Drug-Acceptor Interactions

Bindslevs book A"Drug-acceptor interactionsA" represents a renewal of the common toolbox in use to quantify biochemical parameters from dose-response measurements at equilibrium. Assessment of constants in such systems is frequently done by various types of transformations of variables, aimed to make the data look more or less linear, followed by an estimation of parameter values from slopes and intersects. Bindslev argues that methods based on linearizing transformations, as well as other widespread techniques such as the Hill exponential, is outdated and inadequate for modern biological research. Nevertheless, these methods are relied heavily on in many labs. As a replacement to the established toolbox Bindslev provides a novel modelling kit that the reader can use to design specialized models for the particular ligand-receptor system at hand, together with instructions on how to pull out parameter values from the non-linear models that appear. The underlying mathematics in Bindslevs book is well known; assuming that reactions follow the laws of mass action and that they are in equilibrium yields an algebraic relationship between the concentrations of the freely diffusing ligand, the ligand bound to receptors and the unbound receptors. All of the above molecular species can potentially appear in one-state models, in which the corresponding interaction schematics (i.e. arrow diagrams) become quite simple, or in multi-state models that result in more complex schemes. Either way, the ligand-receptor interaction models translate into mathematical expressions that are relatively simple, yet more complex than what typically appears when using traditional tools. This increased level of complexity does of course resemble the fact that a more detailed description of the biochemical interactions is captured. However, it is important to note that Bindslev is not introducing bigger and better mathematical models for the sole purpose of capturing more detail, but more so because it is naive to trust several of the traditional transformations or models to be appropriate for analyzing many types of ligand-receptor systems. Instead Bindslev suggests that the mathematical models used to quantify system parameters from dose-response experiments should be constructed around the unique properties of the particular system under investigation. In addition to a more dynamic approach to model construction, Bindslev is asking his readers to accept that non-linear models are, in fact, non-linear. He argues that with today's ubiquitous presence of fast computers and accommodating advanced software it is time for medical scientist to leave old pen-and-paper methods behind and replace them with modern techniques for fitting of non-linear data to non-linear models. At a first glance the models and methods Bindslev is presenting might appear to be cumbersome compared to the traditional toolbox. Undoubtedly, this is partly correct, as building specialized models and performing non-linear regressions do require a closer integration of some advanced maths and stats skills with the wet-work. However, unlike many traditional tools, the new ones do not involve unnecessary layers of abstraction, or A"meaningless mathematical manipulationsA", as Bindslev puts it in his preface. In such, the book adds significant momentum to an exciting development in biomedical research. A development directed away from the use of opaque quantification tools and pointed towards the use of tailor-made mathematical models, built to emulate the specific biochemistry of interest, to derive estimates of system constants.Hakon Kirkeby, Norwegian University of Science and TechnologyIt really is a most impressive bit of work! Arthur Christopoulos, Monash University, Australia