Residual analysis in determining the error structure in enzyme kinetic data. Simulation experiments and observations on Carcinus maenas phosphofructokinase.

The nature of the experimental error in the initial velocities of an enzyme-catalysed reaction is required if meaningful least-squares regression is to be applied. When a rate equation more complex that that of Michaelis and Menten is to be solved least-squares techniques are the method of choice and so determination of the error structure becomes mandatory. The use of residual analysis and Tukey's T statistics to determine the weights to use are described. This method has the advantage of requiring no additional experimentation over that required for the primary investigation. Using data obtained for Carcinus maenas phosphofructokinase the variance was found to increase with velocity and was approximated by either an empirical power function, var (vi) alpha vi1.8 or by the function, var (vi) alpha 0.007 + vi2. The latter function is preferred and suggests that the data contains both a constant absolute error and a constant percentage error component.