BBP Somatosensory Cortex model - SONATA version

This dataset contains the SONATA version of the BBP Somatosensory cortex model. The model itself is described in the following manuscripts: Modeling and Simulation of Neocortical Micro- and Mesocircuitry. Part I: Anatomy and Modeling and Simulation of Neocortical Micro- and Mesocircuitry. Part II: Physiology and Experimentation We ask you to please cite these papers if you use the model for a publication! A subvolume of this model containing only ~210,000 neurons has been published here. If the ~750GB download of the full model seems too much for you, the subvolume may be your choice instead. The format of the model is described in the SONATA documentation The dataset describes geometry, anatomy and physiology of the cortical neurons and their synaptic connections. The model can be simulated under in vitro and in vivo-like conditions. For details on what to do with the model and how, see the list of recommended tools in the notes below. To use the model, you will first have to uncompress the archives to the same directory. For "synapses.xz", you will first have to concatenate the individual parts, followed by decompressing the result: cat synapses.xz* > synapses.tar.xz xz -d synapses.tar.xz tar -xf synapses.tar The other files can be directly uncompressed using unzip. ARCHIVES We have split the model into several archives that contain different aspects of the model. Individual archives are split into files of 45GB or less. Some of the archives / aspects of the circuit are useful in isolation, others probably only together with the rest. circuit.zip: Contains descriptions of the neurons in the model and their physiology. Contains a reference to the name of the morphology assigned to each neuron, but not the files describing the morphologies themselves. Most importantly, this archive contains the circuit_config.json, which lists references to all parts of the model and is therefore the starting point for all circuit analyses and simulations. For details, see the BluepySNAP documentation. With this archive alone, you will be able to perform analyses of the neuronal composition of the circuit model, e.g. how many neurons of each type exist and where they are placed. This can provide important context for the analyses of result of simulation campaigns, such as the ones we publish here. The results of a simulation campaign will for example tell you the spike times of each individual neuron, and the contents of this archive allow you to look up the location and type of that neuron. For examples of simulation analyses, see here. However, with this archive alone you will not be able to analyze neuron morphologies nor their connectivity. Calling the related functions will lead to errors. It will also not allow the execution of simulations. To that end, you will have to extract the contents of the other archives to the same location. morphologies_swc_fmt.zip: Contains descriptions of the dendritic and axonal morphologies of the neurons in the model. Given in the .swc format. This format is the standard expected by SONATA. For working with the morphologies, see the recommended tools listed in the notes below. morphologies_asc.zip: Contains descriptions of the dendritic and axonal morphologies of the neurons in the model. Given in the Neurolucida .asc format. There is a one-to-one correspondence between these morphologies and the ones in morphologies.swc.zip. We provide these copies of the morphologies because the .asc format is required to simulate the model with NEURODAMUS. synapses.xz: Contains the description of the synaptic connectivity of the model. Four connectomes are contains: Local connectivity within the model, long-range connectivity within the model, extrinsic inputs from VPM, and extrinsic inputs from POm. See note above for uncompressing the files! We recommend you also download and uncompress into the same directory the brain atlas used for the construction of the model. While the atlas is not part of the SONATA format and is not required to run simulations, it enables some advanced analysis in Connectome Utilities (see below).