A logarithmic approximation to initial rates of enzyme reactions.

A straightforward and empirical regression method based on a logarithmic approximation has been developed to accurately estimate initial rates from nonlinear progress curves of enzyme reactions. The principle of this parametric approach is to use a relatively large number of observations, while averaging out random errors, to predict the curvature at time zero, which has the highest rate of change. The usual linear regression of a few initial time points lacks prediction power at time zero and therefore underestimates the true initial rate. Application of this nonlinear regression approach to enzyme reactions demonstrated satisfactory results. This approach is less subjective in choosing initial time points to be used for rate determination, and much more robust to random errors. Moreover, it is relatively easy to realize with commonly available software.