cam a22 7 4500 316985481 CHVBK 20201029052737.0 cr ## ######## 140620s2014 sz sm 00 0 eng d oai:infoscience.epfl.ch:198684 urn:nbn:ch:bel-epfl-thesis6168-9 urn 10.5075/epfl-thesis-6168 doi (SNL)991017980981303976 (NEBIS)010138238 (Sz)991017980981303976 (Sz)001871454 Sz ger rda 610 23sdnb Reimann Michael 1981- (DE-588)1116584069 Verfasser aut The In-Silico Neocortical Microcircuit From Structure to Dynamics Michael Reimann Lausanne 2014 1 Online-Ressource Thèse no 6168 EPF Lausanne, 2014 application/pdf The neocortex is one of the most important brain regions, occupying more than 80% of the brain volume in mammals and contributing massively to its capability of perceiving and interpreting then environment. While it has been extensively studied, it is still largely unclear how it accomplishes the enormous computational tasks it is faced with. A number of structural features are conserved across the cortex, such as its vertical organization in six individual layers with a large number of morphologically unique neuron types. The exceptional horizontal uniformity, lead to the hypothesis of horizontally repeated unitary microcircuits, the idea that the cortex consists of a number of relatively uniform horizontally repeated modules, each with their own function in the concert of neural activity. While a rough mapping of function to cortical region is virtually identical between individuals and very similar between species, the extraordinary plasticity of the cortex allows functions of damaged parts to be taken over by neighboring regions. This indicates that flexible, "reprogrammable" modules implement the functions of the cortex. The Blue Brain Project (BBP) aims to build a facility for the collection and integration of available data on the fundamental structure and operating principles of the cortex to build and continuously refine a detailed, bottom-up model of the unitary microcircuit. It can then be used as a building block for the construction and simulation of larger cortical regions. A bottom-upmodel such as the one of the BBP aims to match the object being represented primarily on an underlying physical level. As such it is guaranteed to be anchored in real world physical constraints, this however comes with the disadvantage that on a higher level the behavior of the model is undetermined. The goal of this thesis is to study and characterize the model to learnmore about its function and what determines it. In order to ensure the biological relevance, this endeavour was to be tightly linked with comparisons to and validations against relevant experimental data. I concluded that the simulation of extracellular electrical signals are the right tool to link characterization and validation, due to their great importance in in-vitro work. I authored several improvements to the model to ensure the biological accuracy of the computed extracellular signals, the most important being the derivation of the local connectome inside the modeled microcircuit. Next, I implemented software for the calculation of extracellular signals during the course of a simulation on a BlueGene/P supercomputer. The method for the calculation is flexible and can discern between the contributions of individual cells. In addition to validating the signal against in-vitro data, this allowed us to analyze the composition of the signals in a way that is impossible outside of a simulation. The method revealed the crucial role of active currents in the local field potential, i.e. the relatively slow deflections in extracellular signals. We also extensively analyzed and quantified the varying contributions of individual cell types and membrane mechanisms to extracellular signals at different frequencies. MIKROMODULE + MINIATURISIERUNGSTECHNIKEN (MIKROELEKTRONIK) ger (ETHUDK)000028105 ethudk NEOCORTEX (NEUROLOGIE) ger (ETHUDK)000040171 ethudk NEURONALE PLASTIZITÄT (NEUROLOGIE) ger (ETHUDK)000039298 ethudk SYNAPTISCHE ÜBERTRAGUNG (NEUROLOGIE) ger (ETHUDK)000043609 ethudk Hochschulschrift (DE-588)4113937-9 gnd-content Hochschulschrift gnd-content xehelv xediss 2017 610 snl ehelv sb 2017/26 610 snl-sb u NEOCORTEX (NEUROLOGIE) ger 612.825*1 nebis E1 u MIKROMODULE + MINIATURISIERUNGSTECHNIKEN (MIKROELEKTRONIK) ger 621.3.049.76 nebis E1 u SYNAPTISCHE ÜBERTRAGUNG (NEUROLOGIE) ger 611-018.8-018.1,117.016.7 nebis E1 u NEURONALE PLASTIZITÄT (NEUROLOGIE) ger 611-018.81,2.03 nebis E1 u NÉOCORTEX (NEUROLOGIE) fre 612.825*1 nebis E1 u NEOCORTEX (NEUROLOGY) eng 612.825*1 nebis E1 u TECHNIQUES DE MINIATURISATION + MICROMODULES (MICROÉLECTRONIQUE) fre 621.3.049.76 nebis E1 u MINIATURIZATION + MICROMODULES (MICROELECTRONICS) eng 621.3.049.76 nebis E1 u SYNAPTIC TRANSMISSION (NEUROLOGY) eng 611-018.8-018.1,117.016.7 nebis E1 u TRANSMISSION SYNAPTIQUE (NEUROLOGIE) fre 611-018.8-018.1,117.016.7 nebis E1 u NEURAL PLASTICITY (NEUROLOGY) eng 611-018.81,2.03 nebis E1 u PLASTICITÉ NEURONALE (NEUROLOGIE) fre 611-018.81,2.03 nebis E1 École polytechnique fédérale de Lausanne (DE-588)1003500-X Grad-verleihende Institution dgg SNL NB001 NB999 - chb99 - Record https://infoscience.epfl.ch/record/198684 Thesis https://infoscience.epfl.ch/record/198684/files/EPFL_TH6168.pdf URN https://nbn-resolving.org/urn/resolver.pl?urn=urn:nbn:ch:bel-epfl-thesis6168-9 http://www.e-helvetica.nb.admin.ch/directAccess?callnumber=bel-epfl-thesis6168-9 BK020353 XK020053 XK020000 1 Hochschulschrift = Thèse/Mémoire 2 CF Elektron. Daten Fernzugriff=Fichier online EPF nebis E2 e-library 2014 these nebis EC 133 E01-20140623 NEBIS E16 E16 RH HB 16888 NEBIS E02 E02 E02SP ZTH 6168 SNL 100 1- Reimann Michael 1981- (DE-588)1116584069 Verfasser aut SNL 710 2- École polytechnique fédérale de Lausanne (DE-588)1003500-X Grad-verleihende Institution dgg SNL 856 40 Record https://infoscience.epfl.ch/record/198684 SNL 856 40 Thesis https://infoscience.epfl.ch/record/198684/files/EPFL_TH6168.pdf SNL 856 4- URN https://nbn-resolving.org/urn/resolver.pl?urn=urn:nbn:ch:bel-epfl-thesis6168-9 SNL 856 42 http://www.e-helvetica.nb.admin.ch/directAccess?callnumber=bel-epfl-thesis6168-9 NEBIS 700 1- Reimann Michael NEBIS 700 1- Hill Sean Lewis Akademischer Betreuer dgs NEBIS 700 1- Markram Henry 1962- (DE-588)1089464851 Akademischer Betreuer dgs NEBIS 856 -- https://dx.doi.org/10.5075/epfl-thesis-6168 Online via EPFL E-Library E02 SWISSBIB 316985481 ORANGE 316985481