CortexDynamicsNCB

Agent based simulation for generating tension in an actin cortex.

Dependencies

These are the build dependencies required for building and testing the project.

• Java At least version 8.
• ImageJ For saving images.
• LightWeightGraphing For creating graphs.
• Junit 4 For running tests.

Running a Simulation

There are three ways to run the simulation.

• With a graphical display, primarily for debugging.
• Headless for long term and remote execution.
• Restarting a previously executed simulation.

Headless simulation

Simulations can take a long time to run, so the primary method to run a simulation is to use the HeadlessSimulation class.

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.simulations.HeadlessSimulation parameters.txt

The parameters.txt is optional, it is used to initialize the simulation parameters. We have included a sample parameters.txt.

The program will immediately try to create two files: 12345678.lock and 12345678-simulation.txt.* After the simulation has been running a sufficient amount of time, two more files will be written: 12345678-rods.dat and 12345678-interactions.dat. An explanation of the output files can be found below.

*The 12345678 is a label and will most likely be a different number.

Once the interactions and rod files have been written the simulation can be interrupted or waited until completion.

Graphical Simulation.

To start a graphical simulation use the ActinMeshApp class.

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.ActinMeshApp

From the GUI you can load simulations, and or run simulations with graphics.

To run a simulation you first need to initialize it, click on the button initialize. Then to simulated, click the button simulate.

Previously saved simulations can be loaded by selecting 'load simulation' from the 'file' menu. A file dialog will open and allow you to select a .lock file and view the simulation.

Restarting a Simulation.

To restart or continue a previously executed simulation. First delete the lock file, then run:

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.simulations.RestoreAndWrite 12345678-simulation.txt

That will recreate the lock file and start running the simulation from where it left off.

If there was an error in writing the files originally RemoveBrokenTimePoints might be able to repair the file.

Analysing Simulation Output

Create a movie of a saved simulation, run:

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.analysis.CreateMovieFromLock

The user will be prompted to select a .lock file.

Interactively plot values from a directory of saved simulation, run

 java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.analysis.GraphingDirectoryCollector

The user will be prompted to select a .lock file in a directory that they wish to analyse. All of the lock files in that directory will be analysed and the output will be plotted.

Tension for a collection of directories.

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.analysis.HeadlessSubdirectoryCollector

This will look for directories in the current directory that contain lock-files and perform analysis and write the results to genera-output.txt

Analyse the tension by parts.

java -cp CortexDynamics-1.0.jar org.paluchlab.agentcortex.analysis.TensionByParts directory1 directory2

Each argument, eg directory1, directory2, will be scanned for directories that contain lock-files in them. The output will be written to directory1-parts.txt or dot-parts.txt if the argument is .

Navigating the source code

The simplest way to understand the flow of the code is to start from HeadlessSubdirectoryCollector.

The main method for that programs goes through the following steps.

• Creates a new model.
• Loads a parameter file (if provided) which sets constants to the desired output.
• Creates a simulation writer.
• Initializes the simulation.
• Writes the simulation data, now that actin filament lengths are known.
• Loops through the simulation a number of times determined by the parameter file.
  • Steps the simulation one time step.
  • Saves the interaction and rod data.
• exits.

Output Files

The naming of the files are important, a simulation has four file and a tag for an identifier.

• {tag}.lock
• {tag}-simulation.txt
• {tag}-rods.dat
• {tag}-interactions.dat

In the previous example the tag was 12345678. The files can be renamed by changing the tag name in each file.

The .lock file just indicates the simulation is running/has been run.

The -simulation.txt is a text file that contains the parameters used for the running simulation, and the length of actin filaments.

The -interactions.dat file is a binary file that contains the information for interactions.

The -rods.dat file is a binary file that contains the positions and directions of all the actin and myosin motors.