The power of modeling

For all the complex calculations and mechanisms behind a model, it is without much worth if it cannot produce useful results. In general, 'useful results' are defined as successful predictions about the effects of modifying some parameter - if we can use a model to determine the effect of each variable upon the outcome, we can better design our system in the real world.

The results on this page represent a tiny fraction of the power of our model. Since any variable can be inspected at any time during the simulation, we can generate tables of data for any aspect of the model. Here, we provide a few examples of data output from our model.

Signal concentration throughout the brain after one simulation

In the image to the right is a diagram of a 2cm diameter spherical section of the brain, showing both signal concentrations once the simulation has finished. Red indicates a high concentration and green a low concentration. Signals are found in higher concentrations where more plaques are located, since the signals are only secreted once a threshold level of oxidative stress is reached. Notice that MMP-9 concentrations are higher than IP-10 concentrations - this is to be expected since the threshold oxidative stress level for MMP-9 is lower than that for IP-10.

These images were generated by firstly writing signal concentration data to a spreadsheet, then taking the logarithm of these values (since signal concentration varies greatly), and finally by colour-coding these values accordingly.

Microglia effectiveness

Since the entire project is concerned with the clearing of amyloid plaques, the main useful result from the model is the effect of changing each variable upon the ability of microglia to locate and degrade these plaques. Below are three graphs showing this. In the simulations behind these graphs, a number of microglia were introduced into the system at the same location, and the time taken for 50% of the plaques within a 10mm radius to be removed was measured.