Review Article for Expert Review of Neurotherapeutics

Epilepsy is a neurological disorder characterized by the recurrence of seizures. It affects 50 million people worldwide. Although a considerable number of new anti-epileptic drugs with reduced side-effects and toxicity have been introduced since the 1950s, 30% of patients remain pharmacoresistant. Although epilepsy research is making progress, advances in understanding the drug resistance have been hampered by the complexity of the underlying neuronal systems responsible for epileptic activity. In such systems where shortor long-term plasticity plays a role, pathophysiological alterations may take place at sub-cellular (membrane ion channels, neurotransmitter receptors), cellular (neurons), tissular (networks of neurons) and regional (networks of networks of neurons) scales. In such a context, the demand for integrative approaches is high and neurocomputational models become recognized tools for tackling the complexity of epileptic phenomena. The purpose of this report is to provide an overview on computational modeling as a way of structuring and interpreting multi-modal data recorded from the epileptic brain. Some examples are briefly described. They illustrate how computational models closely related with either experimental or clinical data can markedly advance our understanding of essential issues in epilepsy like the transition from background to seizure activity. A commentary is also made on the potential use of such models in the study of therapeutic strategies like rational drug design or electrical stimulations.