yitengdang / multicellsim Goto Github PK

With finite Hill coefficient, the colors for simulations with two genes are currently wrong.

For two signals, the "Display signal" planel can be tuned whenever the simulation is not running (e.g. in paused or replay mode). However, when running the simulation, clicking another display signal will only momentarily change the output.

The window may shift from the app to MATLAB or some other window after you save or load a file. This is typically not a sign of an error, and if you go back to the app you'll see that the file has been properly saved/loaded.

If you are unable to run the simulation after installing as app into MATLAB, it could also be due to the following issues:

• Pop-up window displays "Function '(filename)' in the current folder shadows the app. There is a function in the current MATLAB folder that has the same name as the application.
  • Solution: Either change folder or change the name of the function.
• Error in one of the dependent functions (see list of dependent functions). This might be due to having a file of the same name in your MATLAB directory.
  • Solution: Check whether you have any files in the current MATLAB directory which have the same name as one the dependent functions. If so, either change filename or run from a different directory.

The scale of the axes seems hard to fix in the simulation app. For correct positioning of the cells, the ratio between x and y axes should be sqrt(3)/2. I've now used the command
set(ax, 'PlotBoxAspectRatioMode', 'manual');
set(ax, 'PlotBoxAspectRatio', [1+2*d sqrt(3)/2+2*d 1]);
when initializing the lattice. However, this does not always seem to work and resizing the window sometimes have bad effects.

In the preliminary R2.5v2, loading simulations does not work properly. Going to the end of the simulation and continue running it creates an error.
Also, continue running a simulation creates problems.
Probably due to a reorganizing of the main code for running the simulation (after adding the new features).

The initial configuration shows an empty graph rather than a lattice of cells. Is it possible to start with a lattice of cells instead?

Tried loading saved simulations from file. Encountered problems when continuing simulation (periodic motion did not continue as such).

Resizing the GUI window does not appropriately resize the lattice. The cells can become distorted and ugly. Fix graph proportions.

When simulating a single cell (N=1), the app sometimes displays the error:

Warning: Error updating Scatter.
 Color value contains NaN, or element out of range 0.0 <= value <= 1.0

Is there any way to set the initial position of the app (or to maximize it), so that on a laptop the menu bar won't be invisible? When this is the case, clicking on the lower edge of the app moves it down (on my laptop, at least).

The simulation may get slow after some time and skip time steps (e.g. go from t=350 to t=352) in the visualization. The only option for now is to skip through the frames one by one in replay mode.

Changing an Edit Field (parameter input) and immediate clicking "PLAY" will cause the app to malfunction. Features such as plotting and saving may be disabled then. Instead, after changing an input field, make sure to click elsewhere (or on "STOP" to be sure) so that the axes are reset before clicking "PLAY".

Cell states still show up in the four primary colours only.

It seems that generating high spatial order does not work for the system with 2 signals. The generated lattices are still random.

In the continuous system, the colours of the cells are represented as single RGB triplets of the form (1-Xi, 1-Xi, 1-Xi) where Xi is the expression level. This works if there is some linear relation between the 'perceived level of grayness' and the RGB values. Does this really match human perception? See also here.