Computational modelling of physiological systems

点击翻译按钮获取中文摘要

Computer modelling is used in a wide variety of scientific applications, from fluid flow, to species interactions, to diseases and other biomedical applications.  The present research involves modelling of physiological systems with the goal of obtaining a better understanding of the underlying biological processes.  Such research involves techniques and knowledge from the fields of computer science, mathematics, and physiology.  Two particular areas of focus for this research are: (i) neural network modelling with biologically realistic nodes (neurons); and (ii) thrombin generation in vascular and avascular systems.  Interest in the modelling of neural networks has been steadily increasing in recent years, due to both their applicability to pattern recognition and learning, and to the insight they can provide into the pattern storage and retrieval capabilities of the brain.  A biologically motivated model, consisting of small network building blocks found to be present in real brain networks, is important to allow comparison to experimental data as well as to allow prediction of the types of patterns that might be observed physiologically.  In this case, such models may provide a key insight into the mechanisms of biological memory.  Thrombin is the critical enzyme generated in the circulation during blood coagulation, and is also present in avascular systems (ie. those that lack a direct blood supply), such as the nervous system and ovarian follicular fluid (the fluid contained within an ovarian follicle).  By modelling this system we attempt to examine the roles of thrombin in avascular systems, which are not currently fully understood.  Applications of this research include the effects of thrombin on fertility, which is important in in vitro fertilization therapy, the use of small anticoagulant drugs, and in dysfunctional follicles, which in turn is pertinent to the dairy industry.  The importance of physiological modelling such as that presented above is underscored by the difficulty in performing many of the experiments needed to fully elucidate the underlying mechanisms.  A validated model provides a powerful predictive tool that can be used to formulate hypotheses and suggest further experiments, ultimately adding to the level of knowledge.