, and protein content in soybean. The genes or proteins involving seed size (weight) and oil content are shown in red and blue fonts, respectively. The pleiotropic regulators for seed size (weight), oil accumulation, or protein content are indicated in green fonts. The regulatory genes, whose function has been validated only in Arabidopsis but not soybean, are shown in purple fonts (Adopted from Duan et al., 2023) 6 Advances in Biotechnological and Breeding Approaches 6.1 Conventional breeding: exploitation of natural variation Conventional breeding mainly relies on methods such as hybridization, selection and mutagenesis, taking advantage of the natural variations in soybean germplasm resources to cultivate new varieties with high protein, high oil content and strong stress resistance. Both wild soybeans and local varieties worldwide offer rich genetic diversity (Anderson et al., 2019; Kumari et al., 2023; Vargas-Almendra et al., 2024). However, this method has a long cycle and limited genetic progression, making it difficult to meet the demand for rapidly increasing yield and quality (Anderson et al., 2019; Bhat and Yu, 2021). 6.2 Molecular breeding: marker-assisted selection, genomic selection Molecular breeding has greatly enhanced the efficiency of improving complex traits. Marker-assisted selection (MAS) utilizes molecular markers closely linked to target traits to achieve precise aggregation of traits such as proteins, oils, and disease resistance (Bhat and Yu, 2021; Cao et al., 2022; Lin et al., 2022; Vargas-Almendra et al.,
RkJQdWJsaXNoZXIy MjQ4ODYzNA==