So I ran a heap of random bits and pieces and checked them out - looked good. I categorised them according to various behaviours such as cleared infection, persistent infection, oscillating infection etc. Then I realised that I had forgotten to run baselines parameters while investigating variation to a single parameter over a few orders of magnitude. So I quickly added that into the set of parameters - varying a diffusion coefficient, varying two immune response parameters and varying the rate of cell replenishment following death. These are the most speculative of the parameters in the model, so it makes sense to be looking at how changes to them alter the model behaviour.
Nothing much of interest came from the immune parameters or the cell replenishment parameter, and at first I thought the same of the diffusion parameter. Increasing the diffusion coefficient increased the speed at which a certain variable reached further along in space....duh! Then I realised that it actually had the opposite impact on the Chlamydia infection to what I expected - the upshot of all this is (after a few more "no that makes perfect sense...oh wait! no it doesn't!" moments) I have what at least on first glance and to the naive mathematical modeller, looks like a cool result from my simplish model that may have a biologically relevant, counterintuitive statement to make. That is, the results of my model seem to suggest doing something to chlamydial particles or changing them in a particular way, quite the opposite of what you'd expect, that may lead to a reduction in the length of an infection!
Awesome moment - never happened to me before. Even if it turns out to be infeasible in reality and silly, at least I had my one "Homer Simpson, whoop-whoop, spinning around on the floor in my office" moment.
No comments:
Post a Comment