This is where I keep my ug4 project codes and explorations. Below are the specifications of the version of ughub
I recommending cloning ug4/ughub to your ROOT directory and building ug4 there.
📂 Repository Information for ../ughub
-----------------------------------------
Commit Hash: 2143e0d7d900a98f4e997d610062758b626d783d
Version Tag: v2021.amfs
Creation Date: 2015-10-22 11:33:39 +0200
Last Modified: 2024-06-05 08:08:21 +0200
Total Commits: 87
I have provided a simple shell script repo_info.sh which will print the above information for a repo. Its usage is below
./repo_info.sh /path/to/repository
Below are some sample images for the geometry generated in the gen_spine.lua script:
|
Artificial Spine |
Artificial Spine Sliced |
|
Artificial Spine Sliced |
Wire frame geometry |
Here is a video of a Calcium simulation on an artificial spine, includes cytosol and ER calcium dynamics
Calcium Dynamics Simulation (3D)
Here is also a video of a coupled Calcium and Voltage dynamics simulation
Coupled Calcium and Voltage Simulations (1D)
Below is a video of an electrical dynamics simulation on a pyramidal cell
Voltage Simulations (1D)
Below is a video of a coupled ion and electrical dynamics simulation on a 1d cell geometry
Voltage and Ion Coupled Simulation (1D)
I have also provided the file structure for my main folder, the main folder contains this repository and ug4 and ughub, the latter two you will need to build separately, by following the directions on the gitHub page for ug4
├── ug4
│ ├── .ughub
│ ├── CMakeLists.txt
│ ├── apps
│ ├── bin
│ ├── build
│ ├── externals
│ ├── lib
│ ├── plugins
│ ├── tools
│ └── ugcore
├── ug4Projects
│ ├── LICENSE
│ ├── README.md
│ ├── docs
│ ├── images
│ ├── projects
│ └── repo_info.sh
└── ughub
├── .gitignore
├── LICENSE
├── README.md
├── __pycache__
├── autocompletions
├── docu
├── project_templates
├── ughub
├── ughub.bat
├── ughub.py
├── ughubHelp.py
├── ughubHelpContents.py
├── ughubProjectFileGenerator.py
└── ughubUtil.py
You can get the above tree by executing
tree -a -L 2 -I '.git|CMake.*'
Please follow the steps for installing ughub via the github page for ug4. A few things to note because my installation will most likely be different from yours, but below are some of the notes I have for my installation...
- you will need to execute inside the
ug4folder
../path/to/ughub/ughub init
- then execute inside
ug4
../path/to/ughub/ughub install Examples
- you will need to make a
builddirectory and thencdin to it. Once inside thebuilddirectory execute
cmake -DDEBUG=ON -DPARALLEL=OFF -DCMAKE_BUILD_TYPE=Debug -DUSER_LAPACK_LIBRARIES=/usr/lib/x86_64-linux-gnu/lapack/liblapack.so -DUSER_BLAS_LIBRARIES=/usr/lib/x86_64-linux-gnu/libblas.so -DConvectionDiffusion=ON ..
- then execute
make -j2 - the above steps will build
ugshellinsideug4/bin/folder
If you want to use the mpi compilers (should already be part of your Linux OS), then inside build
cmake -DCMAKE_C_COMPILER=/usr/bin/mpicc -DCMAKE_CXX_COMPILER=/usr/bin/mpicxx ..
Inside the build folder execute
cmake -DProMesh=ON .. && make -j2
I plan on using ProMesh features in my explorations, ProMesh has several levels of functionality.
I am also using codes developed by my gradute school colleagues and doctoral advisor, to include them we execute in ug4 folder
../path/to/ughub/ughub addsource neurobox https://github.com/NeuroBox3D/neurobox-packages.git
then execute
../path/to/ughub/ughub install neuro_collection cable_neuron
this will install two of the projects to the to ug4 build
next go back into build and execute
cmake -Dneuro_collection=ON -Dcable_neuron=ON .. && make -j2
I also wanted to use the tetrahedralize features of ug4, to build that plugin you need to execute:
cmake -Dtetgen=ON -DLINK_TETGEN=ON ..
Note If you are building neuron_collection there appears to be an error with the tetrahedralize test, the error may look like below
/root/ug4/plugins/neuro_collection/test/tetrahedralize_util.cpp: In function ‘bool ug::neuro_collection::Tetrahedralize(ug::Selector&, ug::Grid&, ug::ISubsetHandler*, number, bool, bool, ug::APosition&, int)’:
/root/ug4/plugins/neuro_collection/test/tetrahedralize_util.cpp:181:56: error: cannot convert ‘char*’ to ‘tetgenbehavior*’
181 | tetrahedralize(const_cast<char*>(ss.str().c_str()), &in, &out);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| |
| char*
In file included from /root/ug4/plugins/neuro_collection/test/tetrahedralize_util.cpp:46:
/root/ug4/tools/tetgen/tetgen.h:2233:37: note: initializing argument 1 of ‘void tetrahedralize(tetgenbehavior*, tetgenio*, tetgenio*, tetgenio*, tetgenio*)’
2233 | void tetrahedralize(tetgenbehavior *b, tetgenio *in, tetgenio *out,
| ~~~~~~~~~~~~~~~~^
make[2]: *** [ugcore/plugins/neuro_collection/CMakeFiles/neuro_collection.dir/build.make:622: ugcore/plugins/neuro_collection/CMakeFiles/neuro_collection.dir/test/tetrahedralize_util.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [CMakeFiles/Makefile2:580: ugcore/plugins/neuro_collection/CMakeFiles/neuro_collection.dir/all] Error 2
make: *** [Makefile:156: all] Error 2
You can disregard this error, tetgen execuate is still built in bin folder.
NOTE - If you plan on using tetrahedralize then you will need to comment out the line that the error is referencing, I tried this and it appears to work now.
To use superlu you will need to execute in the root directory of your ug4 installation
../ughub/ughub install SuperLU6
this will install the plugin folder SuperLU6, you will need to go into that directory, then navigate to external sub folder and then execute
git clone https://github.com/xiaoyeli/superlu.git
which will clone the superlu repo. Note external already contains superlu folder, just delete that folder.
For superlu you will need to execute in the build directory
cmake -DSuperLU6=ON .. && make -j2
I also use ParMetis with some of my .lua scripts, it is a little tricky, I have include a Parmetis.tar which needs to be extracted. Drop the root directory of Parmetis into ug4/plugins/ folder.
Go back into build and execute
cmake -DParmetis=ON -DPARALLEL=ON -DDEBUG=OFF -DPCL_DEBUG_BARRIER=ON .. && make -j2
I turned off DEBUG because I noticed error would arise, and to use Parmetis you need PARALLEL on.