This is how I run MD simulations on Slurm. There are two directories:
- array- the files I use to set up and start 3 MD simualtions of the same starting structure
- Normal- starts one simualtion for a larger system with simulations broken into 10ns dependencies
- Normal Improved - has more changable variables and instructions for running simulations with ligands though more needs to be changed in the tleap.input file
Generally:
- Create a directory with your starting structure
- copy the whole directory of one of the above places
- Make changes in
tleap.input
if running with ligand, or other not just protein - run
bash setup.bash NAME
- run
sb run_system.slurm
Improvements to the array one are still in the works, but the single one is nice and I can easily change the number of nanoseconds in a section.