episode

installation

git clone https://github.com/elmbeech/episode.git
cd episode
make
./project

changes compare to current PhysiCell version 1.14.1

1. PhysiCell pull request 298 merged ( MathCancer/PhysiCell#298 )
2. main.cpp adapted to run multiple episodes
3. custom_modules/custom.cpp function create_cell_types morphed into function generate_cell_types and reset_cell_types.
4. config/PhysiCell_settings.xml morphed into config/PhysiCell_settings_episode000.xml, config/PhysiCell_settings_episode001.xml, config/PhysiCell_settings_episode002.xml.

test flight

• 2024-12-29: run 3 episodes, 7200 min, omp_num_threads 1, random_seed 0 => check for same output. ok.
• 2024-12-29: run 3 episodes, 7200 min, omp_num_threads 16, random_seed 0 => check for different output. ok.
• 2024-12-29: run 9 episodes, 7200 min, omp_num_threads 16, random_seed system_clock => check for different output. ok.
• 2024-12-29: run 9 episodes, random max time between 1440 min and 10080 min (1 day and 7 days), omp_num_threads 16, random_seed system_clock => check for different output. ok.

SOMETIMES THE EPOCH WORKS FINE. SOMETIMES THE EPOCH RUNS INTO A CORE DUMPED (always at delete cells: custom_modules/custom.cpp ~ line 72.

• 2024-12-30: run 1024 episodes, random max time between 1440 min and 10080 min (1 day and 7 days), omp_num_threads 16, random_seed system_clock => no core dumped! ok.
• 2025-01-02: add and update parameter worx.

CELLTYPE AND DENSITY do not yet update the names. The problem is within BioFVM/BioFVM_microenvironment.cpp.

using the gnu debugger (on a linux)

Amazing blog post: https://askubuntu.com/questions/1349047/where-do-i-find-core-dump-files-and-how-do-i-view-and-analyze-the-backtrace-st

1. in the Make fiel add -ggdb -O0 flag to the compile command to generate a debug symbol table (-g optimized for gdb, code compiled with zero optimization) .

analyze a core dump

1. at the command line run ulimit -c to see the max dumpfile size
2. at the command line run ulimit -c unlimited on the terminal you run the code that dumps.
3. at /var/log/apport.log figure out the corde dumped file name.
4. at /var/lib/apport/coredump/ is the file to find.
5. fire up gdb with executable and dumpfile like this: gdb ./project /var/lib/apport/coredump/<filename>

alternatively, run c++ executable inside gdb

1. gdb ./project
2. r (run)
3. bt (back trace)
4. bt full (back trace)

generating core dumps.

1. 2024-12-30: running custom.cpp setup_microenvironment(); is NOT causing core dump.
2. 2024-12-30: running custom.cpp setup_tissue(); can cause core dump (PhysiCell::add_PhysiCell_cells_to_open_xml_pugi_v2())!
3. 2024-12-30: running custom.cpp reset_cell_types()d; can cause core dump (BioFVM::Basic_Agent::release_internalized_substrates())!
4. 2024-12-30: running mani.cpp create_cell_container_for_microenvironment(microenvironment, mechanics_voxel_size) and custom.cpp setup_microenvironment(); can causes the core dump!
5. 2024-12-30: at BioFVM/BioFVM_microenvironment.cpp add update_microenvironment parameter to initialize_microenvironment function solves the core get dumped issue!

notes

things i noticed and i am not sure if physicell is supposded to run like this.

1. random seed can nonly be set once. if set a second time (by relloading the xml file) we ge the waring.

WARNING: Setting the random seed again.
	You probably have set a user parameter called random_seed.
	Here, we will use the random seed set in user parameters.
	HOWEVER, as of PhysiCell 1.14.0, you should set the random seed in the <options><random_seed> element in the config file.
	Future versions of PhysiCell may throw an error here. Kindly remove the user parameter and just use the <options><random_seed> element.