GROMACS version:2024.4
GROMACS modification: Yes/No
Hi, I am currently trying to form an initial complex by splicing two proteins and simulating its dynamic trajectory. Among them, his initialization state is very important for the later trajectory simulation, is there any attention and suggestions for the part of this operation?
I have tried to assemble the initial state of two protein complexes, but there are some errors in energy minimization.
When i tired to run NVT, it showed me this:
Or the energy minimization process is directly interrupted.
We can’t help you without knowing what the actual error message was. Scroll up in the terminal output and paste it here. Please note that copying and pasting is better than embedded images, especially for users who use screen readers.
Well, I will complete the text content. Are there any requirements for the initial position of the two proteins?
Position them in whatever way you want the initial state to be. GROMACS doesn’t care at all about the initial positioning, per se.
Currently I’m trying to run the NVT:
command: gmx mdrun -v -deffnm **_nvt -nb gpu
Here are some problematic outputs:
"
Reading file 7jlv_xopq_nvt.tpr, VERSION 2024.4 (single precision)
Changing nstlist from 20 to 100, rlist from 1.222 to 1.343
On host server01 2 GPUs selected for this run.
Mapping of GPU IDs to the 8 GPU tasks in the 8 ranks on this node:
PP:0,PP:0,PP:0,PP:0,PP:1,PP:1,PP:1,PP:1
PP tasks will do (non-perturbed) short-ranged and most bonded interactions on the GPU
PP task will update and constrain coordinates on the GPU
Using 8 MPI threads
Using 16 OpenMP threads per tMPI thread
NOTE: DLB will not turn on during the first phase of PME tuning
starting mdrun ‘7jlv_H in water’
5000000 steps, 10000.0 ps.
step 600: timed with pme grid 84 84 84, coulomb cutoff 1.200: 182360.8 M-cycles
step 800: timed with pme grid 72 72 72, coulomb cutoff 1.362: 183853.1 M-cycles
step 1000: timed with pme grid 64 64 64, coulomb cutoff 1.532: 183617.4 M-cycles
step 1200: timed with pme grid 56 56 56, coulomb cutoff 1.751: 185130.7 M-cycles
step 1200: the maximum allowed grid scaling limits the PME load balancing to a coulomb cut-off of 1.886
step 1400: timed with pme grid 52 52 52, coulomb cutoff 1.886: 186134.2 M-cycles
step 1600: timed with pme grid 56 56 56, coulomb cutoff 1.751: 184561.5 M-cycles
step 1800: timed with pme grid 60 60 60, coulomb cutoff 1.635: 184059.2 M-cycles
step 2000: timed with pme grid 64 64 64, coulomb cutoff 1.532: 185294.8 M-cycles
step 2200: timed with pme grid 72 72 72, coulomb cutoff 1.362: 184226.3 M-cycles
step 2400: timed with pme grid 80 80 80, coulomb cutoff 1.226: 184434.0 M-cycles
step 2600: timed with pme grid 84 84 84, coulomb cutoff 1.200: 149775.9 M-cycles
step 2800: timed with pme grid 84 84 84, coulomb cutoff 1.200: 117507.2 M-cycles
step 3000: timed with pme grid 84 84 84, coulomb cutoff 1.200: 119385.4 M-cycles
optimal pme grid 84 84 84, coulomb cutoff 1.200
NOTE: DLB can now turn on, when beneficial
step 4800, will finish Thu Feb 13 02:42:47 2025imb F 1%
"
And I have changed some parameters in nvt.mdp:
"
; Run parameters
integrator = md ; leap-frog integrator
nsteps = 5000000 ; 2 * 50000 = 100 ps
dt = 0.002 ; 2 fs
; Output control
nstenergy = 50000 ; save energies every 1.0 ps
nstlog = 50000 ; update log file every 1.0 ps
nstxout-compressed = 50000 ; save coordinates every 1.0 ps
"
Is the reason for the problem here because the parameters of the nvp file are set too large?
There is nothing abnormal here. All of the output is standard nonbonded cutoff tuning done by the Verlet neighbor searching method.
Ok, thanks for your reply, I will continue to run to see the results.
