Predictive and reconstructive modelling for rare disease etiology: demystifying disordered hemato-immune interactions during cyclic thrombocytopenia

Morgan Craig
Département de mathématiques et de statistique, Université de Montréal

Our understanding of how blood cells are produced and function and how the hematopoietic and immune systems are altered during pathogenesis impacts directly on our ability to diagnose and treat serious hemato-immune diseases. When dysregulation occurs within the hemato-immune hierarchy, a variety of serious pathologies may ensue. A number of hemato-immune disorders, including the rare oscillatory disease cyclic thrombocytopenia (periodic fluctuations in megakaryocyte and platelet numbers, and thrombopoietin, the principal cytokine regulating thrombopoiesis) results from communication breakdowns and dysfunctionality within regulatory cytokine networks. In some cases, diseases arise from specific mutations, however some healthy individuals can carry mutations without impact to their health, while others have the disease in absence of mutation. Unfortunately, broadly speaking, how hemato-immune disorders arise and the consequences of disrupted or broken cytokine networks remains poorly understood.
Here I will discuss two methodologies - predictive modelling using delay-differential equations and reconstructive modelling using convergent cross-mapping, a dynamical systems technique - to understand the mechanisms driving cyclic thrombocytopenia, and to identify causal, dynamic relationships between the cells and cytokines interacting within hemato-immune regulatory circuits.