Skip to main content.
(up)
(home) > Tutorials

V.6 Tutorial 6: Dealing with the 3D visualization

This tutorial explains how the 3D visualization works.

Run NMLPlay as explained in section II.B of this user guide. Then click on "File", then "Open", and then load the file called BIG_Test.xml(Figure 23).

 


Figure 23 : BIG_Test.xml

 

Double-click on the population HodgkinHuxleyNeuron2. On the right side of the screen, the structure of the population is a ModulePosition. This allows loading a module which defines the number of neurons and their positions. The module selected allows this by selecting the number of neurons (Lines) and subsequently define the 3D coordinates of each of them.

 


Figure 24 : Defining the positions of neurons

Actually the HodgkinHuxleyNeuron2 population is a population of neurons depending on the Hodgkin-Huxley neural model. The ExternalIntensity panel below permits selecting the current which goes through the neurons (Figure 25). There should be as many lines in the intensity array as in the position array. The number of columns corresponds to the number of time steps. dt is the duration of each time step. The checkbox Step permits specifying whether the current changes linearly during a step or if it switches suddenly to the next intensity (in this case it should be ticked).

 


Figure 25 : External Intensities

In order to run the 3D simulations, a few other parameters are required. Click on the Settings button (see Tutorial 3). The NEOSIM Settings window will appear (Figure 26). The runtime parameter is the duration of the simulation. The stepping is the duration of each step for the computation of the neural activity for each neuron. Varying it allows slowing down the refreshment of the simulation. In order to run a simulation quickly, the duration step should be quite big. However, if the duration of a simulation would exceed the duration of a number of steps, then this number of steps won't be achieved, hence the duration of the time steps needs to be defined carefully in accordance with the total simulation duration.

The visualization delay also allows slowing down the simulation. In order to have the simulation displayed "in real time", it should be really low. However, if it is quicker than the quickest axonal delay, it will be the determining factor in preventing a fast computation, especially during a parallel simulation. During a parallel simulation, it is recommended that this visualization delay is slower than the quickest axonal delay. The checkbox "Visualiz. On/Off" allows switching off the simulation. It is recommended to switch off the simulation if no 3D population structure has been defined (so to speak the population structure is either neuroml.model.network.Grid3D or neuroml.model.network.ModulePosition).

 


Figure 26 : the NEOSIM Settings window.

You can now run the simulation as explained in Tutorial 3. While the progress is being showed in the progress bar, click on the "3D" tab (the default active tab is the "2D" tab). The network should look like Figure 27.

 

Figure 27 : the 3D rendering of the neural activity

Basically the balls represent neurons while the lines represents axons departing from the neurons. When a neural spike is triggered, the neuron and its output lines turn red for a short time. You can alter the rendering with the following actions :

In there are several subnetworks nested in one another, the "Network Navigation" panel in the top left of the screen allows selecting the root of the network to be displayed. All the neurons in that subnework will be displayed. All neurons in other networks which receive spikes from the neurons in that subnetwork will also be displayed.

In order to switch from a subnetwork to another, pause the simulation first and select the subnetwork (not to be done in this tutorial since there is only one subnetwork).