This is an interactive visualization of the synaptic connectivity of the rat non-barrel somatosensory cortex, as modeled by Blue Brain. It depicts the strengths of the various synaptic pathways in terms of the number of connection, number of synapses and the total physiological strength, calculated as the sum of maximal conductance multiplied with the utilization parameter.

Note: The visualization can take a while to load, please be patient!

Usage

First, select which type of pathway strength you want to consider from the dropdown menu labelled "Data set" in the bottom left.

Next set up how the grouping of synaptic connections. You can select up to four groupings using the controls at the top. Each grouping is then indicated in the main plot by a number of boxes. Hover your mouse over a box to find out the total strength of that class of connections. Links between adjacent groupings indicate the strengths of connections belonging to both the class on the left and right. Possible groups are:

By presynaptic neuron layer (Pre layer)

By presynaptic neuron type (Pre mtype)

By postsynaptic neuron layer (Post layer)

By postsynaptic neuron type (Post mtype)

By connection mechanism (Range), i.e. local short-range, mid-range or thalamo-cortical connections

Any combination of the above

Optionally, you can select up to two filters using the controls in the middle. For each filter select a grouping and any number of valid values and only connections of that class will be considered

Finally, you can optionally normalize the data. Select a grouping using the "Normalize by group" control at the bottom right. The data will be normalized such that total strength of each class within that grouping will sum up to exactly 1. This is useful for studying for example what fractions of synapses from a type of neuron are formed onto neurons in each layer. Alternatively, you can normalize by neuron count. Use the "Normalize by external data" dropdown to select normalization by pre- or postsynaptic neuron count. The width of each link is then divided by the number of neurons relevant for the link. For example, if normalization by presynaptic neuron count is selected, a link from pre layer 5 to post mtype LBC will by divided by the number of neurons in layer 5; a link from pre mtype SBC to pre layer 5, by the number of SBCs in layer 5. You can use both forms of normalization together, but the result will be hard to interpret.

As you can see, the layout of the plot can be fully configured. Some layouts, especially when normalization is selected, can be hard to interpret or downright misleading. Please be careful when interpreting the results!