Engineering Neuron Models

The neocortex is one of the most evolved and complex region of the brain. For more than a century scientists have been curious about the neocortex, to identify the anatomical blueprint of its cellular organization and to understand its role in higher brain functions. The Blue Brain Project aims to study the neocortex of the rat by developing its basic anatomical unit, the cortical column, in a simulation-based, data driven research environment. This requires 10,000 biologically accurate neuron models and a combination of the full spectrum of genetically determined ion channels, to capture the complete electrical diversity in a biologically constrained manner. This dissertation, which was carried out within the scope of the Blue Brain Project, illustrates a framework to build neuron models and its integration for neuron network simulation. Building biophysically accurate neuron models requires realistic ion channel models of around 200 of the different types of ion channels expressed in the neocortex. Although a significant amount of experimental data has been gathered over the past 30 years, consolidation of these findings into an easily accessible online resource is still missing. Moreover, the differences in experimental conditions make it difficult to faithfully create models for these ion channels. Therefore, to consolidate existing ion channel literature, a knowledge base system Channelpedia (www.channelpedia.net) has been developed. Equipped with 187 annotated ion channels with 50 Hodgkin-Huxley(HH) models, Channelpedia provides an ideal discussion platform, for researchers to collaborate and synthesize information from literature. To address the unavailability of experimental data, a high-throughput ion channel screening method was developed. This approach involves ion channel gene transfection in Chinese Hamster Ovarian (CHO) cell lines, automated voltage-clamp experiments, and an automated HH model fitting routine. Finally, to validate the role of ion channels in a model neuron, I perform a series of in-vitro and in-silico dynamic clamp experiments on layer 2/3 pyramidal neurons.