Personalization of a computational systems biology model of blood platelet calcium signaling

Abstract

Anuclear blood cells, platelets, are the basis for the formation of blood clots in human vessels. While antiplatelet therapy is most often used after ischemic events, there is a need for its personalization due to the limited effectiveness and risks of bleeding. Previously, we developed a series of computational models to describe intracellular platelet signaling and a set of experimental methods to characterize the platelets of a given patient. To build a personalized model of platelet signaling, we also conducted research on platelet proteomics. The aim of this study was to personalize the central module of intracellular platelet signaling responsible for the formation of calcium oscillations in response to activation. The model consists of 26 ordinary differential equations. To personalize the model, proteomics data were used and unknown model parameters were selected based on experimental data on the shape and frequency of calcium oscillations in single platelets. As a result of the study, it has been shown that the key personalized parameters of the platelet oscillatory response are the degree of asymmetry of a single calcium spike and the maximum frequency of oscillations. Based on the listed experimentally determined parameters and proteomics data, an algorithm for personalization of the model has been proposed. Here we considered three healthy pediatric donors of different ages. Based on the models, personal curves of platelet calcium response to activation were obtained. The analysis of the models has shown that while there is a large heterogeneity of individual indicators of intracellular signaling, such as the activity of calcium pumps (SERCA) and inositoltriphosphate (IP₃) receptors (IP₃R), these indicators compensate each other in each donors. This observation is confirmed by the analysis of proteomics data from 15 healthy patients: this analysis demonstrates a correlation between the total amount of SERCA and IP₃R. Thus, several new features of human platelet calcium signaling are shown and an algorithm for personalizing its model is proposed.