Literature data, reporting the kinetic reaction rate order of the thermal denaturation of β-lactoglobulin (β-lg), were reviewed using two statistical approaches including R2 and the Box-Cox transformation. Although R2, the coefficient of determination, has been used by many researchers as a criterion to compare the fit of data to various kinetic reaction rate orders, this statistic should not be considered appropriate for transformations of the same data set. The Box-Cox transformation, however, was constructed to systematically approach the problem of choosing a data transformation, which is the basis for the different reaction rate orders. The results of the analysis showed that, for some data sets, the reaction rate order determined using R2 was not in agreement with the Box-Cox determination. Just as importantly, the Box-Cox transformation resolved that for most data sets analyzed, the reaction rate order was indeterminate and that more than one reaction rate order could represent the data. The conflicting results reported in the literature on the thermal denaturation of β-lg could result from the statistical approach used to analyze the data and determine a reaction rate order.
Bibliographical noteFunding Information:
Published as Paper No. 98-1-18-0013 of the Contribution Series of the Minnesota Agricultural Experiment Station based on research conducted under Project 18–24, supported by University of Minnesota Computer Center and National Dairy Promotion and Research Board funds. Mention of companies and products does not constitute endorsement by the University of Minnesota or Minnesota Agricultural Experiment Station over similar products not mentioned.