1. Questions to work on
-
Why do only mrph(0) and mrph(1) appear in any of the rules? Bug!
^ Not a bug. We just had the number of morphogens set to two at the top. Change it to 4, and morphogens 0-3 will appear in the rules. (Causing the rules to get very boring for some reason.)
I suspect that the involvement of more morphogens makes interesting rules rarer. (TW)
-
What happens if baseline morphogens don't change concentration? (Including diffusion-driven changes.) We should try turning their diffusion off and making them unable to change in concentration. (But see category-idea below!)
-
Should morphogens diffuse across veins?
-
There should be three kinds of morphogens, and we should deal with them systematically.
-
baseline morphogens or gradient morphogens. These set up A/P and P/D polarities before patterns even start to develop.
-
vein morphogens, which come from the veins.
-
diffuse morphogens or everywhere morphogens, which can come from any cell, depending on rules.
* We know gradient morphogens are being applied correctly, because they show up very prettily on the morphogen maps. However, the vein morphogens fail to show up on their morhpogen maps. This could be a problem with the maps, but I suspect instead that it is a problem with the morphogen generation, since the maps seem to work fine for thr gradients. So...
-
Why don't the veins produce morphogens? There is nothing obviously wrong with the code that should generate them.
-
If the vein morphogens are not actually being generated, why do the patterns become more boring once more morphogens are added? (Or perhaps this _is_ why; the patterns are now being determined by morphogens that are uniformly absent.)
2. Notes -- erase after a while
What's going on now (evening 6/24):
-
Five computers running three different sizes (200 x 200, 300 x 300, 400 x 400), generating wings and storing time-steps 3, 5, 7, 10 as postscript files. (PS files can be viewed in Photoshop.)
-
Problem: forgot to save rules along with wings. Should modify graphics output so that rules show up next to (in same window as) wing.
-
Problem installing Tkinter on machines that don't yet have it. Problem is that Package Manager crashes. Macpython FAQ gives method
![[WWW]](/wiki/classic/img/moin-www.png)
here for diagnosing problem but not for solving it. Seems there's a module called tarfile which is expected by Package Manager, but which isn't there. Odd, bc tarfile means something else.
-
BTW, when downloading things (like Tcl/Tk Aqua) from Sourceforge mirrors, choose the one in Reston VA instead of the one in Chapel Hill NC. It's much, much, much faster.
What's going on now (7/1)
-
Version 9.0 is now up and running (the one with the new user interface). Along with slight code clean-up to reduce the frequency of hard-coding, it features pausing and the ability to generate morphogen maps. It also features resizing to scale it to 800x600 resolution, which may make it annoying to use for Ted. Oh well, I'm blind. Anyway, the morphogen mapping has raised an interesting question. (see above)
