Analysis of cows', ewes’ and goats’ milk mixtures by capillary electrophoresis: quantification by multivariate regression analysis

Abstract

Capillary electrophoresis was used to separate and analyze the casein fraction in mixtures of cows', goats’ and ewes’ milk. The differences between the capillary electrophoresis patterns of the casein fraction from the whole milk of each species allowed identification and quantitative determination of milk in binary and ternary mixtures. Principal components regression and partial least-squares regression were used to predict the percentages of milk of each species included in a mixture. The predictor variables were the peak areas of bovine and ovine αs1-casein; bovine, ovine and caprine κ-casein; bovine β-casein A1 and A2; and ovine and caprine β2-casein and β1-casein. The root mean square errors in prediction were lower than 2.4% in all cases.