Soybean (Glycine max (L.) Merrill) is a key source of plant protein, with 7S and 11S globulins being the primary fractions. This study investigated the protein content, protein composition, and 11S/7S globulin ratios of 411 soybean samples, and then selected six varieties (S1, S2, S3, S4, S5, and S6) for the analysis of the protein structures and functional properties. The results revealed that varieties with low 11S/7S ratios (S1, S2, and S3) exhibited relatively high random coil contents (20.11–22.94%) and lower β-sheet contents (34.17–38.37%), suggesting the presence of more loosely structured proteins. S2 showed good solubility (73.21%) and water-holding capacity (WHC) (2.73 g/g), which can improve product quality and yield. In contrast, varieties with high 11S/7S ratios (S4, S5, and S6) demonstrated more compact protein structures, increased surface hydrophobicity, larger particle sizes, lower absolute zeta potential values, and greater oil-holding capacity (OHC) values (7.58–8.48 g/g). S4, in particular, demonstrated superior emulsification properties, with emulsion activity index (EAI) (4.71 m²/g) and emulsion stability index (ESI) (58.73 min), which are widely used in the food industry such as in cake, ice cream, and bread. This study provides valuable information for the selection of soybean varieties with optimal 11S/7S ratios for processing soybean products.

To understand the mechanism of sweetness formation in vegetable soybean seeds, an integrative transcriptomic and metabolomic analysis was conducted using high sweetness (HS) and low sweetness (LS) varieties selected from 287 resources based on electronic tongue evaluation. The HS variety exhibited significantly higher levels of soluble sugars (58.91 mg/g) and free amino acids (54.15 mg/g). Transcriptomic results indicated that DEGs correlated with glycolysis/gluconeogenesis, starch and sucrose metabolism, and biosynthesis of amino acids pathways were significantly up-regulated in the HS variety. Metabolomic analysis showed that DAMs were significantly enriched in the biosynthesis of secondary metabolites, the amino acid biosynthesis, and the pentose phosphate pathway. Co-expression network analysis further demonstrated correlations between DEGs and DAMs related to glycolysis/gluconeogenesis and amino acid biosynthesis. Eight candidate genes related to sweetness formation were identified through transcriptomic data and validated by RT-qPCR. The present findings represent a fundamental advance in understanding the regulatory mechanisms underlying the sweetness of vegetable soybeans.