Skip to content

Challenge Level 1

🚧 Under Construction

This page is actively being developed. Content, specifications, and data sets are subject to change. Some sections may be incomplete or contain placeholders marked with ??. Check back soon for updates!

📋 Overview

Challenge Level 1 represents the first tier of the Whole Brain Emulation (WBE) Challenge series. This level introduces fundamental abstractions and simplifications that enable initial exploration of virtual brain data while maintaining biological plausibility.

Abstractions and Simplifications

The first sophistication level of the WBE Challenge features these characteristics and simplifications of virtual brain data:

  • Empty extracellular space: There are no glia, vasculature or other components in the exracellular space. Only neural cell bodies and their processes are present.

  • Very small sample size: The number of neurons in the sample is very small.

  • Two types of neurons: Only 2 types of neurons are present, a generic principal neuron and a generic interneuron. The two types are easily distinguished.

  • Zeroth order morphology: Neuron morphologies are simplified to the point where neurons are composed of only 1 soma compartment and 1 axon compartment. Synaptic connections are made directly onto the soma of a target cell.

  • Two receptor types: Connections use one of 2 receptor types, one used in inhibitory connections from presynaptic interneurons (a GABA type), and one used in excitatory connections from presynaptic principal neurons (an AMPA type).

  • Zeroth order synaptic connections: Synaptic connectivity between any pair of neurons, where present, is achieved by a single synapse (seen in EM as a single postsynaptic density).

Challenge Data Set

📌 Status

All challenge data sets are currently in preparation. The EM image stack will be released soon.

EM Images

The EM image stack was produced using a virtual FIBSEM process.

  • There is 1 imaged sample region.
  • The total sample volume imaged is ??.
  • The z-distance between successive imaging depths is ??.
  • The x-y pixel resolution is ?? on each axis.
  • Each image contains 1024x1024 pixels.
  • Image overlap in both x and y is ??.
  • The data set comprises a total of ?? images with a total size of ??.

Calcium Imaging

🚧 Coming Soon

Calcium imaging data specifications are currently being developed. This section will be updated with detailed information upon completion.

Extracellular Electrode Recording

📊 Data Format

Standardized recording data format for all electrode measurements.

The data from each recording electrode is contained in a JSON file with the following format:

{
  "Name": string,         # Unique name of the electrode.
  "TipPosition": list,    # X,Y,Z coordinates of the electrode tip in the sample.
  "EndPosition": list,    # X,Y,Z coordinates of the end of the electrode.
  "Sites": list of lists, # Relative X,Y,Z locations of each recording site on the electrode (see below).
  "TimeStamp_ms": list,   # List of recording time points.
  "ElectricField_mV": list of lists, # For each site, a list of recorded voltages at each time point.
}

The relative X,Y,Z location of a recording site on an electrode is expressed as a ratio of the electrode X,Y,Z dimensions. For example, a site at [0, 0, 0.5] is located at an X,Y edge of an electrode, halfway up its Z length.

  • This data set contains JSON files with recordings from ?? electrodes.
  • Each recording was made over a time interval of ?? ms.

Minimum electrode spacing is 250 µm and was based on the spacing between shanks of a four-shank Neuropixels electrode 1.

🚧 Not Yet Available

The complete Level 1 data set will be available for download upon release. Please check back later or subscribe to our newsletter for announcements.

📧 Stay Updated

Want to be notified when the data set is released? Subscribe to our newsletter or contact us at wbechallenge@carboncopies.org.

Download the complete data set as a compressed file at this link:

something-something-something


🤝 Collaboration Welcome

The WBE Challenge is developed and supported by the Virtual Brain Platform (VBP) and Neural Emulation System (NES) groups of the BrainGenix division of the Carboncopies Foundation R&D.

The Carboncopies Foundation (CCF) is a 501©(3) non-profit research and education organization.

For questions or comments, please write to wbechallenge@carboncopies.org.