GMX_MMPBSA decomposition error

[Mary-shrmi]

Hi, I am performing a MMPBSA calculation on a biological simulation system, which is a steered molecular dynamics simulation of molecule transport through a nanotube. When I perform pb and decomposition via gmx_mmpbsa, I get this error: "The atom 59465 in the complex index is not found in the topology file. Please check that the files are consistent."
While the files are all the original ones that I used for MD and SMD simulations, I cannot get the problem... all topology files are present in the directory as well.
I performed pb calculation alone, and I did not face this error...
BTW, the input file is like this:

Decomposition namelist variables

&decomposition
idecomp = 4 # Which type of decomposition analysis to do
dec_verbose = 3 # Control energy terms are printed to the output
print_res = "within 6" # Which residues to print decomposition data for
csv_format = 1 # Write decomposition data in CSV format

[Valdes-Tresanco-MS]

This error is related to a mismatch between the index file and the topology. It is normally related to errors in some groups' definitions or using an index file from other complexes. Check if you can save the groups you selected using trjconv

[Mary-shrmi]

Actualy I think it cannot detect the ligand atoms from my input files, but I do not know whether it is the xtc or gro file. I have performed molecular dynamics between a cyclic peptide nanotube and a drug. The starting input file of the steered molecular dynamics was a docking result coordinates between the drug and nanotube, which I manually added at the end of the system gro file to introduce the drug as well as its coordinates. Then, I made index and tpr files and ran the production MD. The thing is that the name of the drug group in the gro file and index or topology file is different. I tried to change it in the gro file like it is in the topology or index file, but I failed, and it said something is wrong with the gro file. Therefore, now gmx_mmpbsa cannot detect the atoms of my ligand and says that the atom number in the index file is not found in the topology file. Since I got the topology files from charmm-gui server, their format is like that; it addresses each itp file corresponding to molecules in the system in another folder named toppar, and only the number of molecules is written in the topol.top. I have checked the number of atoms in index, tpr, and gro files; they are the same and correct.

[Valdes-Tresanco-MS]

If you could run MD with the tpr you built, gmx_MMPBSA can probably process it without problems. In any case, send me the files so I can check what's going on.

[Mary-shrmi]

Hello, unfortunately I cannot provide the files because of the size issue with them.
The only thing that comes to mind is the difference between the names of the molecule in topology, .gro file, and the index file. in the .gro file since I manually added the atom coordinates to the input file of smd, which was a ,gro file, the molecule is recognized by "UNL" (it is the name), while in the index and topology file it is named as the drug's name, so now I do not know which name is recognizable in the .xtc file of my simulation, and maybe this can be the issue.

[Valdes-Tresanco-MS]

The xtc file doesn't contain names, it only contains the atom number and the coordinates by frames. You can send me a sample of your file (only 5-10 frames)