Johnson Use of least squares techniques in biochemistry, Analytical Biochemistry 206英文电子书.pdf

[1] NONLINEAR LEAST-SQUARES 1 [1] U s e of L e a s t - S q u a r e s T e c h n i q u e s in B i o c h e m i s t r y * By MICHAEL L. JOHNSON Introduction There are relatively few methods available for the analysis of experi- mental data in the biochemical laboratory. Graphical methods and least- squares (regression) methods are by far the most common. Unfortunately, both classes of analysis methods are commonly misused. The purpose of this chapter is to explain why, when, and how a biochemist should use least-squares techniques and what confidence can be assigned to the result- ing estimated parameters. One classic group of biochemical experiments involves measuring the response of a system to an external perturbation. Temperature-jump ex- periments perturb the chemical equilibrium of a solution by rapidly increas- ing the temperature of the solution and subsequently monitoring an observ- able, like absorbance, as a function of time. Here the absorbance is the observable (i.e., the variable) that is dependent on the experiment, and time is the variable that can be independently controlled by the experimen- tal protocol. Another example of this general class is the ligand-binding titration experiment. The investigator measures the amount of a ligand bound (the dependent variable) by fluorescence, absorbance, or radioactive counting. To do so, the investigator titrates the ligand concentration (the indepen- dent variable). Note that the ligand concentrations might be either the total or the free ligand concentration, depending on the experimental protocol. In these examples, and all others of this class, the investigator has measured a response caused by a perturbation of the system. The next step is to obtai