ABSTRACT
There are four general designs of assay for the characterization of labeled and unlabeled ligands: dissociation, association, saturation, and inhibition. The assays are interpreted initially by fitting the data to a mechanistic model, the Law of Mass Action or, in the case of inhibition curves, to an empirical model, a logistic function. Before computers were generally available, it was necessary to transform the data with a function that yields a straight line when applied to Law of Mass Action data, and then use linear regression analysis with the transformed data. The most commonly used transformations will be presented for historical reasons, and because the linear plots are useful intuitively and for detecting deviations from the model. Wherever possible, however, parameter estimates should be obtained by fitting the untransformed data directly to the model using nonlinear regression analysis. The appropriate equations will be presented, although most curve-fitting programs have the equations already set up.
