I’m interested in calculating the number of hbonds between 2 groups, a protein and a membrane. Does this require a tpr file? I only have an xtc file and gro/pdb files.
The documentation looks like other inputs are accepted:
Error in user input:
Invalid input values
In option s
Required option was not provided, and the default file ‘topol’ does not
exist or is not accessible.
The following extensions were tried to complete the file name:
.tpr
If you are using gmx hbond before 2024, you can see from the manual page that the .tpr file is not optional but required. I think it is needed to understand bonds and topologies and so to define properly donors/acceptors etc.
Hmm I’m using the version from the Gromacs intro tutorial called “GROMACS = 2023.3=nompi*” in the YAML file. Has this changed in recent versions or do they all need tpr?
I have no idea about the installation you are referring to specifically, but as far as I understand the gmx hbond tool for GROMACS versions < 2024 requires a .tpr file, as the manual says that it is not optional. In GROMACS 2024 versions the old hbond tool has been renamed hbond-legacy, so in these newer versions if you use gmx hbond-legacy you still need a .tpr file, while if you use the new gmx hbond tool in the new 2024.X version it shouldn’t be mandatory (as you can see from the manual where it is listed as ‘Optional’). However, pay attention that the newer hbond tool in 2024.X distros may have some bugs (specifically check here and here).
This is super helpful, thank you. I might try out a 2024 version. Have there been any improvements with the bugs or are they still working on it? I see something here…
Yeah, The error is still in the conda version. I’m guessing we have to download the most recent version from the github to get the bug fix
I don’t actually have a tpr file. I’m trying to reproduce a paper and they uploaded pdb, gro and xtc files: KRAS4b simulation data | Biochemical and Biophysical Systems Group
I’m trying to make the tpr from the pdb but I’m not confident I’ll be able to replicate everything perfectly
I think your best shot might be writing to the authors and ask for the tpr file or, even better, ask them for the topology files so that you can compile your own tpr files for analysis. Eventually, worst case scenario, if the tpr is used only in analysis tools such as gmx hbond where the tpr is used to get the topology (bonds, charges, etc) of the molecules in the box, you should be able to build your own topology as well if you know the force field they used (e.g., if it was CHARMM36m, you can upload the .pdb of the system in CHARMM-GUI, get your force field parameters and topology, and from there compile the .tpr from those files + the pdb and do the analysis you want to do).