Incorrect periodicity error in MMPB(GB)SA calculations

[victorprincipe]

When the MMPBSA script is run using the ambertools20 package, the following output with error message appears:

Loading and checking parameter files for compatibility...
cpptraj found! Using /home/victor/miniconda3/envs/ambertools20/bin/cpptraj
mmpbsa_py_energy found! Using /home/victor/miniconda3/envs/ambertools20/bin/mmpbsa_py_energy
Preparing trajectories for simulation...
231 frames were processed by cpptraj for use in calculation.

Running calculations on normal system...

Beginning GB calculations with /home/victor/miniconda3/envs/ambertools20/bin/mmpbsa_py_energy
  calculating complex contribution...
Found an invalid periodicity in the prmtop file: 23
  File "/home/victor/miniconda3/envs/ambertools20/bin/MMPBSA.py", line 100, in <module>
    app.run_mmpbsa()
  File "/home/victor/miniconda3/envs/ambertools20/lib/python3.9/site-packages/MMPBSA_mods/main.py", line 218, in run_mmpbsa
    self.calc_list.run(rank, self.stdout)
  File "/home/victor/miniconda3/envs/ambertools20/lib/python3.9/site-packages/MMPBSA_mods/calculation.py", line 82, in run
    calc.run(rank, stdout=stdout, stderr=stderr)
  File "/home/victor/miniconda3/envs/ambertools20/lib/python3.9/site-packages/MMPBSA_mods/calculation.py", line 156, in run
    raise CalcError('%s failed with prmtop %s!' % (self.program,
CalcError: /home/victor/miniconda3/envs/ambertools20/bin/mmpbsa_py_energy failed with prmtop tyk2_10_unsolvated_nona.prm7!
Exiting. All files have been retained.

The input command (while in the ambertools20 environment and in the following directory: ML_bind/productions/trial_tyk2/mmpbsa_test/prepared_files) is: $AMBERHOME/bin/MMPBSA.py -O -i mmpbsa.in -o FINAL_RESULTS_MMPBSA.dat -sp tyk2_10.prm7 -cp tyk2_10_unsolvated_nona.prm7 -rp tyk2_vacuum.prm7 -lp tyk2_lig10_vacuum.prm7 -y ~/project2021/ML_bind/productions/trial_tyk2/production.dcd (Please note that your path to the final 'production.dcd' file may be different)

The "prepared files" for this calculation involve:

• The solvated complex, vacuum receptor and vacuum ligand from the ML_bind/data/ folder.
• Using parmstrip within the cpptraj package, the complex was desolvated and Na ions were removed to form the vacuum complex without Na counter ions.
• An earlier error that said there were inconsistent radii set between topology files. Using changeRadii within the ParmEd package, all topology files were set to amber6.

Other earlier (resolved) issue: Another earlier error message stated that there was an inconsistent number of SOLTY variables (expected=0) in the desolvated complex. These variables in the topology file were all set to zero. On the Amber file formats webpage, it stated that SOLTY is "currently unused (reserved for future use)". Removing these values in the topology file resolved the issue.

[jjuarez84]

Hi!

• Can you clarify how did you obtained the original topology file (i.e: the solvated one).
• Have you tried to generate the topfiles with the ante-MMPBSA.py script? I have previously struggled to use the topology files generated by BioSimSpace to do MM-PB(GB)SA purely for formating reasons.
• Not sure how important this is, but have you tried using ante-MMPBSA.py to set up the files? When this fails I usually would advice using parmed (instead of cpptraj) to strip the solvent from the topology file (I am pretty sure there used to be a note on the parmstrip keyword in cpptraj about the topology files generated that way been only useful for visualisation, not calculation, but it seems gone in the latest version of the reference guide, so maybe is OK now).
• To me using amber6 radii is unconventional, unless you have a very good reason to do so I will be very careful. See pages 67 and onwards of the AMBER20 manual for a detailed explanaition of the diferent radii sets.

[victorprincipe]

Hi Jordi,

• The solvated topology file was from BioSimSpace
• I have not tried using the ante-MMPBSA.py script to generate topology files (I have been using cpptraj to strip the solvated topology). I will try using the ante-MMPBSA.py script and see if that works.
• I will look at the manual and figure out which radii to use instead of amber6

Thanks for the help!