Connectivity

In this area you can edit two types of TVB connectivity objects:

  • long-range connectivity and,

  • local connectivity.

You can also download connectomes from the Allen Mouse Brain Connectivity Atlas or create structural and functional connectivities with data from the EBRAINS Knowledge Graph using the siibra library.


../../_images/connectivity_area.jpg

Preview for Connectivity Area

Long Range Connectivity

../../_images/connectivity_large_scale.jpg

Large Scale Connectivity configuration page

This page is split in two columns.

The left View column contains several Long Range Connectivity visualizations:

  • a 3D view of the nodes and edges

  • 2D Projections of the connectivity graph

    • Left

    • Right

    • Top

  • a 3D view showing the time evolution of the connectivity matrix

The right column contains the connectivity matrix editor.

Connectivity Matrix Editor

../../_images/connectivity_editor.jpg

Preview for the Matrix Editor

The matrix editor allows you to :

  • easily edit the connectivity weights or tract lengths

  • select a subset of the available nodes

  • perform basic algebraic operations on that group; and

  • save the new version as a new connectivity matrix.

The Connectivity datatype will be available in the Simulator area.

Hint

In the Connectivity Editor only one quadrant is displayed at a time. You can select which quadrant is shown by accessing the quadrant selector button in the upper left corner of the matrix display.

Assuming that the connectivity matrix is sorted such that the first half corresponds one single hemisphere:

  • quadrants 1 and 4 will represent the intra-hemispheric connections,

  • and quadrants 2 and 3 will be the inter-hemispheric connections.

../../_images/connectivity_quadrants.jpg

Preview for Quadrant Selection

You can create a smaller selection using three methods:

  1. Click on the Quick-select button and edit the list of node names.

  2. Click on the node labels in the matrix to toggle nodes.

  3. Use the node selection dropdown by clicking the Select Nodes button.

../../_images/connectivity_quick_select.jpg

Preview for Quick-select list

../../_images/connectivity_select_nodes.jpg

Preview for Select Nodes list

TVB enables you to save a new Connectivity object by clicking on staricon. This entity can be used later on in TheVirtualBrain Simulator.

You can save a particular selection. Click the Select Nodes button and the selection component will be shown. Enter a name for the selection and click save.

The Weights button opens a menu to perform basic algebraic operations on a group of edges. You can select multiple nodes from the current connectivity (by default all nodes are selected); thus you will end up with two sets of nodes: the set of selected nodes and the set of un-selected nodes. These two sets of nodes, determine four categories of edges:

  • In –> In: are only the edges connecting the nodes of the selected set.

  • In –> Out: are the edges that connect nodes in the selected set (rows) to nodes in the unselected set (columns).

  • Out –> In: are the edges connecting nodes in the unselected set (rows) to nodes in the selected set (columns).

  • Out –> Out: are edges connecting pair of nodes in the ‘unselected set’.

../../_images/connectivity3d_edges_operations.jpg

Preview for Bulk Operations on edges

Note

Available operations are:

  • Assignation (set): assigns the given numeric value to all the edges within the set.

  • Addition (add): adds the new value to the current value in the connectivity weights matrix.

  • Subtraction (decrease): subtracts the new value to the current value in the connectivity matrix of weights.

  • Multiplication (multiply): multiplies the current value in the connectivity matrix of weights by the given numeric value.

  • Division (divide): divides the current value in the connectivity matrix of weights by the given numeric value.

Click on the Apply weight change button to perform the selected operation on a group of edges.

Example: HOW TO REMOVE INTER-HEMISPHERIC CONNECTIONS

  1. Using the Quick select remove all nodes from the right hemisphere.

    ../../_images/connectivityeditor_SelectASetOfNodes_a.jpg

    Node selection

  2. Apply the changes. The selected nodes appear in green.

    ../../_images/connectivityeditor_SelectASetOfNodes_b.jpg

    Node selection

  3. Save the selection to make it easier later.

    ../../_images/connectivityeditor_SaveSelection.jpg

    Save node selection

  4. Move to the third quadrant (Q3).

    The Connectivity editor will be aware of two sets of nodes: the ones in your selection (green nodes) and the ones that are not selected (white nodes).

    ../../_images/connectivityeditor_ShowConnections.jpg

    3D visualizer zoom-in to show the interhemispheric connections

  5. Then you can proceed to perform some operations on the edge values.

    The four categories of edges in this particular case are:

  • edges IN-IN: intrahemispheric edges from the left hemisphere.

  • edges OUT-OUT: intrahemispheric edges from the right.

  • edges IN-OUT: interhemispheric edges in quadrant 2 (Q2)

  • edges OUT-IN: interhemispheric edges in quadrant 3 (Q3)

  1. Select operation “Set(n)” for edges OUT-IN, set the value to 0 and then press Apply.

    ../../_images/connectivityeditor_SetOutIn.jpg

    Set OUT-IN edges to 0

  2. Repeat for edges IN-OUT .

    The inter-hemispheric connections are gone. Do not forget to select all the nodes again before saving your new matrix. To do so click the select all button in the selection dropdown.

    ../../_images/connectivityeditor_NewMatrix.jpg

    Select all nodes.

  3. Save your new matrix

    ../../_images/connectivityeditor_SaveNewConenctivity.jpg

    Save new matrix

9. Once you have your new matrix, you can launch the connectivity visualizers and check that these connections are not there any more.

../../_images/connectivityeditor_ReloadView.jpg

Reload view

Note

TVB is designed to handle connectivity matrices whose values are:

  • positive real values, meaning that there is a connection, or

  • zero values, meaning the absence of a connection

Warning

  • TVB does not handle unknowns such as NaNs or Infs.

  • If your connectivity matrix contains negative values, such as -1 values you should either set these values to zero or an estimated value based on your research assumptions.



Viewers

Connectivity 3D Edges

This connectivity visualizer allows you to see the structural information as a base model part of TVB.

../../_images/connectivity3d.jpg

Preview for Connectivity Viewer 3D Edges

The 3D semi-transparent surface around the connectivity nodes, whether it is the cortical surface or the outer-skin, is used just for giving space guidance.

You can select an individual node and right-click on it to activate the incoming or outgoing edges.

For each node you can choose a different color to apply to its edges.

../../_images/connectivity3d_coloredges.jpg

Preview for Connectivity Viewer 3D Edges - Coloring incoming / outgoing edges



Connectivity 2D Viewer

A 2D representation of the connectivity matrix nodes and edges.

There are three main views (projections):

  • Left sagittal view

  • Transverse view

  • Right sagittal view

../../_images/connectivity2d_left.jpg
../../_images/connectivity2d_top.jpg
../../_images/connectivity2d_right.jpg

Preview for Connectivity 2D Viewer


Nodes are drawn as circles and the connections as lines. Only the selected nodes are shown.

Visualizing Connectivity Measures

The 3D and 2D Views can be used to visualize two ConnectivityMeasure datatypes. These measures can be the output of a BCT Analyzer. If given, they will determine the size and colors of the nodes in the views.

You can choose these connectivity measures before launching the Large Scale Connectivity visualizer, or from the brain menu (see tip below).

To display the measures in the 3D view check the Metrics details checkbox. Nodes will be displayed as colored spheres. The size of the sphere is proportional to the measure labeled Shapes Dimensions. The color comes from the current color scheme and is determined by the measure labeled Node Colors.

../../_images/connectivity3d_metrics.jpg

3D view of a connectivity measure. Node size is defined by the Indegree. Node color is defined by node strength.

To display the measures in the 2D views click the Show all button.

Nodes are draws as circles, their size proportional to the measure labeled Shapes Dimensions. Their color is determined by a threshold and the measure labeled Node Colors. Nodes with values above the threshold will be red and those whose value are below the threshold will be green.

../../_images/connectivity2d_left_metrics.jpg

Preview of 2D Connectivity Viewer (left lateral view). Node size is defined by the Indegree. Node color is defined by node strength, threshold is 40.

Tip

If you wish to change:

  • the color threshold,

  • the metrics used to define the node features,

  • the colormap used in the Connectivity Matrix Editor, or

  • the Connectivity entity

go to the brain menu on the top right corner

../../_images/connectivity_context_menu.jpg


Space-Time

This is a three-dimensional representation of the delayed-connectivity structure (space-time) when combined with spatial separation and a finite conduction speed. The connectome consists of the weights matrix giving the strength and topology of the network and the tract lengths matrix gives the distance between pair of regions. When setting a specific conduction speed, the distances will be translated into time delays. The space-time visualizer disaggregate the weights matrix and each slice corresponds to connections that fall into a particular distance (or delay) range. The first slice is the complete weights matrix. Click on any of the subsequent slices to see the corresponding 2D matrix plot.

../../_images/connectivityspacetime_main.jpg

Preview for the space-time display

../../_images/connectivityspacetime_fullmatrix.jpg

The first slice is the full weights matrix

../../_images/connectivityspacetime_slice_a.jpg

Connections that are between 0 and 2.84 ms, for a conduction speed of 9 mm/ms

../../_images/connectivityspacetime_slice_b.jpg

Connections that are between 5.68 and 8.53 ms, for a conduction speed of 9 mm/ms

Local Connectivity

In this page, you can generate the spatial profile of local connectivity that will be used in surface-based simulations.

../../_images/connectivity_local.jpg

Local Connectivity editing page

On the lower right of the browser you will have access to different functionalities by clicking on:

  • Create new Local Connectivity button: to generate the Local Connectivity entity.

  • View Local Connectivity button: to launch a 3D brain visualizer displaying the spatial profile of the newly generated entity.

    ../../_images/local_connectivity_viewer.jpg

    Local Connectivity Viewer

  • Edit Local Connectivity button: to go back to the main Local Connectivity editing page.

On the right column there is a display showing different estimations of the spatial profile based on the length of :

  • Theoretical case: is the ideal case.

  • Most probable case: resolution is based on the mean length of the edges of the surface mesh.

  • Worst case: resolution is based on the longest edge in the surface mesh.

  • Best case: resolution is based on the shortest edge in the surface mesh.

    ../../_images/local_connectivity_estimations.jpg

    Local connectivity profile estimations.

and the red-dotted vertical line represents the cut-off distance.

The x-axis range is automatically set to two times the cut-off distance.

Allen Connectome Downloader

From this page you can initiate an operation which will download data from The Allen Mouse Brain Connectivity Atlas. See http://connectivity.brain-map.org

This operation needs an internet connection and it will take many minutes to complete. It will produce a Structural Connectivity in TVB format and a compatible brain Volume object.

Check the Project –> Operations page to see when the import from Allen is done. You can also find your resulted Connectivity in Project –> Data Structure area.

Siibra Connectivity Creator

In this page you can use the siibra library to create structural and functional connectivities using data from to the EBRAINS Knowledge Graph. This knowledge graph brings together information from different data sources regarding brain atlases, parcellations and their associated features for multiple species. To be able to access this information, you will need a special token provided by the EBRAINS team. For more information, please refer to this page.

Currently, the structural connectivities are stored as TVB Connectivities and the functional connectivities are stored as TVB Connectivity Measures.

../../_images/siibra_connectivity.jpg

Siibra Connectivity page

In order to use the functionalities provided in this page, an internet connection is required. Four parameters need to be configured in order to obtain structural and functional connectivities, all of which come with default values for first time users.

For the atlas and parcellation parameters, you can see all the available options through siibra. Selecting an atlas and a parcellation which are incompatible with each other will result in an error after launching the operation.

The subjects field lets you specify the ids of the subjects for which you wish to create the connectivities. The ids can be specified in 3 ways:

  • Individual ids, separated by ‘;’. For example: “000;001” will create connectivities for subjects 000 and 001.

  • Range of ids, using ‘-’. For example: “000-002” will create connectivities for subjects 000, 001 and 002.

  • Combination of the 2 aforementioned methods. For example: “000-002;100” will create connectivities for subjects 000, 001, 002 and 100.

The last parameter, Compute Functional Connectivities, which comes in form of a checkbox, lets you decide whether or not you wish to also extract the functional connectivities from the Knowledge Graph. In case the box is checked, 5 functional connectivities will be created for each selected subject.

After pressing the Launch button, check the Project –> Operations page to see the status of this operation. If it finished without an error, you will see there the results: Connectivities and, optionally, Connectivity Measures. You can also access the results from the Project –> Data Structure area.