This study compared the protein profile of 15 different commercial soy milks using a label-free quantitative proteomics approach. Proteins related to nutrient reservoir activity, endopeptidase inhibitor activity, lipid binding, and seed maturation contribute the most in terms of percentage mass. Their associated Gene Oontology terms are also enriched. Samples could be clustered into three groups based on their protein composition, with glycinins and beta-conglycinins being the most influential for determining the clustering. Amino acid composition estimated from the proteomics data also reflects the clustering of samples. Twenty allergenic proteins varying in abundance were identified, with Gly m 5 and Gly m 6 being the predominantly abundant allergens.