Differentiation of Soy Sauce Produced from Whole Soybeans and Defatted Soybeans by Pattern Recognition Analysis of HPLC Profiles

Abstract

Nonvolatile components in commercial fermented soy sauce produced from whole soybeans and defatted soybeans were analyzed by reversed phase HPLC (RP-HPLC). Differentiation of soy sauce produced from two types of soybeans was attempted by pattern recognition analysis of HPLC profiles. Two distinct groups were observed in both the factor score plot and the clustering dendrogram of HPLC profiles. Two soy sauce groups were clearly differentiated by soft independent modeling of class analogy (SIMCA), where ferulic acid was identified as one of the key components for the differentiation. A fractional factorial experimental design combined with multiple linear regression analysis was applied to determine components associated with types of soybeans. Daidzein and three isoflavone derivatives besides ferulic acid were identified as typical components that increased when soy sauce was produced from whole soybeans. Keywords: Soy sauce; pattern recognition; cluster analysis; SIMCA; fractional factorial design