The design of pharmacological experiments using unbiased models of agonist-antagonist interaction

The methods which exist to represent agonist-antagonist interactions, and to distinguish between them are sufficient and effective. Their effective use depends on a set of criteria which are not always fulfilled in actual experiments. In some situations it is difficult or impossible to meet these criteria. An alternative representation is proposed, using straightforward geometry, which evades this problem. It also allows experimental data points to be selected without bias and makes it possible to apply orthodox criteria to more restricted sets of data than is otherwise the case. To do this, it is necessary to reverse the ordinary procedure in antagonist assays, by adding successive concentrations of antagonist to a preparation so that a constant response is evoked by squared multiples of the original agonist dose. This can only be done with rapidly reversible antagonists, but the method has been shown to be effective for such drugs.