error running gmx_MMPBSA using multiple trajectories

[Bejo Ropii]

Hi, I am performing binding energy calculation using gmx_MMPBSA for double stranded nucleic acid using multiple trajectories approach but I got this error:

        Loads the amber20

        Loads the gromacs-2020-i7

[INFO   ] Starting

[INFO   ] Command-line

  gmx_MMPBSA -O -i mmpbsa.in -cs complex.tpr -ci complex.ndx -cg 1 9 -cp complex.top -cr complex.pdb -ct complex.xtc -rs Probe.tpr -ri Probe.ndx -rg 1 -rt Probe.xtc -rp Probe.top -ls miR21.tpr -li miR21.ndx -lg 1 -lt miR21.xtc -lp miR21.top

[WARNING] protein_forcefield and ligand_forcefield variables are deprecate since version 1.4.1 and will be remove in the next version. Please, use forcefield instead.

[WARNING] entropy_seg variable is deprecate since version 1.4.2 and will be remove in v1.5.0. Please, use ie_segment instead.

[INFO   ] Checking external programs...

[INFO   ] cpptraj found! Using /share/apps/amber20/bin/cpptraj

[INFO   ] tleap found! Using /share/apps/amber20/bin/tleap

[INFO   ] parmchk2 found! Using /share/apps/amber20/bin/parmchk2

[INFO   ] sander found! Using /share/apps/amber20/bin/sander

[INFO   ] Using GROMACS version > 5.x.x!

[INFO   ] gmx found! Using /share/apps/gromacs-2020-i7/bin/gmx

[INFO   ] Checking external programs...Done.

[INFO   ] Building AMBER Topologies from GROMACS files...

[INFO   ] Checking gmxMMPBSA data folder exists in Amber data...

[INFO   ] Get PDB files from GROMACS structures files...

[INFO   ] Making gmx_MMPBSA index for complex...

[INFO   ] Normal Complex: Saving group 1_9 in _GMXMMPBSA_COM_index.ndx file as _GMXMMPBSA_COM.pdb

[INFO   ] A receptor structure file was defined. Using MT approach...

[INFO   ] Normal receptor: Saving group 1 in Probe.ndx file as _GMXMMPBSA_REC.pdb

[INFO   ] A ligand structure file was defined. Using MT approach...

[INFO   ] Normal Ligand: Saving group 1 in miR21.ndx file as _GMXMMPBSA_LIG.pdb

[INFO   ] Building Normal Complex Amber Topology...

[INFO   ] Writing Normal Complex Amber Topology...

[INFO   ] A Receptor topology file was defined. Using MT approach...

[INFO   ] Building AMBER Receptor Topology from GROMACS Receptor Topology...

[INFO   ] A Ligand Topology file was defined. Using MT approach...

[INFO   ] Building AMBER Ligand Topology from GROMACS Ligand Topology...

  File "/share/apps/amber20/miniconda/bin/gmx_MMPBSA", line 8, in <module>

    sys.exit(gmxmmpbsa())

  File "/share/apps/amber20/miniconda/lib/python3.8/site-packages/GMXMMPBSA/app.py", line 95, in gmxmmpbsa

    app.make_prmtops()

  File "/share/apps/amber20/miniconda/lib/python3.8/site-packages/GMXMMPBSA/main.py", line 576, in make_prmtops

    self.FILES.mutant_receptor_prmtop, self.FILES.mutant_ligand_prmtop) = maketop.buildTopology()

  File "/share/apps/amber20/miniconda/lib/python3.8/site-packages/GMXMMPBSA/make_top.py", line 117, in buildTopology

    tops = self.gmxtop2prmtop()

  File "/share/apps/amber20/miniconda/lib/python3.8/site-packages/GMXMMPBSA/make_top.py", line 377, in gmxtop2prmtop

    lig_top = self.cleantop(self.FILES.ligand_top, self.indexes[1])

  File "/share/apps/amber20/miniconda/lib/python3.8/site-packages/GMXMMPBSA/make_top.py", line 602, in cleantop

    if rtemp_top.atoms[i-1].residue.number + 1 not in res_list:

IndexError: list index out of range

Error occured on rank 0.

Exiting. All files have been retained.

Kindly help me, please.

Thank you.

Best,

Bejo

[Mario Sergio Valdes]

Hi Bejo and welcome!

Thanks for report this problem...

If you can send me a sample of the files to do the pertinent tests (ndx, groups, top with the parameters folder if is charmmff, xtc [maximum 10 frames] for the complex, receptor, and ligand respectively) it would be of great help.

We are currently working hard on version 1.5.0 which brings major changes to gmx_MMPBSA_ana, so we will try to resolve this for that release.

Cheers!

Mario S

[Mario Sergio Valdes]

Hi Bejo...

I did some testing and it is definitely a bug.

I need to consult with Mario E. about how we will handle the receptor and ligand topologies. Right now, if the -rp and -lp options are defined, the user-defined topologies are converted. This is in a sense redundant concerning how the receptor and ligand topologies are generated if these options were not defined. That is, generating these (Amber) topologies from the complex should give the same results as converting the user-defined topologies, as long as they match as expected.

However, to do this, we must do several tests before. We will try to do it as soon as possible.

Cheers!

Mario S.

[Bejo Ropii]

Thank you!