Tutorial 5 Protein Ligand Complex Production MD job finish inquiry?

User discussions
1

GROMACS version: 2020.3
GROMACS modification: Yes/No
Here post your question:

Hello I am up to the Production MD step 8 of Protein-Ligand Tutorial, my first command:

gmx grompp -f md.mdp -c npt.gro -t npt.cpt -p topol.top -n index.ndx -o md_0_10.tpr

was successful generating the proper o md_0_10.tpr file, however my subsequent command:

gmx mdrun -v -deffnm md_0_10 -nt 8

is exhibiting a substantial delay time of a month to complete, is this amount of time natural for running this simulation on my MacOS Catalina Version 10.15.6 exhibiting a 1,4 GHz Quad-Core Intel Core i5? (When running the similar Production MD for Tutorials 1 and 2 I did not have this substantial delay.). Below is a paste of this substantial delay wait time starting at the exhibited command line:

Command line:
gmx mdrun -v -deffnm md_0_10 -nt 8

Back Off! I just backed up md_0_10.log to ./#md_0_10.log.4#
Compiled SIMD: AVX_256, but for this host/run AVX2_256 might be better (see
log).
Reading file md_0_10.tpr, VERSION 2020.3-MODIFIED (single precision)
Changing nstlist from 20 to 80, rlist from 1.222 to 1.32

Using 1 MPI thread

Non-default thread affinity set, disabling internal thread affinity

Using 8 OpenMP threads

Back Off! I just backed up md_0_10.xtc to ./#md_0_10.xtc.4#

Back Off! I just backed up md_0_10.edr to ./#md_0_10.edr.4#
starting mdrun ‘LYSOZYME in water’
5000000 steps, 10000.0 ps.
step 1700, will finish Thu Oct 29 06:31:36 2020

(I re-input this command quite a few times including without -v and -nt 8)

Thanks:)
Joel

2

MD is computationally intensive, and if you’re running on a desktop or laptop with only a few cores, you should expect the job to take a long time. These machines are not suited to doing such calculations for a long time. If you’re using the machine for anything else while the simulation is running, then you’re fighting for CPU time with yourself. This is why supercomputing clusters have dedicated compute nodes that only run jobs and little else.

3

Hi Justin and thank you for your kind update, why did tutorials 1 and Production MDs complete so much faster (maybe just a few days)? Is it because the ligand together is causing this delay?

For learning purposes would you suggest to lower the steps towards completion? If this is ok how many steps would you then suggest instead of the 5000000?

Kind Regards,
Joel

4

The first two tutorials run 1 ns of production MD. The complex tutorial runs 10 ns. If you’ve gotten to this point, you’ve largely learned all you need to learn. Running the MD is no different than any other system. The challenge in the complex is dealing with the ligand. Once you’ve done that, the lesson is largely over besides some rudimentary analysis that the tutorial shows you.

5

Hi OK thanks a bunch:)