How the work has been tested/reviewed:
Hi Proffessor Justin, I found good tool for convert CHARMM stream files produced by CGenFF to GROMACS topologies, named Charmm2gmx. It likes the python file of charmm2gmx.py. But I use it has problem. I used the example file of charmm2gmx_lipid_ljpme for test.
Hence, I
Fisrt, I successed installing charmm2gmx in virtual environment by anaconda.
Above progress I just want to study using of charmm2gmx and I wonder is that above progress was right for charmm2gmx?
By the way, I want to build poly(ethylene-glycol) 200 as a ligand, but CGenFF only can deal with 384 atoms. Is that charmm2gmx can convert poly(ethylene-glycol) 200 produce by Charmm-gui to toplogy?
The Charmm-gui produced ff field was conflict to charmm36-jul2022.ff due to [default] was defined twice when gmx grompp. Then, I copy relative element such as [bondtype] et al. to ffbond and nonffbond files in charmm36-jul2022.ff, and exclude ff field in topolgy file. But it shew a lot of trouble atomtype over defined. Hence, I hope charmm2gmx can help.
the official CHARMM36 FF for GROMACS (available on Alexander MacKerell’s page) already contains residue topologies for PEG (PEGM in ethers.rtp) and the corresponding termini entries. I see you already started making your own topology. Do not overcomplicate it, it should be pretty simple.
Take the .itp file from your CHARMM-GUI output which contains the [moleculetype] definition of your PEG molecule
Get similar .itp files for all your other molecules (proteins, ligands etc.): you can even take it out from the topol.top file made by pdb2gmx.
Make a new topology by “#include”-ing the forcefield.itp file from the above mentioned “official” GROMACS port of CHARMM
#include all your ITP files with the [moleculetype] definitions
#include your water model and ions (see the topol.top from pdb2gmx how it does that)
Add a [system] clause at the end of your topology and list all molecule types you use with the respective instance count.
You should be then good to go.
Further notes:
CGenFF should only be used for small molecules which are not already parametrized in CHARMM: large polymers are usually constructed from monomers, which are typically available as residue topologies in the core CHARMM.
CHARMM-GUI does a good job at preparing topologies and coordinate files. However, it outputs only a limited set of the force field parameters (bond types etc.).
the recently published charmm2gmx software you mention is for porting CHARMM force field data, and not actual molecular topologies. However, the authors (Justin Lemkul and I) strove to keep the nomenclature as close to the original CHARMM as possible, so replacing the “parameter” part of the CHARMM-GUI output with a charmm36*.ff directory should be straightforward.
for GROMACS (grompp, essentially), the force field parameters are just a part of the topology. Grompp only gets the topology file (typically topol.top), into which all auxiliary files are included using “#include” directives. These are very much like in the C programming language: when processing the topology, grompp silently replaces such lines with the contents of the referenced files (you get the processed topology with the -pp switch to grompp).
the topology files in GROMACS are user editable: pdb2gmx and other tools do a very good job at coming up with a basic version, but the user is free to modify or reconstruct them (at their own risk, however :-)).
Hi @awacha,
Would it be helpful if I have CHARMM formatted residues and patches which are not available in GROMACS.
Means I would want to convert those rtf and prm files to maybe stream(.str) or maybe itp files,
Would it be wise to use this charmm2gmx tool for the above mentioned purpose?
And if yes, the documentation is very sparse and could not wrap my head around the input file
