Request help to fix this in some detail.
I was following the tutorial: ligand_tutorial
At the step doing eq_nvt, I believe I am encountering two errors.
I do not know how to fix this and move forward. I am enclosing the details below (after ******).
Error1 (?): ‘Update groups can not be used for this system because an incompatible virtual site type is used’
Error2: ‘Fatal error:’
‘There is no domain decomposition for 20 ranks that is compatible with the’
‘given box and a minimum cell size of 0.94825 nm’
Running: LIG edge_18625-1_18626-1 stateA run1
b’ GROMACS - gmx mdrun, 2023.3-Ubuntu_2023.3_1ubuntu3 (-:’
b’’
b’Executable: /usr/bin/gmx’
b’Data prefix: /usr’
b’Working dir: /mnt/g/GROMACS/pmx’
b’Command line:’
b’ gmx mdrun -s /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/tpr.tpr -e /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/ener.edr -c /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/confout.gro -x /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/traj_comp.xtc -o /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/traj.trr -cpo /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/state.cpt -g /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/md.log -dhdl /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/dhdl.xvg’
b’’
b’Compiled SIMD is SSE4.1, but AVX_512 might be faster (see log).’
b’Reading file /mnt/g/GROMACS/pmx/workpath/edge_18625-1_18626-1/water/stateA/run1/eq_nvt/tpr.tpr, VERSION 2023.3-Ubuntu_2023.3_1ubuntu3 (single precision)’
b’Changing nstlist from 10 to 80, rlist from 1.1 to 1.231’
b’’
b’Update groups can not be used for this system because an incompatible virtual site type is used’
b’’
b’-------------------------------------------------------’
b’Program: gmx mdrun, version 2023.3-Ubuntu_2023.3_1ubuntu3’
b’Source file: src/gromacs/domdec/domdec.cpp (line 2104)’
b’MPI rank: 0 (out of 28)’
b’’
b’Fatal error:’
b’There is no domain decomposition for 20 ranks that is compatible with the’
b’given box and a minimum cell size of 0.94825 nm’
We’ve modified the tutorial and now you can change the rank with “-ntomp” flag. You can modify the default ntomp=‘8’ to others (‘4’ or ‘16’, …) if you like in the fe.run_simulation_locally( simType=‘em’, bProt=False, ntomp=‘8’, bVerbose=True ) step. The default should also work fine.
Request Clarification: tutorial ligand_tutorial
I observe at nvt step: fe.prepare_simulation( simType=‘eq_nvt’, bProt=True)
[please review the terminal output…]Getting:
Fatal error:
number of coordinates in coordinate file
(/mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/protein/stateA/run1/em/confout.gro,
71537)
does not match topology
(/mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/water/topol.top,
5605)
The generated gmx grompp is as follows:
gmx grompp
-f /mnt/g/pmxvnb/tutorial_files/input/mdppath/eq_nvt_l0.mdp
-c /mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/protein/stateA/run1/em/confout.gro
-r /mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/protein/stateA/run1/em/confout.gro
-p /mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/water/topol.top
-o /mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/protein/stateA/run1/eq_nvt/tpr.tpr
-maxwarn 4
-po /mnt/g/pmxvnb/tutorial_files/workpath/edge_18625-1_18626-1/protein/stateA/run1/eq_nvt/mdout.mdp
It is observed for protein “water/topl.top” used which is giving fatal error. HOW TO CORRECT THIS IN THE SCRIPT. PLEASE HELP
Thanks a lot for the new updated protein-ligand tutorial. I started using the new link: ligand_tutorial
This is great and working. I could do till the step mdrun [ run1, run2, run3 for State A and State B for Ligand & Protein-Ligand Complex]. At present I a running this in my small computer locally [fe.run_simulation_locally( simType=‘eq’, bProt=True, ntomp=‘8’, bVerbose=True )]. I am very happy that I could reach up to this point. Thanks again for the splendid help. I would greatly appreciate your kind attention and help on how to use the results of all these runs to the next step. Kindly provide methods to generate scripts for running the next step, and running these scripts which I believe, selecting for the mdrun 80 or 100 starting points from Ligand perspective and Protein-Ligand perspective both from State A and State B [for all the run1, run2, run3]. After selecting starting points how to do the computations (lamda =0 to 1 or lamda 1 to 0?) Forward and Reverse transitions. Once this is complete how to analyze these results to arrive at the delta_delta_G’s. In advance, thanking you for your efforts and educating me and others. Happy holidays to you all.
For the transiction computations after mdrun, I executed the follwoing in the ligand-tutorial:
fe.JOBsimcpu = 2
fe.JOBqueue = ‘SGE’
fe.JOBsource = [‘/etc/profile.d/modules.sh’,‘/usr/local/gromacs/GMXRC2024’]
fe.JOBmodules = [‘shared’,‘owl/intel-mpi-default’,‘cuda91’]
fe.JOBgpu = True
fe.JOBgmx = ‘mdrun_threads’
fe.prepare_jobscripts(simType=‘transitions’, bProt=True)
…
Can some one help me either using the script on how to submit jobs or how to do
it interactively. Thanks a lot
Sorry to bother you with a dumb and silly question. In the area: tutorial_files\workpath_precalculated\edge_18625-1_18626-1\water\stateA\run1\transitions
we see a set of 80 files with extension .gro.
I had a look at them in pymol. Many of the ligands are extended outside the boundary of water box. Please help me understand if this is ok to use.
In my fresh computation for State A (for water + ligand, run1), I observe that most of molecules are not at center of the water box. It is close surface. Many atoms of ligands are outside the water box. For clarity I have removed the water and showing the 80 *.gro files in the following attachment.
GROMACS - gmx mdrun, 2023.3-Ubuntu_2023.3_1ubuntu3 (-:’
