11-8 |
Modeling functional properties of food proteins from amino acid composition |
K. J. SIEBERT, Food Science & Technology, Cornell Univ., New York State Agricultural Experiment Station, 630 W. North St., Geneva, NY 14456-1371 Justification: It has long been of interest to model functional properties of proteins (e.g. their contribution to foaming, lipid binding, gel formation, etc.) as a function of their physicochemical properties (e.g. hydrophobicity, surface activity, viscosity, etc.). It was shown that, at least in some special cases, it was possible to model protein and peptide properties (Coomassie blue dye binding and UV molar absorptivity) from their contents of relevant amino acids and the amino acid principal properties (obtained from principal components analysis of amino acid chemical and physicochemical property data). The principal properties were expressed as three ‘z-scores’. Since peptide physicochemical properties stem from their contents of particular classes of amino acids (acidic, basic, nonpolar, etc.) it should be possible to model functional properties directly from amino acid composition. Objectives: Attempt to model protein functional properties from their complete amino acid composition and amino acid principal property data (three term z-scores or five term zz-scores). Methods: Published measurements of protein physicochemical and functional properties were modeled from the sums of the principal property contributions of each amino acid to each principal property (for each of the principal properties the moles of each amino acid in the protein were multiplied by the corresponding principal property score for that amino acid and summed) using Partial Least Squares Regression. Results: Successful models based on the contribution of each of the 20 coded amino acids to each of the five zz-score sums were developed to predict protein hydrophobicity (R=0.940), viscosity (R=0.738), and foam capacity (R=0.942). Significance: It was demonstrated that protein functional and physicochemical properties can be successfully estimated from amino acid composition using five (but not three) principal properties.
Session 11, Food Chemistry: Proteins I
|