Data-Driven Modelling of the Inositol Trisphosphate Receptor (\(\text {IP}_3\text {R}\)) and its Role in Calcium-Induced Calcium Release (CICR)

We review the current state of the art of data-driven modelling of the inositol trisphosphate receptor (\(\text {IP}_3\text {R}\)). After explaining that the \(\text {IP}_3\text {R}\) plays a crucial role as a central regulator in calcium dynamics, several sources of relevant experimental data are introduced. Single ion channels are best studied by recording single-channel currents under different ligand concentrations via the patch-clamp technique. The particular relevance of modal gating, the spontaneous switching between different levels of channel activity that occur even at constant ligand concentrations, is highlighted. In order to investigate the interactions of \(\text {IP}_3\text {R}\)s, calcium release from small clusters of channels, so-called calcium puffs, can be used. We then present the mathematical framework common to all models based on single-channel data, aggregated continuous-time Markov models, and give a short review of statistical approaches for parameterising these models with experimental data. The process of building a Markov model that integrates various sources of experimental data is illustrated using two recent examples, the model by Ullah et al. and the “Park–Drive” model by Siekmann et al. (Biophys. J. 2012), the only models that account for all sources of data currently available. Finally, it is demonstrated that the essential features of the Park–Drive model in different models of calcium dynamics are preserved after reducing it to a two-state model that only accounts for the switching between the inactive “park” and the active “drive” modes. This highlights the fact that modal gating is the most important mechanism of ligand regulation in the \(\text {IP}_3\text {R}\). It also emphasises that data-driven models of ion channels do not necessarily have to lead to detailed models but can be constructed so that relevant data is selected to represent ion channels at the appropriate level of complexity for a given application.